Rural Built Environment of Sichuan Province, China [1st ed.] 9789813342163, 9789813342170

Table of contents :
Front Matter ....Pages i-xvii
The Current Status of Research in the Field of the Rural Built Environment and its Impact on the Daily Activities of Residents (Ao Yibin, Igor Martek)....Pages 1-33
Comprehensive Evaluation of Changes to the Rural Built Environment of China (Ao Yibin, Igor Martek)....Pages 35-55
Public Satisfaction-Based Performance Appraisal of Rural Infrastructure Construction (Ao Yibin, Igor Martek)....Pages 57-83
Indicators Impacting Farmers’ Satisfaction in the Use of Rural Facilities (Ao Yibin, Igor Martek)....Pages 85-109
Indicators Impacting Rural Residents’ Satisfaction in Household Latrines (Ao Yibin, Igor Martek)....Pages 111-137
The Relationship Between the Rural Built Environment and Household Vehicle Ownership (Ao Yibin, Igor Martek)....Pages 139-164
The Impact of the Rural Built Environment on Household Car Ownership, Adjusted for Preference Bias (Ao Yibin, Igor Martek)....Pages 165-194
The Impact of the Rural Built Environment on the Travel Mode Preferences of Rural Residents (Ao Yibin, Igor Martek)....Pages 195-229
The Effects of the Rural Built Environment on Travel-Related CO2 Emissions, Adjusted for Travel Preferences (Ao Yibin, Igor Martek)....Pages 231-262
The Impact of the Built Environment on Bicycle Use Behavior of Rural Residents (Ao Yibin, Igor Martek)....Pages 263-288
The Impact of Building Features and Attitudes Regarding Water Conservation on the Water Use Behavior of Rural Residents (Ao Yibin, Igor Martek)....Pages 289-316
Seismic Evacuation Preparedness Behavior of Rural Residents (Ao Yibin, Igor Martek)....Pages 317-340
The Experience and Attitude of Rural Residents with Regard to Flood Disaster Preparedness (Ao Yibin, Igor Martek)....Pages 341-374
Final Reflections on Current Research Contributions, Limitations and Future Research Directions (Ao Yibin, Igor Martek)....Pages 375-378
Back Matter ....Pages 379-426

Citation preview

Yibin Ao Igor Martek

Rural Built Environment of Sichuan Province, China

Rural Built Environment of Sichuan Province, China

Yibin Ao · Igor Martek

Rural Built Environment of Sichuan Province, China

Yibin Ao Chengdu University of Technology Chengdu, China

Igor Martek Deakin University Geelong, VIC, Australia

ISBN 978-981-33-4216-3 ISBN 978-981-33-4217-0 (eBook) https://doi.org/10.1007/978-981-33-4217-0 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface by Yibin Ao

In 1978, China began its domestic reform of the countryside. More than forty years have passed and great change has taken place. Rapid urbanization is gradually bringing rural and urban China closer together, which has directly impacted the lives of rural residents. This brings with it challenges as people and nature strive to coexist, and so sustainable development must to be prioritized. With this change, the imperative to investigate the relationship between the rural built environment and the impact this has on the daily activities of local residents emerges as a significant concern. How effective has the progress been to date, what can we learn, and what directions should development of the rural built environment take from here? This compilation of research is intended to provide a foundation of understanding aimed at informing theoretical development into questions of progressing the quality of life for rural villagers. I was born in a rural village in Sichuan province, and have witnessed first-hand the dramatic changes that have taken place in rural China. I have seen how these changes have influenced rural residents’ beliefs, values and behaviors. The developments, however, have not always delivered the expected outcomes. Extensive infrastructure investments have resulted in all-weather roads now reaching and linking every settlement within the province, no matter how remote. Yet, economic prosperity has not necessarily followed, with a majority of able-bodied people voting to migrate to big cities, leaving villages to the very old and very young. Seeing all this, I have always wanted to do something to promote rural residents’ quality of life and help construct a new countryside which better serves the needs and expectations of rural residents. To that end, I believe it important to understand the lived conditions experienced in rural China if one wants to effect meaningful change, either as a rural planner or as a policy maker. The way forward came to me during my secondment as visiting scholar to the Department of Built Environment at Eindhoven University of Technology in Netherlands, July 2017 to July 2018. There I began work with Dr. Dujuan Yang on a comprehensive research strategy, and on my return to China, began fieldwork, moving from village to village, household to household, interviewing people and recording their views and experiences. The importance of this work was recognized by various government bodies and agencies who generously offered financial support, without which none of this would v

vi

Preface by Yibin Ao

have been possible. I wish to acknowledge their contribution here. In chronological order, the projects and their sponsors are:- Sichuan University Graduate Student Research and Innovation Project: The Influence of Changes in Rural Built Environment on Rural Residents’ Traveling Activities and Energy Consumption (NO. 2012017yjsy215), 2017/09-2018/12; Chengdu University of Technology Philosophy and Social Sciences Project: The Influence of Changes in Rural Built Environment on Rural Residents’ Domestic Consumption of Energy (NO. YJ2017-NS011), 2018.01-2018.12; Department of Education of Sichuan Province: A Study on How to Improve the System of Infrastructure Construction for Agricultural Production (NO. 18ZA0048); 2018.01-2020.12. The Research Center for Governing Rural Communities: A Study of the Influence of the Living condition of Rural Communities on the Rural Residents’ Sense of Satisfaction and their Relationship (NO. SQZL2019C01), 2019.01-2020.12; The Study Center for the Improving Customs in Sichuan’s New Countryside: A Study on the Mechanism of How Rural Built Environment Influences the Healthy Lifestyle of Rural Residents in the New Era (SCXN2019-004), 2019.01-2021.12; Sichuan Province Social Science Project: A Study on the Influence of Spatial Heterogeneity in Rural Built Environment on Aged Rural Residents’ Quality of Life and their Relationship (SC19TJ030), 2019.08-2020.08; Major Project of Science & Technology Department of Sichuan Province: A Study of the Mode of Construction and Management of Infrastructure for the Disposal of Household Waste from the Perspective of Public Participating Behaviors (2020JDR0177), 2020.01-2021.12. As an interdisciplinary research field, this collection of studies covers urban and rural planning, human geography, human behavior, and management. I would like to offer my sincere gratitude to all my colleagues, without whose effort and expertise this book would not be possible. I would also like to extend my thanks to the staff at Springer, who helped and supported the publication of this book. I also owe a great debt to all the students who participated in this study. The field research volunteers, drawn from the master’s and bachelor’s programs in Engineering Management at Chengdu University of Technology, proved to be rigorous and thorough in garnering the extensive data resource of questionnaires surveys. Trekking across mountains and enduring the tough conditions days on end, testifies to their dedication. Last but not least, I want to thank my wife, Yan Wang, whose untiring support made this book possible. Finally, I should point out that this set of studies remains preliminary. This book takes the rural regions of Sichuan as its research object, but China is vast, with significant differences between regions. Much more work remains to be done if there is to be a full understanding of the development of rural China and the impact that progress has made to the daily lives of rural residents. I welcome all interested researchers to join me in progressing this important work, and in so doing, further contribute to the development of rural China, and the world beyond. Chengdu, China September 2020

Yibin Ao

Preface by Igor Martek

I first visited Sichuan Province, China, in the winter of 1992. My travels began with a bus ride across the border from Hong Kong to Guangzhou. The 170 kilometers that separated these two cities was strewn with farm plots, red earth roads, canals, and a continuous vista of shabby brick dwellings. From Guangzhou, my journey began in earnest. I boarded a soviet style train with the luxury of a sleeper carriage. My bed was the bottom bunk in a vertical column of three. The mattress was a thin sheet of bamboo, the floor was wooden boards, and for heating and hot water there was a coal fired furnace at the end of the carriage. I perched there for three days and three nights, enjoying the company of a rotating melee of travelers dressed in blue Mao suits and shared the hours playing cards, chewing sunflower seeds, and inhaling second-hand cigarette smoke. Outside, the view shifted slowly from endless expanses of rice fields as we rose up the center of China to Wuhan, to the undulating grey terrain of the Yangtse River basin. The morning of day three greeted me with rising mountainscapes so high and so close that their peaks hovered out of view above my window. By the time I reached Chengdu, Sichuan’s capital, I felt I had been transported to another time and place. A short walk from the ice-covered concreted plaza and dim electric lights of the main station, and I was absorbed into a maze of narrow earthen alleyways bustling with people; trading, eating, gossiping, and living out their lives, as people might have done a century earlier. Contrast this with my last China trip, January, this year. I stepped onto the G90 high-speed train, late afternoon at Chengdu East, with my Starbucks coffee in hand, and 2000 kilometers and eight hours later I stepped out into the evening at Beijing West. The point to this anecdote is to confirm what everybody with any interest in world events knows; China has arguably progressed faster and more dramatically than any society in history. By the time I began my annual visits to China, the excesses of the Cultural Revolution and Great Leap Forward were past. Following Mao Zedong’s death in 1976, Deng Xiaoping outmaneuvered Mao’s chosen successor, Hua Guofeng, and ensconced himself de facto leader, in December 1978. Deng is credited with “opening up China.” Though an avowed communist, he famously proclaimed “No matter the cat be black or white, so long as it catches mice.” This portended a more practical approach to China’s development, combining socialist ideology with free enterprise; vii

viii

Preface by Igor Martek

the so-called “Socialism with Chinese characteristics.” Deng proposed a three-step development strategy for China, to be delivered over seventy years. First, the population had to be assured of sufficient food and clothing. This meant doubling the 1980 GDP, and this was achieved by the end of the decade. The second step was to quadruple that GDP, giving citizens some discretionary spending power. That happened ahead of schedule, in 1995. Deng’s third goal was to lift China to the status of a medium developed country, by 2050. This milestone has been taken up by successive leaders, and remains China’s ambition to this day. Reinvigorating the Chinese economy was predicated on improving relations with the West, and through that engagement, delivering the “four modernizations” of agriculture, industry, science and the military. In practice, this manifested as Special Economic Zones (SEZs) which offered favored conditions to foreign investors. Western enterprises were promised access to the potentially vast Chinese market in exchange for their managerial expertise and technology transfers. Between 1980 and 1984, China created SEZs in Shenzhen, opposite Hong Kong, Zuhai, opposite Macau, as well as in Shantou, Xiamen and over the entire island of Hainan. Fourteen other coastal cities opened up in 1984, from Dalian in the north, to Guangzhou in the south. In 1985, whole regions were liberalized, from the Shandong Peninsula to the Pearl River Delta. In 1992, this was extended to provincial capitals, including Chengdu. Consequently, China’s per capita GDP has exploded. In 1950 it was $50, in 1970 it was $100, and in 1990 it was $350. Today, China as a whole enjoys a per capita GDP of $10,000; a huge gain, though still short of America’s lofty $60,000. While SEZs did indeed attract foreign know-how that gave China the growth it sought, there have been some unintended consequences. First, growth has been differential, with coastal cities far outperforming the Chinese hinterland. The city of Shanghai, alone, has a GDP equivalent to that of Indonesia. Neighboring Zhejiang Province, to the south, has a GDP equal to Australia’s, while Jiangsu, to the north, has a GDP on par with Russia’s. Per capita GDP in Shanghai is $23,000, but in more remote regions it hovers at about $5,000. Sichuan sits at $8,000. Second, coastal cities like Shanghai have greater connectivity with the rest of the world than they do with China’s interior. Goods can be exchanged between Shanghai and Los Angeles, for example, faster and at less cost, than between Shanghai and the western provinces. Third, the inter-province economic disparities between coast and interior also extend intra-province, between urban and rural communities. These factors lead to a fourth concern; the rural regions of China’s poorer provinces are being hollowed out. In order to capture the fruits of China’s economic growth, villagers have felt compelled to abandon rural life and seek their fortunes in the cities. In May, 1980, at the time Shenzhen was designated a SEZ, it was a simple fishing village with a population of a few hundred thousand. By 2019, Shenzhen had ballooned to 12 million people, of which more than 10 million were transient migrant workers with none of the rights of local residents. More to the point, as China’s population has swollen from 600 million in the 1950s to its current 1.4 billion, China’s rural population has remained a stable 550 million. That is, China’s rural demographic has reduced from 86% to just 39%. The people left in the countryside are the very old and very young; grandparents and their grandchildren. The

Preface by Igor Martek

ix

majority of able-bodied workers have moved off to the factories and construction sites of the eastern sea-board metropolises. This disparity has not gone unnoticed by the authorities. Raising the living standards of the hinterland provinces, and of rural communities in particular, is an emerging priority. Afterall, the Chinese communist revolution, unlike the Russian revolution, was achieved in the countryside on the back of farmers and rural laborers. Mao came from rural Hunan, and Deng himself was born in Sichuan. What is not always appreciated about communist party officials is that their standing within the political system is hugely competitive. Promotion within the ranks is dependent on performance, while rivalry between officials at all administrative levels—provincial, county, township—is rife. The pinnacle measure of success is economic growth, closely followed by social stability. As investment in cities realize diminishing returns, elevating the standard of living among country folk is gaining increasing attention. The rural outflow of human capital has to be curbed, rural incomes and work opportunities must be raised, and the amenities and infrastructure necessary to support economic growth and improve quality of life needs to be lifted. In short, if Deng’s third goal of advancing China to the status of a middle-income country is to be achieved, the time is ripe for corralling rural China within the boundary of prosperity enjoyed by China’s first and second tier cities. It is within this context that this present work should be appreciated. The rural regions of Sichuan Province stand as a quintessential exemplar of the challenges China faces nationwide in bringing its remaining underprivileged rural communities to an acceptable standard of living. Indeed, much work has been done; yet much work remains. This collection of research papers documents the state of play regarding the conditions experienced by people in rural China, and offers important insights into where future efforts must be focused. It represents the first systematic body of research documenting the impact of the rural built environment on the daily life of Chinese villagers. In working through this book, I would draw the reader’s attention to the following points. First, and I have the non-Chinese reader in mind here, it helps to know a little about Sichuan. Superficially, it is just one of China’s twenty-three provinces, along with other municipalities and administrative regions, situated in the southwest, and home to the giant panda and hotpot cuisine. Considered alone, however, Sichuan is as large as France, with a population of 82 million people. It was even bigger prior to Chongqing and its 16 million people being carved away into a separate municipality, in 1997. Sichuan is ringed by mountains, rising as high as 7500 meters, with its capital, Chengdu, situated within a fertile basin. The Qingling mountains straddle its norther frontier, to the south are the Himalayan mountains of Yunnan, and heading west takes you up into the Tibetan Plateau. Sichuan literally means “four rivers,” and the melting waters of the high plateau along with the subtropical monsoonal rains endow the region with bumper harvests and immense biodiversity before its plentiful runoff feeds into the mighty Yangtze River to the east. Sichuan’s relative isolation also gives it a distinctive identity. It was home to the ancient kingdoms of Ba and Shu and hosts present-day ethnic minorities of the Yi and Chiang. In fact, Sichuan’s family of dialects diverge significantly from standard Mandarin in

x

Preface by Igor Martek

phonology, vocabulary and grammar, and can be mutually unintelligible. Sichuanese is spoken by 120 million people, and if it were classified as a distinct language it would rank 10th most spoken in the world, just behind Japanese. The second point to raise relates to the research itself. A glance at the contents will reveal this book is concerned with the behavior of rural people in response to developments within their built environment; matters of transport, vehicle ownership, sanitation and water use, etc. These might seem ubiquitous themes, but one should keep in mind that these items represent benchmarks by which rural people measure their status in relation to those in urban centers. In other words, these themes are proxies for measuring progress in closing the quality-of-life gap between the lived experiences of country and city people. Additionally, Chap. 12 is concerned with earthquakes, and Chap. 13, with floods. Sichuan sits at the interface of the Tibetan plate and the Yangtze plate. It is subject to severe flooding, landslides and earthquakes. The Wenchuan Earthquake of May 2008 resulted in 70,000 deaths, leaving 500,000 homeless. Understanding the disaster preparedness capacity of villagers further informs administrators of social stability risks and the potential for unexpected calamities to undo progress in the effort to bring rural citizens closer to economic parity with city dwellers. Finally, it is also worth appreciating the manner in which this series of studies has been conducted. This is itself a valuable point of reflection. The regions visited are extensive, with as many as ten or more remote localities examined, often far apart and difficult to reach. Access to communities within these localities was comprehensive, with large survey samples being garnered, and as many as 700 questionnaires being administered. (This is despite the self-deprecating claim by the lead author that the efforts were modest.) The dedication of the research teams should also be noted, with dozens of student researchers recruited to the tasks. And significantly, these studies were generously supported by a wide array of interested organizations. All this attests to the great unity of purpose between government agencies, the educational institutions, and rural villages themselves in progressing the cause of lifting the living conditions of China’s rural communities. I commend this series of studies on the rural built environment of Sichuan Province, China. Geelong, Australia September 2020

Igor Martek

Contents

1

2

The Current Status of Research in the Field of the Rural Built Environment and its Impact on the Daily Activities of Residents . . . 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Literature Review: Introductory Comments . . . . . . . . . . . . . . . . . . 1.2.1 Traffic Behavior: A Theoretical Examination . . . . . . . . . 1.2.2 Built Environment: Measurement Indexes . . . . . . . . . . . . 1.2.3 Built Environment: Car Ownership . . . . . . . . . . . . . . . . . . 1.2.4 Residential Self-Selection Effect . . . . . . . . . . . . . . . . . . . . 1.2.5 Built Environment, Travel Behaviors and Carbon Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.6 Seismic Disaster Emergency Evacuation Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.7 Flood Disaster Preparedness Behaviors . . . . . . . . . . . . . . 1.2.8 Water Saving Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.9 Literature Review: Final Comments . . . . . . . . . . . . . . . . . 1.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comprehensive Evaluation of Changes to the Rural Built Environment of China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Indicator, Data Sources and Descriptive Analysis . . . . . . . . . . . . . 2.2.1 Data Sources and Content Analysis . . . . . . . . . . . . . . . . . . 2.2.2 Trend Analysis of Rural Built Environment . . . . . . . . . . . 2.3 Comprehensive Evaluation of Rural Built Environment Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Factor Analysis of Multi Index Panel Data . . . . . . . . . . . . 2.3.2 Empirical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 4 4 6 8 9 10 12 20 21 22 23 24 35 35 38 38 39 42 42 44 48 51 52

xi

xii

3

4

5

Contents

Public Satisfaction-Based Performance Appraisal of Rural Infrastructure Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Research Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Customer Satisfaction Index (CSI) Model . . . . . . . . . . . . 3.2.2 Interpretive Structural Modeling (ISM) . . . . . . . . . . . . . . 3.2.3 Analytic Network Process (ANP) . . . . . . . . . . . . . . . . . . . 3.2.4 Performance Appraisal System Method . . . . . . . . . . . . . . 3.3 Empirical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1 Revising the ACSI to Determine the Evaluation Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.2 Preliminary Evaluation Index Selection . . . . . . . . . . . . . . 3.3.3 Evaluation Index Selection and Determination . . . . . . . . 3.3.4 ISM Index Relationship Determination . . . . . . . . . . . . . . 3.3.5 Establishing an ANP Model to Determine the Weight of Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.6 Performance Appraisal of Rural Infrastructure . . . . . . . . 3.4 Discussion and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57 57 60 60 61 61 62 63 63 63 66 67 67 70 78 81 81

Indicators Impacting Farmers’ Satisfaction in the Use of Rural Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Research Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 Logit Regression Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Empirical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Evaluation Index Selection . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.2 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.3 Farmers’ Satisfaction Factor Analysis . . . . . . . . . . . . . . . . 4.4.4 Logit Regression Analysis Based on Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix: Variable Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

101 102 105 106 107

Indicators Impacting Rural Residents’ Satisfaction in Household Latrines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Method and Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Research Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

111 111 112 115 115 117

85 85 86 89 89 90 91 91 93 93

Contents

xiii

5.4

118 118 122 125

6

7

Empirical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Evaluation Index Selection and Variable Definition . . . . 5.4.2 Descriptive Analysis of Rural Household Latrines . . . . . 5.4.3 Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.4 Logistic Regression Analysis Based on Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

126 132 134 135

The Relationship Between the Rural Built Environment and Household Vehicle Ownership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Data and Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.1 Rural Context in Sichuan Province, China . . . . . . . . . . . . 6.3.2 Rural Household Survey and Sampling . . . . . . . . . . . . . . 6.3.3 Measurement of the Rural Built Environment . . . . . . . . . 6.4 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.1 Variable Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.2 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.1 Household Structure Attributes . . . . . . . . . . . . . . . . . . . . . 6.5.2 Personal Skills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.3 The Built Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6 Conclusion and Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

139 139 141 143 143 144 147 149 149 153 154 155 158 159 160 161

The Impact of the Rural Built Environment on Household Car Ownership, Adjusted for Preference Bias . . . . . . . . . . . . . . . . . . . . 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Built Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.2 Socio-Demographic Factors . . . . . . . . . . . . . . . . . . . . . . . . 7.2.3 Perception, Preferences and Attitudes . . . . . . . . . . . . . . . . 7.3 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.2 Data Collection and Descriptive Analysis . . . . . . . . . . . . 7.3.3 Variable Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

165 165 167 167 168 169 170 170 172 176 182 191 192

xiv

8

9

Contents

The Impact of the Rural Built Environment on the Travel Mode Preferences of Rural Residents . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.1 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.2 Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.3 Data and Sample Collection . . . . . . . . . . . . . . . . . . . . . . . . 8.3.4 Variables Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.5 Multicollinearity of Variables . . . . . . . . . . . . . . . . . . . . . . . 8.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Sociodemographic Variables . . . . . . . . . . . . . . . . . . . . . . . 8.4.2 The Objective Built Environment . . . . . . . . . . . . . . . . . . . 8.4.3 Built Environment Perceptions and Preferences . . . . . . . 8.4.4 Travel-Related Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Conclusion and Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Effects of the Rural Built Environment on Travel-Related CO2 Emissions, Adjusted for Travel Preferences . . . . . . . . . . . . . . . . . 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.1 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.2 Sample Selection and Data Collection . . . . . . . . . . . . . . . 9.3.3 Variable Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.1 Goodness-of-Fit for SEM . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.2 Effects of the Rural Built Environment on Endogenous Variables . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.3 Effects of Individual Travel Attitudes on Endogenous Variables . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.4 Effects of Socio-Demographic Attributes on Endogenous Variables . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 Conclusion and Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 The Impact of the Built Environment on Bicycle Use Behavior of Rural Residents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.1 Built Environment Factors . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.2 Psychological Factors Impacting Bicycling . . . . . . . . . . . 10.2.3 Socio-Demographic Factors . . . . . . . . . . . . . . . . . . . . . . . . 10.3 Materials and Research Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.1 Model Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

195 195 198 200 200 202 202 203 213 213 216 223 224 225 225 227 231 231 233 235 235 237 241 247 247 248 254 255 256 258 263 263 265 265 266 267 268 268

Contents

xv

10.3.2 Data Collection and Descriptive Analysis . . . . . . . . . . . . 10.3.3 Variable Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4.1 Rural Residents’ Attitudes Regarding Bicycling Infrastructure Conditions, in Sichuan . . . . . . . . . . . . . . . . 10.4.2 Rural Residents’ Bicycling Use Motivation . . . . . . . . . . . 10.4.3 Rural Residents’ Purposes for Bicycling . . . . . . . . . . . . . 10.4.4 Preferences for Riding Bicycles, Motorcycles, and Electric Bicycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4.5 Analysis of the EFA Results . . . . . . . . . . . . . . . . . . . . . . . . 10.4.6 Multivariate Models of Acceptable Bicycling Distances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.5 Strengths and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.6 Conclusion and Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

268 269 273

278 284 284 285

11 The Impact of Building Features and Attitudes Regarding Water Conservation on the Water Use Behavior of Rural Residents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.1 Social Demographic Factors . . . . . . . . . . . . . . . . . . . . . . . . 11.2.2 Water Conservation Attitudes . . . . . . . . . . . . . . . . . . . . . . . 11.2.3 Building Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 Research Design and Data Collection . . . . . . . . . . . . . . . . . . . . . . . 11.3.1 Sample Village Selection . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.2 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.3 Descriptive Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4.1 Exploratory Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . 11.4.2 Multicollinearity of Variables . . . . . . . . . . . . . . . . . . . . . . . 11.4.3 Binary Logistics Regression . . . . . . . . . . . . . . . . . . . . . . . . 11.4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5 Conclusion and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

289 289 290 291 291 292 293 293 294 298 304 304 307 307 307 311 313

12 Seismic Evacuation Preparedness Behavior of Rural Residents . . . . 12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2 Materials and Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2.1 Questionnaire Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2.2 Sample and Data Collection . . . . . . . . . . . . . . . . . . . . . . . . 12.2.3 Variable Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2.4 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.3.1 Reliability Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.3.2 Exploratory Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . .

317 317 320 320 321 323 326 327 327 328

275 276 277 277 278

xvi

Contents

12.3.3 Multicollinearity of Variables . . . . . . . . . . . . . . . . . . . . . . . 12.3.4 Binary Logistic Regression Analysis . . . . . . . . . . . . . . . . . 12.4 Conclusion and Policy Recommendations . . . . . . . . . . . . . . . . . . . . 12.5 Limitations and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 The Experience and Attitude of Rural Residents with Regard to Flood Disaster Preparedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.1 Effects of Previous Experiences on Disaster Preparedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.2 Effects of Attitudes on Disaster Preparedness . . . . . . . . . 13.1.3 Effects of Social Demographic Characteristics on Disaster Preparedness . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.4 Rationale of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2.1 Sample Selection and Data Collection . . . . . . . . . . . . . . . 13.2.2 Model Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2.3 Variable Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.1 Effects of Control Variables on Disaster Preparedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.2 Effects of Experiences on Disaster Preparedness . . . . . . 13.3.3 Effects of Attitudes on Disaster Preparedness . . . . . . . . . 13.4 Conclusion and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Final Reflections on Current Research Contributions, Limitations and Future Research Directions . . . . . . . . . . . . . . . . . . . . . 14.1 Research Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2 Research Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3 Future Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

330 330 336 337 338 341 341 343 344 346 347 348 348 350 351 361 361 367 367 368 370 375 375 376 377

Appendix 1: Peasant Households Satisfaction Questionnaire on Rural Infrastructure Construction . . . . . . . . . . . . . . . . . . . 379 Appendix 2: Questionnaire on the Construction of Rural latrines in Sichuan Province . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385 Appendix 3: Household Questionnaire on Rural Built Environment and Rural Residents’ Travel behavior . . . . . . . . . . . . . . . . . . . 391 Appendix 4: Travel Satisfaction of Rural Residents . . . . . . . . . . . . . . . . . . . 407 Appendix 5: Questionnaire on the Influencing Factors of Water Saving Behavior of Rural Residents . . . . . . . . . . . . . . . . . . . . . 413

Contents

xvii

Appendix 6: Questionnaire for the Study of the Impact of Built Environment or Risk Perception on Evacuation Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas . . . . 421

Chapter 1

The Current Status of Research in the Field of the Rural Built Environment and its Impact on the Daily Activities of Residents

Abstract This chapter systematically overviews the current state of research on the built environment and its impact on residents’ daily activity behavior. It comprises an overview of the research background of this book, the theoretical basis of traffic behavior and the concept of built environment and measurement indexes. It considers the impact of the built environment on household car ownership, residential selfselection, travel preferences and water conservation behaviors. Additionally, the emergency preparedness of residents with regard to seismic and flood disasters are provided. These themes constitute a research framework from which further detailed research analysis proceeds in the subsequent chapters of this book. Keywords Built environment · Travel behavior · Water conservation behavior · Disaster preparedness · Seismic evacuation behavior

1.1 Introduction Accumulated scientific evidence shows that climate change has given rise to real and colossal threats to global human development (Stocker et al. 2013). The global climate change caused by energy consumption generated by human activities and related carbon dioxide emission has drawn wide concern of the international community (Ou et al. 2013; Solomon et al. 2007). According to the data from the International Energy Agency (IEA), the global transportation department has generated 7001.1 Mt carbon dioxide in 2011, which accounted for 22.3% of the total emissions, becoming the second largest carbon dioxide emission source, wherein the emission of road transport has taken up around three fourths (73.9%) of the total carbon dioxide emission of the transportation (Statistics 2011). Transportation department grows the fastest in terms of global energy consumption and carbon dioxide emission (Agency 2009; Yan and Crookes 2009). Reducing carbon dioxide emissions from transportation is seen as the main way to meet climate change mitigation targets (Ma et al. 2015), while transportation department is considered as the hardest department regarding emission reducing (Brand et al. 2012; Marsden and Rye 2010). In the meantime, lives of residents are closely related to energy. Energy restricts the development of social

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_1

1

2

1 The Current Status of Research in the Field …

economy and is an important guarantee for improving people’s life quality. Affected by a slower economic growth and slower declines in energy intensity (the amount of energy consumed per unit of output), the global energy demand grew 2.2% in 2017, a growth rate higher than the 1.2% in 2016, the fastest growth since 2013, and was higher than the ten-year average (1.7%); wherein the natural gas consumption has increased by 96 billion cubic meters, or 3%, the fastest growth of a year since 2010; the coal consumption has increased by 25 Mt oil equivalent, or 1%, the first increase since 2013, while the coal proportion in primary energy has declined to 27.6%, the lowest since 2004; and the generated energy has increased by 2.8%, approaching the average level of the decade (BP 2018a). The energy consumption in China has increased by 3.1%, taking up 1/3 of the global increment. The chief executive officer Dudley of British Petroleum (BP) has pointed out that in the future 25 years, along with the sustained growth of energy consumption, the demand thereof would also increase by one third or so (BP 2018b). In some developed countries with a higher urbanization level, household energy consumption has exceeded the industrial sector and become a significant carbon source (Li and Zhang 2016). With the sustained growth of household energy demand, people start to realize that the direct and indirect carbon emissions caused by household living consumption may become new growth points of carbon emissions (Fan and Zhou 2018). The energy consumption and growth rates in China in the past sixteen years are as shown in Fig. 1.1. In 2016, the country’s total consumption was 4358.19 Mt standard coal, wherein the household living energy consumption was 542.09 Mt standard coal. That year per capita nationwide household living energy consumption has been 2.9 times that of 2001, wherein per capita coal household

Fig. 1.1 2001–2016 Chinese energy/ residential energy consumption and the growth rates thereof (Data sources China energy statistical yearbook [2002–2017])

1.1 Introduction

3

living consumption has increased from 66 to 69 kg, per capita electric power household living consumption has increased from 127 to 611 KWh, and per capita natural gas household living consumption has increased from 3.3 to 27.5 m3 (Energy Statistics Department of National Bureau of Statistics 2012–2017). In household energy consumption, the leading role of energy types has been gradually shifting from coal to electric power and the energy structure has been changing. Chinese rural population has taken up nearly 50% of the country’s total population and residential energy is an essential material basis for the residents’ daily lives. In a broad sense, energy consumption of rural residential buildings comprises production energy and household living energy; and in a narrow sense, energy consumption of rural residential buildings merely comprises the energy used by the families, i.e., the energy needed and consumed by the rural families in daily lives (such as refrigerating, heating, hot water, cooking and lighting) (Jiang et al. 2015). As is shown in Fig. 1.2, the rural per capita residential energy consumption was only 44.3% that of the cities and towns in 2001, while up to 2016, the rural per capita residential energy consumption almost equaled that of the cities and towns (Energy Statistics Department of National Bureau of Statistics 2012–2017). The rural per capita residential energy consumption has increased by 3.2 times from 2001 to 2016, and the average growth rate per annum of the rural per capita residential energy consumption is far higher than that of the per capita residential energy consumption in cities and towns. For the moment, a large number of document researches change personal travel behaviors by urban traffic planning, while some further discuss the impact on reducing trip carbon emission (Grazi et al. 2008; Qin and Han 2013). Scholars found out that higher population density, mixed land utilization and walking-friendly street designs had been negatively correlated with household car ownership, trip distance and drive frequency (Ewing and Cervero 2010; Ewing et al. 2015; Khattak and

Fig. 1.2 2001–2016 Per capita residential energy consumption in cities and towns/rural areas (Data sources China energy statistical yearbook [2002–2017])

4

1 The Current Status of Research in the Field …

Rodriguez 2005; Krizek 2003a). Even so, theoretical controversy has never stopped regarding the impact of built environment on travel behaviors, especially after taking individual psychological determinants (such as attitudes, preferences, motivations and purposes) into consideration (Bagley and Mokhtarian 2002; Cao et al. 2007, 2009b; Cao and Yang 2017; Wang and Lin 2014). For example: residents fond of walking may consciously choose communities suitable for walking. Therefore, they walk more and discharge less carbon dioxide (Handy and Clifton 2001). By this time, the relationships between built environment, individual psychology determinants, travel behaviors and trip CO2 emission become more complicated, while systematic researches exploring the impact of built environment on indoor and outdoor daily activities of the residents are rare and the research on the country’s rural area is even more ignored (Ao et al. 2018, 2019a, 2020b, b). China’s rapid urbanization of the rural area had given rise to immense changes of the Chinese rural area built environment, which has directly influenced the daily activity behaviors of the rural residents. This research not only enriches the empirical research perspectives but also provides a theoretical basis for the country’s further urbanization of the countryside and the new rural construction, and provides the world’s other developing countries with a countryside construction reference.

1.2 Literature Review: Introductory Comments 1.2.1 Traffic Behavior: A Theoretical Examination In general, human behaviors are influenced by various different behavioral factors. Therefore, in 1991, psychologist Ajzen developed “the Theory of Planning Behavior” (TPB) (Ajzen 1991). As an extension of “the Theory of Reasoned Action” (TRA) (Ajzen and Fishbein 1980), the Theory of Planning Behavior (TPB) is a broad sense social psychology theory explaining the internal relationships between human behaviors and perception predictive factors as well as intentions. Since mobility is a significant constituent part of human behavior, it might be influenced by different psychological factors. Several empirical researches concerning mobility have demonstrated the effectiveness and significance of these relationships (Bamberg et al. 2003; Haustein and Hunecke 2007; Heath and Gifford 2002; Schoenau and Müller 2017; Thorhauge et al. 2016). Therefore, scholars have found out that the Theory of Planning Behavior (TPB) is a theoretical basis for the analysis of sustainable traffic behaviors and movement behaviors. Built environment, social demographic statistics attributes, resident self-selections and household car ownership have become the principal factors of relevant empirical researches on sustainable travel behaviors (Cao et al. 2009a; Cao and Yang 2017; Ding et al. 2017; Handy et al. 2005; Li and Zhao 2017; Liu et al. 2016; Sun et al. 2017). Car ownership is a significant intermediary factor studying built environment and resident travel behaviors (Ding et al. 2016, 2017; Jiang et al. 2017), social

1.2 Literature Review: Introductory Comments

5

demography characteristic attributes usually serve as beforehand controlled variables to analyze the impact of built environment on residents travel behaviors (Ding et al. 2017; Li et al. 2016), the effect of individual psychological determinants is often considered when exploring the casual relationships between built environment as well as social demographic statistics attributes and residents travel behaviors, and travel behaviors ultimately directly influence the level of trip carbon dioxide emission. The theory of reasoned action and the theory of planning behavior are comprehensive theories of many behaviors, and they have indicated a limited number of psychological variables that can affect behaviors (Albarracín et al. 2001). They are: (a)Intention; (b) Attitude; (c)Subjective Norms; (d) Behavior Control Perception; and (e)Belief Salience, Past Behavior and Habit, Self-Efficacy, Moral Norms and Self-Identity, etc. (Fishbein et al. 2000). As basic theories studying travel behaviors, the relationships among the theory of reasoned action, the theory of planning behavior and their extension, travel behavior theory and travel behaviors and the affecting factors thereof are summed up as is shown in Fig. 1.3. Built environment

Social demographic E-TPB TPB

TRA Subjective norms

Companies Institution Household Individual

Attitudes

Travel behavior Intention Perceived behavior control Behavior

Selfefficacy

Selfidentity

Moral norms

Belief salience

Travel Tools

Travel mode Travel distance Travel frequency Travel duration Travel purpose Travel chain

Density Diversity Design Distance to transit Destination accessibility Demand management Polycentric Urban scale

Other factors Culture Habit External costs

Psychological factors Travel emissions

Travel satisfaction

Quality of life

Other daily activity

Perception Attitudes Preference Motivation

Fig. 1.3 The theoretical development framework of the theory of reasoned action, the theory of planning behavior, and travel behaviors

6

1 The Current Status of Research in the Field …

1.2.2 Built Environment: Measurement Indexes The phrase “built environment” originated from “urban form” and has been gradually defined by its exclusive connotation and denotation in the middle and later 1990s, i.e., emphasizing the urban physical environment depended on by the spatial, temporal and social cultural background with the core of human activities, comprising urban land, building environment and the human activities therein; and traffic infrastructure and its service and initiative design and organizing of the physical elements (Handy 2005; Handy et al. 2005; Saelens and Handy 2008). The land utilization mode means different distributions of space of different social activities, i.e., different divisions of the space region, such as: residential land, commercial land, industrial land and the land for greening; the traffic systems represent different traffic infrastructures including cycle tracks, sidewalks, public transport lanes and so on and the service provided by all traffic infrastructures; and urban design denotes the spatial arrangements of different physical elements in the city and the presented appearance thereby, and the attraction and functions of the spatial arrangements (streets, public spaces). In a great many theoretical and empirical researches, built environment has been gradually accepted as 5D, comprising: Density, Diversity, Design, Destination accessibility and Distance to transit/Transit accessibility (Ewing and Cervero 2001, 2010; Ewing and Handy 2009; Ewing et al. 2015; Vance and Hedel 2007). But existing researches have not unified the measurement indexes of built environment, and different research regions exist differences in built environment, for built environment exists differences in spatial and temporal data acquisition channels and types of data available. The author has summarized 172 built environment measurement indexes of the western developed countries and 98 built environment measurement indexes of Chinese cities through combing nearly 100 research papers about built environment and travel behaviors. Through comparison, it has been found out that: built environment at home and abroad has relatively consistent measurement indexes, such as the aforementioned 5D; and China also has unique built environment elements, which is mainly reflected in the community types aspect, such as: reformed housing from public-owned housing (Wang et al. 2011a), commercial housing (Li and Zhao 2017; Wang et al. 2011b; Zhao and Chai 2013), social welfare housing (Wang et al. 2011b) and housing placement (Feng). Therefore, when studying different type regions’ built environment, according to the availability of spatial geographic information and built environment data, the determining of built environment measurement indexes can be made or added on the basis of actual conditions. Built environment index document summary (statistics) is as shown in Table 1.1. It can be seen from the limited document summary that 5D indexes appear comparatively most frequently in empirical researches at home and abroad, and the next is community type. These six built environments have been expressed roughly the same in research papers at home and abroad, but the specific measurement indexes and the number of indexes still vary. Other built environments and the measurement indexes

1.2 Literature Review: Introductory Comments

7

Table 1.1 Built environment index document summary No

Built environment

The number of times

Frequency (%)

Measurement index number

Index number ratio (%)

1

Density

41

62.12

29

16.86

2

Diversity

21

31.82

16

9.30

3

City Design

39

59.09

41

23.84

4

Destination accessibility

27

40.91

31

18.02

5

Transit accessibility

14

21.21

16

9.30

6

Community type

9

13.64

7

4.07

7

Others

4

6.06

32

18.60

7.1

Public transportation service level

2

3.03

3

1.74

7.2

Traffic/personal safety

1

1.52

9

5.23

7.3

Connection

1

1.52

6

3.49

7.4

Comfort and attractiveness

1

1.52

4

2.33

7.5

Maintenance social capital

1

1.52

2

1.16

7.6

Demand management

1

1.52

5

2.91

7.7

Location

1

1.52

3

8

The total number of documents/indexes

66

–

172

No

Built environment

The number Frequency of times (%)

1

Density

18

66.67

6

6.06

2

Diversity

17

62.96

18

18.18

3

City Design

8

29.63

17

17.17

4

Destination accessibility

11

40.74

15

15.15

5

Transit accessibility

11

40.74

10

10.10

6

Community type

9

33.33

12

12.12

7

Others

3

11.11

21

21.21

7.1

City form

2

7.41

4

4.04

7.2

Infrastructure characteristics

1

3.70

4

4.04

7.3

General index

1

3.70

2

2.02

European and American countries

1.74

Measurement index number

100.00 Index number ratio (%) China

(continued)

8

1 The Current Status of Research in the Field …

Table 1.1 (continued) No

Built environment

The number Frequency of times (%)

Measurement index number

Index number ratio (%)

7.4

Transport level

1

3.70

5

5.05

7.5

Malls

1

3.70

2

2.02

7.6

Recreational facilities

1

3.70

3

3.03

7.7

Multi-center

2

7.41

1

8

The total number of documents/indexes

27

–

99

1.01 100.00

have been expressed not exactly the same at home and abroad, which are mainly reflected as personally characteristic built environments of the research region. The number of related papers of these built environments is comparatively less but the number of specific measurement indexes is, on the contrary, more.

1.2.3 Built Environment: Car Ownership We now already have quite a few relevant household car ownership research documents. Relevant empirical researches have proved that the increase of the number of vehicles has led to the increase of vehicle usage (Shekarchian et al. 2017), and as the vehicle usage and vehicle miles traveled (VMT) increase, the energy consumption and air pollution proportions of the transportation department notably go up (Xue et al. 2015). Therefore, car ownership is a significant intermediary factor of built environment on influencing travel behaviors and trip CO2 emission (Ding et al. 2017, 2016; Jiang et al. 2017), and a reasonable built environment would reduce the car ownership and effectively lessen the trip energy consumption and CO2 emissions. For instance: the increase of built environment population density would lessen household car ownership (Chatman 2013; Hess and Ong 2002; Keller and Vance 2013; Potoglou and Kanaroglou 2008; Van Acker and Witlox 2010; Zegras 2010), and diversity (land utilization mix degree) is negatively correlated with car ownership (Ewing et al. 2015); Ewing et al. (2015) and Hong et al. (2014) have found out that road network density (Design) had been negatively correlated with household car ownership; but Design had been less influential on household car ownership than Density and Diversity (Holtzclaw et al. 2002). Destination accessibility is a regional level built environment, generally comprising the Distance to Central Business District (CBD), Job accessibility and so on (Krizek 2003b; McCormack et al. 2001). Empirical research has shown that the closer the Distance to CBD, the less the family would be inclined to purchase a car (Miller and Ibrahim 1998; Van Acker and Witlox 2010); similarly, a closer Distance to CBD can notably decrease the drive mileage and also the family car ownership (Ewing and Cervero 2010; Potoglou and Kanaroglou 2008);

1.2 Literature Review: Introductory Comments

9

and on the contrary, the increase of Distance to CBD would augment the demand of and reliance on cars (Robert Cervero and Arrington 2008).The Distance to transit is a medium Build Environment of community location level and would influence household car ownership. For instance: Potoglou and Kanaroglou (2008) have found out that if there had been public transport station within the walking scope of the residence, the household car ownership would decrease; and better public transport service would also lessen the car ownership (Chatman 2013). In addition, Demand management has been discussed by the scholars as a built environment dimension in the recent research documents. Demand management usually refers to the distance from the residence to the park, the number of parks or park service level and so on. Some researches have found out that providing low-cost parks in the communities can prompt the increase of car ownership (Guo 2013; Tyrinopoulos and Antoniou 2013), and therefore, Chatman (2008) have suggested lessening community parks to impede the increase of household car ownership. Therefore, Built Environment directly affects the residential household car ownership. The usability of household vehicles directly affects the usage of private cars and residents trip CO2 emission. Household car ownership is a crucial intermediary factor studying built environment and resident travel behaviors and trip CO2 emission.

1.2.4 Residential Self-Selection Effect Recent relevant researches have turned their attention to the effect of individual psychological determinants, such as personal attitudes and preferences on travel behaviors and trip CO2 emission (Ajzen 1991). Vehicle possession (purchase) decision makings and housing location selections are the principal intermediary factors affecting travel behaviors and related carbon emission. In the relevant researches, residential self-selection represents preferring to select residential environment due to personal preference and thus affects travel behaviors and trip CO2 emission. People choose the resident location based on their personal psychological determinants (such as: attitudes and preferences), and thus affect their travel behaviors and trip CO2 emission (Cao et al. 2009a; Handy et al. 2005). Handy and Clifton (2001) discovered that residents fond of walking may consciously choose to live in communities suitable for walking. Therefore, they had walked more and discharged less carbon dioxide. People having sustainable consumption concept had been more likely to choose to live in “neo-traditional” or “transit-oriented” communities (Cao and Cao 2014; Cao et al. 2009b; Cervero 2007; Robert Cervero and Arrington 2008). And the researches about the effect of self-selections on commuting have mostly centered on travel mode selections and trip distances. Taking residential selfselection effect into consideration, Built Environment had been usually deemed to have appreciable impact on commuting mode selection behavior (Cervero 2007; Chen et al. 2008). But Scheiner (2010) found out in research that neither of residents self-selection and lifestyle had strong impact on commuting distance. Though some researches discussed the influence of land utilization features and residents

10

1 The Current Status of Research in the Field …

self-selection on commuting repeated trips (Jahanshahi and Jin 2016; Jahanshahi et al. 2015), few explored the influence of residence self-selection on the CO2 emission generated by commuting trips, and a majority of empirical studies about residents self-selection theory come from western countries (Cao 2014). Meanwhile, researches have shown that even after controlling the remarkable effect of resident self-selection, the influence of built environment variables had been still statistically significant, and this has made the relationship between Built Environment and travel behavior more complicated (Cao 2015a, b; Hong and Shen 2013). The attitudes and preferences of people towards car possession are also psychological determinants. Steg (2005) deemed that people sometimes drove because they like driving instead of necessity. Her research showed that the status symbol of cars had been a key factor when people choose transportation modes. Regarding the employing of private cars, Van and Fuji (2006) found out through the study of six Asian countries that the situation that the attitude variable had notable influence on driving commuting behaviors had only happened in Japan, China and Vietnam while in Indonesia, Thailand and The Philippines, the influence had been not that significant. Individual self-selection also plays an important role in car purchase decision making, Belgiawan et al. (2014) conducted research on the self-selection factors of car purchasing of Chinese mainland and Taiwan, Indonesian, Japanese, Lebanese, Dutch and American undergraduates, and the results showed that developed countries and regions had had significant differences from the developing ones: students in developed countries and regions had been less willing to purchase car. The expectation from others seemed to have been significant determinants of purchase intentions, while income and the symbolic meaning of cars had been less relevant to purchase intention. Besides, environmental awareness can also influence household car ownership and trip energy consumption. Flamm (2009) has found out that families with higher environmental awareness had had less private cars and had been more willing to purchase energy-efficient cars ending up with more energy-conserving and emission-reducing trips. Residential self-selection and car purchase self-selection sometimes mutually affect each other. For instance: people choose the residence location on the basis of their psychological determinants, which further affects their car purchase decisions (Handy et al. 2005). Therefore, the relationships among residents individual selfselection, Built Environment (BE), resident travel behavior and CO2 emission often intersect and appear more complicated.

1.2.5 Built Environment, Travel Behaviors and Carbon Emissions In western developed countries’ researches about the influence of built environment on travel behaviors, trip energy consumption or CO2 emission, scholars have accumulated abundant research results (Boarnet 2011; Cao et al. 2009b; Ewing and

1.2 Literature Review: Introductory Comments

11

Cervero 2001, 2010; Ewing et al. 2015). By the end of 2010, research reports of this topic have been more than 200 (Ewing and Cervero 2010). Chinese scholars’ research of city built environment and residents travel behavior is at an exploration and beginning stage (Cao 2015b) and recent research document number shows the strong interest of the scholars in this research field (Wang and Zhou 2017). Though empirical researches of Chinese scholars have mainly centered on first-tier cities such as Beijing, Shanghai, Guangzhou and Nanjing (Wang and Zhou 2017), this has undoubtedly huge promoting effects on Chinese city development, countryside urbanization and new rural construction. Built environment has tremendous influence on travel behaviors and transport CO2 emission (Hankey and Marshall 2010), and these travel behaviors are measured in multiple ways, including travel mode selection, trip distance, trip frequency, trip purpose or trip time (Boarnet 2011; Ewing and Cervero 2001, 2010; Handy et al. 2005). According to the summary of Ewing and Cervero (2001), trip frequency had been mainly determined by social demographic statistics attributes and built environment. Built environment had had more decisive effect on trip distance, and the next effective had been social demographic statistics attributes. Travel mode selection had rested with built environment and social demographic statistics attributes, but the latter had been more influential. For vehicle-miles of travel (VMT) or vehicle-miles of hours (VHT), the effect of built environment had been more notable. And a similar research conducted by the author ten years later has found out that the comprehensive influence of the several built environment variables had been huge (Ewing and Cervero 2010). Since 1980s and 1990s, relevant researches put forward that traffic energy consumption and related CO2 emissions had been negatively correlated with population density (Brownstone and Golob 2009; Modarres 2013), for the compact and high-density city form had prompted the development of public transport and diminished the use of private cars (Ewing 1997; Kenworthy and Laube 1996; Newman and Kenworthy 1989); and in comparison, the density of working spaces had played a bigger role in reducing energy consumption and greenhouse gas emission than that of family residences (Ding et al. 2014). But Hughes et al. (2004) held the view that city size had been the most powerful determinant in traffic energy use rather than population density. Hong (2017) found out that there had been a non-linear relationship between density and transport CO2 emission: the emission reduction effect of population density had been not that significant to some extent. And in some other researches, the correlation between residences density and CO2 emissions generated by trips has not been notable (Barla et al. 2011; Jiang et al. 2011; Xiao et al. 2011). Improving the road capacity has been considered as a feasible way to enhance traffic energy consumption efficiency and reduce related emissions. Shim et al. (2006)’s research of Korean 61 middle sized and small cities showed that traffic energy consumption had an inverse relationship with the road density. However, the improvement of road capacity may encourage more driving, which might increase the drive mileages. Therefore, public transport plays a key role in slowing down the transport carbon emission (Yang et al. 2015a). Ma et al. (2015) found out that subway accessibility had been negatively correlated with the commuting time carbon

12

1 The Current Status of Research in the Field …

dioxide emissions in Beijing. Xiao et al. (2011)’s research showed that bus accessibility had a slight positive effect on trip CO2 emission. And a research of Ribeiro and Balassiano (1997) found out that as everyday commuting traffic tools, the CO2 emission of private car is 8 times higher than that of city public transportation. And another research in China also showed that the proportion of public transport trips had notable negative impact on transport CO2 emission (Su et al. 2011). The research of Zahabi et al. (2012) showed that 10% increase of Density, Transit accessibility and Land use mixed respectively means 0.5%, 5.8% and 2.5% trip related greenhouse gas emissions reduction. Similarly, the research of Hong and Goodchild (2014) in Puget Sound region showed that 100% increase of Residential density, Land use mixed and Intersection density can reduce 31.2–34.4% transportation emissions. And Zhao (2010) found out that the city expansion and extension of Beijing’s city margin had increased the trip distance and vehicle usage, leading to an increase of the emissions. As for the quantitative research of Built Environment (BE) to travel behaviors, trip energy consumption and related carbon emission, the measurement indexes of Built Environment have experienced the process that the dimension changes from big to small and the index contents gradually enriches. Yang (2013) sorted the relevant documents into three types according to the distinctions of the quantitative methods of studied objects’ dimensions and built environment: One is city scale built environment research; the second is the impact of simple residential area scale built environment on travel behaviors; and the third is the impact of complex residential area scale built environment on travel behaviors. Tables 1.2 and 1.3 have presented representative documents at home and abroad researching built environments’ impacts on travel behaviors and CO2 emission.

1.2.6 Seismic Disaster Emergency Evacuation Behaviors Though people’s behaviors in case of emergencies are likely to be irrational and complex, many experiments and statistical researches have examined emergency evacuation behaviors (Trakoli 2015). Relevant documents at home and abroad have presented that gender difference causes different evacuation behavior choices. (Goltz and Bourque 2017)’s research showed that females had been inclined to look for shelters, while overseas research of human behaviors during 2012 north Italy earthquake emphasized the inadequate behaviors of males during the earthquake (Prati et al. 2012). The researches after Hurricane Ivan (2004) and Hurricane Ike (2008) showed that factors like residing types, home ownership status, education level and income level had been related to the evacuation decision-making process of the affected dwellers (Hasan et al. 2010; Huang et al. 2012). Some researchers have also studied emergency evacuation behaviors through watching video tapes of the crowds’ evacuation when the disasters happen, each video tape is divided as one or more scenes to be researched, and each scene has similar evacuation conditions and is related to only one evacuation (Bernardini et al. 2016; Yang et al. 2011). Shapira et al. (2018)

Document

The relationships between simple residential area scale Built Environment (BE) and residents travel behaviors

Friedman et al. (1994) USA

Europe, Canada and USA

van de Coevering and Schwanen (2006)

DD

AD

AD

DD

USA

45 cities globally

AD

Data type

32 cities globally

Region

Cameron et al. (2003)

The relationships between Newman and city scale built environments Kenworthy (1989) and residents travel behaviors as well as energy consumption Cervero (1996)

Research content

Yes

Yes

Yes

Yes

No

V

DA

MLR

DIA

Logit

DA

Main method

No

No

No

No

No

S

Table 1.2 Foreign representative documents of built environments and travel behaviors as well as CO2 emission researches

(continued)

Residents in suburb residential areas rely more on private car going outs than those in traditional residential areas

Average trip distance and average trip time consumption are both positively correlated with city population; and average trip time consumption is negatively correlated with rail traffic density

City trip demand and private car ownership rate are positively correlated with city vehicle trip mileages

The ratio of public transport, walking and bicycle adopted by families would increase if there is any commercial facilities within the scope of 300 feet from the residence

City population density is negatively correlated with the gasoline consumption per capita

Mainly considered built environments and principal conclusions

1.2 Literature Review: Introductory Comments 13

USA

USA

USA

Cervero and Gorham (1995)

Handy and Clifton (2001)

Handy et al. (2005)

USA

Region

Document

The relationships between Cao et al. (2009b) Complex residential area scale Built Environment (BE) and residents travel behaviors

Research content

Table 1.2 (continued)

DD

DD

DD

AD

Data type

Yes

Yes

No

Yes

V

SUR FA

OLM FA

DA

MSR

Main method

Yes

Yes

Yes

No

S

(continued)

Community features affect trip frequency. Traditional communities walk/ride bikes more. The number of commercial types within the scope of 400 m is negatively correlated with the vehicle trip frequency

Suburb residential areas have a farther driving out distance than traditional residential areas

Outcome research shows that families in traditional residential areas choose to walk out more when purchasing necessities

Families in transit-oriented type residential areas have higher frequency and ratio of walking trips, and are more inclined to transit trips

Mainly considered built environments and principal conclusions

14 1 The Current Status of Research in the Field …

Research content

Table 1.2 (continued)

DD

DD

New Zealand

USA

Badland et al. (2012)

Diao and Ferreira (2014)

Larranaga et al. (2016) Brazil

DD

DD

Chile

Zegras (2010)

Data type

Region

Document

No

Yes

Yes

Yes

V

MIMIC OLM

SRM OLS FA

LRA

MNL OLS

Main method

Yes

Yes

Yes

Yes

S

(continued)

Street connection is positively correlated with walking and Built Environment (BE) affects trip attitude

One standard difference of Built Environment (BE) factor is related to the difference of 5000 miles household vehicle miles traveled (VMT) per annum

Employed adults living in blocks less suitable for walking have longer commuting distance than those living in blocks more suitable for walking

The Distance to CBD is negatively correlated with family vehicle miles traveled (VMT), and three routes intersection density is positively correlated with vehicle usage

Mainly considered built environments and principal conclusions

1.2 Literature Review: Introductory Comments 15

Region

USA

Document

Ding et al. (2017)

DD

Data type

Yes

V

SEM DCM

Main method

No

S

Density and land utilization mixed are negatively correlated with trip distance, and accessibility is positively correlated with trip distances

Mainly considered built environments and principal conclusions

Note V—Vehicle ownership; S—Self-selection; Yes (No)—represents that the document has considered (has not considered) this type factors; SRM—Spatial Regression Model; MLR—Multiple Linear Regression Model; SUR—Seemingly Unrelated Regression Model; LRA—Logistic Regression Analysis; MIMIC— Multiple Index Multiple Cause Model; OLM—Ordered Logit Model; DIA—Dimensional Analysis; DA—Descriptive Analysis; MSR—Modal Decomposition Regression Model; AD—Aggregate Data; and DD—Disaggregate Data

Research content

Table 1.2 (continued)

16 1 The Current Status of Research in the Field …

1.2 Literature Review: Introductory Comments

17

Table 1.3 Domestic representative documents of Built Environments (BE) and travel behaviors as well as CO2 emission researches Research content

Document

Region

Data type

V

Main method

S

Principal conclusions

The relationships between City scale Built Environment (BE) and travel behaviors or energy consumption

Ran et al. (2006)

22 Chinese cities

AD

No

URM

No

City scale is positively correlated with city residents trip distances and time consumption

Wan et al. (2007)

17 Chinese cities

AD

No

DA

No

Non-cluster arrangement cities have larger family car trip time consumption than cluster-arrangement cities, and as city scale increases, family car time consumption would notably increase

Sun et al. (2008)

28 Chinese cities

AD

No

MLR

No

High density city development lowers average trip distance and average trip time consumption

Ma et al. (2009)

Beijing

DD

Yes

OLS WLS No

Suburb residents have longer average shopping trip distances than downtown residents

Chen et al. Shenzhen (2011)

DD

Yes

MNL

No

In contrast with cluster centers and non-center areas, travelers in city centers and city sub-centers are more inclined to choose public transit trips

Chai et al. (2011)

DD

No

MSLB

No

Higher land mixed utilization degree, closer commuting distance and more complete public service facility would decrease household trip carbon emission

Relationships between residential area scale Built Environment (BE) and travel behaviors or energy consumption

Beijing

(continued)

18

1 The Current Status of Research in the Field …

Table 1.3 (continued) Research content

Document

Region

Data type

V

Main method

S

Principal conclusions

Jiang et al. Jinan (2011)

DD

No

MSLB

No

Extra-large blocks style’s high-density development goes against reducing household trip energy consumption

Chai et al. (2012)

Beijing

DD

Yes

SEM

No

The distance to downtown and transit-accessibility are positively correlated with household trip distances; and street population density and commercial facilities supply are negatively correlated with household trip distances

Zhao (2014)

Beijing

DD

Yes

MNL

No

Accessibility, cycle tracks number, environment mix and streets connection are positively correlated with bicycle usage; and public transport service level is negatively correlated with bicycle usage

Zhao et al. Shenzhen (2015)

DD

Yes

PLS-SEM

Yes

Good residential build environments are positively correlated with car usage

Feng (2016)

DD

Yes

BLR

No

Blocks design is more influential to active travel than community forms

Nanjing

(continued)

1.2 Literature Review: Introductory Comments

19

Table 1.3 (continued) Research content

Document

Region

Data type

V

Main method

S

Principal conclusions

Li and Zhao (2017)

Beijing

DD

Yes

MLM

Yes

The diversity and accessibility of communities near subway stations are negatively correlated with car ownership and trip distance

Cao and Yang (2017)

Guangzhou

DD

Yes

SEM

Yes

Land utilization mix, subway station density and road network density are negatively correlated with CO2 emission; and bus stations density, the distance to downtown and parks availability are positively correlated with CO2 emission

Note V—Vehicle ownership; S—Self-selection; Yes (No)—represents that the document has considered (has not considered) this type factors; MSLB—Modeled Settlement Type Comparison; MLM—Multivariable Logistic Model; BLR—Binary Logistic Regression Model; CJM—Copulabased Joint Model; PLS-SEM—Structural Equation model based on Partial Least - Squares; URM— Unitary regression model; DA—Descriptive Analysis; AD—Aggregate Data; and DD-Disaggregate D

conducted questionnaire surveys to 306 residents with an average age of 35 of the earthquake stricken areas and found out that 12% respondents had had no idea what emergency actions to take when the disaster took place, and among the 88% respondents knowing what actions to take, 43% had chosen to immediately evacuate to the outdoors, taking up the majority. Ao et al.’s (2020a) research found out that Built Environment (BE) and risk perception had simultaneously affected the emergency action choices of the residents of regions liable to disasters. Disaster risk perception is usually a significant predictive index of disaster evacuation behaviors and researches found out that individuals with higher risk perception levels had presented higher evacuation rates (Peacock et al. 2005; Vitek and Berta 1982).

20

1 The Current Status of Research in the Field …

1.2.7 Flood Disaster Preparedness Behaviors Chen Yulin (2019) found out in the investigation of D town rural families’ disaster preparedness status that the disaster preparedness consciousness had been out of step with the preparedness actions. During the enacting of contingency plans, communities had been organized to carry out disaster prevention and mitigation or drills activities, but escape plans at the family level had not been involved, and the approaches of the farmers’ getting medicines had been too complex and time and energy wasting, which was detrimental to the economy, energy and illnesses state of the farmers. Self-aid and mutual aid competences are very important in disaster evacuation and rescue. Liu Yajuan (2015) thought that primary level disaster management research work had guiding functions of improving the people’s life confidence and opening up a good life, and summarized 6 items of experience worth promoting. Yehong (2010) found out in research that foreign individuals, communities and governments had participated jointly in the management of disasters prevention and mitigation and these had gradually developed into a system, especially in developed countries and a handful of Southeast Asia countries, but China has just put forward in the National Comprehensive Disaster Mitigation “11th Five-Year” Plan to construct 1000 comprehensive disaster mitigation demonstration communities. When facing real disasters, the government had mainly functioned in disaster warning and emergencies meeting, so conducting disasters prevention and mitigation by the family units had had comparatively large room of disasters preventing. Wang Heng and Hu Xiuying (Wang 2014) found out in the research of undergraduates disasters preparedness that disasters preparedness lectures had positive influential relationships with the students’ disasters preparedness behaviors and knowledge. Wang Xingping (2012) analyzed the related issues of the 8.8 Quzhou county, Gansu Province catastrophic debris flow family disaster preparedness and pointed out that the local family disaster preparedness condition had been very poor and family disaster preparedness plans had been nearly none. Local warnings had been not in time and disasters prevention and mitigation knowledge had been deficient, with the lack of evacuation and escaping drills answering emergencies and the family emergency supplies not stored. In a word, the number of published papers of “flood” research keeps going up but a majority of them have fastened on the forming features, disaster mechanism and engineering treatment measures aspects of flood disasters (Cao et al. 2018), and the number of published papers about natural disasters after the signing of 2015–2030 Xiantai disaster risks reducing frame has appeared even more obvious (Raikes et al. 2019). The past scientific development and progress of flood damage had been mainly formed by civil engineers, and they had placed extra emphasis on techniques and funds while ignoring the importance of socioeconomic factors and social sciences method aspect (Messner and Meyer 2006). Structural flood prevention measures can not completely eliminate flood risks, and sometimes they are only a transfer of the risks. For instance, the constructed flood control dam can withstand the floods within the flood control capacity, but caused dyke breaching when the flood control capacity is gone beyond would make the downstream farmers suffer from sudden

1.2 Literature Review: Introductory Comments

21

floods. The pattern of taking active non-structural disaster preparedness measures can reduce flood losses, and has gradually become a significant constituent part of modern flood management (Osberghaus 2015). However, the domestic behavioral studies about natural disasters are immature and the frameworks have not been formed (Sun et al. 2018). Researches about disaster risk attitudes of rural farmers are rare, while disasters prevention and mitigation behaviors are affected by attitudes, and therefore, the exploring of disaster risk attitudes is the foundation of studying the farmers’ decision behaviors (Li 2015). Under the characteristic social, economic and cultural background of our country, it is necessary to carry out researches of these aspects, and the researches taking the farmers of the village flood-frequent regions as the subjects are especially few.

1.2.8 Water Saving Behaviors The researches of water resources by domestic scholars are mostly targeted at urban areas, while the researches of rural areas are comparatively less, and the research interests concentrate mainly on residents water consumptions and urban residents water for life quotas (Choi and Lee 2016; Jiang 2015, 2019; Wang 2015, 2018; Xia 2018; Xiong 2018; Zhang 2016; Zhao 2017). Through document research, it is found that residents’ water resources consuming behaviors are very complex and are directly or indirectly affected by many factors. (1) The price of water resources directly affects the residents’ water using behaviors, and water prices, as the tool of water supply relevant department to manage the water resources, have been widely accepted (Chicoine and Ramamurthy 1986). Research shows that adopting the fee form of progressively increasing water prices can, after meeting the water demand of low income families, limit the unreasonable water using behaviors of high water consumption families using price leverages and prompt the high water consumption families to take water saving actions (W 1989). Progressively increasing water prices have been widely confirmed efficient in lowering the water consumption for life (Nieswiadomy and Molina 1991). (2) Family income is a notable affecting factor to family water consumption. Families with higher income use more water (Arbués et al. 2000; Dandy et al. 1997). High income group have higher bearing capacity of water prices. The water fees they pay only account for a fraction of their total income. They are not sensitive to the regulatory effects of water prices and are likely to have water wasting behaviors (Fan 2014). (3) Climate conditions also influence residents’ water using behaviors, which first directly affects the outdoor water consumption including swimming pool water consumption and vegetable plots water consumption, etc. (Hewitt and Hanemann 1995). Different seasons and different precipitations affect the number of times of showering and cloth washing. Regions with longer rainy seasons are more likely to bring about water-sufficient subconsciousness to people than regions with longer drought seasons, impairing the water saving consciousness (Nyong and Kanaroglou 2001). (4) Researches show that the essential characteristics of families and individuals would appreciably influence the water consumption for

22

1 The Current Status of Research in the Field …

life, and the affecting factors include age, belief and family population size, etc. Families constituted by more adolescents generally have higher household water consumption (Nauges and Thomas 2000). The aged are usually willing to more positively take water saving measures (Zhao 2019). Families with collectivism ideas consume relatively less water than other families (Liping Yan 2017). Researches also show that the bigger the number of family resident population, the less the family water consumption per capita. Water consumption for life per capita would increase as the users’ education levels increase, and the family’s total water consumption is also notably relevant to the ratio of the family’s educated members (Chen et al. 2005). (5) Different environment consciousness or water saving consciousness would lead to different water using and saving behaviors. With unaltered other conditions, the stronger the environment consciousness or water saving consciousness, the lower the water consumption (Corral-Verdugo et al. 2003). In addition, in contrast with the influence of water prices on water saving behaviors, water saving propagation usually has more obvious influence (Hu 2006). (6) Housing traits like the age of the building, house density, living area, house location and the number of faucets, etc. all have big influence on water consumption. The larger the living area is, the more the water consumed (Renwick and Archibald 1998). In addition, the urbanization level of the places have significant influences on local residents’ water for life consumption behaviors (Jin et al. 2018; Wang 2011; Yang et al. 2018; Zhang et al. 2009), and using water saving instruments would notably decrease the family’s water consumption (Geller et al. 1983; Inman and Jeffrey 2006). Meanwhile, residents’ water for life consumption is further influenced by many other factors. For example, local water using limiting strategies, local codes and regulations (Kenney et al. 2008; Renwick and Green 2000), incentive systems like price systems or water price subsidies (Chu et al. 2007; Guo et al. 2011; Kong et al. 2011), the fairness of water prices and the confidence in water supply institutions all influence residents’ water for life consumption to some extent (Li et al. 2010; Zhang 2011).

1.2.9 Literature Review: Final Comments All in all, the research of the impact of built environments on residents’ daily activity behaviors have not arrived at unanimous conclusions. Chinese energy consumption and related emission rates are generally lower than those of European and American countries (Dodman 2009; Yang et al. 2015b), and the social developments of developing countries and developed countries have tremendous discrepancies (Sperling and Salon 2002; Wright and Fulton 2005). What’s more, China has peculiar behavior attitudes and preferences compared to western countries (Wang and Lin 2014). In the meantime, built environments in Chinese cities and countryside have huge differences, and the impacts of residents’ self-selections on built environments are not exactly the same. For instance, the relatively settled homesteads of rural resident families have determined that they cannot choose the residential area location and environment according to their self-attitudes and preferences, while city dwellers

1.2 Literature Review: Introductory Comments

23

are exactly the opposite. Therefore, the studies of the impact of Chinese rural built environment on rural residents’ daily activity behaviors considering the intermediary variables may arrive at different conclusions. Seen from the presently collected document materials at home and abroad, Chinese researches of the impact of city built environment on daily activity behaviors are at the beginning stage, while research documents studying the relationships of city built environment and residents daily activity behaviors are even less developed, and scholars have often ignored the related researches of Chinese rural areas. On the basis of the above, the author has conducted a series of related researches of Chinese rural areas trying to provide a rural planning and construction theoretical basis for further improving the rural residents’ life quality. The fruits of that effort are compiled here in this book, the research framework of which is summarized as Fig. 1.4.

1.3 Summary China is in the process of rapidly urbanizing the countryside with massive construction initiatives delivering huge changes to the way people live. This has dramatically affected values, daily life behaviors and the life quality of rural residents. But the inter-relationship and the mediating and moderating mechanisms between the built environment and rural residents’ daily activity behaviors remains unclear. On the basis of the above, the author has conducted extensive fieldwork investigations to assess the temporal and spatial differences of the built environment from a macroscopic perspective and to empirically study rural residents’ expressed views in regard to daily activities impacted. Policy makers will benefit for the understanding generated with regard to both the lived experiences arising from changes in the built environment, and the consequent shifts in preferences, behaviors and needs of rural residents. This understanding is a necessary prerequisite for developing policy in order to further improve the lived conditions of farmers and villagers. The research contained in this book involves a plurality of subjects, which specifically comprise urban and rural planning, sociology, human geography and administration. Cross-over studies of subjects are beneficial in better clarifying the relationship between rural the built environments, and local residents’ daily activity behaviors. For all that, different regions and countries have notable differences, and therefore more relevant theories and empirical studies are needed to enrich and complete this research framework. Therefore, the final chapter of this book explains its research shortcomings and provides suggestions for further carrying out in-depth studies.

24

1 The Current Status of Research in the Field …

Comprehensive evaluation on rural built environment in China

Evaluation of rural built environment from the perspective of rural residents Public satisfaction-based performance appraisal of rural infrastructure construction

Acceptable bicycling distance

……

Indicators impacting rural residents’ satisfaction in household latrines

Indicators impacting farmers’ satisfaction in rural facilities

Objective built environment

Vehicle hold combination

Travel attitudes of rural residents Daily average travel frequency

Social demographic factors Car ownership

Travel CO 2 emissions

Perceived built environment

Daily average travel distance

Individual attitudes and preference

Travel mode choice

Purpose of bicycling

The impact of building features and attitudes to water conservation on the water use behavior of rural residents

Motivation of bicycling

Conditions of rural facilities

Built environment preference

The impact of the built environment and risk perceptions on the seismic evacuation preparedness of rural residents

The experience and attitude of rural residents with regard to flood disaster preparedness

Some other daily activity behavior of rural residents

Final reflections on research contribution, limitations and future research directions

Fig. 1.4 Research framework of this book

References Agency, I. E. 2009. Transport Energy and CO2 : Moving Towards Sustainability: OECD Publishing. Ajzen, I. 1991. The Theory Of Planned Behavior. Organizational Behavior And Human Decision Processes, 50 (2): 179–211. Ajzen, I., & Fishbein, M. 1980. Understanding attitudes and predicting social behaviour. Albarracín, D., Johnson, B. T., Fishbein, M., & Muellerleile, P. A. 2001. Theories of reasoned action and planned behavior as models of condom use: A meta-analysis. Psychological Bulletin, 127 (1): 142–161.

References

25

Ao, Y., Chen, C., Yang, D., & Wang, Y. 2018. Relationship between Rural Built Environment and Household Vehicle Ownership: An Empirical Analysis in Rural Sichuan, China. Sustainability, 10 (5): 1–18. Ao, Y., Yang, D., Chen, C., & Wang, Y. 2019a. Effects of rural built environment on travelrelated CO2 emissions considering travel attitudes. Transportation Research Part D: Transport and Environment, 73: 187–204. Ao, Y., Yang, D., Chen, C., & Wang, Y. 2019b. Exploring the effects of the rural built environment on household car ownership after controlling for preference and attitude: Evidence from Sichuan, China. Journal of Transport Geography, 74: 24–36. Ao, Y., Huang, K., Wang, Y., Wang, Q., & Martek, I. 2020a. Influence of built environment and risk perception on seismic evacuation behavior: Evidence from rural areas affected by Wenchuan earthquake. International Journal of Disaster Risk Reduction, 46. Ao, Y., Zhang, Y., Wang, Y., Chen, Y., & Yang, L. 2020b. Influences of rural built environment on travel mode choice of rural residents: The case of rural Sichuan. Journal of Transport Geography, 85: 102708. Arbués F, Barberán Ortí R., Villanúa I. 2000. Water price impact on residential water demand in the city of Zaragoza. Arabia[J]. Badland, H. M., Oliver, M., Kearns, R. A., Mavoa, S., Witten, K., Duncan, M. J., & Batty, G. D. 2012. Association of neighbourhood residence and preferences with the built environment, work-related travel behaviours, and health implications for employed adults: Findings from the URBAN study. Social Science & Medicine, 75 (8): 1469–1476. Bagley, M. N., & Mokhtarian, P. L. 2002. The impact of residential neighborhood type on travel behavior: A structural equations modeling approach. Annals Of Regional Science, 36 (2): 279– 297. Bamberg, S., Ajzen, I., & Schmidt, P. 2003. Choice of travel mode in the theory of planned behavior: The roles of past behavior, habit, and reasoned action. Basic and applied social psychology, 25 (3): 175–187. Barla, P., Miranda-Moreno, L. F., & Lee-Gosselin, M. 2011. Urban travel CO2 emissions and land use: A case study for Quebec City. Transportation Research Part D-Transport And Environment, 16 (6): 423–428. Belgiawan, P. F., Schmocker, J. D., Abou-Zeid, M., Walker, J., Lee, T. C., Ettema, D. F., & Fujii, S. 2014. Car ownership motivations among undergraduate students in China, Indonesia, Japan, Lebanon, Netherlands, Taiwan, and USA. Transportation, 41 (6): 1227–1244. Bernardini, G., Quagliarini, E., & D’Orazio, M. 2016. Towards creating a combined database for earthquake pedestrians’ evacuation models. Safety Science, 82: 77–94. Boarnet, M. G. 2011. A Broader Context for Land Use and Travel Behavior, and a Research Agenda. Journal Of the American Planning Association, 77 (3): 197–213. BP Group. 2018a. Statistical Review of World Energy, https://wenku.baidu.com/view/dfd81e278 14d2b160b4e767f5acfa1c7ab008238.html ed. BP Group. 2018b. BP World Energy Outlook, https://wenku.baidu.com/view/b715f63c30b765ce05 08763231126edb6f1a7618.html?rec_flag=default&sxts=1550734639506 ed. Brand, C., Tran, M., & Anable, J. 2012. The UK transport carbon model: An integrated life cycle approach to explore low carbon futures. Energy Policy, 41: 107–124. Brownstone, D., & Golob, T. F. 2009. The impact of residential density on vehicle usage and energy consumption. Journal Of Urban Economics, 65 (1): 91–98. Cameron, I., Kenworthy, J. R., & Lyons, T. J. 2003. Understanding and predicting private motorised urban mobility. Transportation Research Part D-Transport And Environment, 8 (4): 267–283. Cao, X. Y. 2014. Residential self-selection in the relationships between the built environment and travel behavior: Introduction to the special issue. Journal Of Transport And Land Use, 7 (3): 1–3. Cao, X. 2015a. Examining the Relationship Between Neighborhood Built Environment and Travel Behavior: A Review from the US Perspective, 30 (4): 46–52. Cao, X. Y. 2015b. Examining the impacts of neighborhood design and residential self-selection on active travel: a methodological assessment. Urban Geography, 36 (2): 236–255.

26

1 The Current Status of Research in the Field …

Cao, J. S., & Cao, X. S. 2014. The Impacts of LRT, Neighbourhood Characteristics, and Selfselection on Auto Ownership: Evidence from Minneapolis-St. Paul. Urban Studies, 51 (10): 2068–2087. Cao, X., & Yang, W. 2017. Examining the effects of the built environment and residential selfselection on commuting trips and the related CO 2 emissions: An empirical study in Guangzhou, China. Transportation Research Part D: Transport and Environment, 52: 480–494. Cao, X., Mokhtarian, P. L., & Handy, S. L. 2007. Do changes in neighborhood characteristics lead to changes in travel behavior? A structural equations modeling approach. Transportation, 34 (5): 535–556. Cao, X., Mokhtarian, P. L., & Handy, S. L. 2009a. Examining the Impacts of Residential SelfSelection on Travel Behaviour: A Focus on Empirical Findings. Transport Reviews, 29 (3): 359–395. Cao, X. Y., Mokhtarian, P. L., & Handy, S. L. 2009b. The relationship between the built environment and nonwork travel: A case study of Northern California. Transportation Research Part a-Policy And Practice, 43 (5): 548–559. Cao, Y., Yuan, L., Wang, F. 2018. Research status and development trend of flood disaster based on Bibliometrics, 36 (4): 33–39. Cervero, R. 1996. Mixed land-uses and commuting: Evidence from the American housing survey. Transportation Research Part a-Policy And Practice, 30 (5): 361–377. Cervero, R. 2007. Transit-oriented development’s ridership bonus: a product of self-selection and public policies. Environment And Planning A, 39 (9): 2068–2085. Cervero, R., & Gorham, R. 1995. Commuting In Transit Versus Automobile Neighborhoods. Journal Of the American Planning Association, 61 (2): 210–225. Chatman, D. G. 2008. Deconstructing development density: Quality, quantity and price effects on household non-work travel. Transportation Research Part a-Policy And Practice, 42 (7): 1008– 1030. Chatman, D. G. 2013. Does TOD Need the T?: On the Importance of Factors Other Than Rail Access. Journal Of the American Planning Association, 79 (1): 17–31. Chai, Y., Xiao, Z., & Liu, Z. 2011. Comparative Analysis on CO2 Emission Per Household in Daily Travel Based on Spatial Behavior Constraints (07). Chai, Y., Xiao, Z., & Liu, Z. 2012. Generation mechanism and control strategy of carbon emissions from household daily travel: a case study of Beijing (02). Chen, Y. 2019. Investigation on disaster preparedness of rural families in D town. Study on agricultural disasters (01). Chen, X., Xu., Ji, Y. 2005. Study on Influencing Factors of water demand of urban residents. Chen, C., Gong, H. M., & Paaswell, R. 2008. Role of the built environment on mode choice decisions: additional evidence on the impact of density. Transportation, 35 (3): 285–299. Chen, Y., Song, Y., Zhang, Y., Peng, K., Zhang, L., & Jin, X. 2011. The impact of urban land use characteristics on Residents’ travel modes: a case study of Shenzhen City (05). Chicoine D L, Ramamurthy G. 1986. Evidence on the specification of price in the study of domestic water demand[J]. Choi, S.-M., & Lee, Y.-S. 2016. Economic Analysis for Setting Appropriate Repair Cycles on the Fixed Materials and Facilities in the Public Rental Housing. Advances in Materials Science and Engineering 2016: 1–10. Corral-Verdugo V, Bechtel R. B., Fraijo-Sing B. 2003. Environmental Beliefs and Water Conservation:An Empirical Study[J]. Chu, J., Chen, J., & Wang, C. 2007. Simulation and analysis of household water use pattern of urban residents. Chinese Environmental Science (02): 273–278. Dandy G, Nguyen T., Davies C. 1997. Estimating residential water demand in the presence of free allowances[J]. . Diao, M., & Ferreira, J. 2014. Vehicle miles traveled and the built environment: evidence from vehicle safety inspection data. Environment And Planning A, 46 (12): 2991–3009.

References

27

Ding, C., Wang, Y. W., Xie, B. L., & Liu, C. 2014. Understanding the Role of Built Environment in Reducing Vehicle Miles Traveled Accounting for Spatial Heterogeneity. Sustainability, 6 (2): 589–601. Ding, C., Wang, Y., Tang, T., Mishra, S., & Liu, C. 2016. Joint analysis of the spatial impacts of built environment on car ownership and travel mode choice. Transportation Research Part D: Transport and Environment. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. 2017. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100: 65–80. Dodman, D. 2009. Blaming cities for climate change? An analysis of urban greenhouse gas emissions inventories. Environment and Urbanization, 21 (1): 185–201. Ewing, R. 1997. Is Los Angeles-style sprawl desirable? Journal of the American planning association, 63 (1): 107–126. Ewing, R., & Cervero, R. 2001. Travel and the built environment - A synthesis. Land Development And Public Involvement In Transportation (1780): 87–114. Ewing, R., & Cervero, R. 2010. Travel and the Built Environment. Journal Of the American Planning Association, 76 (3): 265–294. Ewing, R., & Handy, S. 2009. Measuring the Unmeasurable: Urban Design Qualities Related to Walkability. Journal of Urban Design, 14 (1): 65–84. Ewing, R., Tian, G., Goates, J. P., Zhang, M., Greenwald, M. J., Joyce, A., Kircher, J., & Greene, W. 2015. Varying influences of the built environment on household travel in 15 diverse regions of the United States. Urban Studies, 52 (13): 2330–2348. Fan, L. 2014. Study on the behavior of rural residents in Guanzhong area of Weihe River Basin. Fan, J., & Zhou, L. 2018. A comparative study on carbon emissions of urban and rural residents in China. Chinese environmental science, 38 (11): 4369–4383. Feng, J. X. 2016. The Built Environment and Active Travel: Evidence from Nanjing, China. International Journal Of Environmental Research And Public Health, 13 (3). Fishbein, M., Triandis, H. C., Kanfer, F. H., Becker, M., & Middlestadt, S. E. 2000. Factors influencing behavior and behavior change. Flamm, B. 2009. The impacts of environmental knowledge and attitudes on vehicle ownership and use. Transportation Research Part D-Transport And Environment, 14 (4): 272–279. Friedman B, Gordon S. P., Peers J B. 1994. Effect of neotraditional neighborhood design on travel characteristics. Transportation Research Record 1466: 63–70 Geller E S, Ericson J. B., Brenda A B. 1983. Attempts to promote residential water conservation with educational, behavioral, and engineering strategies[J]. Goltz, J. D., & Bourque, L. B. 2017. Earthquakes and Human Behavior: A Sociological Perspective. International Journal of Disaster Risk Reduction. Grazi, F., van den Bergh, J. C. J. M., & van Ommeren, J. N. 2008. An empirical analysis of urban form, transport, and global warming. Energy Journal, 29 (4): 97–122. Guo, Z. 2013. Does residential parking supply affect household car ownership? The case of New York City. Journal Of Transport Geography, 26: 18–28. Guo, R., Wu., Yu, H., Li, Y. 2011. Emergy calculation method and application of domestic water price of urban residents_. Handy, S. 2005. Critical assessment of the literature on the relationships among transportation, land use, and physical activity. Transportation Research Board and the Institute of Medicine Committee on Physical Activity, Health, Transportation, and Land Use. Resource paper for TRB Special Report, 282 (1): 1–81. Handy, S. L., & Clifton, K. J. 2001. Local shopping as a strategy for reducing automobile travel. Transportation, 28 (4): 317–346. Handy, S., Cao, X., & Mokhtarian, P. 2005. Correlation or causality between the built environment and travel behavior? Evidence from Northern California. Transportation Research Part D: Transport and Environment, 10 (6): 427–444.

28

1 The Current Status of Research in the Field …

Hankey, S., & Marshall, J. D. 2010. Impacts of urban form on future US passenger-vehicle greenhouse gas emissions. Energy Policy, 38 (9): 4880–4887. Hasan, S., Ukkusuri, S., Gladwin, H., & Murray-Tuite, P. 2010. Behavioral Model to Understand Household-Level Hurricane Evacuation Decision Making. Journal of Transportation Engineering, 137 (5): 341–348. Haustein, S., & Hunecke, M. 2007. Reduced use of environmentally friendly modes of transportation caused by perceived mobility necessities: An extension of the theory of planned behavior. Journal of Applied Social Psychology, 37 (8): 1856–1883. Heath, Y., & Gifford, R. 2002. Extending the theory of planned behavior: Predicting the use of public transportation. Journal Of Applied Social Psychology, 32 (10): 2154–2189. Hess, D. B., & Ong, P. M. 2002. Traditional neighborhoods and automobile ownership. Travel Demand And Land Use 2002 (1805): 35–44. Hewitt J A, Hanemann W M. 1995. A discrete/continuous choice approach to residential water demand under block rate pricing[J]. . Holtzclaw, J., Clear, R., Dittmar, H., Goldstein, D., & Haas, P. 2002. Location efficiency: Neighborhood and socioeconomic characteristics determine auto ownership and use - Studies in Chicago, Los Angeles and San Francisco. Transportation Planning And Technology, 25 (1): 1–27. Hong, J. 2017. Non-linear influences of the built environment on transportation emissions: Focusing on densities. Journal Of Transport And Land Use, 10 (1): 229–240. Hong, J., & Goodchild, A. 2014. Land use policies and transport emissions: Modeling the impact of trip speed, vehicle characteristics and residential location. Transportation Research Part DTransport And Environment, 26: 47–51. Hong, J., & Shen, Q. 2013. Residential density and transportation emissions: examining the connection by addressing spatial autocorrelation and self-selection. Transportation Research Part D: Transport and Environment, 22: 75–79. Hong, J. H., Shen, Q., & Zhang, L. 2014. How do built-environment factors affect travel behavior? A spatial analysis at different geographic scales. Transportation, 41 (3): 419–440. Hu, F. 2006. Research on Influencing Factors of domestic water demand of urban residents. Unpublished Master, Zhejiang University Huang, S.-K., Lindell, M. K., Prater, C. S., Wu, H.-C., & Siebeneck, L. K. 2012. Household evacuation decision making in response to Hurricane Ike. Natural Hazards Review, 13 (4): 283– 296. Hughes, B., Chambers, L., Lansdell, H., & White, R. 2004. Cities, area and transport energy. Road & Transport Research, 13 (2): 72. Inman D, Jeffrey P. 2006. A review of residential water conservation tool performance and influences on implementation effectiveness[J]. Jahanshahi, K., & Jin, Y. 2016. Trend-Breaking Influences of Built Form on Travel in UK Cities: Evidence from New Quantifications of Within-and Between-Built-Form Variations. Transportation Research Record: Journal of the Transportation Research Board (2564): 31–40. Jahanshahi, K., Jin, Y., & Williams, I. 2015. Direct and indirect influences on employed adults’ travel in the UK: New insights from the National Travel Survey data 2002–2010. Transportation Research Part A: Policy and Practice, 80: 288–306. Jiang, H. 2015. Investigation and analysis of water use behavior of Beijing residents and water saving measures. Water resources protection, 31 (5). Jiang, W. 2019. Investigation on current situation of residential water consumption in Chongqing and Study on water consumption quota. Water supply and drainage, 0 (2). Jiang, Y., Gu, P., Chen, Y., He, D., & Mao, Q. 2017. Influence of land use and street characteristics on car ownership and use: Evidence from Jinan, China. Transportation Research Part D: Transport and Environment, 52: 518–534. Jiang, Y., Peng, C, & Hu, S. 2015. Classification of building energy consumption in China. Construction technology (14): 22–26. Jiang, Y., He, D, & Christopher, Z. 2011. Study on the influence of urban block form on Residents’ travel energy consumption. Urban traffic (04).

References

29

Jin, W., Zhang., Wang, H., Shang, Z., Liu, S. 2018. Urbanization_Dynamic evolution and threshold effect of water resources consumption. Keller, R., & Vance, C. 2013. Landscape pattern and car use: Linking household data with satellite imagery. Journal Of Transport Geography, 33: 250–257. Kenney D, Goemans C., Klein R, Lowery J, Reidy K. 2008. Residential water demand management: lessons from Aurora, Colorado[J]. Kenworthy, J. R., & Laube, F. B. 1996. Automobile dependence in cities: an international comparison of urban transport and land use patterns with implications for sustainability. Environmental impact assessment review, 16 (4–6): 279–308. Khattak, A. J., & Rodriguez, D. 2005. Travel behavior in neo-traditional neighborhood developments: A case study in USA. Transportation Research Part a-Policy And Practice, 39 (6): 481–500. Kong, K., Xu., Tian, S. 2011. Step water price optimization design of < rural drinking water safety project. Krizek, K. J. 2003a. Neighborhood services, trip purpose, and tour-based travel. Transportation, 30 (4): 387–410. Krizek, K. J. 2003b. Residential relocation and changes in urban travel - Does neighborhood-scale urban form matter? Journal Of the American Planning Association, 69 (3): 265–281. Larranaga, A. M., Rizzi, L. I., Arellana, J., Strambi, O., & Cybis, H. B. B. 2016. The influence of built environment and travel attitudes on walking: A case study of Porto Alegre, Brazil. International Journal Of Sustainable Transportation, 10 (4): 332–342. Li, J. 2015. Research Progress on measurement methods of farmers’ attitude towards natural disasters at home and abroad. Anhui Agricultural Science, 43 (12). Li, S., & Zhao, P. 2017. Exploring car ownership and car use in neighborhoods near metro stations in Beijing: Does the neighborhood built environment matter? Transportation Research Part D: Transport and Environment, 56: 1–17. Li, Y., & Zhang, H. 2016. Regional differences in the impact of urbanization on household CO2 emissions, 38 (3): 545–556. Li, C., Tao., Liu, S. 2010. Price elasticity of domestic water consumption in Suzhou. Li, D., Guan, M. L., & Li, Y. J. 2016. Self-control and factors based on the Theory of Planned Behavior: Predicting adolescent pedestrians’ unsafe road-crossing behaviors in China. International Journal Of Psychology, 51: 1147–1148. Liu, Y. 2015. Grassroots disaster management research: participation status, experience characteristics and working mechanism——Taking Mianyang disaster management research as a sample. Research on economic and social development, 0 (2). Liping Yan, P. M. a. E. D. 2017. Understanding ethnic differences in perceptions, attitudes, and behaviours: a study of domestic water use in Sydney[J]. Liu, Z., Ma, J., & Chai, Y. 2016. Neighborhood-scale urban form, travel behavior, and CO2 emissions in Beijing: implications for low-carbon urban planning. Urban Geography, 38 (3): 381–400. Ma, J., Chai, Y., & Zhang, Y. 2009. Spatial differentiation of influencing factors on shopping trip of Beijing residents (12). Ma, J., Liu, Z. L., & Chai, Y. W. 2015. The impact of urban form on CO2 emission from work and non-work trips: The case of Beijing, China. Habitat International, 47: 1–10. Marsden, G., & Rye, T. 2010. The governance of transport and climate change. Journal Of Transport Geography, 18 (6): 669–678. McCormack, E., Rutherford, G. S., & Wilkinson, M. G. 2001. Travel impacts of mixed land use neighborhoods in Seattle, Washington. Land Development And Public Involvement In Transportation (1780): 25–32. Messner, F., & Meyer, V. 2006. Flood damage, vulnerability and risk perception - challenges for flood damage research. Ufz Discussion Papers, 67: 149–167. Miller, E. J., & Ibrahim, A. 1998. Urban form and vehicular travel - Some empirical findings. Land Use And Transportation Planning And Programming Applications (1617): 18–27.

30

1 The Current Status of Research in the Field …

Modarres, A. 2013. Commuting and energy consumption: toward an equitable transportation policy. Journal Of Transport Geography, 33: 240–249. Nauges C, Thomas A. 2000. Privately-operated water utilities, municipal price negotiation, and estimation of residential water demand: the case of France[J]. . Newman, P. W., & Kenworthy, J. R. 1989. Gasoline consumption and cities: a comparison of US cities with a global survey. Journal of the American planning association, 55 (1): 24–37. Nieswiadomy M L, Molina D. J. 1991. A note on price perception in water demand models[J]. Nyong A O, Kanaroglou P. S. 2001. A survey of household domestic water-use patterns in rural semi-arid Nigeria[J]. Osberghaus, D. 2015. The determinants of private flood mitigation measures in Germany—Evidence from a nationwide survey. Ecological Economics, 110: 36–50. Ou, J. P., Liu, X. P., Li, X., & Chen, Y. M. 2013. Quantifying the relationship between urban forms and carbon emissions using panel data analysis. Landscape Ecology, 28 (10): 1889–1907. Peacock, W. G., Brody, S. D., & Highfield, W. 2005. Hurricane Risk Perceptions Among Florida’s Single Family Homeowners. Landscape & Urban Planning, 73 (2): 120–135. Potoglou, D., & Kanaroglou, P. S. 2008. Modelling car ownership in urban areas: a case study of Hamilton, Canada. Journal of Transport Geography, 16 (1): 42–54. Prati, G., Catufi, V., & Pietrantoni, L. 2012. Emotional and behavioural reactions to tremors of the Umbria-Marche earthquake. Disasters, 36 (3): 439–451. Qin, B., & Han, S. S. 2013. Planning parameters and household carbon emission: Evidence from high- and low-carbon neighborhoods in Beijing. Habitat International, 37: 52–60. Raikes, J., Smith, T. F., Jacobson, C., & Baldwin, C. 2019. Pre-disaster planning and preparedness for floods and droughts: A systematic review. International Journal of Disaster Risk Reduction, 38. Ran, X., Zhou, S., Mao, J. 2006. Transportation system and land use of high density development cities: a case study of Guangzhou. Renwick M E, Archibald S. O. 1998. Demand side management policies for residential water use: who bears the conservation burden? [J]. Renwick M E, Green R. D. 2000. Do residential water demand side management policies measure up? An analysis of eight California water agencies[J]. Ribeiro, S. K., & Balassiano, R. 1997. CO2 emissions from passenger transport in Rio de Janeiro. Transport policy, 4 (2): 135–139. Robert Cervero, G. B. Arrington. 2008. Vehicle Trip Reduction Impacts of Transit-Oriented Housing. Journal of Public Transportation, Vol. 11, No. 3,: 1–17. Saelens, B. E., & Handy, S. L. 2008. Built environment correlates of walking: a review. Med Sci Sports Exerc, 40 (7 Suppl): S550–566. Scheiner, J. 2010. Social inequalities in travel behaviour: trip distances in the context of residential self-selection and lifestyles. Journal Of Transport Geography, 18 (6): 679–690. Schoenau, M., & Müller, M. 2017. What affects our urban travel behavior? A GPS-based evaluation of internal and external determinants of sustainable mobility in Stuttgart (Germany). Transportation Research Part F: Traffic Psychology and Behaviour, 48: 61–73. Shekarchian, M., Moghavvemi, M., Zarifi, F., Moghavvemi, S., Motasemi, F., & Mahlia, T. M. I. 2017. Impact of infrastructural policies to reduce travel time expenditure of car users with significant reductions in energy consumption. Renewable and Sustainable Energy Reviews, 77: 327–335. Shim, G.-E., Rhee, S.-M., Ahn, K.-H., & Chung, S.-B. 2006. The relationship between the characteristics of transportation energy consumption and urban form. The Annals of Regional Science, 40 (2): 351–367. Solomon, S., Qin, D., Manning, M., Averyt, K., & Marquis, M. 2007. Climate change 2007-the physical science basis: Working group I contribution to the fourth assessment report of the IPCC: Cambridge university press. Sperling, D., & Salon, D. 2002. Transportation in developing countries: an overview of greenhouse gas reduction strategies.

References

31

Statistics, I. 2011. CO2 emissions from fuel combustion-highlights. IEA, Paris https://www.iea.org/ co2highlights/co2highlights.pdf. Cited July. Steg, L. 2005. Car use: lust and must. Instrumental, symbolic and affective motives for car use. Transportation Research Part a-Policy And Practice, 39 (2–3): 147–162. Stocker, T., Qin, D., Plattner, G., Tignor, M., Allen, S., Boschung, J., Nauels, A., Xia, Y., Bex, V., & Midgley, P. 2013. IPCC 2013: summary for policymakers in climate change 2013: the physical science basis, contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change: Cambridge University Press, Cambridge, New York, USA. Su, T., Zhang, J., Li, J., & Ni, Y. 2011. Empirical Study on Influencing Factors of urban traffic carbon emissio—Evidence from panel data of Beijing, Tianjin, Shanghai and Chongqing. (05). Sun, B., Ermagun, A., & Dan, B. 2017. Built environmental impacts on commuting mode choice and distance: Evidence from Shanghai. Transportation Research Part D: Transport and Environment, 52: 441–453. Sun, S., Qi, W., Deng, Y. & Wei, B. 2018. Research on disaster behavior abroad: origin, topics and discoveries: 27–34. Thorhauge, M., Haustein, S., & Cherchi, E. 2016. Accounting for the Theory of Planned Behaviour in departure time choice. Transportation Research Part F: Traffic Psychology and Behaviour, 38: 94–105. Trakoli, A. 2015. Risk Communication: A Handbook for Communicating Environmental, Safety, and Health Risks. Australian & New Zealand Journal of Public Health, 39 (3): 297–298. Tyrinopoulos, Y., & Antoniou, C. 2013. Factors affecting modal choice in urban mobility. European Transport Research Review, 5 (1): 27–39. Van Acker, V., & Witlox, F. 2010. Car ownership as a mediating variable in car travel behaviour research using a structural equation modelling approach to identify its dual relationship. Journal Of Transport Geography, 18 (1): 65–74. van de Coevering, P., & Schwanen, T. 2006. Re-evaluating the impact of urban form on travel patterns in Europe and North-America. Transport Policy, 13 (3): 229–239. Van, H. T., & Fuji, S. 2006. A Cross Asian Country Analysis In Attitudes toward Car And Public Transport. Transportation Studies: Sustainable Transportation, Proceedings Of the 11th International Conference Of Hong Kong Society for Transportation Studies: 803–803. Vance, C., & Hedel, R. 2007. The impact of urban form on automobile travel: disentangling causation from correlation. Transportation, 34 (5): 575–588. Vitek, J. D., & Berta, S. M. 1982. Improving perception of and response to natural hazards: The need for local education. Journal of Geography, 81 (6): 225–228. W, C. R. 1989. Effectiveness of conservation—oriental water rates in Tucson[J]. Wan, X., Chen, J., & Wang, W. 2007. Study on the travel characteristics of group cities in China. Journal of urban planning (03). Wang, X. 2012. Research on family disaster preparedness strategy from the perspective of vulnerability analysis. Wang, H. 2014. The influence of “disaster preparedness education” on College Students’ disaster preparedness awareness, behavior and knowledge 30 (3). Wang, X. 2015. Study on the source of “waste” of domestic water for urban residents in the process of urbanization in China, 0 (11). Wang, J. 2018. Analysis and Prospect of domestic water quota for urban residents. Municipal Technology, 36 (5). Wang, D. G., & Lin, T. 2014. Residential self-selection, built environment, and travel behavior in the Chinese context. Journal Of Transport And Land Use, 7 (3): 5–14. Wang, D., & Zhou, M. 2017. The built environment and travel behavior in urban China: A literature review. Transportation Research Part D: Transport and Environment, 52: 574–585. Wang, D., Chai, Y., & Li, F. 2011a. Built environment diversities and activity–travel behaviour variations in Beijing, China. Journal of Transport Geography, 19 (6): 1173–1186.

32

1 The Current Status of Research in the Field …

Wang, Y., Feng., Zhao, D., Mao, D. 2011b. Correlation analysis on Influencing Factors of water consumption of residents in Jinhua City. Wright, L., & Fulton, L. 2005. Climate change mitigation and transport in developing nations. Transport Reviews, 25 (6): 691–717. Xia, L. 2018. Water resources assessment of Jiangxi Province, 44 (6). Xiao, Z., Chai, Y., & Liu, Z. 2011. Quantitative distribution and influencing factors of carbon emissions from household daily travel in Beijing (09). Xiong, S. 2018. Empirical Analysis on Influencing Factors of domestic water consumption of urban residents -- Taking Chengdu as an example, 18 (1). Xue, X. Z., Ren, Y., Cui, S. H., Lin, J. Y., Huang, W., & Zhou, J. 2015. Integrated analysis of GHGs and public health damage mitigation for developing urban road transportation strategies. Transportation Research Part D-Transport And Environment, 35: 84–103. Yan, X., & Crookes, R. J. 2009. Reduction potentials of energy demand and GHG emissions in China’s road transport sector. Energy Policy, 37 (2): 658–668. Yang, W., Li, T., & Cao, X. 2015a. Examining the impacts of socio-economic factors, urban form and transportation development on CO 2 emissions from transportation in China: A panel data analysis of China’s provinces. Habitat International, 49: 212–220. Yang, W. Y., Li, T., & Cao, X. S. 2015b. Examining the impacts of socio-economic factors, urban form and transportation development on CO2 emissions from transportation in China: A panel data analysis of China’s provinces. Habitat International, 49: 212–220. Yang, X., Wu, Z., & Li, Y. 2011. Difference between real-life escape panic and mimic exercises in simulated situation with implications to the statistical physics models of emergency evacuation: The 2008 Wenchuan earthquake. 390 (12): 2375–2380. Yang, X., Li., Tian, Z., Nie, E., Fang, S. 2018. Research on Farmers’ domestic water consumption in the process of Urbanization. Yang, Y. 2013. Quantitative study on the relationship between built environment and household travel energy consumption in Jinan. Ye, H. 2010. Community disaster management” is the basis of disaster prevention and mitigation. (07): 26–27. Zahabi, S. A. H., Miranda-Moreno, L., Patterson, Z., Barla, P., & Harding, C. 2012. Transportation Greenhouse Gas Emissions and its Relationship with Urban Form, Transit Accessibility and Emerging Green Technologies: A Montreal case study. Proceedings Of Ewgt 2012 - 15th Meeting Of the Euro Working Group on Transportation, 54: 966–978. Zegras, C. 2010. The Built Environment and Motor Vehicle Ownership and Use: Evidence from Santiago de Chile. Urban Studies, 47 (8): 1793–1817. Zhang, L. 2016. Study on water saving and reclaimed water reuse of urban domestic water, 0 (8). Zhang, H., Dong., Yan, J., Run, N. 2009. Response analysis of urban domestic water to climate change in Xi’an. Zhang, L., Liu., Liu, T., Qian, S., Lin, L. 2011. Evaluation and revision analysis of urban domestic water quota in Sanya City. Zhao, P. 2010. Sustainable urban expansion and transportation in a growing megacity: Consequences of urban sprawl for mobility on the urban fringe of Beijing. Habitat International, 34 (2): 236–243. Zhao, P. J. 2014. The Impact of the Built Environment on Bicycle Commuting: Evidence from Beijing. Urban Studies, 51 (5): 1019–1037. Zhao, G. 2017. Prediction and analysis of domestic water consumption based on grey system, 31 (6).

References

33

Zhao, Y., & Chai, Y. 2013. Residents’ activity-travel behavior variation by communities in Beijing, China. Chinese Geographical Science, 23 (4): 492–505. Zhao, H., Zhong., Ao, Y. 2019. Analysis on Influencing Factors of water saving behavior of residents in rural areas of Sichuan Province. Zhao, L., Gong, Y. X., Lin, Y. Y., Zhang, N., & Duan, Z. Y. 2015. The Effects of Residential Quarters’ Built Environment on Travel Behavior Using PLS-SEM Approach: A Case Study of Nanshan District, Shenzhen. 2015 23rd International Conference on Geoinformatics.

Chapter 2

Comprehensive Evaluation of Changes to the Rural Built Environment of China

Abstract This chapter presents a comprehensive evaluation of rural built environment developments through the collection and analysis of rural data from statistical yearbooks. Macro perspective trends within the rural built environment in China are identified using factor analysis of multi index panel data from a total of 24 samples. It is found that the rural built environment in China presents dynamic fluctuations, with an irregular distribution of public factors. Significantly, the overall ranking of built environmental scores across provinces does not accord with the level of economic development. Evens so, the rural energy and ecological environment remain relatively steady. These findings coincide with the current stage and policies of rural urbanization and new rural construction in China. China’s rural areas are generally experiencing rapid development, but development is variable across provinces. Findings suggest that the government should first prioritize the continuous promotion of those projects that meet the needs of the people and which support their livelihood. Keywords Rural built environment · Comprehensive evaluation · Multi index panel data · Factor analysis

2.1 Introduction In 1978, China implemented the internal reform starting from the countryside. After 38 years, Great changes have taken place in China’s rural areas especially in the past ten years, Chinese government have attached great importance to the construction of rural infrastructure. Since 2005, the central “No.1 Document” continuously proposed to strengthen the construction of agricultural and rural infrastructure and increase local governments’ support for rural infrastructure construction. Correspondingly, China’s rural fixed asset investment increased significantly in the past ten years. The investment of rural fixed assets in 2015 was 2.64 times that of 2005, and rural infrastructure construction increased significantly, growth rate in 2015 was 1.86 times

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_2

35

36

2 Comprehensive Evaluation of Changes …

that of 2005.1 And the investment in fixed assets of rural households in 2015 was 2.64 times as many as in 2005.2 While the built environment in rural China has changed greatly, scholars have either studied the urban built environment or studied the rural infrastructure construction in china, and no scholars have studied the environmental changes in China’s rural areas. Studies of built environment mainly focus on the impact of urban built environment on travel behavior. At present, a number of literature reviews on this topic have been published (Boarnet 2011; Cao et al. 2009b; Ewing and Cervero 2001, 2010; Ewing et al. 2015), and more than 200 papers on this topic had been published by 2010 (Ewing and Cervero 2010). Chinese scholars’ research on built environment and travel behavior is still in the initial stage of exploration and development (Cao 2015) (only two public publications in this field were found before 2005), but the motive force of scholars’ interest in this topic is unprecedented in the development of the city (Wang and Zhou 2017). Chinese scholars have used various methods to study the impact of built environment on traffic behavior in several cities of China (Cao 2014; Ettema and Nieuwenhuis 2017; Pan et al. 2009; Wang and Lin 2014; Zhang et al. 2014; Zhao 2014). More than 60% of existing literatures regard the relationship between work and housing as the main research content of the built environment (Wang and Zhou 2017). Employment, transportation, neighborhood relations and daily living facilities are also often considered in studies (Feng et al. 2014; Lin and Wang 2015; Ma et al. 2015; Pan et al. 2009; Zhao and Lu 2011; Zhao et al. 2010). It is not difficult to find that the case studies of Chinese scholars mainly focus on such first-tier cities as Beijing, Shanghai, Guangzhou and Nanjing (Wang and Zhou 2017). Few studies have been done in undeveloped areas such as western China, and the research on rural areas in China has not been found so far. Through the literature research, it is not hard to find that the research on residents’ travel behavior theory mostly comes from the theory of planning behavior (TPB) and the theory of reasoned action (TRA) (Schoenau and Müller 2017; Van Acker et al. 2007). In the study of the relationship between the built environment and travel behavior, the influencing factors are constantly expanding. Based on the comprehensive literature analysis (Cao et al. 2009a; Cao and Yang 2017; Ding et al. 2017; Handy et al. 2005; Li and Zhao 2017; Liu et al. 2016; Sun et al. 2017), the theoretical framework and relationship between the built environment and travel behavior is shown in Fig. 2.1. At present, there is no literature related to China’s rural built environment, but there is a large amount of literature on the construction of China’s rural infrastructure. Scholars’ researches related rural infrastructure are mainly focused on four aspects: the sustainable development of infrastructure construction (Shen et al. 2011; Zhi et al. 2016; Gan et al. 2009), the performance evaluation of government investment (Li et al. 2014; Luo 2014), the farmers’ participation in infrastructure construction and farmers’ satisfaction (Fan and Luo 2009; Li 2011, 2012; Tang et al. 2010), and the investment efficiency and capital stock of rural infrastructure (Xu 2010, 2011). There 1 “China 2 “China

Statistical Yearbook” (2006–2016). Rural Statistical Yearbook” (2006–2016).

2.1 Introduction

37

1500

80 60

1000

40 20

500

0

10

200

0

150

-10

100

-20

50

-30

0

-40 2000 2002 2004 2006 2008 2010 2012 2014

2014

2012

2010

2008

2006

2004

2002

(20) 2000

0

250

rural hydropower installed capacity growth rate %

50

400

100

0

50

(10)

0

(20)

0 Township Cultural StaƟon growth rate % AdopƟve Welfare InsƟtuƟons growth rate % Year-end adopƟon rate growth rate %

80

(50)

4

50

2

0

2014

2010

2012

2006

2008

2004

2000 (50)

2002

2014

2012

2010

2008

2006

2004

2002

200 2000

0

Bike growth rate % Motorcycle growth rate %

(30)

Health personnel growth rate % Bed growth rate % Township(town) hospital growth rate %

50

(50)

0 2014

2012

2010

2008

2006

2004

2002

2000

2014

2012

2010

2008

30 20 10

(2)

SanitaƟon penetraƟon rate growth rate % Rural water improvement growth rate %

2006

0

20

2004

2000

40

2002

40

60

(100)

0

Primary school student number growth rate % Primary school teachers growth rate % Student/teacher raƟo growth rate %

Fig. 2.1 Rural built environment changes of China

is still a blank in the impact of rural infrastructure construction on rural residents, especially in rural residents’ travel behavior and energy consumption. The main reason for the lack of research on rural built environment may be that the huge rural population, the wide coverage area, the large geographical regional gap, the decentralized rural household layout, and data collection difficulties. Meanwhile,

38

2 Comprehensive Evaluation of Changes …

China’s rural areas are in a rapid development stage of new rural construction and rural urbanization, and the rural built environment is in a dynamic change process. This paper attempts to give an answer to following questions by sort out China’s relevant statistical year books: (1) What indicators does China’s rural built environment have? (2) What trends and laws does China’s rural built environment have? (3) what is the integrated development level of the China’s rural built environment and other provinces (municipalities)? (4) Are there any internal laws among indicators that affect the changes of rural built environment? (5) What is the practical significance of statistical analysis conclusion? This article is organized as following: the second part includes the process of data collection and descriptive analysis of data; the third part includes introductions factor analysis of multi index panel data and parameters; the fourth part includes conclusion and discussion of the comprehensive factor analysis evaluation; and the last part is the summary, deficiencies and prospects of the research.

2.2 Indicator, Data Sources and Descriptive Analysis 2.2.1 Data Sources and Content Analysis A large number of literatures about the relationship between urban built environment and travel behavior mainly reflect the urban built environment through density, diversity, urban design, city size, polycentrism and accessibility (as shown in Fig. 2.1). Indicators of urban built environment may be a reference for rural built environment, but it is difficult for the rural built environment research to consider with the urban micro environment index system due to rural backward state of development, the wide regional coverage, obvious geographical differences, and different characteristics of rural built environment. Meanwhile, the indicators for the built environment are not uniform. According to the comprehensive analysis of the literature, there are as many as 171 built environmental measurement indicators in the international sample (excluding the Chinese sample), and as many as 99 indicators of the built environment from Chinese sample, as detailed in the supplementary materials. This paper attempts to construct an index system of China’s rural built environment start from the rural education, culture, health, welfare, ecological energy, transportation facilities and other basic needs of the people’s livelihood environment, combined with the “Chinese Statistical Yearbook”, “China Rural Statistical Yearbook”, “Chinese Auto Market Statistical Yearbook”, “Sichuan Statistical Yearbook” and other provinces Statistical Yearbook. Details of rural built environment and indicators are shown in Table 2.1.

2.2 Indicator, Data Sources and Descriptive Analysis Table 2.1 Contents and indicators of built environment

39

Built environments

Indicators

Rural traffic facilities environment

a. The number of motorcycles per 100 households in rural areas b. The number of bicycle per 100 households in rural areas

Rural cultural welfare environment

a. Number of township cultural stations b. Number of elderly adoption welfare institutions c. Number of adoptions at the end of the year

Rural ecological environment

a. The cumulative benefit rate of rural water improvement b. Prevalence rate of sanitary latrines

Health environment

a. Number of township (township) health centers b. Number of health personnel c. Bed number

Rural energy environment a. Number of rural hydropower stations b. Rural installed capacity c. Rural generating capacity Rural educational environment

a. The ratio of teachers and students in rural primary schools

2.2.2 Trend Analysis of Rural Built Environment In order to analyze the overall change trend of rural environment built, this paper takes two samples from China and Sichuan as an example, and uses fixed growth ratio to calculate change trends in the past 16 years of each built environment indicator according to the continuity and integrity of the data. The following is the formula for calculating the growth rate of indexes in each period. ni =

xni − x2000 (i = 2000, . . . , 2015; n = 1,2, . . . ,14) x2000

China’s rural power generation growth rate and the rural hydropower installed capacity growth rate continued to have the fastest grow speed in the past 16 years, with the overall growth rate of 1046.98 and 985.61% respectively, while the growth rate of was not so obvious, and the overall growth rate was 58% by the end of 2015, which shown that rural hydropower stations are expanding scale rapidly rather than increasing quantities. From the overall growth trend chart, the growth rate of rural power generation is consistent with the trend of the installed capacity of rural

40

2 Comprehensive Evaluation of Changes …

hydropower stations, indicating that the annual increase of installed capacity was put into production (Fig. 2.1a). With the development of rural road construction, the number of motorcycle of rural residents had also continued to increase. The overall growth rate of motorcycles per 100 households was 208.22%, and by the end of 2015, the proportion of rural residents with motorcycles was 68%. This corresponds to a continued decline in rural bicycle ownership in China. By the end of 2012, the overall growth rate of bike number 100 households was −34.44%. There were no statistics of bicycle ownership in 2013–2015, but data of bicycle ownership per 100 households began to appear in statistics, which were 9.90, 11 and 13.30 respectively (Fig. 2.1b). Access to roads will increase the number of cars holding by residents and reduce the number of physical activities such as cycling (Aziz et al. 2017; Heesch and Langdon 2016; Jain and Tiwari 2016; Tiwari et al. 2016). With the aspect of social and cultural welfare, the township cultural station and the elderly adoption welfare institutions showed a downward trend, with overall growth rates of −12.98 and −39.06% respectively. But at the end of the year, the number of adoptions continued to grow, with an overall growth rate of 169.05%, indicated that pension welfare institutions exist the situation of merger and expansion in the process of rural urbanization (Fig. 2.1c). Similarly, the number of medical institutions in rural areas is decreasing, but the number of beds is increasing in total, which were −25.21 and 62.78% respectively (Fig. 2.1d). In the aspect of rural ecological environment, the growth of rural water supply total benefit rate fluctuated in X axis, shown that the rural water improvement project is basically universal and it would change with the rural resident population changes. The growth rate of rural sanitary toilets, which was 75% by the end of 2015, continued to increase year by year, however, it was not obvious compared with other data (Fig. 2.1e). The growth rate of rural primary school students and full-time teachers continued to decline, and the overall growth rate were −44.64 and −65.12% respectively by the end of 2015. There was a phenomenon of merging and relocation in the process of urbanization, leaded to a continued decline of the number rural schools, students and teachers, but, the number of teachers decrease slower than the number students and the ratio of teachers to students in rural primary schools showed an increasing trend (Fig. 2.1f). Figure 2.1 show that China’s rural built environment presents a regular change, basically no big fluctuations, indicating that the overall control of China’s rural areas, which was in an orderly development process, was well in the past 16 years. Compared with the built environment in rural China, the environmental changes in rural areas of Sichuan province fluctuated greatly, as shown in Fig. 2.2. The largest growth of rural built environment in Sichuan was still in the rural energy environment, and the overall growth rate of rural hydropower station installed capacity was as high as 1323.46, exceeding the national data with 337.85. The overall growth rate of rural power generation in Sichuan, 1113.15, was 66.17 more than the national data (Fig. 2.2a). The rate of change of bicycle and motorcycle numbers per 100 households in rural areas of Sichuan is higher than that of the whole country. The overall growth rate of bicycles number was −66.63, the absolute ownership rate of which was 16.45%, and the overall growth rate of motorcycles number was 422.84, 214.62 higher than that of the whole country, at the end of 2013. There was

2.2 Indicator, Data Sources and Descriptive Analysis

2000

50 1000

(50)

20

400

0

300

-20

200

-40

100

-60

0

-80 2000 2002 2004 2006 2008 2010 2012 2014

2000 2002 2004 2006 2008 2010 2012 2014

0

500

0 rural hydropower staƟon growth rate %

50

1000 500 2000 2002 2004 2006 2008 2010 2012 2014

0 -50

Motorcycle growth rate % Bike growth rate %

0 -500

40

150

20

100

0

50

(20) Township Cultural StaƟon growth rate % AdopƟve Welfare InsƟtuƟons growth rate % Year-end adopƟon rate growth rate %

50

(40)

0

2000 2002 2004 2006 2008 2010 2012 2014

100

41

(50)

Township(town) hospital growth rate % Health personnel growth rate % Bed growth rate %

60

5

40

2000 2002 2004 2006 2008 2010 2012 2014

-5

-50

0

20 -5

0 -10

Primary school student number growth rate % Student/teacher raƟo growth rate % Primary school teachers growth rate %

Fig. 2.2 Rural built environment changes of Sichuan

-20

2000 2002 2004 2006 2008 2010 2012 2014

0

0

-10

Rural water improvement growth rate % SanitaƟon penetraƟon rate growth rate %

42

2 Comprehensive Evaluation of Changes …

no bicycle statistics in 2014 and 2015, but car statistics began to appear in that time, and the car ownership rate was 7.54 and 8.56% respectively (Fig. 2.2b). From the overall growth trend, there is no significant fluctuations in rural energy indicators, while growth rates of other built environment indicators fluctuated obviously, such as rural elderly adoption welfare institutions, township cultural stations (Fig. 2.2c), toilet popularity and rural health personnel (Fig. 2.2d). But the overall trend is that the number of institutions is decreasing, such as the elderly welfare institutions and health centers, and the overall level of accommodation is still growing, such as the number of elderly institutional adoptions and the number of beds in health centers. Situation of Sichuan is consistent with the overall trend of the country. In the process of urbanization, Sichuan province had the situation of service institutions merger and expansion. For rural education, the change is generally consistent with the country. The number of students and teachers in rural primary school were continued to decline, and the number of teachers decreased faster than number of students. The overall ratio of teachers to students had increasing (Fig. 2.2e). The rural ecological environment, different with other environmental indicators, fluctuated among X axis. Rural water supply benefit rate is particularly evident, indicating that rural water supply tends to be popularized. It mainly becaused by the resident population number (Fig. 2.2f).

2.3 Comprehensive Evaluation of Rural Built Environment Change To comprehensively evaluate the changes of the built environment in China’ rural and the differences among provinces, this paper uses the multi index panel data factor analysis to study data from statistical yearbooks through statistical analysis. Indicators totally meets the specifications listed in Table 2.1. Data and provinces were chosen by the principle that data missing values cannot be more than three consecutively. Finally, data of eight years from 23 provinces (including two municipalities, Chongqing and Tianjijn) and China was chosen. Missing values are supplemented by interpolation, and data from statistical yearbooks are homogenized by population or household.

2.3.1 Factor Analysis of Multi Index Panel Data Factor analysis is a multivariate statistical analysis method which converts multiple measured variables into a few irrelevant comprehensive indexes. The few comprehensive indexes, namely the factor, can reflect the main information represented by the multiple measured variables, and explain the dependency relationship between these measured variables. That is to say, factor analysis is a method to study how to

2.3 Comprehensive Evaluation of Rural Built Environment Change

43

condense a large number of measured variables into a few factors with minimal information loss. Steps of multivariable panel data factor analysis are shown by following formulas. 1. Calculating public factor score of each year

Zit (t) = μ + ai1 (t)Fi1 (t) + ai2 (t)Fi2 (t) + · · · + ain (t)Fin (t) + εi (t) (i = 1, 2, . . . , 14; t = 2008, 2009, . . . , 2015)

(2.1)

Among them, Zit (t) is the observed random variable, which represents the individual characteristics of the sample; Fi (t) is the common factor, i, of the t year; air (i = 1,2,3, . . . , p) is factor loadings; εit is the special factor part that is not included in the first n common factors. 2. Calculating the comprehensive score of rural built environment of each province in each year Environmental change formula of province j in the t year is:  N

ϕi (t) · Fij (t)   N i=1 ϕi (t)



i=1

Sj (t) =

(2.2)

ϕi (t) is the contribution ratio of the factor, i, in the t year; Fij (t) is the common factor score of the factor, i, of province j (Dong et al. 2009). 3. Calculating the total score of common factor of panel data Comprehensive calculation formula of factors in the i year of province j is: T Gij =

ϕi (t) · Fij (t) N i=1 ϕi (t)

t=1

(2.3)

4. Calculating panel data comprehensive score of each province Calculation formula of comprehensive score of built environment for province J in eight years is:  T S*j =

σ(t) · Fij (t)   T t=1 σ(t) t=1

 (2.4)

σ(t) represents the cumulative contribution rate of variance in the factor analysis of the province j in the t year.

44

2 Comprehensive Evaluation of Changes …

Table 2.2 KMO and Bartlett’s test

Kaiser-Meyer-Olkin measure of sampling adequacy Bartlett’s Test of Sphericity

.526

Approx. χ 2

279.574

df

91

Sig.

.000

2.3.2 Empirical Analysis The evaluation index of rural built environment in this paper is 14 indicators with different dimension. The dimension inconsistency is eliminated first before analysis, and the data is standardized by Z-score. KMO and Bartlett sphericity test were chosen to the applicability of the factor analysis. Taking the data of 2008 as an example, the KMO statistic is .526, satisfy the minimum standard, .5, and the Bartlett spherical test corresponding probability Sig is .00, rejecting the original assumption of the unit matrix, indicating the significant correlation between variables and that the data of 2008 is suitable for factor analysis. And the data of other years is also suitable for factor analysis through KMO and Bartlett sphericity test. Table 2.2 is based on the sample data of 2008 as an example. SPSS23.0 is chosen to conduct factor analysis for 14 indicators. Common factor was chosen by the principle that the eigenvalue is more than 1, and Maximum variance rotation method is used to extract common factors. The Total Variance Explained is derived from SPSS23.0 is shown in Table 2.3. From Table 2.3, eigenvalues of the first four common factors are more than 1, and the rate of cumulative variance interpretation have reached 78.910%. So, the first four common factors are extracted, with few information losses. The initial factor load matrix is orthogonally rotated by the maximum variance method, and the component score coefficient matrix is estimated by the regression method. The Component Score Coefficient Matrix is obtained, as shown in Table 2.4. Score of four common factor is calculated according to formula (2.1) (taking sample data of China in 2008 as an example): ⎧ F1 = 0.092 × X1 + 0.072 × X2 + · · · + 0.065 × X14 = −0.1239 ⎪ ⎪ ⎨ F2 = 0.225 × X1 + 0.146 × X2 + · · · − 0.091 × X14 = 0.0771 ⎪ = −0.088 × X1 − 0.061 × X2 + · · · + 0.061 × X14 = −0.3039 F 3 ⎪ ⎩ F4 = 0.038 × X1 + 0.200 × X2 + · · · + 0.424 × X14 = −0.1375 According to the formula (2.2), the comprehensive score of each province is calculated as follows (taking the data of Chinese sample in 2008 as an example): 23.789 ∗ F1 + 23.314 ∗ F2 + 16.990 ∗ F3 + 14.817 ∗ F4 78.9091 = −0.1058

SChina (2008) =

.074

.038

.003

13

14

.368

8

12

.467

7

.087

.646

6

11

.842

5

.277

1.425

4

.150

2.418

3

10

3.325

9

3.878

2

.019

.272

.530

.619

1.075

1.979

2.630

3.337

4.616

6.013

10.181

17.273

23.753

27.703

100.000

99.981

99.709

99.179

98.560

97.485

95.506

92.876

89.539

84.923

78.910

68.729

51.456

27.703

1.425

2.418

3.325

3.878

10.181

17.273

23.753

27.703

78.910

68.729

51.456

27.703

Cumulative

Total

Variance

Extraction sums of squared loadings (%) Cumulative

Total

Variance

Initial eigenvalues (%)

1

Component

Table 2.3 Total variance explained

2.074

2.379

3.264

3.330

Total

14.817

16.990

23.314

23.789

Variance

78.910

64.093

47.103

23.789

Cumulative

Rotation sums of squared loadings (%)

2.3 Comprehensive Evaluation of Rural Built Environment Change 45

46

2 Comprehensive Evaluation of Changes …

Table 2.4 Component score coefficient matrix Component 1

2

3

4

X1

.092

.225

−.088

.038

X2

.072

.146

−.061

.200

X3

.297

.055

.035

.103

X4

.297

.014

.022

.009

X5

.283

.002

.005

−.048

X6

.012

.061

.420

.118

X7

.008

.232

.102

.211

X8

−.010

.241

.031

−.122

X9

.035

.296

.097

−.194

X10

−.137

.188

−.060

−.074

X11

.050

.020

.389

.045

X12

−.044

.009

.319

.373

X13

−.095

−.021

−.135

.249

X14

.065

−.091

.061

.424

Similarly, the comprehensive score of each year in each sample is calculated by eight factor analyses on eight years’ statistical data respectively. Table 2.5 lists only the comprehensive score of national samples in each year. At the same time, the Total Variance Explained could be estimated. Which was shown in Table 2.6. Table 2.5 The total score of national rural built environment in each year China

2008

2009

2010

2011

2012

2013

2014

2015

−.1058

−.0412

−.1071

−.0963

−.0266

−.0145

−.0305

.0280

Table 2.6 Common factor contribution rate in each year Contribution 2008 rate

2009

2010

2011

2012

2013

2014

2015

F1

23.789

23.471

24.074

23.225

24.030

24.947

23.122

22.782

F2

23.314

23.397

23.645

21.503

18.786

22.663

23.038

22.470

F3

16.990

16.993

19.734

21.263

18.380

21.285

17.412

19.020

F4

14.817

15.536

11.799

11.636

18.062

11.257

14.562

15.082

Cumulative contribution rate

78.9097 79.3963 79.2526 77.6261 79.2583 80.1523 78.1344 79.3532

2.3 Comprehensive Evaluation of Rural Built Environment Change

47

According to formula (2.3), the composite score of common factors of panel data was calculated (taking the first factor of the sample data of China as an example). T

ϕi (t) · Fi j (t) N i=1 ϕi (t) −0.1058 ∗ 23.789 − 0.0412 ∗ 23.471 + · · · + 0.0280 ∗ 22.782 = 23.789 + 23.471 + · · · + 22.782 = −0.0494

G 1China =

t=1

The panel data comprehensive score of provinces is calculated according to formula (2.4) (taking the sample data of China as an example). T

σ (t) · Fi j (t) T t=1 σ (t) −0.1058 ∗ 78.9097 − 0.0412 ∗ 79.3963 + · · · + 0.0280 ∗ 79.35 = 78.9097 + 79.3963 + · · · + 79.3532 = −0.0491

∗ = SChina

t=1

In the same way, other public factors, score of each year and 8 years’ comprehensive score of other provinces can be calculated. The statistical analysis results are shown in Figs. 2.3 and 2.4. 2.0000 1.5000 1.0000 0.5000

-0.5000 -1.0000 -1.5000 2008 2011 2014

Chian Sichuan Liaoning Jilin Heilongjiang Zhejiang Anhui Fujian Shandong Henan Hunan Guangdong Guangxi Hainan Chongqing Guizhou Yunnan Shanxi Gansu Xinjiang Tianjin Ningxia Neimenggu Jiangsu

0.0000

2009 2012 2015

Fig. 2.3 Comprehensive score in each year of provinces

2010 2013 comprehensive score

48

2 Comprehensive Evaluation of Changes … 1.4 1.2 1 0.8 0.6 0.4 0.2 -0.2 -0.4 -0.6 -0.8

Chian Sichuan Liaoning Jilin Heilongjiang Zhejiang Anhui Fujian Shandong Henan Hunan Guangdong Guangxi Hainan Chongqing Guizhou Yunnan Shanxi Gansu Xinjiang Tianjin Ningxia Neimenggu Jiangsu

0

F1

F2

F3

F4

comprehensive score

Fig. 2.4 Common factors and comprehensive scores in provinces

2.4 Results and Discussion From Fig. 2.4, provinces show irregular fluctuation on the time axis, and significant changes occurred among years, indicating that the rural built environment in China is in a period of rapid change and has not yet entered a mature and steady period, and this is consistent with the actual situation of rural urbanization and new rural construction in China. From the horizontal comparison of the overall level of the built environment in each province, we can see that the overall level of China’s rural built environment is in the middle level of provinces, and it is in line with the actual situation. Compared with other provinces, Chongqing, Fujian and Sichuan have the highest level of rural built environment, while Ningxia, Guizhou and Hunan are located in the latter three. This is not the same as the level of rural economic development in various provinces of China (Niu et al. 2010; Yang et al. 2011a). The main reason may be that rural built environmental indicators in this research do not covered all indicators of rural economic development. Meanwhile, some provinces, like Sichuan, have a well rural built environment order with a backward economic development (Niu et al. 2010), indicating that these provinces have a greater investment in rural built environment in past eight years. On the contrary, other provinces, such as Jiangsu, Guangdong, have a backward rural environment order with a greater economy (Yang et al. 2011a), showing shows that the rural built environment is relatively mature and steady in such developed regions, and the subsequent input do not change significantly. From the comprehensive score of common factors, score of the four extracted common factors from provinces is abnormal steady, showing that selection and description of China’ rural built environment evaluation indexes index is relatively steady in the perspective of statistics (shown in Fig. 2.5). The comprehensive score of each factor is basically consistent with the final comprehensive score of each

2.4 Results and Discussion

49

Fig. 2.5 Common factor distribution

province, which indicates that the evaluation results based on this index system are reliable. However, it is not difficult to find that the distribution of common factors has no steady law in the case of the common factor scores in different years. The distribution of public factors in each year is shown in Table 2.7. The main reason for this phenomenon is still the statistical dynamic changes of China’s rural built environment. Provinces promote the rural construction with different levels and inputs under the background of vigorously developing the rural economy, leading to differences in provinces’ rural environment in different years (Yang et al. 2011b). It also fully shows that China’s rural areas are in a period of rapid development, and there would not appear regular changes in a long period. Table 2.7 Distribution of rural built environment in each year F1

F2

F3

F4

2008

X3 , X4 , X5

X1 , X2 , X7 , X8 , X9 , X10

X6 , X11

X12 , X13 , X14

2009

X1 , X2 , X7 , X10

X3 , X4 , X5

X8 , X9 , X14

X6 , X11 , X12

2010

X3 , X4 , X5

X1 , X2 , X7 , X8 , X9 , X10

X6 , X11 , X13

X12 , X14

2011

X3 , X4 , X5

X1 , X7 , X8 , X9

X2 , X6 , X11 , X13

X10 , X12 , X14

2012

X3 , X4 , X5

X6 , X11 , X12

X8 , X9 , X10 , X14

X1 , X2 , X7 , X13

2013

X3 , X4 , X5

X1 , X2 , X9 , X10 , X14

X6 , X11 , X13

X7 , X8

2014

X2 , X6 , X11 , X12 , X13

X3 , X4 , X5

X9 , X10 , X14

X1 , X7 , X8

2015

X1 , X3 , X4 , X5

X6 , X11 , X13

X2 , X9 , X10 , X14

X7 , X8 , X12

50

2 Comprehensive Evaluation of Changes …

In the case of seemingly random distribution of factors, it is not difficult to find that there is also a relatively steady distribution of index in common factors with a thorough statistics for the frequency of the environmental indicators to common factors. As is shown in Fig. 2.5, X1 , X2 and X3 , which are hydropower stations numbers of ten thousand people, rural installed capacity of ten thousand people and rural power generation of ten thousand people, are relatively steady attributable to F1 . These three indicators can be well explained as the rural energy built environment. Rural energy construction environment is concentrated in F1 , and the cumulative contribution rate of variance have reach 23.67%, showing that China’s provinces have common input in the field of rural energy construction in the past eight years (2015). Combined with the statistical data of rural household electricity consumption, rural power generation is very consistent with rural household electricity consumption. Provinces’ investment in rural energy construction is to meet basic energy needs of rural residents, with the improvement of rural residents’ living conditions, the increase of household energy equipment and the continuous increase of rural residents’ demand for energy are becoming internal driving forces of rural energy construction investment. At the same time, the installed capacity of rural hydropower station and power generation tend to be consistent, which also fully shows that the energy construction investment are effective, installed equipment operation effectively, and the investment—production is reasonable under the impact of these internal driving forces (2015). According to the principle that the built environment index belongs to the public factor that the frequency is absolutely the highest, it is not difficult to find that X1 (the beneficiary population per ten thousand rural people) and X2 (population that know sanitary toilet per ten thousand rural people) belong to F2 . X1 and X2 could be regarded as rural ecological environment, and the average contribution rate of variance of F2 is about 22.35%. The distribution of F2 is relatively steady, fully shows that the rural ecological environment of provinces has relative a steady improvement in the past eight years. Based on the statistical data, in the provinces of rural water supply engineering table showed a steady trend, no obvious growth, the rural water supply engineering in many provinces show a steady trend, no obvious growth, with very small fluctuations accompanied by changes in the resident population of rural population (Xiang and Zhong 2017), indication that the water supply project have mostly been completed (Xu et al. 2017) and it can meet the demand of rural residents in a period of time. In the next period of time. However, the renovation of rural household toilets shows a steady and increasing trend, but the increase is small. It shows that the rural household latrine renovation project has been completed in a fairly large area, and the government will continue to invest in some parts that without renovation in the future (Xu et al. 2017). According to the characteristics of wide distribution of China’s rural households, further renovation projects of household latrines will be remote mountainous areas, and the regional difficulty and regional characteristics should be taken into full consideration in this area. According to the frequency that built environment attributable to common factor, X6 (rural township hospital number per ten thousand residents in rural), X9 (number of elderly adoption welfare institutions per ten thousand residents), X11 (rural township

2.4 Results and Discussion

51

cultural station number per ten thousand residents), X13 (number of bicycle per 100 rural households) and X14 (the number of motorcycles per 100 households) belong to F3 , which could be explained as rural related institutions and transportation of rural residents. X7 (health personnel ratio per 10,000 residents) and X12 (rural primary school teacher student ratio) belong to F4 , which could be explained as medical and education service ratio. The mean variance contribution rates of F3 and F4 are 18.88 and 14.09% respectively. The distribution of these two factors were unsteady during the past eight years, indexes in F3 and F4 of provinces showed irregular changes. With the construction of China’s rural urbanization and new rural areas, institutions in rural exist situation of merger, expansion and relocation, and these situation exists differences in provinces (Yu 2013). Irregular changes in the number of institutions lead to irregular changes in the staff of the institutions, such as the number of medical staff, the number of rural teachers and the number of students in rural primary schools (Yang 2009). With the continued improvement of rural facilities and area differentiation in rural facilities, rural residents’ dependence on cars have increased significantly in the past eight years, but the two indexes of vehicles are unsteady and the variance contribution is not big enough because of regional differences. As it comes to X8 (number of beds in township hospitals per ten thousand rural residents) and X10 (adoption of elderly welfare institutions at the end of the year per ten thousand rural residents) cannot be extracted according to the principle that the built environment belongs to the highest absolute frequency of the public factor. With the development of rural urbanization and new rural construction, the number of institutions is in a dynamic and adjustive process_ENREF_50 (Yu 2013), and the optional nature of medical and elderly adoption services would be affected while the demand for medical and elderly adoption services would have no influence. These two indicators represent basic needs of life, we should pay attention to this type of residents needs in the next rural construction under the dynamic process.

2.5 Conclusion This paper analyzes the built environment literature, summarizes the main micro indicators reflected in the city built environment, and establishes a macro indicator system for the comprehensive evaluation of the rural built environment. Data is drawn from the “China Statistical Yearbook”, “Chinese Rural Statistical Yearbook”, “China automobile market Statistical Yearbook” along with the statistical yearbooks of the various provinces. According to the availability, continuity and integrity of the data, descriptive analysis and multi index panel data factor analysis was carried out for the study of China’s rural built environment, and the following conclusions were obtained:

52

2 Comprehensive Evaluation of Changes …

a. The built environment of China’s rural areas presents an upward trend on the time axis, and the fluctuation is not significant among years, showing that the changes of China’s rural built environment are under the overall control; b. The comprehensive score of national rural built environment is in the middle level of the comprehensive score of rural built environment in all provinces, and this accords with the objective law of statistics; c. The rural built environment in provinces have irregular changes among years, and the comprehensive sorting fluctuates greatly. It shows that the rural areas in provinces are in dynamic development, and the development level and speed are different among years; d. The common factor scores and the comprehensive score remained abnormal consistency among provinces, indicating that factor analysis and factor extraction were steady in provinces, and the result of factor selection and factor analysis is reliable; e. The overall distribution of common factors in each year is inconsistent, indicating that the input and output of each province in different years are inconsistent, and the rural built environment in each province is in dynamic development with inconsistent level and speed; f. The distribution of indicators in rural energy and ecological environment is relatively steady, which is in line with the actual needs of rural residents for energy and also reflect the effectiveness of rural water supply and toilet renovation; g. Distribution of number of rural institutions and transport holdings show regularity, but they can be determined with the F3 and F4 according to the principle of the highest frequency, which is consistent with the phenomenon of merger, relocation and residents’ dependence on cars in the process of rural urbanization and new rural construction; h. The number of beds and the number of adoptions at the end of each year per capita could not be attach to any common factor. One of the reason is the dynamic change in institution number, and another reason is the steady demand of medical and geriatric adoption services. Therefore, this type of residents’ needs should be satisfied first in the next step of rural construction. On the one hand, this paper tries to fill the gap in the study of rural built environment, and on the other hand, uses the simplest research methods for preliminary exploration and expands the indicators of China’s rural built environment. The overall conclusion shows that China’s rural built environment is in dynamic development, and it has a strong theoretical and social significance for the continuous study of China’s rural built environment.

References Aziz, H. M. A., Nagle, N. N., Morton, A. M., Hilliard, M. R., White, D. A., & Stewart, R. N. 2017. Exploring the impact of walk–bike infrastructure, safety perception, and built-environment on

References

53

active transportation mode choice: a random parameter model using New York City commuter data. Transportation. Boarnet, M. G. 2011. A Broader Context for Land Use and Travel Behavior, and a Research Agenda. Journal Of the American Planning Association, 77(3): 197–213. Cao, X. Y. 2014. Residential self-selection in the relationships between the built environment and travel behavior: Introduction to the special issue. Journal Of Transport And Land Use, 7(3): 1–3. Cao, X. 2015. Examining the Relationship Between Neighborhood Built Environment and Travel Behavior: A Review from the US Perspective, 30(4): 46–52. Cao, X., & Yang, W. 2017. Examining the effects of the built environment and residential selfselection on commuting trips and the related CO2 emissions: An empirical study in Guangzhou, China. Transportation Research Part D: Transport and Environment, 52: 480–494. Cao, X., Mokhtarian, P. L., & Handy, S. L. 2009a. Examining the Impacts of Residential SelfSelection on Travel Behaviour: A Focus on Empirical Findings. Transport Reviews, 29(3): 359– 395. Cao, X. Y., Mokhtarian, P. L., & Handy, S. L. 2009b. The relationship between the built environment and nonwork travel: A case study of Northern California. Transportation Research Part a-Policy And Practice, 43(5): 548–559. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. 2017. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100: 65–80. Dong, F., Tan., Zhou, D. 2009. Factor analysis of enterprise R & D capability based on multi index panel data, 21(3): 50–56. Ettema, D., & Nieuwenhuis, R. 2017. Residential self-selection and travel behaviour: What are the effects of attitudes, reasons for location choice and the built environment? Journal Of Transport Geography, 59: 146–155. Ewing, R., & Cervero, R. 2001. Travel and the built environment - A synthesis. Land Development And Public Involvement In Transportation (1780): 87–114. Ewing, R., & Cervero, R. 2010. Travel and the Built Environment. Journal Of the American Planning Association, 76(3): 265–294. Ewing, R., Tian, G., Goates, J. P., Zhang, M., Greenwald, M. J., Joyce, A., Kircher, J., & Greene, W. 2015. Varying influences of the built environment on household travel in 15 diverse regions of the United States. Urban Studies, 52(13): 2330–2348. Fan, L., & Luo, Y. 2009. Analysis on the influencing factors of farmers’ satisfaction with rural infrastructure, 09: 51–59. Feng, J. X., Dijst, M., Wissink, B., & Prillwitz, J. 2014. Understanding Mode Choice in the Chinese Context: The Case of Nanjing Metropolitan Area. Tijdschrift Voor Economische En Sociale Geografie, 105(3): 315–330. Gan, L., Shen, L., & Fu, H. 2009. Study of the infrastructure project evaluation factors based on sustainable development, 42(11): 133–138. Handy, S., Cao, X., & Mokhtarian, P. 2005. Correlation or causality between the built environment and travel behavior? Evidence from Northern California. Transportation Research Part D: Transport and Environment, 10(6): 427–444. Heesch, K. C., & Langdon, M. 2016. The usefulness of GPS bicycle tracking data for evaluating the impact of infrastructure change on cycling behaviour. Health Promot J Austr. Jain, D., & Tiwari, G. 2016. How the present would have looked like? Impact of non-motorized transport and public transport infrastructure on travel behavior, energy consumption and CO2 emissions – Delhi, Pune and Patna. Sustainable Cities and Society, 22: 1–10. Li, W. 2012. Research on Farmers’ satisfaction evaluation of rural public infrastructure, 18(4): 9–15. Li, W., &Xu. 2011. Research on satisfaction model and evaluation index construction of rural public infrastructure, 28(5): 74–98. Li, S., & Zhao, P. 2017. Exploring car ownership and car use in neighborhoods near metro stations in Beijing: Does the neighborhood built environment matter? Transportation Research Part D: Transport and Environment, 56: 1–17.

54

2 Comprehensive Evaluation of Changes …

Li, W., Jiang, L., & Hua, J. 2014. Performance analysis of provincial financial support to agriculture based on three stage DEA model, 1: 69–77. Lin, T., & Wang, D. G. 2015. Tradeoffs between in- and out-of-residential neighborhood locations for discretionary activities and time use: do social contexts matter? Journal Of Transport Geography, 47: 119–127. Liu, Z., Ma, J., & Chai, Y. 2016. Neighborhood-scale urban form, travel behavior, and CO2 emissions in Beijing: implications for low-carbon urban planning. Urban Geography, 38(3): 381–400. Luo, S. 2014. Performance evaluation of small-scale irrigation and water conservancy investment based on Farmers’ satisfaction. Ma, J., Liu, Z. L., & Chai, Y. W. 2015. The impact of urban form on CO2 emission from work and non-work trips: The case of Beijing, China. Habitat International, 47: 1–10. Niu, J., Yang, C., & Bai, Y. 2010. Analysis on the regional economic development of Hunan Province, 30 (3). Pan, H. X., Shen, Q., & Zhang, M. 2009. Influence of Urban Form on Travel Behaviour in Four Neighbourhoods of Shanghai. Urban Studies, 46(2): 275–294. Schoenau, M., & Müller, M. 2017. What affects our urban travel behavior? A GPS-based evaluation of internal and external determinants of sustainable mobility in Stuttgart (Germany). Transportation Research Part F: Traffic Psychology and Behaviour, 48: 61–73. Shen, L. Y., Wu, Y. Z., & Zhang, X. L. 2011. Key Assessment Indicators for the Sustainability of Infrastructure Projects. Journal Of Construction Engineering And Management-Asce, 137(6): 441–451. Sun, B., Ermagun, A., & Dan, B. 2017. Built environmental impacts on commuting mode choice and distance: Evidence from Shanghai. Transportation Research Part D: Transport and Environment, 52: 441–453. Tang, J., Zhu, Y., & Liu, C. 2010. Rural public service satisfaction and its influencing factors, 32(1): 110–128. Tiwari, G., Jain, D., & Ramachandra Rao, K. 2016. Impact of public transport and non-motorized transport infrastructure on travel mode shares, energy, emissions and safety: Case of Indian cities. Transportation Research Part D: Transport and Environment, 44: 277–291. Van Acker, V., Witlox, F., & Van Wee, B. 2007. The Effects of the Land Use System on Travel Behavior: A Structural Equation Modeling Approach. Transportation Planning and Technology, 30(4): 331–353. Wang, D. G., & Lin, T. 2014. Residential self-selection, built environment, and travel behavior in the Chinese context. Journal Of Transport And Land Use, 7(3): 5–14. Wang, D., & Zhou, M. 2017. The built environment and travel behavior in urban China: A literature review. Transportation Research Part D: Transport and Environment, 52: 574–585. Xiang, J., & Zhong, N. 2017. Re-estimation of urban rural migration and floating population in China—Based on the analysis of the whole village survey, Vol. 2: 3–18. Xu, S. 2011. Construction of contribution model of rural infrastructure investment. Journal of engineering management, 25(3): 288–291. Xu, S., & Wang. 2010. China’s 29 provinces and cities rural infrastructure investment efficiency measurement, 2: 112–114. Xu, J., Zhang, H., Zhou, C., Wang, W., Zhang, Y., & Jin, B. 2017. Investigation and analysis of rural environmental sanitation in Duyun minority mountainous areas in 2015, Vol. 28 No. 6: 483–486. Yang, Z. 2009. Analysis of rural primary education in 5 provinces in Southwest China, 12: 68–72. Yang, D., Liu, S., & Zhang, Z. 2011a. Comprehensive evaluation of rural economic development level in 31 regions of China, 39(26): 16277–16279. Yang, R., Liu, Y., & Liu, Y. 2011b. Regional Difference and Patterns of China’s Rural Development in the New Era, vol.30, No.10: 1247–1254. Yu, J. 2013. Chongqing strives to deepen the comprehensive reform of rural institutions, 6: 22–23. Zhang, Y., Wu, W., Li, Y., Liu, Q. X., & Li, C. Y. 2014. Does the Built Environment Make a Difference? An Investigation of Household Vehicle Use in Zhongshan Metropolitan Area, China. Sustainability, 6(8): 4910–4930.

References

55

Zhao, P. J. 2014. The Impact of the Built Environment on Bicycle Commuting: Evidence from Beijing. Urban Studies, 51(5): 1019–1037. Zhao, P. J., & Lu, B. 2011. Managing urban growth to reduce motorised travel in Beijing: one method of creating a low-carbon city. Journal Of Environmental Planning And Management, 54(7): 959–977. Zhao, P. J., Lu, B., & De Roo, G. 2010. Urban expansion and transportation: the impact of urban form on commuting patterns on the city fringe of Beijing. Environment And Planning A, 42(10): 2467–2486. Zhi, Y., Yang, Z. F., Yin, X. A., Hamilton, P. B., & Zhang, L. J. 2016. Evaluating and Forecasting the Drivers of Water Use in a City: Model Development and a Case from Beijing. Journal Of Water Resources Planning And Management, 142(1).

Chapter 3

Public Satisfaction-Based Performance Appraisal of Rural Infrastructure Construction

Abstract This study revised the American Customer Satisfaction Index (ACSI) model and selected a performance appraisal index for rural infrastructure based on this revised model. Then, the study adopted an interpretive structural model (ISM), analyzed the influence of each index factor, and developed a hierarchical directed graph. Finally, based on the mutual-influence relationships among the index factors in the hierarchical directed graph, a performance appraisal analytic network process (ANP) model was established. Based on discussions with rural college students and rural households in Sichuan, China, 246 questionnaires were obtained pertaining to rural infrastructure, and an empirical analysis was conducted. The results indicated that the performance of rural infrastructure construction is not very good. In particular, the full use of infrastructure and its role in improving the environment were found to be the worst. Meanwhile, the possibility of building information transparency and the longitudinal comparison of perceived performance appraisal results were the best. The performance of rural infrastructure construction was evaluated based on the perceptions of the direct users of rural infrastructure, and the relationship between the factors and the weight was measured reasonably. The proposed method was found to be workable and the analysis results reliable and effective. Keywords Performance appraisal · Rural infrastructure construction · Public satisfaction · American customer satisfaction index

3.1 Introduction China’s investment in rural fixed assets has grown significantly over the past 10 years. The rate of growth in 2014 was 2.72 times that of 2005. Growth in rural infrastructure is evident as well: 3368.3 billion yuan was invested in 2014, which was two times that of 2005 (2015). With increased investment in rural areas, rural construction has produced certain benefits; however, problems with rural infrastructure construction have been highlighted as well. Environmental pollution, for example, is a major problem drawing global attention. As with rapid urban development, will the pursuit

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_3

57

58

3 Public Satisfaction-Based Performance Appraisal …

of new rural construction come at the expense of the environment? Will rural infrastructure construction meet the needs of agricultural production as well as those of peasants after a certain amount of investment and development? Further investigations are needed to address these questions. As such, there is a need to evaluate the effects of rural infrastructure construction by testing past inputs, which can also provide a theoretical reference for further investment and development in rural areas. Recent studies on the effects of rural infrastructure have mainly focused on three aspects. The first is the assessment of infrastructure sustainability. Boz et al. developed a framework for evaluating sustainable infrastructure projects, proposed corresponding evaluation criteria, and established an evaluation index system (Boz and Eladaway 2015). Domingo assessed the effect of complex medical and health programs on construction-waste generation through interviews with people in the health-care infrastructure as well as questionnaire surveys (Domingo 2015). Analyzing 23 feasibility study reports, Shen et al. identified 30 key indexes for the sustainable development of infrastructure projects and selected 20 key assessment indexes using fuzzy set theory (Shen et al. 2013). Torres-Machi et al. analyzed environmental economic evaluation models as well as the practice of sustainable network pavement management (Torres-Machi et al. 2014). Reza et al. proposed an evaluation method for sustainable infrastructure development based on the energy value of the whole life cycle (Reza et al. 2014). Parrish et al. proposed life-cycle analysis (LCA), which assesses a project’s social, economic, and environmental aspects and requires that all three aspects be balanced to improve sustainability on the basis of life-cycle cost (LCC) (Parrish and Chester 2013). Investigating the infrastructure sustainability of wastewater treatment, Glick et al. evaluated a case using both LCC and LCA, and concluded that most sewage treatment and pollution costs arise from the sewage transport process. They recommended replacing central treatment facility (CTF) technology with community-scale technology (CST) (Glick and Guggemos 2013). Analyzing the environmental, economic, and social aspects of four commonly used underground public infrastructure construction methods, Ariaratnam et al. found that pilot tube micro tunneling (PTMT) had the highest sustainability (Ariaratnam et al. 2013). Bocchini et al. used the comprehensive application of resilience and sustainability to evaluate infrastructure construction (Bocchini et al. 2014). Using system dynamics, Zhou et al. studied the unique characteristics of infrastructure from a microengineering perspective and established a model to analyze sustainable construction and operation (Zhou and Liu 2015). Anagnostopoulos et al. combined geographic information systems (GIS) and spatial fuzzy analytic hierarchy process (SFAHP) to analyze and sort the location of a site (Konstantinos and Athanasios 2012). Second, a number of studies have employed social investigation to examine rural infrastructure construction, use of the situation, willingness to raise funds, sources of funds, and user satisfaction. Wang et al. divided rural infrastructure into two types: production and living. They conducted household surveys in 19 villages in Guangdong and analyzed farmers’ satisfaction with the current infrastructure supply and their willingness to finance based on their needs (Chunchao 2010). Through case

3.1 Introduction

59

studies of three infrastructure projects (rural schools, drinking water, and irrigation), Ma et al. analyzed the government’s role in rural infrastructure investment and the direction of self-financing investment among villagers (Lin-jing 2009). Yi et al. surveyed 101 villages in 5 provinces to analyze famers’ needs and investment behaviors in relation to roads, irrigation facilities, and drinking water facilities (Hong-mei et al. 2008). To analyze the status of private capital involved in rural infrastructure and farmers’ willingness to invest, Gan et al. investigated 31 villages and towns in the resource-rich areas of Shanxi, Shaanxi, and Inner Mongolia (Gan et al. 2011). Using data from 30 provinces, Hao et al. analyzed the effect of increased income on rural infrastructure (Hao et al. 2015). Using 670 questionnaires, Fan et al. established a structural equation model of the factors that influence satisfaction with rural infrastructure construction (Fan and Luo 2009). Using Henan Province to study farmers’ willingness to invest in rural infrastructure, Zhang et al. made recommendations for bottom-up and top-down public decision-making mechanisms (Zhang and Wan 2009). Third, some studies have focused on project performance appraisal. Project performance measurement has been widely applied as an important tool for improving the level of enterprise project management. Although appraising the performance of infrastructure construction can improve the efficiency of public management and decision-making, research in this area is limited. To identify the gap between theory and practice, Bassioni et al. reviewed performance appraisal frameworks and their application among construction companies in the United Kingdom (Bassioni et al. 2004). Ramani et al. used performance management to establish a sustainable improvement framework model for traffic management (Ramani et al. 2011). Conducting questionnaire surveys on the importance of performance indexes for participants in a project in Hong Kong, Lai and Lam found significant differences in the attitudes of participants (LAI and LAM 2010). Existing studies have mainly used the key performance index method to establish a project performance index system. Bassioni et al. argued that the key performance index method pays too much attention to project performance rather than enterprise performance (Bassioni et al. 2004). Discussing project performance measurement from the perspective of contractors, Zhang et al. noted that research on project performance measurement has mainly focused on the perspectives of third parties and owners (Zhang et al. 2012). Compared to general construction projects and urban infrastructure construction projects, rural infrastructure construction has clear and fixed end users. Thus, the main target of infrastructure construction should be meeting users’ needs. Accordingly, this study used a customer satisfaction index (CSI) model to evaluate the performance of rural infrastructure construction. Factors that affect the performance of rural infrastructure can be preliminarily determined on the basis of CSI. However, since the relationships between various factors are different from general merchandise, the relationships among various factors in the CSI model cannot be fully applied. In this study, the internal relations between factors were determined using the interpretive structural model (ISM) theory. Project performance index weight is usually determined using the analytic hierarchy process (AHP) method, which reduces problems to a top-down hierarchical structure and assumes that elements on the same

60

3 Public Satisfaction-Based Performance Appraisal …

level are independent (i.e., no mutual-influence relationship exists). Since factors are not completely independent but mutually influenced, this study employed an analytic network process (ANP) and used Super Decisions (v. 2.8.0) to determine the weight of the indexes in the performance appraisal of rural infrastructure. The purpose of using CSI is to obtain users’ attitudes toward infrastructure. Data were obtained by investigating college students from rural areas in Sichuan Province and then conducting a survey of selected villages. Following analysis and calculation, rural infrastructure construction was evaluated from farmers’ perspectives. On that basis, reasonable, scientific suggestions are proposed for the future planning and construction of rural infrastructure. The remainder of this paper is organized as follows. Section 3.2 presents the methods used in the study. Section 3.3 introduces the empirical study. Next, the empirical study results, discussion, and recommendations are presented in Sect. 3.4. Finally, Sect. 3.5 concludes the paper by providing the key findings of the study.

3.2 Research Methods 3.2.1 Customer Satisfaction Index (CSI) Model Dardozo introduced customer satisfaction into the marketing field in 1965 (Cardozo 1965). Howard and Sheth (1969) suggested that the evaluation of customer satisfaction is restricted to a certain time or occasion after the purchase (Howard and Sheth 1969). Day and Bodur (1977) defined customer satisfaction as a kind of process that is generated by experience and evaluation (Day and Bodur 1977). After that, Hunt proposed a value and satisfied-relationship model (Hunt 1991). In 1989, under the guidance of Professor Fornell at the University of Michigan, the Swedish Customer Satisfaction Barometer (SCSB)—the first national CSI model—was developed (Fornell and Larcker 1981). On this basis, CSI models were established in Sweden, the United States, and other countries. The most widely used model is the American Customer Satisfaction Index (ACSI), which was developed in 1994 and covers 10 economic sectors and 43 industries (https://www.theacsi.org/about-acsi/his tory). In 1999, the China Quality Association, Peking University, Tsinghua University, the People’s University, the Academy of Social Sciences, and other institutions jointly designed a CSI evaluation system. As China’s first standardized evaluation method, it laid the theoretical foundation for future studies of public satisfaction. Since then, scholars have thoroughly studied the customer satisfaction evaluation index system and the customer satisfaction model (Hong-yu et al. 2006; Hsu 2008). Revising the ACSI model, Li applied it to rural public infrastructure satisfaction and developed an evaluation index system for farmers’ satisfaction with rural public infrastructure (Wen-yia and Kun-pengb 2011; Wenyi 2012). In 2013, Ma Jieqiong used the improved ACSI to evaluate the performance of public projects and encourage

3.2 Research Methods

61

the public to participate in performance appraisals of local government projects (J.Q 2013). Based on the literature review, the CSI model includes seven indexes: expected quality, perceived product quality, perceived service quality, perceived value, customer satisfaction, customer complaints, and customer loyalty.

3.2.2 Interpretive Structural Modeling (ISM) Interpretive structural modeling (ISM) was developed by Warfield to analyze complex system structures (N. 1973). The core idea of ISM is to extract the constituent elements of problems and use some auxiliary means—such as matrices, programming, and direct graphs—to deal with the relationships between factors. This allows us to obtain a clear hierarchical structure and hierarchical structure graph, as follows: • Determine the research object and determine the relationships between the various factors of the study object through an ISM analysis team composed of experts in the relevant industry. Compose element set S so that element set S = {S1 , S2 , . . . , Sn }. • After determining the factors affecting the relationship, determine the direct relationship between the two elements according to the following:  Si RS j

= 1, Si has a direct dualistic r elationshi p with S j (i, j = 1, 2, . . . n). = 0, Si does not has a dualistic r elationshi p with S j

The adjacency matrix is established according to the above relation: A = (ai j )n × n. • The reachable matrix is calculated by the adjacency matrix, and the hierarchical structure of the index factor is drawn.

3.2.3 Analytic Network Process (ANP) Analytic network process (ANP) is a decision-making method proposed by Saaty, which is suitable for a nonindependent hierarchical structure. Although based on the analytic hierarchy process, this method is more flexible (Saaty 1987, 2004).

3.2.3.1

Unweighted Super Matrix W

Assuming the control layer of ANP has P1 , P2 , . . . , PN , and the network layer has an element level S1 , S2 , . . . , S N , where there are elements si1 , si2 , . . . , si N in Si , in which the control layer element Pi (i = 1, 2, . . . , m) is the criterion,

62

3 Public Satisfaction-Based Performance Appraisal …

  S jk k = 1, 2, . . . , n j is the substandard, and the element in the element set Si carries the indirect advantage of comparison to draw the matrix according to its influence on T S jk . The characteristic root method obtains the sort vector [w i1 ( jk) wi2 ( jk) . . . win ( jk) ] according to the consistency check and obtains matrix wi j : ⎡

wi1 ( j1) wi2 ( j2) . . . wi2 ( jn j ) ⎢ wi2 ( j1) wi2 ( j2) . . . wi2 ( jn j ) ⎢ Wi j = ⎢ .. .. .. .. ⎣ . . . . wini ( j1) wini ( j2) . . . wini ( jn j )

⎤ ⎥ ⎥ ⎥. ⎦

(3.1)

The column vector Wi j is the ranking vector of the important degree of Si1 , Si2 , . . . , Sini in Ui to Si1 , Si2 , . . . , Sin j in S j . If U j is not affected by Si , then Wi j = 0. Using the above steps, the super matrix W of Ps can be calculated as follows: ⎡ ⎢ ⎢ W =⎢ ⎣

3.2.3.2

W11 W12 . . . W1N W21 W22 . . . W2N .. .. .. .. . . . . WN 1 WN 2 . . . WN N

⎤ ⎥ ⎥ ⎥. ⎦

(3.2)

¯ Weighted Super Matrix W

Comparing the importance of any two elements of Pi , we get the weighted matrix. ⎡

a11 a12 . . . a1n ⎢ a21 a22 . . . a2n ⎢ A=⎢ . .. .. .. ⎣ .. . . . an1 an2 . . . ann

⎤ ⎥ ⎥ ⎥. ⎦

(3.3)

Then, we get the weighted super matrix: W¯ = (W¯ ij ) = ai j wi j , i = 1, 2, . . . , N ; j = 1, 2, . . . , N .

(3.4)

3.2.4 Performance Appraisal System Method This study appraised rural infrastructure construction performance based on public satisfaction using an evaluation index system based on ACSI according to research

3.2 Research Methods

63

American Customer Satisfaction Index (ACSI)

Interpretative Structural Model (ISM)

Network Analytic Process (ANP)

Determine evaluation index

Determine the index relationship

Synthetical evaluation model Performance appraisal Questionnaire design and survey

Fig. 3.1 Performance appraisal process

reviews of the multicriteria assessment of sustainable infrastructures (Sierra et al. 2017a, 2018a, b). Then, the relationship between the indexes was analyzed using structural modeling, and the hierarchical structure of the index factors was obtained. To establish the network hierarchy model and determine the weight of each evaluation index, household survey questionnaires were designed based on performance appraisal indexes, which were selected based on the literature review and expert investigation. Then, the performance of rural infrastructure construction was evaluated based on the household perception data. The performance appraisal process is shown in Fig. 3.1.

3.3 Empirical Study 3.3.1 Revising the ACSI to Determine the Evaluation Index In applying ACSI to the study of rural infrastructure satisfaction, scholars have only used the relevant indexes mentioned in ACSI. There has been no research on the correlation between the indexes. This study suggests that there are internal relations among factors based on the revised ACSI model proposed by Wenyi Li. Although these relationships are not recognized by farmers, they affect farmers’ perceptions of rural infrastructure. Therefore, this study initially assumed that the ACSI indexes for rural infrastructure were interrelated and mutually influenced.

3.3.2 Preliminary Evaluation Index Selection The index design for rural infrastructure construction performance was evaluated based on the revised ACSI model. Twenty-six performance appraisal indexes for

64

3 Public Satisfaction-Based Performance Appraisal …

rural infrastructure construction were identified by studying related documents. Six new factors affecting famers’ evaluations were obtained through discussions with college students from rural areas; thus, the number of performance appraisal indexes for rural infrastructure construction was 32. The initial selection of the index system is shown in Table 3.1. Explanations are given below for the six extra indexes proposed in this study (indexes extracted from the literature are not explained). Impression of full use: Rural infrastructure is the foundation of the development of rural economies. Further, its construction has for a long time been a matter of top-down, government-led supply. Whether rural infrastructure construction meets the real needs of farmers is rarely considered. During the field trip, many farmers suggested that some rural infrastructures are “vanity projects” built for the purpose of local government performance. These are not based on need and are not put into use. Impression of full use is an important part of rural infrastructure performance appraisal and can reflect the construction and use of rural infrastructure. Construction efficiency: Infrastructure construction has several characteristics (large-scale investment, long construction period, etc.). Further, the project-approval cycle is longer since it involves the use of funds. Some farmers mentioned these problems during the field trip. Construction efficiency directly affects the construction and use of infrastructure. Frequency of hearing complaints: Li et al. used complaints as an evaluation index. However, this study found that many farmers did not have a deep sense of selfperception regarding this problem, but they clearly perceived the frequency with which other people complained. Thus, this study used the frequency of hearing complaints as a supplementary index. Investment possibilities: Investment involves risks. Although villagers invest in the construction of some infrastructures for their own use, investment can fail for reasons such as government credit risk, risks in the use of funds, or construction risks. Villagers’ investment possibilities can reflect their confidence in the local government’s ability to deal with risks. The greater the willingness to invest, the greater the confidence in local government. Possibility of building information transparency: Infrastructure construction involves decision-making, planning, bidding, and many other procedures that pertain to the vital interests of farmers. These include the question of whether to solicit the views of farmers or whether to choose a reasonable contractor. All of these aspects directly affect the results of rural infrastructure construction. Some farmers know little or nothing about rural infrastructure construction. This study used these indexes in the performance evaluation of rural infrastructure construction. The smaller the possibility of the result, the smaller the farmers’ trust in local government. Possibility of no construction corruption: Construction corruption is a prominent topic in China. Where it exists, it directly affects the quality, efficiency, and quantity of rural infrastructure construction.

3.3 Empirical Study

65

Table 3.1 The screening results for the evaluation index Performance index

Weight

Source

Result

Code

Overall impression of quality

2581

Wenyi (2012)

✓

C11

Impression of full use

1896

The author

✓

C12

Overall expectation

2366

Wenyi (2012)

✓

C21

Expectation of improving quality of life

2124

Wenyi (2012)

✓

C22

Expectation of increasing production

478

Wenyi (2012)

×

Expectation of improving environment

1764

Wenyi (2012)

✓

C23

Expectation of increasing income

2198

Wenyi (2012)

✓

C24

Total quality perception

2488

Wenyi (2012)

✓

C31

Sense of improving quality of life

2068

Wenyi (2012)

✓

C32

Sense of increasing production

613

Wenyi (2012)

×

Sense of improving environment

1476

Wenyi (2012)

✓

C33

Sense of improving income

1924

Wenyi (2012)

✓

C34

Sense of using security

623

J.Q (2013)

×

Sense of reasonable planning and design

517

J.Q (2013)

×

Horizontal comparison

1971

Fan and Luo (2009)

✓

C35

Longitudinal comparison

2083

Fan and Luo (2009)

✓

C36

Construction efficiency

602

The author

×

Sense of quality under given costs

2137

Wenyi (2012)

✓

C41

Sense of cost under given quality

2049

Wenyi (2012)

✓

C42

Overall satisfaction

2477

Wenyi (2012)

✓

C51

Satisfaction relative to expectation

1956

Wenyi (2012)

✓

C52

Satisfaction relative to ideal condition

1882

Howard and Sheth (1969)

✓

C53

Complaints to others

1328

N. (1973)

✓

C61

Frequency of hearing complaints

216

The author

×

Complaints to relevant departments

2127

N. (1973)

✓

Possibility of no longer using rural infrastructure

395

J.Q (2013)

×

C62

(continued)

66

3 Public Satisfaction-Based Performance Appraisal …

Table 3.1 (continued) Performance index

Weight

Source

Result

Code

Possibility of improving the quality of projects

1279

N. (1973)

✓

C71

Possibility of participating in construction

572

N. (1973)

×

Possibility of investment

2234

The author

✓

C72

Possibility of participating in operation and maintenance

1849

N. (1973)

✓

C73

Possibility of building information transparency

1138

The author

✓

C74

Possibility of no corruption

983

The author

✓

C75

3.3.3 Evaluation Index Selection and Determination First, a questionnaire survey was conducted to screen the primary indexes and avoid selecting indexes that pertained to one-sided, repeated, or related issues. Thirtytwo preliminary evaluation indexes were selected from the relevant literature and discussions with rural students (see Table 3.1). Respondents were asked to select 30 indexes in order from the 32 indexes according to the size of the impact on infrastructure performance appraisal. To avoid misleading respondents with the original target sequence, the order of the 32 indexes was randomized for each questionnaire. The weighted cumulative principle was as follows: respondents selected 30 indexes from the 32 indexes, assigned 30 to the first one, 29 to the second, and so on. Finally, all weights were accumulated for each index selected by all 91 respondents according to the order they chose. For example, if all of the 91 respondents selected “impression of full use” as the first one, the weighted cumulative value was 91 * 30 = 2730. This study excluded weighted cumulative values of less than 900 (a significant fault existed in data below 900) and selected the remaining effective evaluation indexes for the evaluation index system. The screening results for the evaluation index are shown in Table 3.1. To ensure respondents were familiar with rural infrastructure construction, the questionnaire was conducted at a local feasibility report review meeting concerning 10 local infrastructure projects. The participants included leaders of the administrative departments for the construction, review experts, representatives of the owners, and representatives of the unit preparing the feasibility report. One hundred questionnaires were issued; 91 were valid (91% effective recovery rate), and the feedback was good. The final composition of the questionnaire is shown in Table 3.2.

3.3 Empirical Study

67

Table 3.2 Final composition of the questionnaire No

Investigation objects

Number

Ratio (%)

1

Leaders of the administrative departments for construction

13

14.29

2

Review experts

44

48.35

3

Representatives of the owners

16

17.58

4

Representatives of the unit preparing the feasibility report

18

19.78

5

Total

91

100%

3.3.4 ISM Index Relationship Determination As shown in Table 3.1, the final 24 selected indexes were the objects to be studied. Construction administrative departments, experts, owners’ representatives, and representatives of the feasibility report comprised the ISM analysis group. Assuming a relationship existed between the 24 indexes, thus constituting the set of elements S, the number of element set S = {S1 , S2 , …, S24 } = {C11 , C12 , …, C75 }. A two-dimensional questionnaire was established according to the ISM principle using the selected 24 indexes. The 91 respondents were asked to judge the relationship between any two indexes. Using statistics, it was confirmed that the two factors had an impact if more than half of the questionnaire showed that the two factors had relationships. The final determination of the ISM adjacency matrix could be obtained from the statistics. The reachable matrix of the adjacency matrix was solved using MATLAB (Saaty 2004) as shown in Table 3.3.

3.3.5 Establishing an ANP Model to Determine the Weight of Indexes In the network-level analysis model, the two-level subindex system was the control layer, and the three-level index system was the network layer. The index factors of the network layer were determined according to Table 3.3. The performance appraisal model for rural infrastructure construction (Fig. 3.2) was obtained by entering the relationships between the indexes into Super Decisions (v. 2.8.0). The index was established using Super Decisions. First, we needed to compare the indexes of two points. The score value was determined according to the ratio of weighted cumulative values in Table 3.2, reflecting the impact of 91 expert opinions regarding the construction of rural infrastructure. When the ratio was in the 1.0–1.5 range, the score value was 2; in the 1.5–2.0 range, it was 3; and in the 2.0–2.5 range, it was 4. Each index factor and its related factors needed to be compared according to the weighted cumulative values in Table 3.2. After each scoring comparison, a consistency test was conducted. After examination, all of the related factors were compared, and the results were satisfactory at CR < 0.1, which met the requirements

C11

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

C

C11

C12

C21

C22

C23

C24

C31

C32

C33

C34

C35

C36

C41

C42

C51

C52

C53

C61

C62

C71

C72

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

C12

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

C21

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

C22

Table 3.3 Reachable matrix A

C23

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

1

C24

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

C31

0

0

0

0

0

0

0

0

0

0

0

0

1

0

0

0

1

0

1

1

1

C32

0

0

0

0

0

0

0

0

0

0

0

0

1

0

1

0

1

1

1

1

1

C33

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

0

1

1

1

C34

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

0

0

0

0

1

C35

0

0

0

0

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

C36

0

0

0

0

0

0

0

0

0

0

1

0

1

1

1

1

1

1

1

1

1

C41

0

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

C42

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C51

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

C52

0

0

0

0

0

0

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

C53

0

0

0

0

0

0

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

C61

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C62

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C71

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C72

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C73

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

C74

C75

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

(continued)

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

68 3 Public Satisfaction-Based Performance Appraisal …

C11

0

0

0

C

C73

C74

C75

0

0

0

C12

0

0

0

C21

Table 3.3 (continued)

C22

0

0

0

C23

0

0

0

C24

0

0

0

C31

0

0

0

C32

0

0

0

C33

0

0

0

C34

0

0

0

C35

0

0

0

C36

0

0

0

C41

0

0

0

C42

0

0

0

C51

0

0

0

C52

0

0

0

C53

0

0

0

C61

0

0

0

C62

0

0

0

C71

0

0

0

C72

0

0

0

C73

0

0

0

C74

1

0

1

C75

0

0

1

3.3 Empirical Study 69

70

3 Public Satisfaction-Based Performance Appraisal …

Image of rural infrastructure

Trust of farmers

Overall impression of quality

Possibility of improving the quality of projects

Impression of full use

Possibility of building information transparency Possibility of participating in operation and maintaince Possibility of no corruption Possibility of investment

Farmers’ expctation Expectation of increasing income Expectation of improving quality of life Expectation of improving environment Overall expectation

Farmer’s complaint Complain to others Complain to relevant departments

Farmer’s satisfaction

Quality perception

Satisfaction relative to expectation

Total quality perception

Satisfaction relative to ideal condition

Sense of improving environment Sense of improving income Longitudinal comparison

Value perception

Sense of improving quality of life

Sense of quality under given costs

Horizontal comparison

Sense of costs under given quality

Fig. 3.2 ANP model diagram for rural infrastructure construction performance appraisal

of the consistency check. Using Super Decisions, we obtained the rural infrastructure performance appraisal for the specific weights of the 24 factors (Table 3.4).

3.3.6 Performance Appraisal of Rural Infrastructure We used the performance appraisal index system for rural infrastructure to design a questionnaire for farmers. To obtain effective, real data and ensure that respondents understood all of the problems in the questionnaire, the research group organized rural students from engineering management as respondents and visitors to gather the data. A meeting was organized to help students fully understand the questionnaire and to collect data about student attitudes toward rural infrastructure. Then, these students conducted investigations in their hometowns to gather firsthand data. Out of 300 questionnaires issued, 246 were recovered, for a recovery rate of 82%, which is quite high. Among them, 144 respondents were male and 102 were female. The questionnaires covered 23 cities in Sichuan Province and had a wide coverage area. The reliability of the questionnaire was tested using the Cronbach alpha coefficient (α = 0.8103) with the use of SPSS 21.0, indicating that the questionnaire has high reliability. The questionnaire distribution and recovery numbers are shown in Table 3.5.

3.3 Empirical Study

71

Table 3.4 Performance appraisal index system factor weights Performance appraisal index

Normalized by cluster Limiting

Image of rural infrastructure Overall impression of quality (C11 )

0.60000

0.068232

Impression of full use (C12 )

0.40000

0.045488

Overall expectation (C21 )

0.46154

0.045488

Expectation of improving quality of life (C22 )

0.07692

0.007581

Expectation of improving environment (C23 )

0.30769

0.030325

Expectation of increasing income (C24 )

0.15385

0.015163

Total quality perception (C31 )

0.19187

0.041647

Sense of improving quality of life (C32 )

0.21437

0.046531

Sense of improving environment (C33 )

0.03493

0.007581

Sense of improving income (C34 )

0.06986

0.015163

Horizontal comparison (C35 )

0.24449

0.053069

Longitudinal comparison (C36 )

0.24449

0.053069

Sense of quality under given costs (C41 )

0.72125

0.068232

Sense of cost under given quality (C42 )

0.27875

0.026371

Overall satisfaction (C51 )

0.49999

0.041861

Satisfaction relative to the expectation (C52 )

0.33334

0.027908

Satisfaction relative to ideal condition (C53 )

0.16667

0.013954

Complaints to others (C61 )

0.50000

0.068232

Complaints to relevant departments (C62 )

0.50000

0.068232

Possibility of improving the quality of projects (C71 )

0.06667

0.017058

Possibility of investment (C72 )

0.20000

0.051174

Possibility to participate in operation and maintenance (C73 ) 0.20000

0.051174

Possibility of building information transparency (C74 )

0.26667

0.068232

Possibility of no corruption (C75 )

0.26667

0.068232

Farmers’ expectations

Quality perception

Value perception

Farmers’ satisfaction

Farmers’ complaints

Trust of farmers

Twenty-four questions were set for the 24 indexes for the performance appraisal of rural infrastructure construction, each with five options according to the satisfaction situation, degree of implementation, and possibility of improvement. After preliminary statistics, topics were selected as shown in Table 3.6. According to the questionnaire recycling statistics, measurement was carried out using a Likert-type scale. In Table 3.7, not satisfied (very small) (much) = 1, less

Number

11

33

14

9

14

Region

Bazhong

Chengdu

Dazhou

Deyang

Guangan

Meishan

Luzhou

Liangshan State

Leshan

Guangyuan

Region

7

6

21

7

6

Number

Table 3.5 Questionnaire distribution and recovery

Mianzhu

Panzhihua

Neijiang

Nanchong

Mianyang

Region

5

9

7

10

17

Number

Yibin

Ganzi State

Chi Yi Autonomous County

Aba State

Region

15

7

5

7

Number

Zigong

Ziyang

Yaan

Suining

Region

8

9

12

7

Number

72 3 Public Satisfaction-Based Performance Appraisal …

3.3 Empirical Study

73

Table 3.6 Performance appraisal index score table Performance appraisal index

Questionnaire score Model weight Performance score

Image of rural infrastructure Overall impression of quality (C11 )

4.6446

0.0682

0.3169

Impression of full use (C12 )

3.3817

0.0455

0.1538

Overall expectation (C21 )

4.2397

0.0455

0.1929

Expectation of improving quality of life (C22 )

4.9835

0.0076

0.0378

Expectation of improving environment (C23 )

4.9091

0.0303

0.1489

Expectation of increasing income (C24 )

4.6281

0.0152

0.0702

Total quality perception (C31 )

4.6364

0.0416

0.1931

Sense of improving quality of life (C32 )

4.9753

0.0465

0.2315

Sense of increasing production (C33 ) 3.9421

0.0076

0.0299

Sense of improving income (C34 )

4.4815

0.0152

0.0680

Horizontal comparison (C35 )

5.0579

0.0531

0.2684

Longitudinal comparison (C36 )

6.1901

0.0531

0.3285

Sense of quality under given costs (C41 )

4.7190

0.0682

0.3220

Sense of cost under given quality (C42 )

4.4711

0.0264

0.1179

Overall satisfaction (C51 )

4.5885

0.0419

0.1921

Satisfaction relative to the expectation (C52 )

4.4357

0.0279

0.1238

Satisfaction relative to ideal condition (C53 )

4.1276

0.0140

0.0576

Complaints to others (C61 )

4.5270

0.0682

0.3089

Complaints to relevant departments (C62 )

4.4380

0.0682

0.3028

Possibility of improving the quality of projects (C71 )

5.3568

0.0171

0.0914

Possibility of investment (C72 )

6.0413

0.0512

0.3092

Farmers’ expectations

Quality perception

Value perception

Farmers’ satisfaction

Farmers’ complaints

Trust of farmers

(continued)

74

3 Public Satisfaction-Based Performance Appraisal …

Table 3.6 (continued) Performance appraisal index

Questionnaire score Model weight Performance score

Possibility to participate in operation 5.3058 and maintenance (C73 )

0.0512

0.2715

Possibility of building information transparency (C74 )

4.8257

0.0682

0.3293

Possibility of no corruption (C75 )

4.3112

0.0682

0.2942

Table 3.7 Questionnaire statistics No Question

Option Not satisfied (very small) (much)

Less General Satisfactory satisfied (larger) (smaller) (smaller) (larger)

Very Empty Total satisfactory (great) (very small)

1

What do you 15 think of the overall quality of existing infrastructures?

59

125

40

3

4

246

2

Are you satisfied with the village infrastructure’s role?

67

82

75

13

4

5

246

3

What do you 29 expect to be the degree of village infrastructure?

61

128

21

3

4

246

4

To what extent do you expect to improve quality of life by improving infrastructure?

9

53

117

57

6

4

246

5

How much do you hope to improve the village’s appearance through the improvement of infrastructure in the rural environment?

11

62

104

57

8

4

246

(continued)

3.3 Empirical Study

75

Table 3.7 (continued) No Question

Option Not satisfied (very small) (much)

Less General Satisfactory satisfied (larger) (smaller) (smaller) (larger)

Very Empty Total satisfactory (great) (very small)

6

To what extent do you expect to increase household income by improving infrastructure?

23

57

112

42

8

4

246

7

What do you think of the overall quality of rural infrastructure construction?

18

64

110

44

6

4

246

8

How much do you think the construction of rural infrastructure has improved quality of life among villagers?

14

45

118

62

4

3

246

9

What do you think of the effects of infrastructure in terms of improving the village and rural environments?

36

92

83

26

5

4

246

10

How much does infrastructure construction increase farmers’ incomes?

18

70

114

39

2

3

246

(continued)

76

3 Public Satisfaction-Based Performance Appraisal …

Table 3.7 (continued) No Question

Option Not satisfied (very small) (much)

Less General Satisfactory satisfied (larger) (smaller) (smaller) (larger)

Very Empty Total satisfactory (great) (very small)

11

Are you satisfied with infrastructure construction in your village compared to other villages?

23

39

109

50

21

4

246

12

Are you satisfied with the current infrastructure compared to five years ago?

9

19

71

105

38

4

246

13

Are you satisfied with the quality of rural infrastructure under existing infrastructure costs?

16

64

108

46

8

4

246

14

Are you satisfied with the use of infrastructure under the quality of existing infrastructure?

28

64

103

38

9

4

246

15

Are you satisfied with the overall situation of the village’s infrastructure?

18

71

101

49

4

3

246

16

Are you satisfied with the village’s infrastructure construction compared with your expectations?

24

67

107

39

4

5

246

(continued)

3.3 Empirical Study

77

Table 3.7 (continued) No Question

Option Not satisfied (very small) (much)

Less General Satisfactory satisfied (larger) (smaller) (smaller) (larger)

Very Empty Total satisfactory (great) (very small)

17

Are you satisfied with the village’s infrastructure construction compared with your ideal condition?

43

59

104

35

2

3

246

18

How much do you complain about village infrastructure?

23

57

120

36

5

5

246

19

How likely are you to complain to related authorities?

33

49

119

35

6

4

246

20

How likely do you think it is that the overall infrastructure condition will be improved?

14

31

107

76

13

5

246

21

Are you willing 7 to invest in rural infrastructure construction?

23

82

97

33

4

246

22

Are you willing 13 to help operate and maintain rural infrastructure?

25

134

52

18

4

246

23

What do you 24 think about the possibility of village infrastructure construction information becoming more transparent in the future?

48

100

63

6

5

246

(continued)

78

3 Public Satisfaction-Based Performance Appraisal …

Table 3.7 (continued) No Question

24

Option

What do you think about the possibility of no corruption in the infrastructure construction in the future?

Not satisfied (very small) (much)

Less General Satisfactory satisfied (larger) (smaller) (smaller) (larger)

Very Empty Total satisfactory (great) (very small)

41

54

6

99

41

5

246

satisfied (smaller) (larger) = 3, general = 5, satisfactory (larger) (smaller) = 7, and very satisfactory (great) (very small) = 9. Each index corresponds to the final questionnaire to obtain its score (xi j is the number of respondents who selected the score j for each index): yi =

xi1 × 1 + xi2 × 3 + xi3 × 5 + xi4 × 7 + xi5 × 9 . 5 j=1 x i j

The questionnaire design question was made to correspond to the rural infrastructure performance appraisal index for the weight of ωi ; then, the index considered the weight score of µi = ωi × yi . The final scores for the indexes are shown in Table 3.7. For the Sichuan area, the performance of rural infrastructure construction, evaluated from the perspective of farmers, is: U=

24

µi = 4.7606.

i=1

3.4 Discussion and Recommendations This study adopted the American Customer Satisfaction Index (ACSI) as the basis for appraising the performance of rural infrastructure. This research method has excellent applicability. The direct users of rural infrastructure are relatively fixed, and they have an intuitive and objective understanding of rural infrastructure; thus, the results of their appraisals are scientific and reasonable. Construction and investment

3.4 Discussion and Recommendations

79

in rural infrastructure aim to promote rural economic development, production, and livelihoods. As such, the construction of rural infrastructure should meet the needs of farmers for both production and daily life. In recent decades, China’s investment in rural infrastructure has increased, and new rural construction has shown initial successes. However, does such construction really meet the most urgent needs in rural areas? Does it really make farmers satisfied or meet the requirements for rural production? Considering these questions, this study evaluated the effect of such construction from the perspective of farmers. The performance appraisal index for rural construction was based on not only previous studies but also information added through new research. The preliminary determination of the index system is more comprehensive and practical for reflecting farmers’ requirement information for rural infrastructure. To avoid high correlations between indexes, repeated meanings, and poor accuracy measurement, this study administered the index-selection questionnaire to senior professionals. The respondents had high professional levels, including the leaders of construction administration, representatives of rural infrastructure projects, and experts on project evaluation. Thus, the survey results are reliable. Cumulative weight value can comprehensively reflect how important each index is for performance appraisal. This study used cumulative weight value to select the evaluating index. A pairwise comparison of associated factors was performed using an analytic network process (ANP) based on the AHP method used by Jin et al. to select rural building sites (Jeong et al. 2013). In addition, this study used index scoring related to cumulative weight value to avoid one-sidedness and single-expert subjectivity. Thus, the index selection and scoring are reliable. Previous studies have used a customer satisfaction index (CSI) for the performance appraisal of construction projects. However, they only considered seven CSI factors, or just revised the factors, and they did not study the logical relationships between them. This study assumed that farmers’ perception factors are related to each other; thus, it was necessary to study the relationships among them. The relationships among various indexes were studied using an interpretive structural model (ISM). The adjacency matrix of the evaluation index was obtained from the summary statistics by collecting questionnaires on related factors from the abovementioned professionals. MATLAB was used to calculate the reachable matrix; the relationships among the various factors were measured using Super Decisions. Based on participation by professionals, the model is objective, scientific, and reasonable. Based on model operation and household surveys, the performance appraisal score for rural infrastructure construction in Sichuan Province was found to be 4.7606. From the perspective of farmers, the performance of rural infrastructure construction is not very good and is inconsistent with China’s increasing infrastructure investments. This result is consistent with Arjuna and Manoj, who found that a lack of infrastructure affects rural residents’ satisfaction because of physical activity in the rural areas of Sri Lanka (Medagama and Galgomuwa 2018). However, the longitudinal comparison score for farmer satisfaction was 6.1909, which was the highest among all factors investigated. This case study result represents a typical contribution to the method for estimating the social sustainability of infrastructure

80

3 Public Satisfaction-Based Performance Appraisal …

(Sierra et al. 2017b). This indicates that the overall rural infrastructure is better than ever, which is consistent with China’s infrastructure investment. In addition, farmers’ willing to invest in rural infrastructure construction was strong with a score of 6.0413, indicating their eagerness to improve rural infrastructure. This is because infrastructure development has a positive influence on agricultural land and regional sustainable development (Bacior and Prus 2018). China’s rural construction over the past 10 years has caused farmers to gain confidence in the construction of rural infrastructure. However, many farmers do not believe existing rural infrastructures are put to full use. The score for impression of full use was 3.3817, which was the lowest among all factors. This suggests that some existing infrastructures are unnecessary, and some are negatively viewed as “vanity projects.” While investment in rural infrastructure has increased, it is not effectively used where it is most needed. In the future, we should pay more attention to the real needs of rural areas and farmers in the construction of rural infrastructure. Regarding environmental perceptions, farmers generally believed that construction did not improve the rural environment (score: 3.9421). It is clear that damage will occur if China does not pay sufficient attention to the environment during the process of rapid infrastructure construction. The weight of the performance appraisal of rural infrastructure construction reflects the importance of various indexes. This study found that the index weights for overall impression of quality, sense of quality under given costs, complaints to others, complaints to relevant departments, possibility of building information transparency, and possibility of no corruption were the largest at 0.0682. Weight depended on the experience of professionals, and their degree of specialization was high. From their perspective, farmers’ complaints greatly influence the performance appraisal of rural infrastructure construction. Different types of complaints may come from different lifestyles, and they are important in the performance appraisal of rural infrastructure construction (Sierra et al. 2015). The possibility of building information transparency and of no corruption also greatly influence performance appraisal. We should, therefore, focus on information transparency and on preventing corruption in future rural infrastructure construction. This also highlights the need for better supervision. Experts emphasized the quality of rural infrastructure construction, so the overall impression of quality and the sense of quality under given costs had the highest weight. This performance appraisal of rural infrastructure construction fully considered farmers’ perceptions and professional advice. At 0.3293, the possibility of building information transparency had one of the highest scores; it was very large for both farmers and experts. This is in line with the conclusion that information transparency has an important influence on the performance appraisal of infrastructure (Munda and Russi 2008). We can conclude that construction administrative departments should increase the transparency of information regarding rural infrastructure construction in the future. Governments should open all information channels to help villagers understand rural infrastructure construction, and for balance, governments should accept feedback from villagers. The longitudinal comparison score of 0.3285 reflects the fact that rural infrastructure is under constant development. Studying the index of the longitudinal comparison of rural infrastructure from the perspective of experts is

3.4 Discussion and Recommendations

81

very important for performance appraisal and emphasizes the importance of development. The experiences of experts are also very important for urban infrastructure construction (Munda 2006). The performance appraisal indexes were selected from the relevant literature as well as discussions with rural students and experts in Sichuan Province. In addition, the relationships between the indexes were judged by 91 experts in Sichuan Province. As such, the conclusions have strong geographical applicability, and different areas may yield different findings. The performance appraisal of rural infrastructure should also consider aspects such as technology, planning, location, and project management. Since this study evaluated rural infrastructure based only on farmer satisfaction, it has strong subjectivity. In future work, other aspects should be added for a more comprehensive appraisal.

3.5 Conclusion Based on previous studies, this research established a comprehensive ACSI-ISMANP framework for rural infrastructure performance evaluation (Sierra et al. 2017a, b, 2018b; Munda 2006). A systematic integration of the experiences of infrastructure construction experts and the judgments of infrastructure users was used to evaluate rural infrastructure performance. Overcoming the problem of one-sided performance evaluation in the past, this approach not only satisfies the scientific aspect of performance evaluation but also considers the needs of infrastructure end users. The performance appraisal score for rural infrastructure in Sichuan Province was determined to be 4.7606. Based on farmer satisfaction, this score is relatively low. According to the data analysis, farmers are most satisfied with the longitudinal comparative perception of rural infrastructure construction. The possibility of participation in rural infrastructure investment is also relatively high. However, farmers have a very bad impression of the function and application of rural infrastructure. They are also dissatisfied with the effect on the environment. From the index of evaluation weights, this study concludes that the performance of rural infrastructure construction affects farmers’ complaints, the possibility of transparency, and the possibility of no corruption a great deal. Moreover, transparency in building information has the highest value, followed by longitudinal comparison.

References Ariaratnam, S. T., Piratla, K., Cohen, A., & Olson, M. 2013. Quantification of sustainability index for underground utility infrastructure projects. Journal of Construction Engineering and Management, 139 (12): A4013002. Bacior, S., & Prus, B. 2018. Infrastructure development and its influence on agricultural land and regional sustainable development. Ecological Informatics, 44: 82–93.

82

3 Public Satisfaction-Based Performance Appraisal …

Bassioni, H. A., Price, A. D. F., & Hassan, T. M. 2004. Performance measurement in construction. Journal Of Management In Engineering, 20: 42–50. Bocchini, P., Frangopol, D. M., Ummenhofer, T., & Zinke, T. 2014. Resilience and sustainability of civil infrastructure: Toward a unified approach. Journal Of Infrastructure Systems, 20. Boz, M. A., & El-adaway, I. H. 2015. Creating a Holistic Systems Framework for Sustainability Assessment of Civil Infrastructure Projects. Journal Of Construction Engineering And Management: 141. Cardozo, R. N. 1965. An experimental study of customer effort, expectation, and satisfaction. Journal of marketing research: 244–249. Chunchao, W. 2010. Study on the will of peasants participating in the financing of the construction for rural infrastructure—household survey of 19 villages in Guangdong as an example. Shanghai Journal of Economics, 3: 005. Day, R. L., & Bodur, M. 1977. A comprehensive study of satisfaction with consumer services. Consumer satisfaction, dissatisfaction and complaining behavior. 64–70. Domingo, N. 2015. Assessment of the impact of complex healthcare features on construction waste generation. Buildings, 5: 860–879. Fan, L. M., & Luo, Y. M. 2009. Analysis on the influence factors of farmers’ satisfaction to rural infrastructure: Structural equation modeling based on 670 questionnaires. Issues in Agricultural Economy, 9: 51–59. Fornell, C., & Larcker, D. F. 1981. Evaluating structural equation models with unobservable variables and measurement error. Journal Of Marketing Research, 18: 39–50. Gan, J., Zhu, & Y.C. 2011. Study on the participation of private capital in rural infrastructure: Based on an investigation of 31 villages and towns in the resource-rich regions of Shanxi, Shaanxi, and inner Mongolia. Inquiry into Economic Issues, 11: 153–158. Glick, S., & Guggemos, A. A. 2013. Rethinking wastewater-treatment infrastructure: Case study using life-cycle cost and life-cycle assessment to highlight sustainability considerations. Journal of Construction Engineering and Management, 139: A5013002. Hao, E. H., Hu, K., & Chen, X. P. 2015. Income increase effect of rural infrastructures: Based on the data analysis of 30 provinces in China. Rural Economy, 4: 64–68. Hong-mei, Y., Lin-xiu, Z., Hare, D., & Cheng-fang, L. 2008. The provision of infrastructure and farmers demand—evidence from 5 province. China Soft Science, 11: 106–115. Hong-yu, L., Jian, L., & Yun-xian, G. 2006. Design of customer satisfaction measurement index system of ems service. The Journal of China Universities of Posts and Telecommunications, 13: 109–113. Howard, J. A., & Sheth, J. N. 1969. The theory of buyer behavior. Hsu, S.-H. 2008. Developing an index for online customer satisfaction: Adaptation of American customer satisfaction index. Expert Systems with Applications, 34: 3033–3042. Hunt, H. K. 1991. Consumer Satisfaction, Dissatisfaction, And Complaining Behavior. Journal Of Social Issues, 47: 107–117. J.Q, M. 2013. A research on local government project performance evaluation base on public view. Wu Han, China. Wuhan Textile University: 15–17. Jeong, J. S., García-Moruno, L., & Hernández-Blanco, J. 2013. A site planning approach for rural buildings into a landscape using a spatial multi-criteria decision analysis methodology. Land Use Policy, 32: 108–118. Konstantinos, A., & Athanasios, V. 2012. Site suitability analysis for natural systems for wastewater treatment with spatial fuzzy analytic hierarchy process. Journal Of Water Resources Planning And Management-Asce, 138: 125–134. LAI, I. K. W., & LAM, F. K. S. 2010. Perception of various performance criteria by stakeholders in the construction sector in Hong Kong. Construction Management and Economics, 28: 377–391. Lin-jing, M. 2009. Analysis on government’s role and farmers’ will in investment in rural infrastructure: Take drinking water, schools, and irrigation as examples. Finance and Trade Research, 2: 38–42.

References

83

Medagama, A., & Galgomuwa, M. 2018. Lack of infrastructure, social and cultural factors limit physical activity among patients with type 2 diabetes in rural Sri Lanka, a qualitative study. Plos One, 13: e0192679. Munda, G. 2006. Social multi-criteria evaluation for urban sustainability policies. Land Use Policy, 23: 86–94. Munda, G., & Russi, D. 2008. Social multicriteria evaluation of conflict over rural electrification and solar energy in Spain. Environment and Planning C: Government and Policy, 26: 712–727. N., W. J. 1973. Binary matrices in system modeling. IEEE Transactions on Systems Man And Cybernetics SMC, 3: 441–449. Parrish, K., & Chester, M. 2013. Life-cycle assessment for construction of sustainable infrastructure. Practice periodical on structural design and construction, 19: 89–94. Ramani, T. L., Zietsman, J., Knowles, W. E., & Quadrifoglio, L. 2011. Sustainability enhancement tool for state departments of transportation using performance measurement. Journal Of Transportation Engineering-ASCE, 137: 404–415. Reza, B., Sadiq, R., & Hewage, K. 2014. Emergy-based life cycle assessment (Em-LCA) for sustainability appraisal of infrastructure systems: A case study on paved roads. Clean Technologies And Environmental Policy, 16: 251–266. Saaty, R. W. 1987. The analytic hierarchy process—what it is and how it is used. Mathematical modelling, 9: 161–176. Saaty, T.L. 2004. Fundamentals of the analytic network process—dependence and feedback in decision-making with a single network. Journal of Systems science and Systems engineering 13: 129–157. Shen, L., Wu, Y., & Zhang, X. 2013. Key assessment indicators for the sustainability of infrastructure projects. Journal Of Construction Engineering And Management, 137: 411. Sierra, L. A., Victor, Y., & Eugenio, P. 2015. Social sustainability in the lifecycle of Chilean public infrastructure. Journal of Construction engineering and management, 142: 05015020. Sierra, L. A, Yepes, V., & Pellicer, E. 2017a. Assessing the social sustainability contribution of an infrastructure project under conditions of uncertainty. Environmental Impact Assessment Review, 67: 61-72. Sierra, L. A., Yepes, V., & Pellicer, E. 2017b. Method for estimating the social sustainability of infrastructure projects. Environmental Impact Assessment Review, 65: 41–53. Sierra, L. A., Yepes, V., Garcia-Segura, T., & Pellicer, E. 2018a. Bayesian network method for decision-making about the social sustainability of infrastructure projects. Journal of Cleaner Production, 176: 521–534. Sierra, L. A., Yepes, V., & Pellicer, E. 2018b. A review of multi-criteria assessment of the social sustainability of infrastructures. Journal of Cleaner Production, 187: 496–513. Torres-Machi, C., Yepes, V., Chamorro, A., & Pellicer, E. 2014. Current models and practices of economic and environmental evaluation for sustainable network-level pavement management. Journal of Construction, 13 (2): 49–56. Wen-yia, L., & Kun-pengb, X. 2011. Farmers’ satisfaction model and evaluation indexes in rural public infrastructure. Indexes in Rural Public Infrastructure, Journal of Hubei University of Education: 022. Wenyi, L. 2012. Study on evaluation of farmers satisfaction of the rural public infrastructure. Journal of Chongqing University (Social Science Edition), 4 (001). Zhang, K. H., & Wan, M. 2009. Study on the will of peasants participating in the financing of the construction for rural infrastructure. Journal of Zhongnan University of Economics and Law, 176 (no.5): 36–41. Zhang, S. B., Hou, X. W., & Yang, Q. B. 2012. Performance measurement of engineering contractor project based on BSC and ANP. Statistics and Decision. 2012, 1: 175–178. Zhou, J., & Liu, Y. 2015. The method and index of sustainability assessment of infrastructure projects based on system dynamics in China. Journal of Industrial Engineering and management, 8 (3): 1002–1019.

Chapter 4

Indicators Impacting Farmers’ Satisfaction in the Use of Rural Facilities

Abstract The disparity between construction targets and the real needs of farmers in the construction of rural facilities is a problem that has led to a failure in meeting farmers’ demands. This chapter investigates farmers’ satisfaction and the influencing factors of rural facilities through factor analysis and logit regression model. This research led to three key findings: (1) overall satisfaction of farmers of rural facilities is below average level; (2) farmers’ satisfaction is affected mainly by the horizontal comparison, road facilities, electricity and signal facilities, reconstruction of public toilets, irrigation facilities, cultural and recreational facilities, renovation of fuel and kitchen, healthy facilities, village planning and renovation, and income factor; (3) farmers’ needs are shifting from production to life type. This paper is the foundation of further analysis of the effects of significant factors on farmers’ satisfaction, providing a theoretical basis for the construction of a “bottom-up” and “top-down” decision-making mechanism. Keywords Rural facilities · Famer’ satisfaction · Logit regression model · Rural China

4.1 Introduction Rural facilities construction is the foundation to ensure the comprehensive and rapid development of a rural economy, and an important content of new rural construction. This construction also serves as the basis of a harmonious rural environment and overall affluence. Addressing problems in the construction of rural facilities is very crucial to the achievement of sustainable and coordinated urban and rural development. In recent years, some issues such as the incomplete supervision mechanism, limited financing channels, and lack of maintenance, have become increasingly prominent with the rapid development of rural facilities. The real needs of farmers have been ignored all the time in the “top-down” planning pattern, resulting in a disconnection between construction targets and farmers’ real demands.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_4

85

86

4 Indicators Impacting Farmers’ Satisfaction …

With the aim of addressing the aforementioned disconnection, the researchers conducted a study on farmers’ satisfaction of rural facilities and its significant indicators. On the basis of existing studies, the researchers selected evaluation factors of satisfaction assessment to create a comprehensive questionnaire that consists of whole satisfaction, satisfaction of each type of facilities, price attitude, horizontal comparison, vertical comparison, and basic information of interviewees. The reliability of the questionnaire was analyzed to test the validity of the survey. Factor analysis was then conducted to identify 15 factors and weights of indicators to calculate the satisfaction score. Finally, the logit regression model was used to analyze the 15 factors to test their effects on farmers’ satisfaction.

4.2 Literature Review At present, research related to rural facilities focus mainly on sustainable development, existing problems and policy recommendations, and performance evaluation. Previous research on the sustainable development of infrastructure can be divided into research of evaluation index and the establishment of evaluation model. Many foreign studies related are about the sustainability of the whole life cycle of projects; while domestic Chinese research usually focus on a certain stage to study the sustainable development of projects. With the deepening of research, the evaluation index system of infrastructure sustainability is constantly improved. Gan et al. (2009) pointed out the lack of sustainable evaluation index in project decision-making stage in China and developed a questionnaire-based evaluation indicator system of infrastructure projects on the basis of sustainable development. Shen et al. (2011) referenced 23 feasibility study reports and summarized 30 indicators to assess the sustainability of infrastructure projects. They selected 20 key evaluation indicators using the fuzzy set theory, and combined the expert scoring method to assess the sustainability of infrastructure projects. Subsequently, Bocchini et al. (2014) combined resilience and sustainability in the assessment of infrastructure construction. Then, on the basis of previous studies, Boz et al. (2015) proposed three innovative system-based benchmarks and a systematic framework to assess the sustainability of civil facilities projects. There are also other studies discussing project characteristics for specific types of infrastructure, such as, Anagnostopouos et al. (2012) used spatial fuzzy analytic hierarchy process to help with the site selection of wastewater treatment system. Domingo (2015) studied the complexity of medical and health projects and its effects on the generation of construction waste in the life cycle of projects, and Zhou and Liu (2015) studied infrastructure projects from the perspective of micro engineering, with the aim of creating a basic model to analyze sustainable construction and operation of infrastructure projects. With the deepening of study, the evaluation model of infrastructure sustainability is constantly improved. Not only the sustainability of the project life cycle is evaluated from the cost point of view, but also the overall sustainability of the project life cycle is evaluated by using the LCA model. Glick et al. (2013) combined LCC, LCA, and economic input–output to assess

4.2 Literature Review

87

the sustainability of project. Parrish et al. (2014) developed an LCA based on the LCC for sustainability of infrastructure construction. Reza et al. (2014) proposed a sustainability appraisal method of infrastructure named Energy-Based Life Cycle Assessment. There are also studies on project forecasting based on previous studies. Zhi et al. (2015) analyzed the driving factors of water demand in Beijing through IO-IOAT-SDA, and systematically predicted the utilization of water resources in the following time of Beijing. In addition, some scholars have analyzed sustainability evaluation models. Ariaratnam et al. (2013) compared four common underground public infrastructure construction methods from three aspects, namely, the environment, cost, and social influence. Torres-Machi et al. (2014) peormed a comparative analysis to study seven economic evaluation road project models. There are many studies on sustainable infrastructure assessment, and several evaluation systems and index have been established in these studies, however, there is still no widely accepted criteria and system for assessing sustainability of infrastructure. Performance evaluation mostly focuses on the efficiency of investment and farmers’ satisfaction. Summing up previous studies, the construction of rural infrastructure in China has been continuously improved with government’s continuous investment, which has promoted the development of rural economy and farmers’ economic status and living environment. However, the investment efficiency of rural infrastructure still has to be increased, and farmers’ satisfaction on infrastructure construction is unsatisfactory. There is still great need for improvement in the performance of rural infrastructure construction. Many scholars have studied the investment in rural infrastructure by panel data (Xu 2011; Xu and Wang 2010) analyzed the changes in investment efficiency of rural infrastructure and the trends in its changes in 29 provinces and cities. They studied stock of rural roads, running water facilities, and power facilities from 1990 to 2007, and calculated the contribution of infrastructure investment to farmers’ income, expenditure, and the improvement of rural economy. Li (2011) established a fiscal performance evaluation system from economic, social, and ecological effects. The fiscal efficiency of 26 provinces (region) in China was analyzed on the basis of the evaluation system in their research. The investment efficiency of the projects has also been studied. Ansar et al. (2016) analyzed project performance from the perspective of cost performance, schedule performance, and performance by collecting data for 95 railway and highway projects from 1984 to 2008. They found that contrary to previous studies, the performance of Chinese infrastructure construction is unsatisfactory. They pointed out for the necessity of changing the development pattern of China’s infrastructure. Besides, scholars have also analyzed the performance of infrastructure investment through surveys, such as, Peng (2012), Zhang and Wang (2012), and Luo (2014) study the investigation of infrastructure investment performance from the perspective of farmers’ satisfaction. They investigate the influencing factors of farmers’ satisfaction, and draw lessons from domestic and international experience to put forward suggestions to improve the performance of rural infrastructure investment. Studies related to farmers’ satisfaction have focused mainly on the customer satisfaction model and empirical research. For the research based on customer satisfaction, Li and Zeng (2008) performed an empirical analysis using CSI-Probit Regression

88

4 Indicators Impacting Farmers’ Satisfaction …

Model on the satisfaction of rural public goods and its influencing factors. They found obvious common characteristics in the same type of city (state), and influencing factors of CSI consist of farmers receiving education, medical accessibility, the income level of farmers, and effective irrigation rate. Li (2012) and Li and Xu (2011) built a performance evaluation model and evaluation index system of rural public infrastructure based on the American Customer Satisfaction Index (ACSI). Studies on farmers’ satisfaction and influencing factors show that farmers are dissatisfied with the construction of rural infrastructure by the end of 2010 in most areas of China, and farmers are mainly concerned about roads, drinking water, basic education, water conservancy, medical facilities. On the other hand, it is worth noting that farmers began to pay attention to living infrastructure such as waste disposal facilities and sewage treatment facilities. Kong and Tu (2006), Yi et al. (2008), Zhang and Zhang (2009), Wang (2010), and Gan and Zhu (2011) studied farmers’ satisfaction, farmers’ demand, and investment willingness and current situation of rural infrastructure construction through empirical investigation. Fan and Luo (2009) used the structural equation model to analyze 670 questionnaires and found that farmers’ satisfaction is positively correlated with income, village type, price of infrastructure, family structure, and sense of superiority compared with neighboring villages. They also argued that the rural infrastructure construction not only suffers from “scarcity” but also from “inequality”. Tang et al. (2010) and Lei Wang (2013) launched studies of farmers’ satisfaction and its influencing factors based on a survey of 32 villages and towns in Shaanxi Province. The results showed that farmers’ satisfaction is significantly affected by rural roads, rural infrastructure, rural health care, irrigation facilities, drinking water facilities, and government credit. Moreover, Tang pointed out that farmers’ demand for rural public services has a certain level and stage, while farmers’ satisfaction has a certain order according to their demand. Han et al. (2015) built a custom satisfaction-based quality evaluation index and evaluation system. Existing research on farmers’ attitude only stayed at 2010, and there is no further related research after 2010. Problems and policy recommendations in the study showed that main problems in the construction of rural infrastructure consisting of the “top-down” decisionmaking mechanism cannot meet the real needs of farmers, has incomplete maintenance, imperfect laws and regulations, unclear division of responsibilities, imperfect supervision mechanism, lack of farmers’ participation, lack of capital, and lack of investors (Zhang et al. 2004; Li 2008a, 2011; Wang 2008; Xuan 2010; Li et al. 2010; Kang 2012). To solve these problems, researchers suggest establishing a decisionmaking mechanism to combine “bottom-up” and “top-down” and sound decisionmaking information communication mechanism, improving the decision-making, and supervising the mechanism of rural infrastructure and responsibility mechanism and improving the laws and regulations (Li and Zeng 2008; Li and Xu 2011; Li 2012; Kong and Tu 2006; Yi et al. 2008; Zhang and Wan 2009; Wang 2008, 2010; Fan and Luo 2009; Gan and Zhu 2011; Tang et al. 2010; Han et al. 2015; Lei Wang 2013; Zhang et al. 2004; Li 2008a; Xuan 2010), rural infrastructure investment, and financing system innovation (Xuan 2010) and introducing PPP into rural infrastructure (Wang 2008).

4.2 Literature Review

89

Integrated existing research, considerable research has been performed in relation to farmers’ satisfaction and influencing indicators in rural infrastructure. However, no further analysis related to significant indicators exists, such as the reason of their significant impact, influence pattern, measures to improve their performance to help with the improvement of rural infrastructure, and ways to avoid negative effects caused by them. Besides, most existing studies separately focus on farmers’ satisfaction, a certain kind of infrastructure, and the farmers’ satisfaction under horizontal comparison or vertical comparison. No comprehensive consideration of all rural infrastructure satisfaction, horizontal comparison, and vertical comparison of farmers’ satisfaction and farmers’ perception of infrastructure charges and other factors exist. On the other hand, There remains a gap ever since 2010, so that we have no understanding of the current situation and famers’ actual demands of rural facilities. Thus, to investigate the current status of rural infrastructure in Sichuan, this paper takes these aforementioned factors into account, and plans to achieve further analysis for significant indicators. In this paper, factor analysis and Logit Regression Model were used to analyze farmers’ satisfaction and its influencing indicators to know the current situation of rural facilities in Sichuan and prepare for further analysis.

4.3 Research Method Several methods to study farmers’ satisfaction have been utilized, including Analytic Hierarchy Process, factor analysis, Cluster Analysis, and CSI. Results calculated by different methods have some differences, but the general trend is basically the same (Tang et al. 2010). This paper used factor analysis to extract 15 factors, and then used the regression model to study the effects of the 15 factors on farmers’ satisfaction.

4.3.1 Factor Analysis Factor analysis is a multivariate statistical analysis method that can convert measured variables to a small number of non-related comprehensive factors. These comprehensive factors reflect the main information of original measured variables, and explain the relationship between measured variables (Tang et al. 2010). Specifically, factor analysis studies the condensation of a large number of measured variables to a few factors with the least information loss (Li 2008b). In this paper, for as many as 53 indicators, the use of factor analysis to convert indicators into comprehensive factors is necessary. The general form of factor analysis model is X i = μ + ai1 Fi1 + ai2 Fi2 + · · · + ain Fin + εi (i = 1, 2, . . . , p)

(4.1)

90

4 Indicators Impacting Farmers’ Satisfaction …

Among them, X i is a random observed variable, representing farmers’ personal feature and their judgment of infrastructure; Fi is a common factor; ai j (i = 1, 2,…, p) is factor load; εi is special factor part not included in common factors.

4.3.2 Logit Regression Model The dependent variable is the overall satisfaction of farmers with the rural facilities and is divided into two categories: “satisfaction” and “dissatisfaction”. Statistical methods that can be used to handle categorical dependent variables include Discriminant Analysis, Probit Analysis, Logit Regression Model and Log-linear Model. Logit Regression Analysis is an ideal model for analyzing individual decision behavior and is widely used in the analysis of influencing factors. Logit Regression Analysis is divided into Binary Logistic Regression Analysis, which the dependent variables can only be 1 or 0, and Multinomial Logistic Regression Analysis where the dependent variables can take more than two values. In this paper, the dependent variable is divided into two categories, so the Binary Logistic Regression Model is adopted. Variables do not have to meet the normal distribution or equal variance in Logit Model. The probability of occurrence for specimen is P(y = 1|xi ) = pi , and two probability incidents, occurrence and nonoccurrence, are recorded as the following two formulas: pi =

1

m

m

1 + e−(α+

i=1

βi xi )

eα+ i=1 βi xi m = 1 + eα+ i=1 βi xi

(4.2)

m

eα+ i=1 βi xi 1 m m = 1 − pi = 1 − α+ β x α+ i i i=1 i=1 βi x i 1+e 1+e

(4.3)

pi represents the probability of occurrence of events in observation; 1 − pi represents the probability of nonoccurrence of events in observation; both are nonlinear functions formed only by variable xi . pi /(1 − pi ) represents the ratio of the probability occurrence and the nonoccurrence of the event, named occurrence ratio. The logarithmic transformation to the occurrence ratio produces the linear model of the logit regression model.  pi )=α+ βi xi 1 − pi i=1 m

ln(

(4.4)

4.4 Empirical Study

91

4.4 Empirical Study 4.4.1 Evaluation Index Selection On the basis of existing research, we selected evaluation indices related to several types of facilities and several aspects, such as rural roads, drinking water, sewage treatment, renovation and construction of public toilets, village planning and renovation, electricity and communication signal facilities, renovation of fuel and kitchen, irrigation facilities, healthcare facilities, cultural and entertainment facilities, educational facilities, farmers’ satisfaction, satisfaction compared to that five years ago and neighboring villages, and feeling of fees. At the same time, farmers’ individual characteristics indices, such as gender, age, region, education, family population, family structure, family income, source of income, village type, distance from the village to county, were also included. Finally, 53 indices were selected. The sources and specific content of the evaluation index are shown in Tables 4.1 and 4.3. Farmers’ individual characteristics indices are important parts of the questionnaire. We can understand the regional differences of rural infrastructure construction in Sichuan Province, the economic conditions of farmers, and the views and needs of different ideological level on the construction of rural infrastructure through the study of famers’ individual characteristics. Rural roads, drinking water, sewage treatment, renovation and construction of public toilets, village planning and renovation, electricity and communication signal facilities, renovation of fuel and kitchen, irrigation facilities, healthcare facilities, cultural and entertainment facilities, educational facilities are contents of rural infrastructure. The purpose of this study is to fully understand the situation of rural infrastructure construction in Sichuan. Therefore, this study takes into account the above-mentioned infrastructure, and aims to understanding the construction of various types of rural infrastructure in Sichuan. Satisfaction compared to that five years ago: it can reflect the construction of rural infrastructure in Sichuan Province in the past 5 years through famers’ satisfaction compared to that five years ago. Satisfaction compared to neighboring villages: it shows differences in infrastructure construction between adjacent villages, and also studies the psychology of farmers to discuss the influence of comparative superiority on peasant household. Feeling of fees: it helps to understand the charges of infrastructure in the process of using, and understand the economic burden that rural infrastructure construction brings to peasant household. Total satisfaction: it reflects the satisfaction of rural households to the construction of rural infrastructure directly, at the same time as the dependent variable in Logistic Regression Analysis. It is essential in the research index.

92

4 Indicators Impacting Farmers’ Satisfaction …

Table 4.1 Index source Index

Source

Index

Source

Region

Peng (2012)

Gender

Peng (2012), Luo (2014), Li and Zeng (2008), Zhang and Wan (2009), Tang et al. (2010), and Lei Wang (2013)

Age

Peng (2012), Luo (2014), Li and Zeng (2008), Kong and Tu (2006), Zhang and Wan (2009), Tang et al. (2010), and Lei Wang (2013)

Family top education

Peng (2012), Luo (2014), Kong and Tu (2006), Zhang and Wan (2009), Tang et al. (2010), and Lei Wang (2013)

Family size

Luo (2014), Zhang and Wan (2009), Tang et al. (2010), and Lei Wang (2013)

Family structure

Fan and Luo (2009)

Village type

Peng (2012), Fan and Annual income Luo (2009), and Lei Wang (2013)

Peng (2012), Luo (2014), Li and Zeng (2008), Zhang and Wan (2009), Tang et al. (2010), and Lei Wang (2013)

Source of income

Peng (2012) and Kong and Tu (2006)

Peng (2012), Li and Zeng (2008), Fan and Luo (2009), and Lei Wang (2013)

Road

Li and Zeng (2008), Drinking water Yi et al. (2008), Zhang and Wan (2009), Wang (2010), Gan and Zhu (2011), and Tang et al. (2010)

Yi et al. (2008), Zhang and Wan (2009), Wang (2010), Gan and Zhu (2011), Tang et al. (2010), and Lei Wang (2013)

Sewage treatment

Yi et al. (2008), Zhang and Wan (2009), and Wang (2010)

Renovation of public toilets

Wang (2010)

Village planning and Wang (2010) reconstruction

Electricity and communication signal

Zhang and Wan (2009) and Wang (2010)

Renovation of fuel and kitchen

Irrigation facilities

Li and Zeng (2008), Kong and Tu (2006), Yi et al. (2008), Wang (2010), Gan and Zhu (2011), and Tang et al. (2010)

Wang (2010)

Distance from village to County

(continued)

4.4 Empirical Study

93

Table 4.1 (continued) Index

Source

Index

Source

Health facility

Li and Zeng (2008), Yi et al. (2008), Zhang and Wan (2009), and Gan and Zhu (2011)

Culture and entertainment

Zhang and Wan (2009), Gan and Zhu (2011), and Tang et al. (2010)

Education facilities

Li and Zeng (2008), Yi et al. (2008), Zhang and Wan (2009), Gan and Zhu (2011), and Tang et al. (2010)

Infrastructure compared with 5 years ago

Fan and Luo (2009) and Lei Wang (2013)

Infrastructure Compared with neighboring villages

Fan and Luo (2009) and Lei Wang (2013)

Infrastructure charges

Li and Xu (2011), Kong and Tu (2006), and Fan and Luo (2009

Total satisfaction

Li and Xu (2011), Fan and Luo (2009), and Tang et al. (2010)

4.4.2 Data Sources This paper is based on the National Natural Science Foundation of China Youth Fund Project “contribution of rural infrastructure investment and the degree of satisfaction of the government role orientation” (71301151), and data were obtained under the organization of home research with the help of students from the Engineering Management Department of Chengdu University of Technology. To ensure the validity and authenticity of the data, and to guarantee that interviewees can understand the questions well, the research group regarded students from rural area as pre-interviewers. A meeting was held to investigate students’ opinions of rural infrastructure and guarantee the validity of questionnaires been in charge of students. A total of 300 questionnaires were issued, 243 valid questionnaires were recovered, with an effective recovery rate of 81%. The questionnaire covers 23 cities in Sichuan Province, encompassing a wide range of regional representation. A total of 106 women and 137 men were interviewed, accounting for 43.6 and 56.4%, respectively. The basic situation of all investigated objects is shown in Table 4.2.

4.4.3 Farmers’ Satisfaction Factor Analysis The Likert scale was used for the questionnaire analysis (for example: 1 = very dissatisfied; 3 = dissatisfied; 5 = moderate; 7 = satisfied; 9 = very satisfied). The definition of variables, except for 5, is established with linear interpolation method. Parts of variables are shown in Appendix.

94

4 Indicators Impacting Farmers’ Satisfaction …

Table 4.2 Basic information of interviewees Factor Gender

Family size

Age

Proportion (%)

Factor

Female

43.6

Male

56.4

Family top education

Less than 4 people

Proportion (%) Primary school

.8

Junior high school

7.6

30.4

High school

13.9

4 people

30

Bachelor

75.6

5 people

25.8

Master

1.7

More than 5 13.8 people

Doctor

.4

Less than 20 years

15

Annual income

Less than 50,000 Yuan

67.5

From 20 to 30 years

69.1

From 50,000 to 22.8 100,000 Yuan

From 30 to 40 years

5.2

From 100,000 to 200,000 Yuan

6.8

From 40 to 50 years

8.2

Over 200,000 Yuan

2.9

From 50 to 60 years

2.1

Over 60 years

.4

SPSS software was used to analyze the data. The original data are normalized to eliminate the difference in magnitude and dimension. To ensure validity of the questionnaire, the validity and construct validity of the questionnaire were tested by the Cronbach’s coefficient. The coefficient of the questionnaire data was .927, indicating that the questionnaire is reliable. Factor analysis showed that the KMO statistic was .775, and the P value was 0, hence, the test results were significant. These data indicated that questionnaire data had a certain correlation, indicating its suitability for factor analysis. Factor analysis showed that the extractions of 53 indicators are all above .4, and all these extractions of indicators, except for family structure and the average annual income, reached a degree of .5, demonstrating that most of the information in the representation variable is extracted by factors. Using the principal component method to extract 16 factors, the researchers determined that their feature values are greater than 1. The component matrix is rotated using the method of maximum variance orthogonal rotation because the initial loading structure is not clear. Moreover, the load of the village type index is less than .4 on the common factor, and factor analysis is carried out again after the variable is deleted. Fifteen factors were extracted after the move out of village type. The component matrix is rotated using maximum variance orthogonal rotation,

.125

.069 −.079 −.050 −.067

.090 −.190 .893 .064 .580 .007 .045 .135

.324 −.011

.120 −.011

.179 .037

.049 −.023

.196 .003

Road satisfaction (X10 )

Drinking water facilites satisfaction (X11 )

Sewage treatment satisfaction (X12 )

Renovation of public toilets satisfaction (X13 )

Village planning and reconstruction satisfaction (X14 )

.141 .092

.169

.171

.799

.605

−.002 .169

.052

.090

.164

.513

.137 .079

.241 .032

.067 .093

.079 .093

.063 −.062

.083 .063

.032 −.152 −.002 −.011 −.087 .057

.005

.262 .051

.015

Distance from village to County (X9 )

−.072 .060

12

.045 .865

.078 .096

.087 .027

.047 .118

.365 .128

13

14

.638

.033 .673

.044 .653

.078 .375

.037 .094

.193

.601

−.024

−.040 .221

.304

.184

−.037 −.113

.047

.034 −.004 −.016 .013

(continued)

.146

.039

−.047

.006

−.101

.135

.005

.444

.083

.039

.015

.082 −.072 .065

−.132 .092

.065 −.056 −.023 .079

.012 .192

.063

−.041 .141

.064

−.844

.212

15

−.053 −.013

.024

−.083 .715

.082 −.055 .691

.020 .185

.035 −.088 .045

.341 −.174 −.467 −.089

11

.010 −.129 −.013 .079

−.045 .001

.003 −.030

.103 −.053 −.056 .097

.087 .048

.072

.056 .027

.029 −.038 −.047 .069

.104 .005

.173 −.082 −.006 −.021

.039

.099 −.003

.001

Source of income (X8 )

.069 .055

.053 −.228

10

−.049 −.024

9

.066 −.027 −.082 −.176 −.085

.063 .010

.128 −.078

8

.012 −.035

−.031 −.122 −.184 .039

Family structure (X6 )

.040

7

Annual income (X7 )

−.075 .033

Family size (X5 )

.044 .008

.027

−.150 −.042 −.002 −.104

Family top education (X4 )

Age (X3 )

.100

6 .039 −.046

.041 −.166 .001 −.030

5

.093 −.193

4

−.041 −.037

3

Gender (X2 )

2

Region (X1 )

1

Factor

Table 4.3 Rotated component matrix

4.4 Empirical Study 95

.251 .050 .002 .278

.124 −.013

.079 −.013

.157 .051

.098 .015

Irrigation facities satisfaction (X17 )

Health facility satisfaction (X18 )

Culture and entertainment satisfaction (X19 )

Education facilities satisfaction (X20 )

Road compared with 5 years ago (X21 )

.012

.618 −.002 .059 .185

.208 .030

.137 −.032

Sewage treatment compared with 5 years ago (X23 )

Renovation of public toilets compared with 5 years ago (X24 )

.184

.164

.893 .064

.120 −.011

Drinking water facilites compared with 5 years ago (X22 )

.751

.594

.052

.621 −.079

.182 −.141

.133

.310

.124

.125

.070

.165

.059 .030

.119

.157

.108

.227

.799

6

.056

5

.057 .841

.061 .152

.088 .109

.114 .220

4

.137 .009

3

Renovation of fuel and kitchen satisfaction (X16 )

2

.095 −.020

1

Factor

Electricity and communication signal satisfaction (X15 )

Table 4.3 (continued) 7

.189 .009

.114 .107

.079 .093

.073 .094

.128 .077

.758 .086

.161 .069

.572 .275

.140 .873

.072 .197

8

9

12

.054 .038

−.132 .023

11

.089 .236

.096 .036

.047 .118

.205 .285

.244 .104

.100 .171

.816 .015

.002

14

.216

−.005 .078

.059

−.035

.002

(continued)

−.038 −.066

−.047

−.148

−.057 −.085 −.006

.167 −.118 .052

−.067 .082

.050

.198

−.109 −.063

.065 −.055 −.023 .079

.043 .224

.037 .128

.015

−.046 −.037 −.099 −.084 .016 .174

.052

15

−.010 −.011 −.036

.114

13

.433 −.058 −.090 −.142 .123

.159 .076

.028 .099

10

96 4 Indicators Impacting Farmers’ Satisfaction …

.094 .128

.262 .088 .132 .392

.154 −.003

.191 −.033

.143 −.082

.201 .027

.098 .015

.532 .109

Renovation of fuel and kitchen compared with 5 years ago (X27 )

Irrigation facities compared with 5 years ago (X28 )

Health facility compared with 5 years ago (X29 )

Culture and entertainment compared with 5 years ago (X30 )

Education facilities compared with 5 years ago (X31 )

Road compared with neighboring villages (X32 )

.504 −.086

.133

.249

.108

.083

.046

.173

.166

.110 −.111

.124

.116

.106

.222

.812

.209

6

.056

5

.057 .841

.031 .209

.116 .211

.082 −.012

4

Electricity and communication signal compared with 5 years ago (X26 )

3 .150 .104

2

.191 .003

1

Factor

Village planning and reconstruction compared with 5 years ago (X25 )

Table 4.3 (continued) 7

.104 .120

.128 .077

.748 .087

.125 .086

.544 .279

.128 .882

.098 .233

.152 .093

8

9

.051 .169

.244 .104

.112 .216

.753 .115

.466 .012

.137 .081

.034 .073

.060 .862

10

12

13

.013

−.024

14

.027

.035

15

.109

−.031 .135

.037 .128

.048 .181

−.106 −.104

−.059 .129

(continued)

−.006

−.057 −.085 −.006

.007

−.014 −.023 −.094 −.070 −.056

.199

−.017 −.001 −.054

.119

−.063 −.164 .115

.083 .035

−.112 .018

.047 −.021 .017

11

4.4 Empirical Study 97

.625 .024

.401 .069

.541 .112

Electricity and communication signal compared with neighboring villages (X37 )

Renovation of fuel and kitchen compared with neighboring villages (X38 )

Irrigation facities compared with neighboring villages (X39 )

.277 .173

.313 .248

.057 .156

.094 .168

.170 .092

.020 −.036

.123

.147

.485

.051 .532

.032

.096 −.025

.007 −.087

8

.194 −.001 .132

7

.025

.366

.086 .211

.169

−.041 .034

.665 .080

Village planning and reconstruction compared with neighboring villages (X36 )

−.140

.525 −.019 −.011 .214

Renovation of public toilets compared with neighboring villages (X35 )

6 .190 −.073

.202 −.002 −.098

5

.688 −.006

4

Sewage treatment compared with neighboring villages (X34 )

3 .412 .086

2

.536 −.181

1

Factor

Drinking water facilites compared with neighboring villages (X33 )

Table 4.3 (continued) 12

14

.017

−.002 .022

.149

.053

.052

(continued)

−.142 −.035 −.057 −.015

.047 −.011 −.083 −.080 .138

.078

−.031

−.110 −.031

−.019 .087

.024

.006

15

−.066 −.043

−.067 −.101

13

.032 −.006 −.046 .033

.099 .163

.005 .121

11

.073 −.064 −.001 .002

.022 .357

.103 .159

.225 .104

.057 .071

10

−.155 .064

9

98 4 Indicators Impacting Farmers’ Satisfaction …

.149 .671

Fuel charge evaluation (X49 )

−.111 .200

−.052 .596

Communication charge evaluation (X47 )

.096

.011

.145

.136 .051

.033 .016

−.069 .079

−.029 −.038

.150 −.008

−.022 −.010

.181 −.255 −.233 −.065

.013 −.093 .027

13

14

.096

−.076 .052

.021

.002

−.086

−.093

.273 −.101 .020

.336 −.133 .199

.199

.164

(continued)

−.148

−.219

.173 −.078 −.029 −.008 −.020

.013 .113

.071 .048

.174

−.007

.016

15

−.078 −.002 −.095 .107 −.028 −.062 .156

−.116 .098

.073 −.094 −.074 .087

.054 −.067 −.110 .053

12

−.005 .187

11 −.087 −.099 −.056 .108

10 .134 .045

9

.260 −.010 −.150 −.052

.020 −.202 −.053 .017

.162 −.209

.041 −.135

.193 −.103 −.007

.066 .108

.143 .105

.496 .102

.256 .058

8

−.131

.007

.045

.146

.155

7

.165 −.099 .072

6

−.009

.023 −.101 −.014

.147 .088

−.089 .022

−.131 .728

Evaluation of electricity price (X46 )

.097 .605

.051 .016

.196 .672

Evaluation of environmental management fees (X45 )

.054 .159

.130 .658

Evaluation of water charge (X44 )

Broadband charge evaluation (X48 )

5

−.088 −.255 −.006

.073 .603

.522 −.059

Education facilities compared with neighboring villages (X42 )

Bus charge evaluation (X43 ) −.185 .637

.125 .221

.016 .451

4

.532 .095

3

Culture and entertainment compared with neighboring villages (X41 )

2

.489 .059

1

Factor

Health facility compared with neighboring villages (X40 )

Table 4.3 (continued)

4.4 Empirical Study 99

Evaluation of education fees (X52 )

Evaluation of cultural and entertainment charges (X51 )

Evaluation of medical treatment fee (X50 )

Table 4.3 (continued)

2

4

−.119

.010 .226

6

7

9

10

.055 −.006 −.001 −.047 .081

.010 .034

.079 −.017 −.141 −.023

8

.151 −.019 .042

.134 −.168

−.039 .245

5

.033 −.068 −.135

−.006 −.106

3

.064 .249

−.116 .510

1

Factor 12

13

14

.733 .052

.680 .162

.076

15

−.061

−.068 −.008

−.011 .132

.070

.491 −.123 −.002 .043

11

100 4 Indicators Impacting Farmers’ Satisfaction …

4.4 Empirical Study

101

Table 4.4 Indicators in each factor Factor

Index

Factor

Index

F1 (Horizontal comparison)

X32 , X33 , X34 , X35 , X36 , X37 , X39 , X40 , X41

F9 (Health facilities)

X18 , X29

F2 (Charge factor)

X43 , X44 , X45 , X46 , X47 , X48 , X49 , X50

F10 (Village planning and reconstruction)

X14 , X25

F3 (Drinking water facilities)

X11 , X22 , X23

F11 (Charge factor)

X51 , X52

F4 (Education)

X20 , X31 , X42

F12 (Income situation)

X7 , X8

F5 (Road, electric and communication signal)

X10 , X15 , X21 , X26

F13 (Area, age and family highest degree)

X1 , X3 , X4

F6 (Renovation of public toilets)

X12 , X13 , X24

F14 (Family size and distance from village to County)

X5 , X9

F7 (Irrigation facilities, culture and entertainment)

X17 , X19 , X28 , X30

F15 (Gender and family structure)

X2 , X6

F8 (Renovation of fuel and kitchen)

X16 , X27 , X38

and the loads on the common factors are all above .4, whereas the total variance explained is 71.2%. The rotated component matrix is shown in Table 4.3. The distribution of factor indices is shown in Table 4.3. Index situation and factor naming are shown in Table 4.4. Table 4.4 shows that horizontal comparison indices (perceptual evaluation index compared with neighboring villages) are concentrated mainly in factor F1, and the longitudinal comparison index is distributed mostly in the factor of this kind of infrastructure. At the same time, rural infrastructure charge perception indices are concentrated in factors F2 and F11, personal and village characteristics are concentrated in factors F13, F14, and F15. Farmers’ satisfaction on rural infrastructure and vertical comparison index (perception evaluation index compared with five years ago) are relatively concentrated in the infrastructure factor. The comprehensive score of each index was obtained based on the analysis of the factor score coefficient table, and the scores of the absolute values are normalized to obtain the weight of each index. According to the results of the questionnaire, the overall evaluation score of rural infrastructure construction satisfaction was 4.84.

4.4.4 Logit Regression Analysis Based on Factor Analysis The dependent variable of this paper is the total satisfaction of rural households with the infrastructure construction, which is divided into two categories: “satisfaction” and “dissatisfaction". Logit regression analysis was used to analyze 15 factors. The results of the omnibus test are shown in Table 4.5, and the results of Hosmer and Lemeshow test are shown in Table 4.6. Test results show that the model is very

102

4 Indicators Impacting Farmers’ Satisfaction …

Table 4.5 Omnibus test of model coefficients Step 1

χ2

df

Sig

Step

81.505

15

.000

Block

81.505

15

.000

Model

81.505

15

.000

Table 4.6 Hosmer and Lemeshow test

Step

χ2

df

Sig

1

1.818

8

.212

Table 4.7 Model estimation results Variable Parameter Wald

Variable Parameter Wald

Variable

Parameter Wald

Ln(F1 )

1.073***

−.247

2.246

Ln(F2 )

−.091

.292 Ln(F8 )

.570***

11.302 Ln(F14 )

.184

1.383

Ln(F3 )

.214

1.590 Ln(F9 )

.473***

7.763 Ln(F15 )

.043

.069

5.969 Constant

.856***

27.497 Ln(F7 )

Ln(F4 )

.026

Ln(F5 )

.398**

6.053 Ln(F11 )

.027 Ln(F10 )

Ln(F6 )

.643***

11.778 Ln(F12 )

.322*

.407** .020 .467**

3.629 Ln(F13 )

23.710

.015 6.347

Note *, **, *** respectively represent the significant level of 10%, 5% and 1%

significant, and dependent variables had no significant difference with predicted values. The model has a −2Ln likelihood of 237.764, while the observation correct percentage of the model reached 75.3%, indicating that the fitting effect of the model is ideal. Estimation results of the model are shown in Table 4.7.

4.5 Discussion According to the Wald value in Table 4.7, factors with significant effect on the satisfaction of farmers include F1, F6, F8, F9, F12, F10, F5, and F7, and their effect size is decreased in turn. F1 represents the horizontal comparison factor, which reflects mainly the evaluation of infrastructure compared with neighboring villages with a significant level of 1%. It has a very significant positive effect on farmers’ satisfaction in all factors. Specifically, the higher the horizontal comparison, the higher the total satisfaction. This result is consistent with the result of Fan and Luo (2009). F6 refers to the factor of renovation and construction of public toilets, which has a significance level of 1%. It has a positive effect on rural infrastructure satisfaction, meaning the more satisfied the farmers are with public toilets construction and renovation, the more satisfied they are with the overall rural infrastructure. Public

4.5 Discussion

103

toilets construction and renovation is related to the overall health condition and village image, and the positive effect shows that farmers pay attention to the living conditions and environment. F8 represents the renovation of fuel and kitchen factor, which passed the 1% significant level test and also has a positive effect on rural infrastructure satisfaction, indicating that the better constructed the fuel and kitchen renovation is, the more satisfied farmers are with the overall infrastructure. Although the rural economy improved significantly, the main fuel is still coal or firewood in many rural areas, and natural gas is not entirely popular in rural areas. Fuel and kitchen renovation can reduce waste gas pollution caused by the rural kitchen and bring convenience to farmers. The effect of this factor on farmers’ satisfaction shows that farmers pay attention to the improvement of the living environment and life quality. F9 represents healthcare facilities, which reached the 1% level in the significance test, and has a positive effect on farmers’ satisfaction, indicating that the more perfect health facilities and better medical conditions significantly contribute to farmers’ satisfaction of rural infrastructure. F12 is a reflection of the income of farmers, with a 5% level of significance test. It has a positive effect on farmers’ satisfaction, with the higher household annual income, the more income sources, and the higher the rural infrastructure satisfaction evaluation. F10 is a reflection of the village planning and renovation, with a 5% significance level. It has a positive effect on farmers’ satisfaction, consistent with F6, indicating that farmers pay attention to the change of living environment and propose requirements of living conditions. F5 comprehensively represents road, electricity, and communication signals facilities, with a 5% significance level of testing. It also has a positive effect on farmers’ satisfaction. In recent years, the government has increased its investment in rural roads, resulting in their significant improvement. However, judging from the results of this analysis, rural roads construction still does not meet farmers’ expectations, implying that the government should continue to adhere to the construction of rural roads. With the improvement of the economic level of the rural people, an increasing number of families begin to own electrical appliances, demanding higher power supply and telecommunications signals. F7 comprehensively reflects irrigation facilities and cultural and entertainment facilities. It has a significant level test of 10% and also has a positive effect on farmers’ satisfaction. Similar with the rural roads construction, farmland irrigation equipment has received strong support from the government in recent years, but does not meet the needs of farmers. A need to continue to adhere to the construction of irrigation facilities still exists. The positive effect of cultural and entertainment facilities shows that people care about their life quality. The top three factors that affect farmers’ satisfaction are “horizontal comparison”, “renovation and construction of public toilets” and “renovation of fuel and kitchen”. The horizontal comparison factor ranks in the first place, which coincides with the research of Fan and Luo (2009). This shows that there still exists the phenomenon of rural infrastructure construction that is not only “suffering from poverty”, but also

104

4 Indicators Impacting Farmers’ Satisfaction …

“suffering from uneven”. To some extent, this unequal gap grows with the increase of time and the continuous construction of rural infrastructure. “Renovation and construction of public toilets” and “renovation of fuel and kitchen” rank second and third respectively, indicating that they both have a very significant influence on farmers’ satisfaction. Wang (2010) found that farmers’ satisfaction degree with the renovation and construction of public toilets of Guangdong province was only .35 in 2009, which is nearly half below average satisfaction level, and the satisfaction degree with renovation of fuel and kitchen of Guangdong province was only .29 in 2009 which is 58.54% below average. That study indicated that renovation of rural public toilets and the renovation of kitchen fuel in Guangdong province were not optimistic in 2009. Results of Wang (2010) show that farmers were more satisfied with production facilities, such as roads, telecommunications, electricity, and small water conservancy, but less satisfied with living facilities, such as sewage treatment, renovation and construction of public toilets, and renovation of fuel and kitchen. Our finding is consistent with Wang’s research. It is worth noting that Table 4.7 shows that farmers are focusing on living infrastructure rather than production infrastructure which has always been the focus of attention of farmers in existing research. Tang et al. (2010) pointed out that the farmers’ satisfaction of rural public service investment are mainly influenced by rural roads, rural basic education, rural medical treatment, irrigation and water conservation facilities, and drinking water facilities. Kong and Tu (2006) found that farmers from resourced areas of Shanxi, Shaanxi and Mongolia expect private capital to invest in roads, medical care, and education. Fan and Luo (2009) pointed out that farmers’ satisfaction is mostly influenced by their superiority compared with other villages, followed by rural infrastructure prices, family structure, distance from village to county, growth of per capita income, gap between communication supply and demand, and village type. Kong and Tu (2006) found that the most urgent needs of farmers in Fujian province were infrastructure for farmland, water conservancy facilities, roads, medical and health conditions, education and other infrastructure in 2005. Yi et al. (2008) find that what farmers’ of Sichuan need most are roads, irrigation, drinking water, schools and clinics in 2005. However, according to the results of this study, health facility, roads, electricity, communications, electricity and communication signal, irrigation facilities have significant effects on farmers satisfaction of Sichuan province, but their influence rankings are relatively backward except health facility, and the effects of drinking water facilities, educational facilities, infrastructure charges and sewage treatment facilities on the satisfaction of farmers were less significant. Tang et al. (2010) pointed out that farmers’ demand for rural public has a certain level and different phases, while farmers’ satisfaction has a certain order according to their demand. That is to say, farmers’ satisfaction and demand varies from regions, economic conditions and infrastructure construction. Since most of studies have been done for a long time, and not for Sichuan Province, results of this paper are quite different from those of the existing literatures. With continuous improvement of rural infrastructure construction, rural roads, education and health infrastructure can basically meet farmers’ needs. Farmers’ attention to rural infrastructure has gradually shifted from production infrastructure to living infrastructure.

4.5 Discussion

105

At the same time, with the development of rural economy and the increase of government investment, rural infrastructure charges no longer bring economic burden to farmers. Sewage treatment facilities is also related to the living environment and health status of rural areas, such as public toilets and construction of healthcare facilities, but it does not have significant effect on farmers’ satisfaction. Although farmers pay attention to living environment and health status, their understanding of rural environmental pollution is insufficient, indicating the absence of a good understanding of the hazardous effects of domestic sewage to environment and health. Publicity and training should be increased to help farmers become fully aware of the pollution sources and consequences of rural environmental pollution, and to appeal everyone to work together to improve the ecological environment in rural areas. Based on Table 4.7, this study finds that the regions, family size, age, and family top education have little influence on farmers’ satisfaction. This result is basically consistent with the results of Tang et al. (2010) and Fan and Luo (2009). Since the study was carried out only in Sichuan, where regional differences are not obvious, so the regional factor has little influence on farmers’ satisfaction. Family structure, distance between villages and village type have little influence on farmers’ satisfaction, which is inconsistent with the results of Fan and Luo (2009). The reason of this difference may be the greatly improvement of farmers’ collective literacy with the development of rural. The ideological differences between ordinary families and cadres, families and Party members have narrowed. On the other hand, infrastructure of ordinary villages and villages far from the county has also been well improved under continuous construction of the rural infrastructure. According to the above analysis, factors that affect satisfaction of farmers are represented mainly by the rural living infrastructure, confirms the hierarchy and phase of farmers’ demand for rural public services, and order of their satisfaction proposed by Tang et al. (2010). Hence, the demand for basic facilities related to production can be considered as having been basically met, and farmers are beginning to require better living environment, cultural entertainment, and other living-related aspects. Farmers’ demand for rural infrastructure is no longer limited to meeting the needs of life and production, but now include the requirement of living infrastructure.

4.6 Conclusion This paper conducted a comprehensive investigation of the situation of rural infrastructure from the perspective of farmers’ satisfaction. Factor analysis and Logit Regression Model were used to analyze farmers’ satisfaction. The overall satisfaction score of farmers in terms of infrastructure was dissatisfied (4.84 points), indicating that the demand for infrastructure of farmers has not been satisfied, and many problems still need to be addressed in the construction of rural infrastructure. Farmers’ satisfaction is affected mainly by the horizontal comparison factor, road facilities, electricity and communication signal facilities, public toilets renovation, irrigation facilities, cultural and entertainment facilities, fuel and kitchen renovation,

106

4 Indicators Impacting Farmers’ Satisfaction …

village planning, medical facilities, and farmers’ income situation. Area, age, family top education, family size, and distance from village to county have certain effects on the satisfaction of the farmers, but the effects are limited. Drinking water facilities, sewage treatment facilities, educational facilities, charges, gender, and family structure have little effect on farmers’ satisfaction. Farmers’ demand for rural infrastructure is transitioning from production to livelihood. Thus, the government needs to proceed from the actual situation in rural areas and take into account the real needs of farmers to promote the development of rural economy, improve the living standards of farmers, and enable rural infrastructure to truly meet users’ demands. This paper studies the present situation of the construction of rural infrastructure in Sichuan Province, and discusses the construction of various types of rural infrastructure and farmers’ views and needs comprehensively. It lays a foundation for further research on development of rural infrastructure construction and farmers’ demands, and provides a theoretical basis for policy makers. Indexes used in this research are derived from previous studies, most of which are old, and are aimed at other provinces. Therefore, these indexes cannot reflect the current rural infrastructure situation of Sichuan Province. It is hoped that indexes could be further adjusted after this round of research, so that a comprehensive index system could be established to evaluate the current situation of rural infrastructure construction in Sichuan Province.

Appendix: Variable Definition

Variable

Variable definition

Total satisfaction

0 = dissatisfied, 1 = satisfied

Gender

1 = female, 9 = male

Age

1 = over 60 years, 2.6 = from 50 to 60 years, 4.2 = from 40 to 50 years, 5.8 = from 30 to 40 years, 7.4 = from 20 to 30 years, 9 = less than 20 years

Family top education

1 = primary school, 2.6 = junior high school, 4.2 = high school, 5.8 = Bachelor, 7.4 = Master, 9 = Doctor

Family size

1 = more than 5 people, 3.67 = 5 people, 6.33 = 4 people, 9 = less than 4 people

Family structure

1 = ordinary family, 5 = party family, 9 = cadre family

Annual income

1 = less than 50,000 Yuan, 3.67 = from 50,000 to 100,000 Yuan, 6.33 = from 100,000 to 200,000 Yuan, 9 = over 200,000 Yuan (continued)

Appendix: Variable Definition

107

(continued) Variable

Variable definition

Village type

1 = ordinary village, 3.67 = township resident, 6.33 = combination of urban and rural areas, 9 = both township resident and a combination of urban and rural areas

Distance from village to County

1 = more than 100 km, 2.6 = from 50 to 100 km, 4.2 = from 30 to 50 km, 5.8 = from 20 to 30 km, 7.4 = from 10 to 20 km, 9 = less than 10 km

Infrastructure satisfaction

1 = very dissatisfied, 3 = dissatisfied, 5 = moderate, 7 = satisfied, 9 = very satisfied

Infrastructure compared with 5 years ago

1 = worse, 3.67 = almost no change, 6.33 = have certain improvement, 9 = much better

Village infrastructure Compared with neighboring villages

1 = one of the worst, 3 = worse than medium, 5 = medium, 7 = better than medium, 9 = one of the best

Infrastructure charges

1 = no charge, 2.6 = very cheap, 4.2 = cheap, 5.8 = suitable, 7.4 = expensive, 9 = very expensive

References Anagnostopoulos, K., Vavatsikos, A., & Ph.D. 2012. Site suitability analysis for natural systems for wastewater treatment with spatial fuzzy analytic hierarchy process. Journal of Water Resources Planning and Management, 138(2): 125–134. Ansar, A., Flyvbjerg, B., Budzie, A., & Lunn, D. 2016. Does infrastructure investment lead to economic growth or economic fragility? Evidence from china. Oxford Review of Economic Policy, 32(3): 360–390. Ariaratnam, S. T., Piratla, K., Cohen, A., & Olson, M. 2013. Quantification of sustainability index for underground utility infrastructure projects. Journal of Construction Engineering and Management, 139(12): A4013002. Bocchini and M.ASCE, Frangopol, D. M., Dist.M.ASCE, Ummenhofer, T., & Zinke., T. , 2014. Bocchini, P., M.ASCE, Frangopol, D. M., Dist.M.ASCE, Ummenhofer, T., & Zinke., T. 2014. Resilience and Sustainability of Civil Infrastructure: Toward a Unified Approach. Journal of Infrastructure Systems, 20(2): 04014004. Boz, M. A., M.ASCE, El-adaway, I. H., & M.ASCE. 2015. Creating a holistic systems framework for sustainability assessment of civil infrastructure projects. Journal of Construction Engineering and Management, 141(2): 125–134. Domingo, N. 2015. Assessment of the impact of complex healthcare features on construction waste generation. Buildings, 5: 860–879. Fan, L., & Luo, Y. 2009. Influencing factors of farmers’ satisfaction to rural infrastructure—based on the structural equation model analysis of 670 questionnaires. Agricultural Economic Problems(9): 51–59. Gan, J., & Zhu, Y. 2011. Folk capital access to rural infrastructure—based on the investigation of 31 villages and towns in the resource rich region of Shanxi and Shaanxi. Exploration of Economic Problems(11): 153–158. Gan, L., Shen, L., & Fu, H. 2009. Evaluation index system of infrastructure projects based on sustainable development. China Civil Engineering Journal, 42(11): 133–138.

108

4 Indicators Impacting Farmers’ Satisfaction …

Glick, S., Guggemos, A. A., & A.M.ASCE. 2013. Rethinking wastewater-treatment infrastructure: Case study using life-cycle cost and life-cycle assessment to highlight sustainability considerations. 139, 12: A5013002. Han, S., Chae, M. J., Hwang, H., & Choung, Y.-k. 2015. Evaluation of customer-driven level of service for water infrastructure asset management. Journal of Management in Engineering, 31(4): 04014067. Kang, W. 2012. Research on investment and financing system innovation of Chinas’ rural infrastructure. Contemporary Economic Management, 34(9): 91–96. Kong, X., & Tu, S. 2006. Research on the demand preference and influence factors of rural households to public goods in new rural construction—a case study of irrigation and water conservancy facilities. Agricultural Economic Problems (10): 10–16. Lei Wang, Y. Z. 2013. Performance appraisal of rural drinking water supply based on farmers’ perception: An analysis of the data from 803 households. Agricultural Technical and Economic(2): 64–71. Li, C. 2008a. Current problems and countermeasures of rural infrastructure construction. Journal of Shandong Agricultural Administrators College, 23(3): 25–28. Li, W. 2008b. Application of Multivariate Statistical Analysis. Beijing: Peking University Press. Li, Z. 2011. Study on the regulation and control of rural infrastructure in China. Northeast Normal University, 6. Li, W. 2012. Farmers’ satisfaction evaluation of rural public infrastructure. Journal of Chongqing University, 18(4): 9–15. Li, W., & Xu, K. 2011. Research on farmers’ satisfaction model and evaluation index establishment of rural public infrastructure. Journal of Second Hubei Education University, 25(8): 74–98. Li, Y., & Zeng, F. 2008. Farmers’ satisfaction of rural public goods supply and its influencing factors. Research on Quantitative Economic Technology, 8: 3–18. Li, Z., Liu, H., & Liu, J. 2010. Study on the imbalance of China’s rural infrastructure construction. Geographical Science, 30(6): 839–846. Luo, S. 2014. Small farmland water conservancy investment performance evaluation based on farmers’ satisfaction. Hunan: Hunan Agricultural University: 6. Parrish, K., A.M.ASCE, & Chester, M. 2014. Life-cycle assessment for construction of sustainable infrastructure. Practice Periodical on Structural Design and Construction, 19(1): 89–94. Peng, D. 2012. Performance evaluation of rural public investment based on farmers’ satisfaction. Shandong: Shandong Agricultural University, 6. Reza, B., Sadiq, R., & Hewage, K. 2014. Emergy-based life-cycle assessment (Em-LCA) for sustainability appraisal of infrastructure systems: A case study on paved roads. Clean Technologies and Environmental Policy, 16: 251–266. Shen, L., M.ASCE, Wu, Y., & Zhang, X. 2011. Key assessment indicators for the sustainability of infrastructure projects. Journal of Construction Engineering and Management, 137(6): 441– 451. Tang, J., Zhu, Y., & Liu, C. 2010. Analysis of rural public service satisfaction and its influencing factors—based on the survey of 67 villages in 32 towns in Shaanxi province. Contemporary Economic Science, 32(1): 101–116. Torres-Machi, C., Yepes, V., Chamorro, A., & Pellicer, E. 2014. Current models and practices of economic and environmental evaluation for sustainable network-level pavement management. Journal of Construction, 13(2): 49–56. Wang, C. 2008. Study on PPP model of rural infrastructure governance. Agricultural Economic Problems(6): 64–67. Wang, C. 2010. Farmers’ willingness to participate in the financing of rural infrastructure construction—a case study of 19 villages in Guangdong. Shanghai Economic Research(3): 18–26. Xu, S. 2011. Research on the establishment of investment contribution model of rural infrastructure. Project Management Journal, 25(3): 288–291.

References

109

Xu, S., & Wang, Y. 2010. Efficiency measure of rural infrastructure investment in 29 provinces and cities of China. Construction Economy(2): 112–114. Xuan, S. 2010. Research on Decision Making Mechanism of Rural Infrastructure Construction. Nanjing: Nanjing Agricultural University, 12. Yi, H., Zhang, L., & al, D. H. e. 2008. The relationship between the investment of rural infrastructure and the farmers’ investment demands: An empirical analysis of 5 provinces. China Soft Science(11): 106–115. Zhang, K., & Wan, M. 2009. Research on farmers’ willingness to invest in the rural infrastructure construction—a case study of Henan province. Journal of Zhongnan University of Economics and Law(5): 36–41. Zhang, X., & Wang, K. 2012. Analysis on regional difference of China’s rural infrastructure investment. Financial Science(3): 77–84. Zhang, Z., Xia, L., & Zhang, L. 2004. Thoughts on introducing nongovernmental funds into rural infrastructure construction in China. (3): 66–68. Zhi, Y., Yang, Z., Xin-An Yin, B.Hamilton, P., & Zhang, L. 2015. Evaluating and forecasting the drivers of water use in a city: Model development and a case from Beijing. Journal of Water Resources Planning and Management: 04015042. Zhou, J., & Liu, Y. 2015. The method and index of sustainability assessment of infrastructure projects based on system dynamics in China. Journal of Industrial Engineering and Management, 8(3): 1002–1019.

Chapter 5

Indicators Impacting Rural Residents’ Satisfaction in Household Latrines

Abstract This is a second survey, following an earlier study of rural facilities that found that the toilet was the most significant influencing factor affecting farmers’ satisfaction. The quality of rural sanitary latrines in China has been low, and the health of rural residents and the environmental sanitation in rural areas has not been effectively guaranteed. Factor analysis and logistic regression models are used to study the current situation of rural household latrines, rural residents’ satisfaction and its influencing indicators within Sichuan. This research led to three key findings: (1) the present situation of rural household latrine construction in Sichuan is less than ideal; (2) rural residents are relatively satisfied with rural household latrines; (3) rural residents’ satisfaction is affected mainly by village committee performance, response time, the quality of latrine construction, transparency of village affairs, sources and subsidies of funds, construction participants and construction methods. Results of this study lay a foundation for further research regarding rural household latrines, and provide a theoretical basis for the construction and reform of rural household latrines in Sichuan. Keywords Household latrine · Rural residents’ satisfaction · Factor analysis

5.1 Introduction Latrines are an essential part of every family. A latrine with good hygienic condition and complete facilities plays an important role in improving rural residents’ quality of life and rural sanitation environment, and ensures the health of rural residents. According to “Hygienic specification for rural household latrine,” a sanitary latrine should meet the following conditions: walls, roof, no leakage tanks, air-tight lid, cleanliness, not have maggots, odorless, remove feces in time, and have innocent treatment. Innocuous-sanitary latrines have facilities that could reduce the infectivity of biological pathogenic factors in feces. Innocuous-sanitary latrines include three septic latrines, double urn funnel latrine, three unicom biogas pool latrine, sanitary latrine, double pit latrine and water flushing latrine with complete sewer system and sewage treatment facilities.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_5

111

112

5 Indicators Impacting Rural Residents’ Satisfaction …

We have conducted a survey on rural facilities and found that the construction of rural latrines have a great influence on rural residents’ satisfaction among several rural infrastructures (Ao et al. 2017). On the basis of the previous study, a questionnaire was conducted to study the current construction status and rural residents’ attitudes toward rural household latrines in Sichuan. On the basis of existing studies, the researchers selected evaluation indexes to create a comprehensive questionnaire that consists of individual characteristics of interviewees, characteristics of villages, household latrine building conditions, cost of household latrines, household latrine use condition, and attitudes of rural residents. The reliability of the questionnaire was analyzed to test the validity of the survey. Factor analysis was then conducted to identify 13 factors. Finally, the logistic regression model was used to analyze the 13 factors to test their effects on rural residents’ satisfaction. Sichuan province is located in the hinterland of Southwest China, and its area is 484,000 km2 , with a population of 82 million of which rural residents comprise 42 million. Statistics show that in the first three quarters of 2017, Sichuan realized a gross domestic product (GDP) of 27,297 billion yuan, with an increase of 8.1% compared to the same period in the previous year, with a growth rate of 1.2 percentage points higher than the national average. In the first three quarters, the total investment in fixed assets in Sichuan province was 24,000 billion yuan, with an increase of 10.3% over the previous year, and the investment of the first industry (Agriculture) was 1035.9 billion yuan, with an increase of 22.6% compared to the previous year.

5.2 Literature Review In recent years, many studies related to rural latrines have been done in China, and they can be divided into several aspects, such as the situation, effect, and benefits of rural latrine reform along with existing problems in the process of rural latrine reform. Research about the situation of rural latrine reform could be divided into theoretical analysis and empirical study. Guo et al. (2010) proposed that there were problems as low enthusiasm and low participation of rural residents, insufficient attention by leaders and lack of policy support in the reform of rural latrines in Taizhou in 2003, and put forward corresponding suggestions. Jun (2012) argued that rural latrine reform in China have problems such as the masses’ lacking enough understanding, unstandardized construction operation, lack of technical guidance, and funding problems. Jun (2012) analyzed reasons for the obstruction of rural latrine reform from the perspective of cultural distance. Chun et al. (2013) studied problems in the reform of rural latrine from the perspective of rural civilization and put forward corresponding suggestions. Kui-wei et al. (2000) conducted research on existing papers related to rural latrine reform and proposed that rural latrine reform is helpful for the prevention of intestinal infection, the improvement of rural environmental sanitation and the improvement of rural residents’ health awareness.

5.2 Literature Review

113

In the empirical analysis, Cheng-yun et al. (2005) carried out a background investigation, which is the first national survey to study rural latrines and excreta disposal in China, in 1993. The survey results showed that China’s rural latrine proportion sanitary latrine popularity proportion were 85.9 and 7.5% respectively, and the fecal harmless treatment proportion was 13.5% in 1993. Ke-ning and Weidong (2011) adopted a stratified cluster method to investigate the prevalence of rural sanitary latrines in Shandong province at the end of 1999. He pointed out that there were some problems such as low starting points and low construction levels in rural latrines in Shandong. The reform situation of rural latrines in Cangxi and Mabian were investigated by Li-jian et al. (2007) in 2007, and the results showed that the proportion of privy in rural household latrines reached 86.17% in Cangxi and Mabian. Li (2008a, b) pointed out that the whole level of rural latrine reform of family in Anhui was unideal in 2002, and the economy had a significant impact on latrine reform behavior. Zhang et al. (2005) and Cheng-yun et al. (2013) pointed out that the proportion of popularization of sanitary latrines was 33.46%, and the proportion of decontamination of feces was 10.34% in 2003. They reinvestigated the situation of rural latrine reform of Sichuan in 2011, and the results showed that the proportion of innocuous-sanitary latrine was 48.75%, which has got great improvement compared with the proportion of 2003. Lin et al. (2008) investigated the situation of rural latrine reform during 2004–2005, and the results showed that the proportion of popularization of sanitary latrine and the proportion of decontamination of feces were lower than the national average level. Liang et al. (2002) investigated the situation of rural latrine construction of 5 counties in Xianyang in 2004, and they put forward a suitable sanitary latrine type for Xianyang and pointed out that the proportion of popularization of sanitary latrine and the proportion of decontamination of feces were both low, while villagers lacking hygienic knowledge and enthusiasm. Liu et al. (2016) and Ye and An (2013) studied rural latrine reform results during 2005–2009 of Fujian through questionnaires, field observations, and interviews, then analyzed the correlation between rural latrine reform and economy. Investigation results of Yao et al. (2009) showed that the prevalence proportion of rural sanitary latrines in China was 23.83% in 2006, while the prevalence proportion of sanitary latrines in the eastern region was obviously higher than that in other areas. They also pointed out that the construction of sanitary latrine was positively related to economy. Besides, Zhang (2012) have also respectively studied the situation and effect of rural latrine reform of Liaocheng, Tianjin, and Shijiazhuang. They also have investigated the popularization of sanitary latrines, and analyzed problems and experience in their work. On the basis of researches related to the situation of rural latrine, Shanhong (2002) and Yanfen (2004) studied the effect of rural latrine reform. Shanhong (2002) studied the influence on environmental sanitation brought by latrine reform in Tongbai through the comparison between villages without latrine reform and villages with latrine reform. And they pointed out that the reform of rural latrines could improve the sedimentation proportion of parasitic eggs, reduce the density of flies and reduce the pollution of soil and water resources. Yanfen (2004) surveyed the rural latrine status before and after the implementation of the World Bank lending project for

114

5 Indicators Impacting Rural Residents’ Satisfaction …

rural water supply and sanitation. According to the survey, the sanitary conditions of rural latrines have been significantly improved, and proportion of construction and use of sanitary latrines have also increased significantly through the development of the project. Zhan et al. (2011) proposed that the percent of qualified fecal decontamination was only 45.2%, and the effect of fecal decontamination was significantly influenced by latrine construction quality, condition of fecal exposure, and rural residents’ awareness of excreta disposal knowledge. Researches on the benefits of rural latrine reform are mainly focused on environmental benefits, hygienic benefits, economic benefits and social benefits. Zheng-kui et al. (2005) analyzed economic benefits and social benefits brought by the reform of latrines in rural areas of Hunan. Zhang et al. (2008) analyzed the change of rural residents’ hygienic consciousness and behavior through a contrast between reformed villages and villages without reform. Fu et al. (2006) argued that rural latrines reform could bring hygienic benefits, economic benefits, environmental benefits and social benefits. And they also put forward suggestions such as leadership attention, publicity and education, long-term management from the rural latrines reform condition in 2005. As to international, much attention has also been paid to the construction of rural regional toilets in the underdeveloped areas. Gedefaw et al. (2015) studied the latrine utilization and associated factors among rural communities of Northwest Ethiopia, and they pointed that a construction with proper household based health education, good construction, and supportive supervision would help to realize the sustainability of rural toilets. Hussain et al. (2017) did a qualitative study in rural Bangladesh to study advantages and limitations for users of double pit pour-flush latrines. They pointed out that the double pit pour-flush latrine model is feasible to use and maintain, and the double pit pour-flush latrine increased accessibility of a sanitation facility for low-income residents and provided them privacy, convenience and comfort. In addition to the study of the advantages and disadvantages of rural sanitary toilets and traditional toilets, people have also paid attention to the relationship between rural toilets and health. Hiscox et al. (2016) conducted an analysis of the health problems caused by mosquitoes in the Laos latrine, and they suggested that the seal of septic tanks should be kept through the covering layer to prevent mosquitoes from entering. They point out that this simple intervention will have a global impact on the prevention of mosquito bites and the spread of diseases. Beukes et al. (2017) verified existence of MDR Escherichia coli in pit latrines and they argued the presence of MDR E. coli strains in pit latrine samples demonstrates that pit latrines were potential sources for MDR bacteria. Kumwenda et al. (2017) did an analysis for the differences in the prevalence of parasites from the use of an ecological toilet and a traditional pit type toilet in Malawi. They found that Ascaris lumbricoides was significantly higher in households using EcoSan latrines and they advocate EcoSan users to pay attention to safe ways of handling faecal sludge in order to reduce chances of reinfection from Ascaris lumbricoides. Ben Yishay et al. (2017) found that microfinance could greatly promote the willingness of residents to improve their toilets through a randomizedcontrolled trial of the rural Cambodia. The influence of the toilet on the water quality has attracted the attention of people as the research goes in. Back et al. (2018)

5.2 Literature Review

115

appraised the groundwater risk caused by pit toilet policies in developing countries. Ferrante et al. (2018) pointed out that the quality of the water supply was closely related to the distance from the toilet, and stressed the importance of maintaining a sufficient distance between the access of drinking water and toilets. As to developed countries, there are few researches focus on rural infrastructure. External researches are more about wastewater treatment and environmental sanitation. Such as, K. A. and A. V. (2012) used spatial fuzzy analytic hierarchy process to help with the site selection of wastewater treatment system. Glick et al. (2013) combined LCC, LCA, and economic input–output to assess the sustainability of sewage treatment project. Zhou and Liu (2015) studied infrastructure projects from the perspective of micro engineering, with the aim of creating a basic model to analyze sustainable construction and operation of infrastructure projects, and a sewage treatment plant is used as a case study. Integrated existing research, the reform of rural latrine in China mainly involves such problems as rural residents’ lack of awareness, rural residents’ low participation, insufficient capital input, lag in technology, inadequate government attention and lack of incentive mechanism, and corresponding suggestions such as creasing investment, strengthening propaganda, strengthening technical training and formulating incentive mechanism were put forward. What is worth thinking about is even if there are so many researches related to rural latrines, and problems and suggestions were proposed again and again, problems raised still have not been solved effectively.

5.3 Method and Data Sources 5.3.1 Research Method In this paper, descriptive analysis, factor analysis and logistic regression model were used to analyze the satisfaction of rural residents and its influencing factors.

5.3.1.1

Factor Analysis

Factor analysis is a multivariate statistical analysis method that can convert measured variables to a small number of non-related comprehensive factors. These comprehensive factors reflect the main information of original measured variables, and explain the relationship between measured variables (Tang et al. 2010). Specifically, factor analysis studies the condensation of a large number of measured variables to a few factors with the least information loss (Li 2008a). In this paper, for as many as 44 indicators, the use of factor analysis to convert indicators into comprehensive factors is necessary. The general form of factor analysis model is: Xi = μ + ai1 Fi1 + ai2 Fi2 + · · · + ain Fin + εi (i = 1, 2, . . . , p)

(5.1)

116

5 Indicators Impacting Rural Residents’ Satisfaction …

Among them, Xi is a random observed variable; Fi is a common factor; ai j (i = 1, 2,…, p) is factor load; εi is special factor part not included in common factors.

5.3.1.2

Logistic Regression Model

The dependent variable is the rural residents’ overall satisfaction of rural latrines and it is divided into two categories: “satisfied” and “dissatisfied”. Statistical methods that can be used to handle categorical dependent variables include discriminant analysis, probit analysis, logistic regression model and log-linear model. Logistic regression model is an ideal model for analyzing individual decision behavior and is widely used in the analysis of influencing factors. Logistic regression model is divided into binary logistic regression analysis, which the dependent variables can only be 1 or 0, and multinomial logistic regression analysis where the dependent variables can take more than two values (Peng 2012). In this paper, the dependent variable is divided into two categories, so the binary logistic regression model is adopted. Logistic model can be used to handle both continuous and categorical variables unlike in multiple regression analysis, where the variables must be numeric. Also, the variables need not have normal distribution as required in the case of discriminate analysis (CG and CJ 2003). The probability of occurrence for specimen is P(y = 1|xi ) = pi , and two probability incidents, occurrence and nonoccurrence are recorded as the following two formulas: pi =

1

m

m

1 + e−(α+

i=1

βi xi )

eα+ i=1 βi xi m = 1 + eα+ i=1 βi xi

(5.2)

m

eα+ i=1 βi xi 1 m m = 1 − pi = 1 − α+ β x α+ i i i=1 i=1 βi xi 1+e 1+e

(5.3)

pi represents the probability of occurrence of events in observation; 1−pi represents the probability of nonoccurrence of events in observation; both are nonlinear functions formed only by variable xi . pi /(1 − pi ) represents the ratio of the probability occurrence and the nonoccurrence of the event, named occurrence ratio. The logarithmic transformation to the occurrence ratio produces the linear model of the logistic regression model.  pi )=α+ βi xi 1 − pi i=1 m

ln(

(5.4)

5.3 Method and Data Sources

117

5.3.2 Data Sources This paper is based on the Natural Science Foundation of Sichuan Education Department Fund Project “research on system optimization of agricultural production infrastructure construction”, and data were obtained under the organization of home research with the help of students from the Engineering Management Department of Chengdu University of Technology. To ensure the validity and authenticity of the data, and to guarantee that interviewees can understand the questions well, the research group regarded students from rural area as pre-interviewers. A meeting was held to investigate students’ opinions of rural infrastructure and guarantee the validity of questionnaires finished by students. A total of 300 questionnaires were issued, 153 valid questionnaires were recovered, with an effective recovery rate of 51%. A total of 69 women and 84 men were interviewed, accounting for 45.1 and 54.9%, respectively. The basic situation of all investigated objects is shown in Table 5.1. Table 5.1 Basic information of interviewees Index Age

Annual income

Proportion (%)

Index

Less than 20 years

14

Gender

From 20 to 30 years

49.00

From 30 to 40 years

19.60

From40 to 50 years

11.10

Junior high school

17.70

From50 to 60 years

4.60

High school

12.50

Over 60 years

1.30

Bachelor

62.50

Less than 50,000 Yuan

52.90

Master

3.30

From 50,000 to 100,000 Yuan

30.70

Doctor

.70

From 100,000 to 200,000 Yuan

14.40

Over 200,000 Yuan

2.00

Family to education

Proportion (%) Female

45.10

Male

54.90

Primary school

3.30

118

5 Indicators Impacting Rural Residents’ Satisfaction …

5.4 Empirical Study 5.4.1 Evaluation Index Selection and Variable Definition 5.4.1.1

Evaluation Index Selection

On the basis of existing research, we selected evaluation indices related to several aspects, such as interviewees’ individual characteristics indices, village type, construction condition of rural latrine, cost of rural latrine, condition of use and rural residents’ attitudes. Finally, 44 indices were selected. The specific content of the evaluation index are shown in Table 5.2. Individual characteristic indexes of rural residents, such as Gender, age, family top education and annual income, are important parts of the questionnaire survey, which could help us to understand the individual differences, economic conditions of rural households, and views and needs of rural residents in rural Sichuan. Village type and village housing distribution are village characteristic indexes, which show construction and planning differences in rural household toilets among villages with different characteristics. Basic condition and construction condition indexes, such as whether there is a household latrine or not, household latrine location, household latrine type, household latrine use time, household latrine area, whether the household latrine has walls or not, whether the household latrine has a roof or not, whether the household latrine has a door or not, ventilation in the household latrine, whether the household latrine is 10 cm higher than terrace, are necessary information for this investigation. They could help to understand the basic condition and construction condition of household latrines in rural area. Indexes such as whether there is a closet or not, whether there is a decontamination of feces or not, whether there are sanitary fixture or not, whether there is a fly prevention facility or not, sanitary condition of household latrine, frequency of feces cleaning, whether the tank is airtight without leakage or not, odor concentration in household latrine, whether the household latrine is reformed or not, frequency of failure affecting use and failure recovery time could reflect the health allocation, sanitary condition and the use of rural latrines. Indexes related to building/renovation pattern and founding, such as cost of household latrine, households latrine building/renovation pattern, household latrine building/renovation participants, funding sources of household latrine building/renovation, payment time of subsidy fund and subsidy percentage, could reflect the role of the government in the rural latrines construction. Indexes related to village affairs, village planning and publicity of latrines, such as transparency of village affairs and whether there are any publicity and training in the construction, renovation, use and management of rural latrine in the village or not, could reflect the role of village government in rural latrine construction. Indexes related to feelings, attitudes and thoughts of rural residents, such as financial burden caused by household latrine construction, influence on surrounding sanitation of household latrine and satisfaction with the overall status of the household

5.4 Empirical Study

119

Table 5.2 Variable definition Index

Variable definition

Village type (X1 )

1 = ordinary village, 3.67 = township resident, 6.33 = combination of urban and rural areas, 9 = both township resident and a combination of urban and rural areas

Gender (X2 )

1 = Female, 9 = Male

Age (X3 )

1 = over 60 years, 2.6 = from 50 to 60 years, 4.2 = from 40 to 50 years, 5.8 = from 30 to 40 years, 7.4 = from 20 to 30 years, 9 = less than 20 years

Family top education (X4 )

1 = primary school, 2.6 = junior high school, 4.2 = high school, 5.8 = Bachelor, 7.4 = Master, 9 = Doctor

Annual income (X5 )

1 = less than 50,000 yuan, 3.67 = from 50,000 to 100,000 yuan, 6.33 = from 100,000 to 200,000 yuan, 9 = over 200,000 yuan

Village housing distribution (X6 )

1 = very decentralized, 3 = decentralized, 5 = moderate, 7 = centralized, 9 = very centralized

Whether there is a household latrine or not (X7 )

1 = no, 9 = yes

Household latrine location (X8 )

1 = outside the yard, 5 = in the yard, 9 = indoor

Household latrine type (X9 )

1 = others, 2 = pit type, 3 = loop type, 4 = fecaluria diversity, 5 = triplex methane tank, 6 = double urn funnel, 7 = three compartment septic tank, 8 = double pit alternation, 9 = water jet

Household latrine use time (X10 )

1 = 10 years or more, 3.67 = 5–10 years, 6.33 = 2–5 years, 9 = 2 years or less

Household latrine area (X11 )

1 = smaller than 1.2 m2 , 2.6 = 1.2–1.4 m2 , 4.2 = 1.4–2.25 m2 , 5.8 = 2.25–5 m2 , 7.4 = 5–10 m2 , 9 = larger than 10 m2

Whether the household latrine has walls or not 1 = no, 9 = yes (X12 ) Whether the household latrine has a roof or not (X13 )

1 = no, 9 = yes

Whether the household latrine has a door or not (X14 )

1 = no, 9 = yes

Ventilation in the household latrine (X15 )

1 = no ventilation, 5 = natural ventilation, 9 = mechanical ventilation

Whether the household latrine is 10 cm higher 1 = no, 9 = yes than terrace (X16 ) Whether there is a closet or not (X17 )

1 = no, 9 = yes (continued)

120

5 Indicators Impacting Rural Residents’ Satisfaction …

Table 5.2 (continued) Index

Variable definition

Whether there is a decontamination of feces or 1 = no, 9 = yes not (X18 ) Whether there are sanitary fixture or not (X19 ) 1 = no, 5 = having a part, 9 = all have Whether there is a fly prevention facility or not 1 = no, 9 = yes (X20 ) Sanitary condition of household latrine (X21 )

1 = bad, 5 = moderate, 9 = good

Frequency of feces cleaning (X22 )

1 = over 2 years, 3.67 = 1–2 years, 6.33 = 6 months to a year, 9 = within 6 months

Whether the tank is airtight without leakage or 1 = no, 9 = yes not (X23 ) Odor concentration in household latrine (X24 )

1 = obvious odor, 5 = a little odor, 9 = basically tasteless

Whether the household latrine is reformed or not (X25 )

1 = no, 9 = yes

Cost of household latrine (X26 )

1 = under 500 yuan, 3.67 = 500–1000 yuan, 6.33 = 1000–2000 yuan, 9 = over 2000 yuan

Household latrine renovation reason (X27 )

1 = no renovation, 3.67 = people around have reformed their latrines, 6.33 = under government advocacy, 9 = not satisfied with the previous situation

Households latrine building/renovation pattern 1 = villagers are responsible, 3.67 = (X28 ) government advocacy organizations, villagers funded, 6.33 = government guidance, and provide some funds, 9 = government responsible for the construction, and provide subsidies Household latrine building/renovation participants (X29 )

1 = rural residents, 2.33 = government, 3.67 = professional construction company; 5 = rural residents and government, 6.33 = rural residents and professional construction company, 7.67 = government and professional construction company, 9 = rural residents, government and professional construction company

Management and maintenance participants of household latrine (X30 )

1 = no management or maintenance, 2.14 = rural residents, 3.29 = government, 4.43 = professional management company, 5.57 = rural residents and government, 6.71 = rural residents and professional management company, 7.86 = government and professional management company, 9 = rural residents, government and professional management company (continued)

5.4 Empirical Study

121

Table 5.2 (continued) Index

Variable definition

Funding sources of household latrine building/renovation (X31 )

1 = rural residents, 2.33 = government, 3.67 = social capital; 5 = rural residents and government, 6.33 = rural residents and social capital, 7.67 = government and social capital, 9 = rural residents, government and social capital

Payment time of subsidy fund (X32 )

1 = no subsidy, 3 = 6 months later after the construction, 5 = 3–6 months later after the construction, 7 = within 3 months, 9 = before the construction

Subsidy percentage (X33 )

1 = no subsidy, 3 = under 20%, 5 = 20–50%, 7 = 50–80%, 9 = more than 80%

Financial burden caused by household latrine construction (X34 )

1 = heavy, 5 = moderate, 9 = basically no burden

Frequency of failure affecting use (X35 )

1 = under 3 months, 3.67 = 3–6 months, 6.33 = 6–12 months, 9 = over 12 months

Repair approach (X36 )

1 = rural residents, 5 = there is a special person in the village for maintenance, 9 = looking for professional maintenance personnel

Failure recovery time (X37 )

1 = over 36 h, 3.67 = 24–36 h, 6.33 = 12–24 h, 9 = within 12 h

Whether the construction/renovation of the household latrine is worthwhile or not (X38 )

1 = no, 9 = yes

Transparency of village affairs (X39 )

1 = opaque, villagers do not know many cases, 5 = moderate, inform part of the work arrangements and information; 9 = transparent, inform related arrangements and information

Are there any publicity and training in the 1 = no, 9 = yes construction, renovation, use and management of rural latrine in the village? (X40 ) Does the village have a unified plan for the household latrine? (X41 )

1 = no, 9 = yes

Influence on surrounding sanitation of household latrine (X42 )

1 = no influence, 5 = have some influence, 9 = significant influence

Influence on health of family and villagers of household latrine (X43 )

1 = no influence, 5 = have some influence, 9 = significant influence

Satisfaction with the overall status of the household latrine (X44 )

0 = dissatisfied, 1 = satisfied

latrine, could directly reflect rural residents’ attitudes and ideas, and also could reflect their cognition of rural household latrine.

122

5.4.1.2

5 Indicators Impacting Rural Residents’ Satisfaction …

Variable Definition

The Likert scale was used for the questionnaire analysis. As the number of options in this study involves several indexes and the options among these indexes are quite different, among these indexes are quite different, involves several indexes, option number of which is not 5, to ensure parameters of all variables are controlled between 1 and 9 so that we could compare the results of different indicators more intuitively. Meanwhile, parameters, which have a degree of good and bad, are arranged in order from bad to good. As “Satisfaction with the overall status of the household latrine” is only used in logistic regression analysis, so the variable definition is: 0 = dissatisfied, 1 = satisfied. Variables are shown in Table 5.2.

5.4.2 Descriptive Analysis of Rural Household Latrines From the survey, 79 household latrines are located indoor, 50 household latrines are in yard, and 24 household latrines are outside the yard, and the percentage points are 51.6, 32.7 and 15.7% respectively. There are 32 pit latrines, account for 20.9%, 14 loop type latrines, account for 9.2%, 6 fecaluria diversity latrines, account for 3.9%, 10 triplex methane tank latrines, account for 6.5%, 2 double urn funnel latrines, account for 2.0%, 6 three compartment septic tank latrines, account for 3.9%, 15 double pit alternation latrines, account for 9.8%, and 64 water jet latrines, account for 41.8%. However the percentage of decontamination of feces is only 39.2%, which shows that there are still a large number of water flushing latrines which do not meet requirements of innocuous-sanitary latrine. Distribution of rural household toilets type is shown in Fig. 5.1. In order to get a better understanding of the situation in China’s rural toilets, a few photos collected during the investigation are attached in Fig. 5.2 (the first two are water jet and the third is pit type). With regard to the construction condition, 94.1% of rural household latrines have walls; only 5.9% of household latrines do not have walls. 94.8% of rural household latrines have a roof, and 5.2% of rural household latrines do not have a roof. 86.3% of rural household latrines have a door, while 13.7% of rural household latrines do not have a door. 39.9% of tanks are not airtight, and only 58.2% of the tanks are airtight. With regard to the sanitary facilities of rural household latrines, 62.1% of rural household latrines have no toilets, and the missing rate is .7%. 18.3% of the household toilets do not have sanitary facilities such as storage bucket, special cleaning tools, paper containers, 51.6% of household latrines have parts of sanitary facilities, and 30.1% of household latrines do not have sanitary facilities. And only 19.6% of rural household latrines have fly prevention facilities. Basic information of rural latrines is shown Table 5.3. In the process of use, 36.6% of latrines’ hygienic condition is good, 49.0% of latrines’ hygienic condition is moderate, and 14.4% of latrines’ hygienic condition is bad. Only 23.5% of the household toilets are basically tasteless, and 24.8% of the

5.4 Empirical Study 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

123 41.80%

20.90% 9.20% 2.00%

3.90%

9.80%

6.50% 2.00%

3.90%

Fig. 5.1 Distribution of rural household toilets type

Fig. 5.2 Pictures of rural latrines

household latrines have obvious odor. What’s more, 16.3% of the failure frequency is under 3 months, 15.7% of the failure frequency is between 3–6 months, 15.7% of the failure frequency is between 6 and 12 months, 45.1% of the failure frequency is more than 12 months, and the missing percentage is 7.2%. 19.0% of the household latrine cost is under 500 yuan, 27.5% of the household latrine cost is between 500 and 1000 yuan, 19.6% of the household latrine cost is between 1000 and 2000 yuan, 18.3% of the household latrine cost is over 2000 yuan, and the missing percentage is 15.7%. 80.4% of the funding resource is rural residents, 5.2% of the funding resource is government, .7% of the funding resource is social capital, 6.5% of the funding resource is rural residents and government, .7% of the funding resource is rural residents, government and social capital, and the missing percentage is 6.5%. 18.3% of the construction of household latrines cause heavy economy burden to rural residents, 46.4% of the construction of household latrines cause moderate economy burden to rural residents, 26.8% of the construction of

0 (0%)

153

Missing

Total

14 (94.1%)

153

0 (0%)

8 (5.2%)

145 (94.8%)

Number

Number

9 (5.9%)

Incompletely equipped

Fully equipped

Roof

Wall

Unequipped

Equipped

Sanitation facilities allocation

Table 5.3 Basic information of rural latrines

153

0 (0%)

21 (13.7%)

132 (86.3%)

Number

Door

153

1 (.7%)

95 (62.1%)

57 (37.3%)

Number

Closet

153

0 (0%)

28 (18.3%)

79 (51.6%)

46 (30.1%)

Number

Sanitary fixture

153

0 (0%)

123 (80.4%)

30 (19.6%)

Number

Fly prevention facility

153

1 (.7%)

92 (60.1%)

60 (39.2%)

Number

Decontamination of feces

124 5 Indicators Impacting Rural Residents’ Satisfaction …

5.4 Empirical Study

125

household latrines cause little economy burden to rural residents, and the missing percentage is 8.5%. As into village publicity and training, only 20.9% of interviewees chose there are latrine related publicity and training in their villages, 74.5% of interviewees chose there are no latrine related publicity and training in their villages, and the missing percentage is 4.6%. Only 24.2% of interviewees chose there are unified plan for the household latrine in their villages, 71.2% of interviewees chose there are no unified plan for the household latrine in their villages, and the missing percentage is 4.6%. In the aspect of farmers’ satisfaction, 72.5% of the rural residents are satisfied with the overall status of household latrines, and 27.5% of them were dissatisfied with the overall status of household latrines. 17.6% of rural residents believe that household toilet hygiene has no effect on the surrounding environmental hygiene, 46.4% of rural residents believe that household toilet hygiene has a certain effect on the surrounding environmental hygiene, 32.7% of rural residents believe that household toilet hygiene has great effect on the surrounding environmental hygiene, and the missing percentage is 3.3%. 15.0% of rural residents believe that household toilet hygiene has no effect on the health of family and villagers, environmental hygiene, 43.8% of rural residents believe that household toilet hygiene has a certain effect on the health of family and villagers, environmental hygiene, 37.3% of rural residents believe that household toilet hygiene has great effect on the health of family and villagers, environmental hygiene, and the missing percentage is 3.9%. From the descriptive analysis we can see that the current rural latrine construction situation of Sichuan is not ideal. And there are many problems, such as low percentage of feces decontamination, low percentage of sanitation facilities, bad condition of sanitary and odor, and absence of village committee function. But, it is worth noting that even though the status quo of the construction of rural household toilets is not good, the satisfaction percentage of rural residents with household toilets has reached 72.5%.

5.4.3 Factor Analysis 5.4.3.1

Missing Value Analysis

SPSS software is used to analyze the data, and the missing value analysis is carried out first. The index of cost of household latrine has the biggest missing rate, which is 15.7%, in addition, missing rates of subsidy percentage and village latrine related announcements and transparency are also reach 10.5 and 13.7% respectively. Meanwhile, the MCAR test significance (P) value of little was .03, significantly less than .05, so the hypothesis that the missing value is MACR is rejected. Therefore, EM method is used to run missing values. To ensure validity of the questionnaire after it was run, the validity and construct validity of the questionnaire were tested by the Cronbach’s coefficient. The coefficient of the questionnaire data was .812, indicating that the questionnaire is reliable.

126

5.4.3.2

5 Indicators Impacting Rural Residents’ Satisfaction …

Factor Analysis

The original data are normalized to eliminate the difference in magnitude and dimension. Factor analysis showed that the KMO statistic was .657, and the P value was 0, hence, the test results were significant. These data indicated that questionnaire data had a certain correlation, indicating it is suitable for factor analysis. Factor analysis showed that the extractions of 44 indicators are all above .5, demonstrating that most of the information in the representation variable is extracted by factors. Using the principal component method to extract 15 factors, the researchers determined that their feature values are greater than 1. The component matrix is rotated using the method of maximum variance orthogonal rotation because the initial loading structure is not clear. Moreover, loads of the indexes that whether the household latrine is 10 cm higher than terrace (.354) and whether the household latrine is reformed or not (.377) are less than .4 on their common factors, and factor analysis is carried out again after deleting these two variables. Loads of the renovation reason (.384) is less than .4 on the common factor after the deletion of aforementioned indexes, so factor analysis is carried out again after the renovation reason is deleted. Similarly, in subsequent analysis, whether the household latrine is reformed or not, ventilation in the household latrine and repair approach, load of which are .361, .304 and .383 respectively, are deleted in proper order. Finally, indexes that whether the household latrine is 10 cm higher than terrace, whether the household latrine is reformed or not, renovation reason, whether the household latrine is reformed or not, ventilation in the household latrine and repair approach are deleted. Finally we got 13 factors, and loads of indexes on their common factors are all above .4 after the component matrix is rotated, whereas the total variance explained is 69.4%. The rotated component matrix is shown in Table 5.4. We can get the distribution of factor indices from Table 5.4. And the index situation and factor naming are shown in Table 5.5.

5.4.4 Logistic Regression Analysis Based on Factor Analysis The dependent variable of this paper is the total satisfaction of rural households with household latrines, which is divided into two categories: “satisfied” and “dissatisfied”. Logistic regression analysis was used to analyze 13 factors. The results of the omnibus test are shown in Table 5.6, and the results of Hosmer and Lemeshow test are shown in Table 5.7. Table 5.7 shows that the χ 2 is 4.628, indicating that there is no significant difference between the actual value of the variable and the predicted value. And the significance (P) value is .796, showing that the assumption of the model fits the data well cannot be rejected, which means that the model fits the data well. The model has a −2Ln likelihood of 121.476, while the observation correct percentage of the model reached 77.8%. Wang (2008) has summed up the domestic and foreign literature that correctly classify percentages are among 54–90%. And correctly classify percentages in papers of Tang et al. (2010), Wang (2008) and

.066

.130

.220

.555

.780

.583

.145

.055

.048

Village housing distribution (X6 )

Whether there is a household latrine or not (X7 )

Household latrine location (X8 )

Household latrine type (X9 )

Household latrine use time (X10 )

Household latrine area (X11 )

Whether the household latrine has walls or not (X12 )

Whether the household latrine has a roof or not (X13 )

.759

−.183

.112

.017

−.022 .791

.031

−.030

−.326

.010

.124

.072

−.017

−.041

.142

.177

.147

.047

−.084

−.095 .443

.104

−.025

.130

−.100 .048

−.004

.105

.163

.050

.085

−.011

Annual income (X5 )

Family top education (X4 )

−.028

.122

−.019

−.183

.135 −.267

.085

4

3

Age (X3 )

2

Gender (X2 )

1

Component

Table 5.4 Rotated component matrix

.134

.084

−.068

.066

−.027

.073

.336

.102

.134

.014

−.105

−.084

5

−.165

.088

.004

.003

.024

.126

.045

.040

.113

.069

.115

−.030

.557

.073

−.084

.011

−.121

−.008

−.139

.027

−.153

.137

.276

.829

−.080 .113

−.022

.020 .564

.062

9

.013

.126

.526

.195

−.355

8

−.029

.165

−.024

−.059

.342

−.206

.061

−.054

−.166

−.073

7

.050

.045

−.111

6

−.058

.034

.084

.434

.056

−.371

.162

.426

−.094

−.075

.189

.093

10

.045

.119

.166

.019

−.057

.073

−.001

.273

−.176

.019

.043

.122

11

−.047

.127

−.281

.049

−.014

−.273

.521

−.082

.270

.023

−.090

−.095

12

(continued)

−.008

−.110

−.057

.106

−.031

−.034

−.052

−.572

.298

−.022

.003

.797

13

5.4 Empirical Study 127

.341

.664

.187

.720

.434

.772

.076

.211

.756

Whether there is a closet or not (X17 )

Whether there is a decontamination of feces or not (X18 )

Whether there are sanitary fixture or not (X19 )

Whether there is a fly prevention facility or not (X20 )

Sanitary condition of household latrine (X21 )

Frequency of feces cleaning (X22 )

Whether the tank is airtight without leakage or not (X23 )

Odor concentration in household latrine (X24 )

1

Component

Whether the household latrine has a door or not (X14 )

Table 5.4 (continued)

.198

−.052 .060

−.076

−.068

.027 .280

−.077 .038 −.090

−.048

.076

−.139 −.078

−.013 −.169

.060

.048

.145

−.060

−.026

.026

−.037

.102

.592

−.064

−.209

4

3

2

.157

.194

−.206

−.079

−.200

−.086

−.246

−.097

−.244

5

-.035

.001

−.038

.065

−.165

.128

.078

.209

−.118

6

.093

.721

.097

.263

.314

.248

.679

−.074

.056

7

.054

.137

.126

.108

−.013

−.071

.003

−.216

−.007

8

.086

.222

.121

.089

.044

.104

-.057

−.102

−.053

9

.078

.035

−.016

−.072

.037

.163

.182

−.048

−.120

10

.068

.088

.108

.175

−.003

−.026

−.049

−.177

.041

11

.044

.098

.784

.016

.085

.035

−.041

.158

−.248

12

(continued)

-.039

.029

−.041

−.128

.070

.123

−.125

.064

.168

13

128 5 Indicators Impacting Rural Residents’ Satisfaction …

.716 .902 −.147

.123

.052

−.066

.017

Management and maintenance participants of household latrine (X30 )

Funding sources of household latrine building / renovation (X31 )

Payment time of subsidy fund (X32 )

Subsidy percentage (X33 )

Financial burden caused by −.074 household latrine construction (X34 ) .010 −.421

Failure recovery time (X37 ) −.002

.155

.551

.024

Frequency of failure affecting use (X35 )

.806

.105

Household latrine building / renovation participants (X29 )

.744

-.073

.182

Households latrine building / renovation pattern (X28 )

2

.201

1

Component

Cost of household latrine (X26)

Table 5.4 (continued)

−.220

−.038

.001

−.166 .276

.726

.729

−.111

−.065 .154

−.163

.041

−.051

.032

.073

.230

-.013

-.129

5

-.054

.083

-.075

.129

4

.042

.046

−.027

.014

.007

-.019

-.120

-.055

3

.206

−.178

−.046

−.040

.005

.099

.087

.269

.313

.146

6

−.048

−.002

.017

−.036

.147

−.091

.226

.040

.120

-.056

7

.094

.733

−.087

−.104

−.009

.008

.033

−.040

-.083

-.033

8 .037

−.043

−.055

.030

−.087

−.194

.201

.004

.338

-.060

9

.181

−.149

.207

−.015

.177

.114

.693

.157

-.024

-.040

10

.425

.157

−.045

.025

−.127

.162

-.103

-.118

-.027

.063

11

.059

−.114

−.020

−.013

−.162

−.048

.013

-.170

0.035

-.008

12

(continued)

.299

−.160

.027

.010

.105

.068

-.010

-.105

-.020

-.036

13

5.4 Empirical Study 129

1

Component

.242

−.065

.020

.181

−.010

−.017

Does the village have a unified plan for the household latrine? (X41 )

Influence on surrounding sanitation of household latrine (X42 )

Influence on health of family and villagers of household latrine (X43 )

.142

.160

Are there any publicity and training in the construction, renovation, use and management of rural latrine in the village? (X40 )

.175

.423

−.037

2

transparency of village affairs (X39 )

whether the construction / −.049 renovation of the household latrine is worthwhile or not (X38 )

Table 5.4 (continued)

.905

.871

.184

.208

.235

.152

3

−.172

−.065

−.078

−.077

−.011

−.077

−.076

−.012

.013

.040

5

.026

.176

4

.097

.214

.696

.817

.231

−.027

6

−.012

−.118

.132

−.051

.079

.027

7

.133

.024

.045

.098

−.008

−.012

8

.015

.082

.089

.007

.023

−.054

9

−.033

.038

.199

−.093

.144

−.157

10

.018

.160

.004

.064

.459

.808

11

.038

−.033

.095

−.084

−.081

−.006

−.027

.169

−.029 −.102

−.041

13

.093

12

130 5 Indicators Impacting Rural Residents’ Satisfaction …

5.4 Empirical Study

131

Table 5.5 Indicators in each factor Factor

Index

F1 (Basic conditions of rural latrines)

X8 , X9 , X10 , X17 , X19 , X20 , X21 , X24

F2 (Fund & construction participants and pattern)

X28 , X29 , X31 , X32 , X33

F3 (Rural residents’ cognition)

X42 , X43

F4 (Building condition)

X12 , X13 , X14

F5 (Economic burden & failure frequency)

X34 , X35

F6 (Village committee performance)

X40 , X41

F7 (Decontamination of feces & Tank airtightness)

X18 , X23

F8 (Annual income & cost & area)

X5 , X11 , X26

F9 (Age & family top education)

X3 , X4

F10 (Management maintenance participants)

X30

F11 (Recovery time & worth & transparency of village affairs)

X37 , X38 , X39

F12 (whether there is household latrine & frequency of sludge cleaning)

X7 , X22

F13 (Gender & village housing distribution)

X2 , X6

Table 5.6 Omnibus test of model coefficients Step 1

χ2

df

Sig

Step

58.358

13

.000

Block

58.358

13

.000

Model

58.358

13

.000

Table 5.7 Hosmer and Lemeshow test

Step

χ2

df

Sig

1

4.628

8

.796

Langer et al. (2018) are between 66.8 and 83.7%. Hence the correct classification rate in this paper is within the acceptable range. Estimation results of the model are shown in Table 5.8. The Wald test is used to test the significance of Logistic regression, and it is a common test method of statistical software to test significance of Logistic regression. Under the zero hypothesis, each regression coefficient is equal to 0, then the univariate Wald statistic is a gradual χ2 distribution with the degree of freedom equal to 1. Therefore, the significance of corresponding model independent variables can be determined according to whether the value of the Wald statistic is greater than the critical value of the χ2 under the degree of freedom equal to 1.

132 Table 5.8 Model estimation results

5 Indicators Impacting Rural Residents’ Satisfaction … Variable

Parameter

Wald

Ln(F1 )

1.237***

17.915

Ln(F2 )

1.078***

6.928

Ln(F3 )

−.109

.208

Ln(F4 )

.369

2.284

Ln(F5 )

.410*

3.094

Ln(F6 )

.993***

Ln(F7 )

.044

.037

Ln(F8 )

.203

.643

Ln(F9 )

−.383

2.133

10.296

Ln(F10 )

.592**

4.495

Ln(F11 )

.720***

9.037

Ln(F12 )

.050

.050

Ln(F13 )

.268

1.388

constant

1.667

28.034

Note *, **, *** respectively represent the significant level of 10%, 5% and 1%

5.4.5 Discussion According to the Wald value in Table 5.8, factors with significant effect on the satisfaction of farmers include F1 , F6 , F11 , F2 , F10 , and F5 , and their effect size is decreased in turn. F1 represents the basic conditions of rural latrines, which reflects mainly the location, type, facilities and sanitation condition of rural household latrines, with a significant level of 1%. It has a very significant positive effect on rural residents’ satisfaction in all factors. Specifically, the better the basic conditions of rural latrines, the higher the total satisfaction. F6 represents the village committee performance factor, which includes village publicity and training and village unified plan, with a significance level of 1%. It has a positive effect on rural residents’ satisfaction, meaning the more publicity and training for rural residents’ and better unified plan in village, the more satisfied rural residents are with household latrines. F11 represents recovery time & worth & transparency of village affairs. It has a significance of 1%, and has a positive effect on rural infrastructure satisfaction. Household toilets failure recovery faster, the villagers think that it is more worthy of household toilet building cost. And more transparent that village work is, rural residents are more satisfied. F2 represents fund & construction participants and pattern, which has a significance of 1%. It includes building/renovation pattern, building/renovation participants, funding sources, subsidies etc. F2 has a positive effect on rural infrastructure

5.4 Empirical Study

133

satisfaction, indicating that the more government guides, the more outside help, the more funds and subsidy funds, the faster the subsidy funds arrive, and the more satisfied rural residents are. F10 represents management maintenance participants, which has a significance of 5% and has a certain positive effect on rural residents’ satisfaction. It shows that the rural household toilets get more professional maintenance in the process of using, and the more satisfied rural residents are. F5 represents economic burden and failure frequency, which has a significance of 5% and has a certain positive effect on rural residents’ satisfaction. The smaller the economic burden of construction and reconstruction is, the lower the frequency of rural household toilet failure, and the more satisfied rural households are. The top four factors that affect farmers’ satisfaction are “basic conditions of rural latrines”, “village committee performance”, “recovery time &worth & transparency of village affairs” and “fund & construction participants”. The basic condition of rural latrines influences rural residents’ satisfaction most significantly. It shows that convenience, comfort and hygiene could affect users’ feeling directly in the use of toilets. Tang et al. (2003) pointed that there were 40.7% of household latrines indoor, 14.1% of household latrines had closet, and 38.9% of household latrines have bad smell. In this survey, 51.6% of the household latrines are located indoor, 30% of the latrines are pit or loop style, 30.1% of household latrines have equipped facilities, 24.8% of latrines have bad smell. It shows that the situation of rural latrines in Sichuan has got better, but it is still not optimistic. It still needs to increase the intensity of latrine construction to improve the basic conditions of rural household latrines. Guoping (2016) summarized the successful experience of Shijiazhuang as leaders’ serious, well publicity, good technical guide, fund guarantee etc. Zhibang (2017) also pointed out that insufficient attention of leadership is one of the reasons for the slow progress of rural toilet reform in China, and they put forward suggestions such as strengthen health publicity and education. Only 20.9% of interviewees chose the option that their villages have latrines related training and publicity, and only 25.3% of interviewees chose the option that their villages have unified plan for their household latrines. On the other hand, the significant influence of village committees’ performance on rural residents’ satisfaction also shows rural residents’ expectation for government guidance. The survey results show that 55.6% of rural latrines’ failure could get recovered within 12 h, and the failure recovery time of 19.3% household toilet is more than 24 h. A long failure recovery time would seriously affect the normal use of latrines in rural areas and affect the use feeling of rural households. 87.6% of rural residents think that the construction and reform of the household latrine is worth, indicating that they have realized the importance of rural household latrine. Only 24.2% of the rural residents think that the related affairs of the household toilet construction in the village were informed well. The higher transparency of village affairs can help rural residents understand work arrangements and policy conditions of the government, and help them make full use of social and government resources in household latrine construction work. Therefore to improve household latrine construction quality and cost savings.

134

5 Indicators Impacting Rural Residents’ Satisfaction …

The funds of rural household toilet reform directly affect the construction quality of rural household latrine. Multiple sources of funding can ensure rural household latrine construction work go on wheels, and it could help to improve the level and quality of the rural household latrines construction. It will increase the economic burden of village committee, and may even cause a stoppage during the latrine reform (Hang 2008). In view of the fact that many village economies are relatively weak and rural household latrines work is an important part of building a new socialist countryside, all levels of financial subsidies should increase investment to help the rural household latrines reform. On the other hand, the collaboration between government and social enterprises not only could arouse the enthusiasm of rural residents, but also provide professional technical guidance for the construction and renovation of rural household latrines. Biao (2008), Hou (2011) and Zaijin (2008) has pointed out that there are such problems as nonstandard operation of construction team, lack of technical guidance, and the untimely payment of subsidy funds in the reform of rural household toilets in China. And they also proposed suggestions to strengthen government attention, strengthening technical guidance and timely subsidy funds. Problems in above studies coincide with this article. It is worth noting that, although the survey results show that the construction of rural household toilets is not good and there are still many points that need to be improved, the satisfaction rate of rural residents is respectively high. This shows rural residents do not have enough understanding for rural household latrines, and they require few for household toilets. The government still needs more publicity to help rural residents understand the importance of sanitary latrine and its impact on people’s health, therefore to stimulate rural residents’ enthusiasm for building sanitary latrines.

5.5 Conclusion This paper conducted an investigation of the condition of rural household latrines from the perspective of farmers’ satisfaction. Factor analysis and logistic regression model were used to analyze rural household latrines’ situation and rural residents’ attitudes. The condition of rural household latrines and rural residents’ attitudes are analyzed from aspects of building condition, cost, using condition and rural residents’ cognition. The results show that the rural residents’ satisfaction rate of rural household latrines is 72.5%, and their demand for rural household toilet is basically satisfied. On the other hand, the construction of rural household toilet is not ideal. At the same time, the analysis results show that rural residents’ satisfaction of rural household latrine is mainly affected by “basic situation of latrines”, “village committee performance”, “recovery time & worth & transparency of village affairs” and “fund & construction participants”. From the results, we can see that problems still exist in the construction of rural household toilets, such as the unsatisfactory status of household toilet construction, absence of government functions and insufficient understanding of rural residents.

5.5 Conclusion

135

In order to better promote the construction of rural household latrine in Sichuan, the government should give full play to the functions, increase publicity, make good planning, enhance the farmers’ understanding of rural household toilets, therefore, it is to change the rural household latrine construction situation, and to create a better environment for the health of rural residents. This paper studies the present situation of the construction of rural household latrines in Sichuan, and discusses rural residents’ attitudes toward rural household latrines. It lays a foundation for further research on development of rural household latrines construction, and provides a theoretical basis for policy makers. Indexes in this research are derived from previous studies, most of which are aimed at other provinces. Therefore, these indexes may not fully reflect the rural household latrines situation of Sichuan. It is hoped that indexes could get further adjustment after this research, so that a comprehensive index system could be established to study the current situation of rural household latrines construction in Sichuan. At the same time, this paper uses questionnaire survey, which accurately capture rural residents’ views. Thus, the semi structured method could be used in investigations aimed at obtaining a more comprehensive understanding of rural residents’ attitude to rural household latrine construction.

References Ao, Y., Li, J., Wang, Y., Liu, C., & Xu, S. 2017. Rural residents’ satisfaction of rural facilities and its influencing indicatiors: A case study of Sichuan, China. Mathematical Problems in Engineering. Back, J. O., Rivett, M. O., Hinz, L. B., Mackay, N., Wanangwa, G. J., Phiri, O. L., Songola, C. E., Thomas, M. A. S., Kumwenda, S., Nhlema, M., Miller, A. V. M., & Kalin, R. M. 2018. Risk assessment to groundwater of pit latrine rural sanitation policy in developing country settings. Science of the Total Environment, 613–614: 592–610. Ben Yishay, A., Andrew Fraker, Raymond Guiteras, Giordano Palloni, Neil Buddy Shah, Stuart Shirrell, & Paul Wang. 2017. Microcredit and willingness to pay for environmental quality: Evidence from a randomized-controlled trial of finance for sanitation in rural Cambodia. Journal of Environmental Economicsand Management, 86: 121–140. Beukes, L. S., King, T. L. B., & Schmidt, S. 2017. Assessment of pit latrines in a peri-urban community in Kwazulu-Natal (South Africa) as a source of antibiotic resistant E. coli strains. International Journal of Hygiene and Environmental Health(220): 1279–1284. CG, O., & CJ, D. 2003. Using multiple regression and gis technology to predict landslide hazard in northeast Kansas, USA. Engineering Geology(69): 331–343. Cheng-yun, Z., Mei, C., Ying, L., & al, e. 2005. The situation and main problems of rebuilding sanitary toilets in rural areas of Sichuan. Modern Preventive Medicine, 32(2): 153–154. Cheng-yun, Z., Ling, Q., Li-jian, J., Hong-chun, T., Da-jun, T., & Huai-xun, H. 2013. Lavatory Reformation Project in Rural Areas of Sichuan. J Prev Med Inf , 29(6): 450–453. chun, L. Z., Jun, C., lan, D. H., Bin, L., Kui, H., & Xu-fang, G. 2013. Analysis on influence factors of hazard-free effects of improved latrines in rural areas in Chengdu. J Med Pest Control, 29(3): 247–250. Ferrante, M., Signorelli, S. S., Ferlito, S. L., Grasso, A., Dimartino, A., & Copat, C. 2018. Groundwater-based water wells characterization from Guinea Bissau (Western Africa): A risk evaluation for the local population. Science of the Total Environment, 619–620: 916–926.

136

5 Indicators Impacting Rural Residents’ Satisfaction …

Fu, Y., Ji, H., Neil, C., Wang, Z., & Tao, Y. 2006. Status quo of rural toilets in 5 counties of Xianyang. Chin J Public Heatlh, 22(7): 881. Gedefaw, M., Amsalu, Y., Tarekegn, M., & Awoke, W. 2015. Opportunities, and challenges of latrine utilization among rural communities of Awabel District, Northwest Ethiopia. Open Journal of Epidemiology(5): 98–106. Glick, S., Guggemos, A. A., & A.M.ASCE. 2013. Rethinking wastewater-treatment infrastructure:Case study using life-cycle cost and life-cycle assessment to highlight sustainability considerations. Journal of Construction Engineering and Management, 139(12): A5013002. Guo, X., Xu, L., & Zhou, L. 2010. Investigation on sanitary latrines in rural areas of Liaocheng City in 2008. Prev Med Trib, 16(16): 129–130. Hang, B. 2008. Difficulties and countermeasures of latrines reform in rural areas. Chinese Primary Health Care, 22(No.03): 68–79. Hiscox, A., Hirooka, R., Vongphayloth, K., Hill, N., Lindsay, S. W., Grandadam, M., & Brey, P. T. 2016. Armigeres subalbatus colonization of damaged pit latrines: A nuisance and potential health risk to residents of resettlement villages in Laos. Medical and Veterinary Entomology(30): 95–100. Hussain, F., Clasen, T., Akter, S., Bawel, V., Luby, S. P., Leontsini, E., Unicomb, L., Barua, M. K., Thomas, B., & Winch, P. J. 2017. Advantages and limitations for users of double pit pour-flush latrines: A qualitative study in rural Bangladesh. BMC Public Health, 17: 515–524. Jun, Z. 2012. Investigation and analysis of the status quo of rural public toilet in Tianjin. Environment Sanitation Engineering, 20(4): 25–27. K, A., & A, V. 2012. Site suitability analysis for natural systems for wastewater treatment with spatial fuzzy analytic hierarchy process. Journal Of Water Resources Planning And Management-Asce, 138: 125–134. Ke-ning, H., & Wei-dong, W. 2011. Study on rural toilet transformation under the background of rural civilization construction. Journal of Hebei Agricultural Sciences, 15(8): 98–100. Kui-wei, Z., Shao-yong, Z., Rong-sheng, Y., & al, e. 2000. Cross-sectional investigation on household latrines and excreta treatment in countryside of Shandong Province. Journal of Enviroment And Health, 17(6): 20–23. Kumwenda, S., Msefula, C., Kadewa, W., Diness, Y., Kato, C., Morse, T., & Ngwira, B. 2017. Is there a difference in prevalence of helminths between households using ecological sanitation and those using traditional pit latrines? A latrine based cross sectional comparative study in Malawi. BMC Res Notes(10): 200–208. Langer, K., Decker, T., Roosen, J., & Menrad, K. 2018. Factors influencing citizens’ acceptance and non-acceptance of wind energy in Germany. Journal of Cleaner Production, 175: 133–144. Li-jian, J., Chang-zhi, C., Huai-xun, H., & al, e. 2007. Current condition of latrine improvement project in rural area of Sichuan Province. J Prev Med Inf,, 23(4): 430–432. Li, W. 2008a. Application of Multivariate Statistical Analysis. Beijing: Peking University Press. Li, W. 2008b. The study of smes credit scoring based on logistic regression model. Hefei University of Technology. Liang, S., Zhang, Y., & Jiang, H. 2002. Effect of Rebuilding Sanitary Latrines on Environmental Sanitation Quality in Rural Areas of Tongbai County. Henan J Prev Med, 13(2): 104–105. Lin, Z., Chen, G., Guo, J., Lai, S., Chen, X., Chen, L., & Lu, C. 2008. Investigation and analysis of rural toilet in Fujian. Chin Prev Med, 9(7): 645–647. Liu, G., Wang, L., Hou, Y., & al, e. 2016. Practice and experience of rural toilet reform in Shijiazhuang. J Med Pest Control, 32(3): 348–351. Peng, D. 2012. Performance evaluation of rural public investment based on farmers’ satisfaction. Shandong: Shandong Agricultural University, 6. Tang, D., Xian, J., & Wen, S. 2003. Investigation of rural toilet situation in Cangxi county and Mabian county, Sichuan. Chinese Journal of Health Education, 19(1): 49–50. Tang, J., Zhu, Y., & Liu, C. 2010. Analysis of rural public service satisfaction and its influencing factors—based on the survey of 67 villages in 32 towns in Shaanxi Province. Contemporary Economic Science, 32(1): 101–116.

References

137

Yang, C., Lv, S., Ji, J., & al, e. 2005. Status and Suggestions of Toilet Reform in Rural Areas. Chinese Journal of Public Health Engineering, 4(4): 229–231. Yao, W., Qu, X., Li, H., & FU, Y. 2009. Present situation of rural toilets and excreta utilization in China. J Environ Health, 26(1): 12–14. Ye, X., & An, D. 2013. Review of the construction of rural sanitary toilet. Chin J of Public Health Eng, 12(No 1): 79–81. Zhan, X., Chen, L., & Lin, Z. 2011. A Preliminary Study on Rural Latrine Reform and Basic Economic Situation in Fujian During the Past 2005–2009. Chinese Primary Health Care, 25(4): 96–97. Zhang, Y., Bi, W., Yang, Y., & al, e. 2005. Investigation and analysis of the influence of rural latrine reform on farmers’ health awareness and behavior. Chinese Journal of Health Supervision, 12(2): 92–95. Zhang, C., Lu, Y., & Zhao, H. 2008. Cultural analysis of barriers to toilet improvement in rural areas. Chin. J of PHM, 2008, 24(No.1): 22–23. Zheng-kui, W., Yun, Y., & Chuan-ye, W. 2005. Economic and social benefit analysis of improvement of latrines in countryside of Hunan Province, China. J Environ Health, 22(6): 445–447. Zhou, J., & Liu, Y. 2015. The method and index of sustainability assessment of infrastructure projects based on system dynamics in China. Journal of Industrial Engineering and Management, 8(3): 1002–1019.

Chapter 6

The Relationship Between the Rural Built Environment and Household Vehicle Ownership

Abstract In this chapter, a questionnaire survey of 374 rural households was conducted and the built environment data of seven typical rural villages in rural Sichuan were collected using Geographic Information System (GIS) technology and on-site measurement. This study aims to investigate the relationship between the rural built environment and rural household vehicle ownership in rural China through a multinomial logit (MNL) model. Results showe that household structure attributes have the most significant relationship with household vehicle ownership, followed by rural built environment attributes and the respondents’ driving skills. In the process of urbanization, with increases in building density, road density, and destination accessibility, an increase in high-carbon vehicle ownership is an inevitable trend among rural households. However, low-carbon-oriented rural planning can effectively control the increase in high-carbon vehicle ownership. For example, the distance between rural households and important destinations, such as hospitals, schools, and markets, should be shortened, and rural residents should be encouraged to learn to ride bicycles. Moreover, rural residents riding motorcycles can effectively reduce household car ownership. Keywords Rural · Built environment · Vehicle ownership · Sustainable transportation · Transport policy · Multinomial logit model · China

6.1 Introduction In 1978, China launched internal reforms that began with rural areas. After 40 years, rural China has undergone a tremendous change. By the end of 2016, the fixed-asset investment of rural households was 20.83 times that in 1985 and 3.43 times that in 2000; the disposable income of rural residents was 31.10 times that in 1985 and 5.49 times that in 2000 (National Bureau of statistics of the People’s Republic of China 1986–2017). The gap between urban and rural areas is gradually narrowing with the rapid urbanization in China. By the end of 2016, China’s highway mileage reached 4,696,263 km, which was 3.35 times that in 2000 (National Bureau of statistics of the People’s Republic of China 1986–2017). In the process of “rural urbanization” © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_6

139

140

6 The Relationship Between the Rural Built Environment …

and “new rural construction”, the temporal and spatial changes have occurred in built environments, and these changes have directly influenced household vehicle ownership in rural China. For example, in 2012, the number of bicycles per 100 households in rural China (79.00) was 0.66 times that in 2000 (120.50); in 2016, the number of motorcycles per 100 households in rural China (65.10) was 2.97 times that in 2000 (21.90); in 2016, the number of cars per 100 households in rural China (17.4) was 13.18 times that in 2000 (1.32) (National Bureau of statistics of the People’s Republic of Chin 1986–2017). In Western developed countries, the rapid increase in high-carbon vehicle ownership and the reliance on cars have caused many problems. For example, the annual growth rate of fossil fuel demand in the transportation sector has reached 10.56% (Wang et al. 2011). Moreover, automobiles have become the main cause of air pollution and photochemical smog pollution in China (Li and Zhao 2017) (Protection 2016). In addition, traffic congestion and obesity are also caused by people’s dependence on automobiles (Handy et al. 2005; Li and Zhao 2017; Potoglou and Kanaroglou 2008).China is the most populous country in the world; thus, a small increase in automobile ownership per capita will result in huge increases in energy consumption and carbon dioxide emissions. By the end of 2016, China’s rural population accounted for 42.65% of the total population in the country, along with tremendous changes in rural built environments. Thus, studying the relationship between rural built environment and rural household vehicle ownership is of great significance for energy conservation and traffic emission reduction. Household vehicle ownership and related information of rural China are shown in Fig. 6.1. The x-axis reports the years from 2000 to 2016 and they are abbreviated with the last 2 digits. The left Y-axis reports the number of vehicles per 100 rural households, while the right Y-axis reports china’s urbanization rate (%), highway mileage (100,000 km), and rural resident per capita income (thousand

Fig. 6.1 Household vehicle ownership and related information of rural China

6.1 Introduction

141

Yuan). Obviously, all the indicators show an increasing trend except for bicycle ownership in rural china from Fig. 6.1. The present study divides rural household vehicles into low-carbon vehicles (including bicycles and electric bicycles) and high-carbon vehicles (including motorcycles and automobiles). We use a multinomial logit (MNL) model to investigate the effect of the built environment on household vehicle ownership after controlling for household structure attributes and individual driving skills in the context of rural Sichuan, China. The structure of this paper is as follows. Section 6.2 reviews previous studies on the relationship between the built environment and vehicle ownership. Section 6.3 provides the data collected and the results of the descriptive analyses. Detailed explanations of the variables and the MNL model used in this study are presented in Sect. 6.4. The results and discussion of the MNL model are presented in Sect. 6.5. Finally, Sect. 6.6 presents the conclusion and policy implications.

6.2 Literature Review The “6Ds” of built environment, namely, density, diversity, design, destination accessibility, distance to transit, and demand management, has been widely used (Ewing and Cervero 2001, 2010; Ewing and Handy 2009; Ewing et al. 2015; Vance and Hedel 2007).The specific measurement indicators for the built environment variables are continuously accumulated and enriched (Wang et al. 2015; Zhang et al. 2014).The most commonly used measure indicators are shown in Table 6.1. In addition to “6Ds,” other indicators are used to measure built environment, such as traffic or personal safety (Broberg and Sarjala 2015), neighborhood type(Cao et al. 2009; Stewart and Moudon 2014)_ENREF_25,infrastructure characteristics (Zhao 2014), and leisure facilities (Feng 2017). A large number of studies have shown that built environment directly influence vehicle ownership. However, most empirical studies have focused on the relationship between built environment and car ownership. Household car ownership decreases with the increase of built environment density (Chatman 2013; Hess and Ong 2002; Keller and Vance 2013; Potoglou and Kanaroglou 2008; Van Acker and Witlox 2010; Zegras 2010). Diversity is negatively correlated with car ownership (Cervero 1996; Ewing et al. 2015). Hong et al. (2014) found that road network density is also negatively correlated with household car ownership; however, the effect of design on household car ownership is weaker than those of density and diversity (Holtzclaw et al. 2002). Destination accessibility is a built environment index at the regional level and generally includes distance to CBD and job accessibility (Krizek 2003; McCormack et al. 2001). Empirical studies have shown that household vehicle ownership decreases with the distance to CBD (Miller and Ibrahim 1998; Van Acker and Witlox 2010). Similarly, distance to CBD can reduce driving mileage significantly and the number of family vehicles to a certain degree if the residence is close to the job or

142

6 The Relationship Between the Rural Built Environment …

Table 6.1 The commonly used built environment indicators of “6Ds” 6D

Meaning

Density

The variable of interest per unit of Dwelling unit density, area Employment density, Population density, Business density, Job density

Commonly used indicators

Diversity

The number of different land uses Land use mix(entropy index), in a given area and the degree to Jobs-population balance which they are represented Jobs-housing balance

Design

Street network characteristics within an area

Intersection density, street density, Street connective, % 4-way intersections

Destination accessibility Ease of access to destination

Job accessibility, Distance to CBD (Central Business District), Distance to other destination

Distance to transit

The level of transit service at the residences or workplaces

Distance to bus stop, Distance to rail station, Distance to highway exit/subway station, Bus stop density, Walk minutes to transit

Demand management

Residential parking distance, quantity, or parking service level

Avg. price daily parking, Avg. price hourly parking._ENREF_46

business center (Ewing and Cervero 2010; Potoglou and Kanaroglou 2008). On the contrary, the demand for and dependence on cars increases with the distance from the residence to CBD (Robert Cervero and Arrington 2008). Distance to transit also influences the level of household car ownership. For example, Potoglou and Kanaroglou (2008) found that the number of household cars can be reduced if public transportation stations are within walking distance and that excellent public transportation services will also reduce the number of household cars (Chatman 2013). Demand management usually refers to residential parking distance, number of parking lots, or parking service level. Demand management will increase car ownership level if the community has low-cost parking lots (Guo 2013; Tyrinopoulos and Antoniou 2013). Thus, Chatman (2008) suggested reducing the number of car parks to impede the increase in household car ownership. Some empirical studies have focused on bicycle, electric bicycle, and motorcycle ownership. Specifically, the establishment of commercial facilities within 300 ft of the settlement will increase the proportions of using public transport, walking, and cycling, and using bicycles is more feasible (Cameron et al. 2003). Accessibility, number of bicycle lanes, mixed environment, and street connectivity are positively related to bicycle use, whereas the service level of public transport is negatively related to bicycle use (Zhao 2014). With regard to motorcycle ownership, Adults who favorably perceive access to public transport and destinations, presence of sidewalks, and safety from crimes at night are less likely to use motorcycles (Lin and Liao 2017). Oyedepo and Etu (2015) found that the likelihood of owning a motorcycle increases 1.43 times with a unit

6.2 Literature Review

143

increase in the number of household members; by contrast, the likelihood of owning a motorcycle decreases by 1.66 times and 2.17 times with unit increases in the average monthly income and academic qualification of the household head, respectively. The effects of socio demographic characteristics on household vehicle ownership are stronger than that of built environment (Li and Zhao 2017). The main socio demographic characteristics include household size, income, age, education level, occupation, and gender (Bhat and Guo 2007; Cao and Cao 2014; Holtzclaw et al. 2002; Keller and Vance 2013). In the past 16 years, especially after 2010, the number of household private vehicles has remarkably changed due to the rapid urbanization in China (Fig. 6.1). In comparison with Western developed countries, China possesses certain unique factors that affect household vehicle ownership, especially in housing properties such as traditional danwei compounds (Wang et al. 2011), reformed danwei communities (Wang et al. 2011), and commodity housing communities (Li and Zhao 2017; Wang et al. 2011; Zhao and Chai 2013). In the process of urbanization, China’s household hukou is also changing. Although the hukou system is a policy for the distribution of social wealth in the era of planned economy in China (Cheng and Selden 1994), the influence of hukou on Chinese households remains significant. The traffic policy also affects household car ownership. One of the most significant characteristics of the literature on the relationship between built environment and car ownership in China is that the study areas are mainly concentrated in large cities, such as Beijing, Guangzhou, Nanjing, and Jinan (Feng et al. 2014; Jiang et al. 2017; Li and Zhao 2017; Zhao 2011). The current study is one of the first to relate the rural built environment to household vehicle ownership in the Chinese rural context (Zhang et al. 2014). This study can provide policy makers and rural planners with insights into ecological rural construction and low-carbon travel behavior.

6.3 Data and Variables 6.3.1 Rural Context in Sichuan Province, China The total economic output of Sichuan Province ranks sixth in China and first in the West, and its per capita GDP exceeds $4,000. With rural urbanization and new rural construction, the rural built environment and rural household vehicle ownership have changed dramatically in Sichuan. By the end of 2016, the total rural population in Sichuan Province was 41.96 million, and the urbanization rate was 49.12%; the total mileage of the highway was 324,200 km, which was 3.57 times that in 2000; the per capital income was 11,203 Yuan, which was 5.89 times that in 2000; and the number of cars per 100 households in rural areas was 12.5, which was 12.38 times that in 2010. By the end of 2015, the number of motorcycles per 100 households was 51.5, which was 5.23 times that in 2000. By the end of 2013, the number of

144

6 The Relationship Between the Rural Built Environment …

Fig. 6.2 Household vehicle ownership and related information in rural Sichuan

bicycles per 100 households was 16.45, which was 0.33 times that in 2000. The household vehicle ownership and related information of rural Sichuan from 2000 to 2016 are shown in Fig. 6.2. The x-axis reports the years from 2000 to 2016 and they are abbreviated with the last 2 digits. The left Y-axis reports the number of vehicles per 100 rural household, while the right Y-axis reports china’s urbanization rate (%), highway mileage (10,000 km), and rural resident per capita income (thousand Yuan). Obviously, all the indicators show an increasing trend except for bicycle ownership in rural Sichuan from Fig. 6.2.

6.3.2 Rural Household Survey and Sampling The College of Environment and Civil Engineering of Chengdu University of Technology and the Business School of Sichuan University organized a rural household survey and collected GIS data from October 1, 2017 to January 31, 2018 based on the research experiences in 2016 and 2017. The survey sample villages should meet two conditions as follows. (1) Each sample village must have at least one local undergraduate or graduate student, and the local residents are willing to participate in the questionnaire survey. (2) At least one road is directly accessible to the sample villages, and the roads are all hardened. This condition is a basic prerequisite for rural residents to hold cars. From 117 sample villages with local students, 7 sample villages were identified, including 3 concentrated-living new villages (namely, Yan Jing, Dong Xing, and Shang Teng) and 4 scattered-living traditional villages (namely, Da Zhuang, Shuang Yan, Xin Long, and Wu Gang).

6.3 Data and Variables

145

The initial survey questions were developed from previous studies (Fu and Farber 2017; Handy et al. 2005; Ma and Dill 2015), and the survey questionnaire was sent to the 117 rural students who were asked to complete the survey. Thereafter, a meeting was organized with the 117 rural students to discuss the survey questions one by one. On the basis of the discussion, the questionnaire was revised and improved in accordance with the actual situation of rural Sichuan. From the 117 rural students, 30 surveyors, comprising 13 graduate and 17 undergraduate students, were completely recruited. All surveyors were uniformly trained before conducting the survey. It is hard to organize a household survey in rural area and this is a long questionnaire. We must make full preparation to guarantee the smooth progress of the survey. Thus, the pre-investigation is needed to understand the response of rural residents to the questionnaire survey, how long time it will take to complete a questionnaire and what kind of gift can attract rural residents to participate in the survey. In order to well understand the problems that may arise in the survey, we randomly selected two sample villages from the seven sample villages and then randomly investigated five rural households to complete the questionnaire in each selected village. Finally, we completed ten pre-investigated questionnaires and found some problems: (1) Rural residents are lack of patience to complete the questionnaire; (2) Incentives have a significant effect on encouraging rural residents to complete the questionnaire; (3) Rural residents from different sample villages have different preferences for incentives. Finally, we prepared different incentives (mainly rural household goods and food) according to their different preferences for the respondents who would complete the questionnaire. In the formal investigation, each survey group was led by a local student, and each questionnaire was completed by face-to-face question-and-answer method between the surveyor and the respondent. Each session lasted 60–80 minutes. Two types of questionnaires were used in the survey, that is, village and household survey questionnaires. The village survey questionnaire was conducted by the surveyor through an on-site measurement and an interview with the village chief. Rural households were randomly selected by the surveyor to complete the household questionnaire. If the selected household refused to accept the questionnaire, then the questionnaire would be randomly transferred to the next household. Finally, 413 completed questionnaires were returned, and 34 questionnaires were eliminated because of missing data. The effective questionnaire rate was 90.56%. Thus, we obtained 374 and 7 valid household and village questionnaires involving 1,758 and 16,953 respondents, respectively. The actual built environment, the regional location, and the number of valid questionnaires of the seven sample villages are shown in Fig. 6.3. The household structure attributes and vehicle ownership of the rural households in the sample villages were in good agreement with the statistical data of China and Sichuan in the China Statistical Yearbook. However, the questionnaire survey data value on rural household car ownership was significantly higher than the data value in the China Statistical Yearbook. This significant difference was mainly attributed to three reasons. (1) The household cars in this survey included all four-wheeled motor vehicles owned by rural households, including small cars, passenger cars,

146

6 The Relationship Between the Rural Built Environment …

Fig. 6.3 Built environment and regional location a of the sample villages

and small transport vehicles. (2) In China’s rural areas, some rural residents go out to work and do not always live in the village. Therefore, some rural households’ cars are not always in the village. The number of household cars used in this paper includes the cars that are not always in the village. (3) The level of infrastructure construction and economic development in the sample villages was higher than the average levels in China and Sichuan. The sample households in the sample villages were randomly selected; thus, they were adequately representative of the sample villages. A comparison of the sample and population characteristics is shown in Table 6.2. Respondent characteristics, including driving skills, household vehicle ownership, and household structure are summarized in Fig. 6.4. Of the respondents, 25% had a driver’s license. The percentages of the respondents who could ride a motorcycle, electronic bicycle, and bicycle were 39, 63, and 71%, respectively. The average numbers of car, motorcycle, bicycle, and electronic bicycle per household were 0.54, 0.58, 0.60, and 0.72, respectively. Of all the respondents, 83% were rural hukou. Other information about household structure attributes are shown in Fig. 6.4.

6.3 Data and Variables

147

Table 6.2 Sample vs. population characteristics Householda Villagea Rural Sichuanb Rural Chinab Total population (Total permanent residents)

1758 (1388)

16953

419.6 2016:Billion

5897.3 2016:Billion

Total number of households

374

5888

–

–

Average permanent residents

3.71

2.88

3.03(2015)

3.88(2012) 1.24(2016)

Per capita income(10 k yuan)

1.36

–

1.13(2016)

Average household income(10 k yuan)

4.44

–

–

–

Average number of household cars

0.54

–

0.13(2016)

0.17(2016)

Average number of household autobikes 0.58

–

0.50(2016)

0.65(2016)

Average number of household ebikes

0.71

–

0.27(2016)

0.58(2016)

Average number of household bicycles

0.59

–

0.31(2012)

0.79(2012)

a Data

from face to face household survey between 29th December 2017 to 5th January 2018 China Statistical Yearbook (2013, 2016 and 2017)

b Source

Fig. 6.4 The average value of socio-demographic information

6.3.3 Measurement of the Rural Built Environment Rural built environment is different from urban built environment in China. Thus, we regulated the scope of measurement by considering the living style of the sample village residents when measuring the built environment indicators. (1) The scope for calculating the built environment indicators was a circle with a 1 km radius from the village center (village committee or neighborhood committee office) for Huojing and Dongxing Villages. Although the respondents from Shangteng also lived a centralized living style, Shangteng is a new rural village under construction; thus, the village possesses many characteristics of a scattered-living village. (2) The administrative village boundary is the scope for calculating the built environment

148

6 The Relationship Between the Rural Built Environment …

indicators in scattered-living traditional villages. The main reason for this provision is that the administrative area of the scattered-living traditional villages in Sichuan Province varies considerably, and the degree of decentralization is inconsistent. The survey household samples were not completely within the scope of the circle with a 1 km radius from the center of the administrative village. We mainly used two approaches to obtain the basic data of the actual built environment considering the limited geographical information of rural areas in China. First, the researchers conducted an on-site measurement using Baidu Navigation App to search and measure the distance to the nearest bus station, train station, bus stop, main road, market, school, health center (hospital), and the center of the city (county) from the village center. The basic data measured onsite are shown in Table 6.3. Second, the basic data of buildings and roads were coded from Tencent Street View imagery (map.qq.com) using ArcGIS 10.2. The road and building land information from ArcGIS 10.2 is shown in Fig. 6.5. Table 6.3 The basic data measured on-site Bus Station (KM)

Train Station (KM)

18.20

19.90

0.20

70.00

Shuangyan

16.30

13.40

Xinlong

Dazhuang Wugang

Public transport station (KM)

Main road (KM)

Market (KM)

School (KM)

Health Center (hospital) (KM)

City Center (KM)

2.50

2.50

3.00

0.50

0.05

19.60

16.00

0.00

3.50

2.50

0.20

16.00

0.50

0.50

1.60

1.60

0.60

13.50

13.40

13.40

1.20

0.80

0.80

3.00

4.90

4.90

Dongxing

3.90

16.40

3.90

0.50

0.00

2.10

0.00

10.00

Shangteng

22.40

24.80

0.69

0.69

1.50

1.50

1.60

14.00

0.50

125.00

34.00

0.50

1.50

0.50

1.70

35.00

Yanjing

Fig. 6.5 Road and building land information

6.4 Method

149

6.4 Method 6.4.1 Variable Specification 6.4.1.1

Dependent Variables

In the basic statistical analysis, most of the rural households owned only zero or one car, motorcycle, electric bicycle, and/or bicycle in the sample rural villages. Only 6.9, 5.8, 7.8, and 8.8% of the households owned two or more cars, motorcycles, electric bicycles, and/or bicycles, respectively. Therefore, the rural households were classified into two groups, namely, with or without a certain vehicle. In addition, electric bicycles and bicycles were considered as low-carbon vehicles, whereas motorcycles and cars were classified as high-carbon vehicles. This classification was based on the power source and carbon emission level of the vehicles. Finally, the rural household vehicle ownership combination set is “no vehicles, only low-carbon vehicles, only high-carbon vehicles, and high- and low-carbon vehicles,” which is denoted by {0, L, H, H&L}, respectively. Additional information about rural household vehicle ownership from the sample villages is shown in Fig. 6.6.

6.4.1.2

Socio Demographic Variables

Socio demographic characteristics affect household vehicle ownership decision. Thus, we also collected information about the respondents and their households. In

Fig. 6.6 Household vehicle ownership of the sample villages (Note “mean” refers average number of household vehicles “Category” refers vehicle set {0, H&l, H, L})

150

6 The Relationship Between the Rural Built Environment …

Fig. 6.7 The distribution of the main socio-demographic variables (Note “I” refers household income [ten thousand Yuan]; “P” refers the number of permanent residents)

this study, household structure attributes and respondents’ driving skills were selected as the main variables because of their significant effects on household vehicle ownership. These variables include the numbers of permanent residents, workers, household members under 18, driver’s license holders, dwelling units, household parking lots, and rural hukou; household income; and ability of the respondents to ride a vehicle, such as cars, motorcycles, electric bicycles, and bicycles. Household size, the number of household members under 18, household income, and the number of driver’s license holders were always significantly influencing household vehicle ownership. Thus, we coded these four variables as dummy variables on the basis of the basic statistical analysis results to obtain additional detailed information about the relationship between these variables and household vehicle ownership. Further information about these four variables is shown in Fig. 6.7, and the descriptive statistics of these variables are shown in Table 6.4.

6.4.1.3

Built Environment Variables

The built environments in rural areas are simpler than those in cities; however, rural built environments pose more difficulties in data collection. This study is mainly concerned about the “4Ds + 1S” built environment variables on the basis of the on-site measurement basic data and GIS extraction data. These variables are design, diversity, distance to transit, destination accessibility, and living style. Although we investigated the population and the number of households in the sample villages, we were unable to obtain data accurately on the population and households within the

6.4 Method

151

Table 6.4 Definitions and descriptive statistics of variables used in this study Variable

Description

Type

Dependent variable vehicle ownership

No vehicle, low-carbon vehicle, high -carbon vehicle, both low and high- carbon vehicle

Category

Explanatory variables household structure

Mean

S.D.

Min

Max

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

Resident population: ≥ 5

1, if household resident population is 5 or more;0, otherwise

Dummy

0.31

0.46

0.00

1.00

Resident population: 3–4

1, if household resident population is 3 or 4; 0, otherwise

Dummy

0.43

0.50

0.00

1.00

Population under 18: 1

1, if household has one member younger than 18 years of age; 0, otherwise

Dummy

0.40

0.49

0.00

1.00

Population under 18:2+

1, if household has two Dummy or more members younger than 18 years of age; 0, otherwise

0.22

0.41

0.00

1.00

Number of license holders: 1

1, if the number of license holders in the household is 1; 0, otherwise

Dummy

0.42

0.49

0.00

1.00

Number of license holders: 2+

1, if the number of license holders in the household is two or more; 0, otherwise

Dummy

0.32

0.47

0.00

1.00

Household income: 1, if the income of high household is more than RMB 50000 yuan; 0, otherwise

Dummy

0.23

0.42

0.00

1.00

Household income: 1, if the income of medium household is between RMB 2000 and 50000 yuan; 0, otherwise

Dummy

0.48

0.50

0.00

1.00

Household parking lot

1, if the household has Dummy parking lot; 0, otherwise

0.54

0.50

0.00

1.00

Rural hukou

1, if the household is Dummy rural hukou; 0, otherwise

0.83

0.37

0.00

1.00

2.01

1.23

0.00

11.00

Number of workers Number of household ordinal workers between the age of 18 and 65 years

(continued)

152

6 The Relationship Between the Rural Built Environment …

Table 6.4 (continued) Variable

Description

Type

Dwelling units

Number of house units

ordinal

personal skills

Mean

S.D.

Min

1.20

0.54

0.00

–

Max 3.00

–

–

Holding a driver’s license

1, if the respondent has a dummy drive license; 0, otherwise

0.25

0.44

0.00

1.00

Ride motorcycle

1, if the respondent can ride motorcycle; 0, otherwise

dummy

0.39

0.49

0.00

1.00

Ride ebike

1, if the respondent can ride ebike; 0, otherwise

dummy

0.71

0.45

0.00

1.00

Ride bicycle

1, if the respondent can ride bicycle; 0, otherwise

dummy

0.63

0.48

0.00

1.00

–

–

–

–

Building density

Defined in Eq. (6.1)

Continuous

11.81

5.60

4.76

19.5

Road density

Defined in Eq. (6.2)

Continuous

3.33

0.76

2.25

Distance to transit

Defined in Eq. (6.3)

Continuous

1.27

0.36

.67

1.91

Destination accessibility

Defined in Eq. (6.4)

Continuous

1.59

0.42

1.14

2.41

Living style

1, if household lives in in dummy concentration area; 0, otherwise

0.39

0.49

0.00

1.00

Built environment

–

4.74

circle of 1 km radius from the central residential area. Finally, we did not consider the density variables, namely, population density and dwelling unit density. The design in this study denotes road density and is calculated as Design index = Total length of roads (m)/Total survey area (mu)

(6.1)

Land use mix is consistently used in calculating the diversity index in most of the related studies. However, land use in rural areas is relatively single, and we can only read building land by using GIS technology. Thus, in this study, building density was used to calculate the diversity of rural land use. Here mu is a unit of land area in China. Fifteen acres equals one hectare.     Diversity index = Building land area m2 /Total survey area m2

(6.2)

Anowar et al. (2014) and Anowar et al. (2014)used mix index to calculate the distance to transit. We simplified his formula in calculating the distance-to-transit mix index and destination accessibility mix index.

6.4 Method

153

Distance − to − transit mix index =

 {1/(dk + 1)},

(6.3)

k

Where k = 1, 2, 3, 4, and dk represents the distance from the village center to the nearest bus station, train station, public transportation station, and main road. Destination accessibility mix index =



{1/(dk + 1)}

(6.4)

k

Where k = 1, 2, 3, 4, and dk represents the distance from the village center to the nearest market, school, health center (hospital), and city (county) center. As a result of urbanization, the living style of rural residents is gradually shifting from traditional scattered living to urbanized centralized living, and the change in the living style directly influences a household’s decision on vehicle ownership. Thus, aside from the influences of the D variables, the influence of living style on rural household vehicle ownership was also investigated in this study. All the variables used in this study are shown in Table 6.4. Multicollinearity problems may cause low significance levels of various spatial variables (Ding et al. 2017). Therefore, the multicollinearity of the independent variables in this study should be examined. The variable expansion factor (VIF) was used to test for multicollinearity. A larger VIF value indicates that a particular explanatory variable is more likely to be represented by a linear function model with other explanatory variables and that the model may have multicollinearity problems (Yao et al. 2014).Our analysis implied that the VIF values of the explanatory variables were well below 5, indicating that no multicollinearity problem was present.

6.4.2 Model Specification We set the vehicle ownership to four categories, namely, no vehicles (0), owning low-carbon vehicles (L), owning high-carbon vehicles (H), and owning high-and low-carbon vehicles (H&L). The utility functions for vehicle ownership of each household can be expressed as follows (Jiang et al. 2017; Train 2009): 

U (0) = β0 xn0 + εn0 , 

U (L) = β L xn L + εn L , 

U (H ) = β H xn H + εn H , 

U (H &L) = β H &L xn H &L + εn H &L ,

(6.5)

where yn = ii f Un (i) > Un (q)∀q = i, and ε is assumed to be independently and identically distributed with identical extreme value distribution. The distribution function is F(εi ) = ex p(−ex p(−εi )). On the basis of this specification, the choice probabilities are

154

6 The Relationship Between the Rural Built Environment … 

Pr obi = Pr obUn (i) > Un (q) = e(βi xni ) /

J 



e



βq xnq



, ∀q = i, i = 0, L , H, H &L

q=0



Pr ob(yn = i) = 1, 0 ≤ Pr ob(yn = i) ≤ 1, i = 0, L , H, H &L (6.6)

i=0,L ,H,H &l

We usedthe option “H” as a reference option. With the coefficient β H = 0, this equation can be rewritten as Pr obyn = i = Pr obUn (i) > Un (q) = 1/(e

 (β0 xn0 )





+ e(βL xn L ) + e(β H &L xn H &L ) + 1), i = H

Pr obyn = i = Pr obUn (i) > Un (q) = (e

 βi xni )

/e

 (β0 xn0 )

+e

 (β L xn L )

+ e(β H &L  xn H &L ) + 1), i = 0, L , H &L

(6.7)

6.5 Results and Discussion The MNL model of the vehicle ownership of rural households contains all the explanatory variables described in the previous section. We used NLOGIT 5.0 for the model estimation. The explanatory variables were entered into the model one by one, following the categories of household structure attributes, respondent driving skills, and rural built environment. Likelihood ratio tests were performed. The results of these tests are summarized in Table 6.5. The test results show that every variable category has a significant contribution to the model because all likelihood ratio values are well above the critical value. The log-likelihood value increased from −434.20 to −269.33; thus, each category variable should be included in the set of explanatory variables to explain rural household vehicle ownership. However, the relative explanatory power of each category variable cannot be observed from the results in Table 6.5. Accordingly, another set of likelihood ratio tests was conducted; the results of which are shown in Table 6.6. The household structure attributes had the highest likelihood value, which implies that this variable had the highest explanatory Table 6.5 Likelihood ratio index (LRI) test results: addition of explanatory variables K

Log-likelihood L(i)

Specific constant

3

L(c) = −434.20

Household structure

39

L(1) = −340.37

LRI

Critical value*

−2[L(c)−L(1)] = 187.67

62.428

11.345

Driving skills

51

L(2) = −302.98

−2[L(1)−L(2)] = 74.79

30.578

Built environment

66

L(3) = −269.33

−2[L(2)−L(3)] = 67.29

34.805

*Represents 0.01 significant level

6.5 Results and Discussion

155

Table 6.6 Likelihood ratio index (LRI) test results: introduction of single explanatory variables K

Log-likelihood L(i)

Specific constant

3

L(c) = −434.20

LRI

Critical value*

Household structure

39

L(1) = −340.37

−2[L(c)−L(1)] = 187.67

Driving skills

15

L(2) = −385.99

−2[L(c)−L(2)] = 96.42

30.578

Built environment

18

L(3) = −374.47

−2[L(c)−L(3)] = 119.48

34.805

11.345 62.428

*Represents 0.01 significant level

power among all the explanatory variables. Built environment and personal driving skill variables follow the household structure attributes. We set high-carbon vehicles (H) as the reference option to estimate the model parameters. We specified a full set of alternative specific constants corresponding to no vehicles (0), low-carbon vehicles (L), high- and low-carbon vehicles (H&L), and all were statistically significant. Generally, specific constants capture the unobserved information (Moshe and Lerman 1985). In vehicle ownership model, alternative specific constants capture the costs associated with vehicle ownership, namely, purchase, maintenance, and lease costs (Ryan and Han 1999).The cost information of household owned vehicles is difficult to collect accurately; thus, the MNL model in this study does not include specific cost variables, and the relationship between unobserved information and rural household vehicle ownership is reflected by the specific constant items. Under the same conditions, the utility of a rural household for owning no vehicle (0), low-carbon vehicles (L), high-carbon vehicles (H), and highand low-carbon vehicles (H&L) is gradually reduced for the unobserved information related to cost (Table 6.7). This result is easy to understand and fully meets our expectations. It also agrees with the findings of (Choudhary and Vasudevan 2017), who categorized household vehicle ownerships into four, namely, no vehicles (0), two-wheel vehicle (2 W), four-wheel vehicle (4 W), and two- and four-wheel vehicles (2&4 W), the descending order of which is 0 > 2 W > 4 W > 2&4 W.

6.5.1 Household Structure Attributes Nearly all the household structure attributes that were explanatory variables, except for the numbers of dwelling units and household parking lots, had significant influences on household vehicle ownership. Households with a resident population of 5 or more had the highest utility for owning high- and low-carbon vehicles (H&L), followed by owning only low-carbon vehicles (L). For households with a resident population of 3–4, the negative β value indicates that they are more willing to own high-carbon vehicles (H). In addition, in comparison with the household utility for owning only high-carbon vehicles (H), household utility will increase significantly with the increase in the number of permanent residents when they own high- and low-carbon vehicles (H&L), only low-carbon vehicles (L), and no vehicles (0).

–

Dwelling units

–

–

–

Ride motorcycle

Ride bicycle

Ride ebike

Built environment

–

Holing a driver’s license

Personal skills

–

–

Number of workers

–

Household parking lot

Rural hukou

–

–

Household income: medium

–

Number of license holders: 2+

Household income: high

–

–

–

Population under 18:1

Number of license holders: 1

–

Resident population: 3−4

Population under 18:2+

–

Resident population: ≥ 5

0.749 0.523 0.779

0.363 −0.695 −0.383

0.851

0.100

−1.875 −0.241

0.771 0.895

−0.395

0.260

−0.964 −0.051

0.062 0.041

−2.361

0.023

−3.428 −1.685

0.243 0.172

−1.621 −1.194

0.172

1.090

0.712 0.041

0.407

0.006

P

−1.702

13.008

Coefficient

–

–

0

H

Specific constant

Household structural attributes

Variable

**

*

**

**

***

Sig

Table 6.7 MNL estimated parameters of rural household vehicle ownership L

2.229

0.001

0.558

0.001 0.341

0.144

−1.979

0.397

0.017

0.053

0.843

0.668

0.029

0.040

0.877

0.916

0.294

0.462

0.063

0.083

P

−0.924

−0.438

−0.500

−1.359

0.093

−0.208

−1.552

−1.440

0.080

0.063

0.522

−0.367

1.202

3.943

Coefficient

***

***

**

*

**

**

*

*

Sig

H&L

1.835

1.835

−1.439

−1.485

0.234

−0.071

−0.805

0.089

0.614

1.206

−0.195

0.383

1.038

1.000

0.513

1.536

−4.043

Coefficient

0.002

0.002

0.007

0.004

0.536

0.647

0.219

0.833

0.273

0.010

0.745

0.447

0.044

0.024

0.262

0.007

0.044

P

(continued)

***

***

***

***

**

**

**

***

**

Sig

156 6 The Relationship Between the Rural Built Environment …

0.091 0.355

−3.215 −2.709

–

Distance to transit mix index

Living style

L

***

Sig

H&L

1.679

1.625

−0.079

−0.580

0.175

Coefficient

0.380 0.330

Likelihood ratil = −2[L(c)−L(β)]

Rho-squared(R2 = 1−[L(β)/L(c)])

Adjusted rho-squared(Adj-R2 = 1−[(L(β)-M)/L(c)])

*Significant at the 10% level **Significant at the 5% level ***Significant at the 1% level

−269.332 329.745

Log-likelihood model(L(β))

374

0.009

0.489

0.318

0.568

0.132

P

−434.204

1.440

−0.546

−0.908

−0.289

0.108

Coefficient

Log-likelihood with alternate specific constants(L(C))

**

*

*

*

Sig

Number of observation

Related statistics

–

–

Destination accessibility 0.021

0.092

2.583

0.071

−1.332

–

−0.003

–

Coefficient

–

Road density

P

0

H

Building density

Variable

Table 6.7 (continued)

0.001

0.007

0.921

0.176

0.005

P

***

***

***

Sig

6.5 Results and Discussion 157

158

6 The Relationship Between the Rural Built Environment …

Households with members under the age of 18 were inclined to own high- and low-carbon vehicles (H&L); however, the utility for owning high- and low-carbon vehicles (H&L) did not increase significantly with the increase in the number of members under the age of 18. Households with two or more members with a driver’s license are most likely to own only high-carbon vehicles (H), followed by high- and low-carbon vehicles (H&L), only low-carbon vehicles (L), and no vehicles (0); and the utilities of the different vehicle ownership significantly differed. The probabilities of owning vehicles for high-income households in descending order was H&L > H > L > 0. All the β values for high-income households were statistically significant, and the significant difference indicated that the utilities of high-income households with different vehicle ownership categories were significantly different. By contrast, the β values for middle-income households were only statistically significant to no vehicles (0), and negative β values implied that the utility of middle-income households with no vehicles (0) was lower than that of middle-income households with high-carbon vehicles (H). This finding agrees with those of previous studies, which asserted that household income is the key factor influencing household vehicle ownership [6,49], especially for high-income households. All the three estimated parameters (β) for rural hukou were negative and only statistically significant for low-carbon vehicles (L), indicating that rural households with rural hukou were more willing to own high-carbon vehicles (H). This finding on rural hukou is in contrast to the those of previous studies on car ownership in the Chinese urban context. Specifically, Yang et al. Jiang et al. (2017) studied household car ownership in Jinan, China and found that households with rural hukou have a relatively low probability of buying cars. However, an in-depth analysis revealed that this difference was entirely dependent on the actual conditions of urban and rural areas in China. In comparison with urban residents, rural residents who worked in urban areas have lower income and fewer resources. Although some rural residents have been urbanized and the lands of these rural residents have been expropriated in the process of urbanization in China, they rely on government subsidies to live without stable work. Therefore, urbanized rural households in rural areas and rural households in urban areas had lower utilities of for owning automobiles and are less likely to own more automobiles. Households with several workers were willing to own high-carbon vehicles; this finding is consistent with those in other studies on car ownership (Hirota 2010; Potoglou and Kanaroglou 2008).

6.5.2 Personal Skills In addition to household structure attributes, the driving skills of the respondents were considered in this study. Although the individual characteristics of the respondents demonstrated a limited effect on household vehicle ownership, the driving skill variable is directly and significantly related to vehicle ownership. Personal driving skills involved four variables, namely, whether they hold a driver’s license and whether they could ride a motorcycle, bicycle, and/or electric bicycle. As shown in Table 6.7,

6.5 Results and Discussion

159

all the four variables were statistically significant. All the three estimated parameters of holding a driver’s license were negative but only statistically significant for highand low-carbon vehicles (H&L). The β values for riding a motorcycle for low-carbon vehicles (L) and high- and low-carbon vehicles (H&L) were negative and statistically significant. These results show that people with skills in driving high-carbon vehicles were inclined to own high-carbon vehicles. In addition, people who could ride a motorcycle had the lowest utility for owning low-carbon vehicles. This outcome demonstrates that motorcycles are deemed to have the kinetic energy of automobiles and the convenience of low-carbon vehicles. The estimated parameters of cycling and riding an electric bicycle were relatively consistent. For households that owned high- and low-carbon vehicles, both variables (cycling and riding an electric bicycle) had nearly the same positive β coefficient, indicating that both types of household were willing to own high- and low-carbon vehicles. By contrast, those who could ride an electric bicycle (or their household)had the highest utility to own low-carbon vehicles (L). however, those who could ride a bicycle (or their household) had a relatively low utility for owning low-carbon vehicles (L) because of lower kinetic energy and convenience than electric bicycles. Thus, the ability to ride a motorcycle and/or electric bicycle had a significant influence on vehicle ownership decision for households in rural areas.

6.5.3 The Built Environment The effects of the five variables of the built environment on vehicle ownership of rural households are statistically significant. Specifically, building density, road density, and destination accessibility (distance to the nearest hospital/health center, school, market, and city/county center) had significant negative effects on no vehicles (0), indicating that with an increase in road density, building density, and destination accessibility in rural areas, rural households were more willing to own high-carbon vehicles (H). This result is in contrast to previous studies on household car ownership in Chinese urban areas. Building and road densities are generally believed to have a positive influence on walking activities for residents and negative influence on vehicle ownership for their households (Bhat and Guo 2007; Ding et al. 2017; Li and Zhao 2017; Zegras 2010). Two studies indicated that a shorter distance from the CBD and job center implies lower vehicle ownership of households (Ewing and Cervero 2010; Potoglou and Kanaroglou 2008). The main reason behind the contradicting results of the current study and previous studies is the huge gap between the rural and urban built environments in China. For Sichuan rural areas, building and road densities can represent the development level of infrastructure to some extent. Although the rural areas are urbanizing rapidly, the rural built environment indicators still lag far behind cities. For example, the average building density of 20 districts in Jinan City is 0.407 (Yang 2013), whereas that of the sample villages selected in this study is 0.119. In the process of rural urbanization, rural residents are more reluctant to have no vehicles (0) than to have high-carbon vehicles (H) as building density, road

160

6 The Relationship Between the Rural Built Environment …

density, and destination accessibility increase in rural areas. As shown in Table 6.7, destination accessibility had the largest influence on no vehicles (0), followed by road and building densities. Therefore, walking- or cycling-oriented rural planning could effectively reduce rural household automobile ownership and vehicle carbon emissions. Locations of schools and health centers should be planned within walking and riding range for rural residents. In addition, building density and distance to transit positively influenced owning high- and low-carbon vehicles (H&L), with distance to transit having a greater influence on H&L. Finally, the centralized living style had a significant positive effect on no vehicles (0), high- and low-carbon vehicles (H&L), and low-carbon vehicles (L), although the degree of influence weakened in such an order. This order is consistent with our expectation that rural households with a centralized living style are more willing to own no vehicles (0), followed by owning high- and low-carbon vehicles (H&L) and low-carbon vehicles (L), than to own highcarbon vehicles (H). In the process of rural urbanization, the living style of rural household changes from traditional scattered living to urbanized centralized living, and the spatial distance between neighborhoods is gradually shrinking. Although centralized living areas have the characteristics of a city and the comprehensive service level in these areas is higher than that in traditional scattered residential areas, a certain gap exists between the two living styles in the level of urban development. Thus, walking can meet the need for the basic neighborhood interaction; however, important destinations, such as schools, hospitals/health centers, and bazaars, are still outside the walking distance. Finally, no vehicles (0) and high- and low-carbon vehicles (H&L) had the highest probabilities of selection for centralized-living rural households. In the estimation of the standardized parameters of all variables, household structure variables had the most significant influence on rural household vehicle ownership. This result is consistent with existing research findings (Cullinane 2003; He and Thøgersen 2017).The household structure variables were followed by built environment variables, indicating that the rapid and considerable changes in rural built environment have a significant effect on household vehicle ownership in rural China.

6.6 Conclusion and Policy Implications With the fast pace of construction and urbanization of rural China, tremendous changes are occurring along with considerable increase in energy consumption of rural households. One of the key factors is the change in rural household vehicle ownership. The use of GIS technology and a discrete choice model allows scholars to investigate the relationship between built environment and household vehicle ownership. This study is one of the first to investigate household vehicle ownership in the rural built environment context. The MNL model of rural household vehicle ownership was derived from the data collected through a rural village and rural household survey and with the use of GIS technology.

6.6 Conclusion and Policy Implications

161

The results show that all the household structure attributes, personal skills and rural built environment variables have significant influence on household vehicle ownership. The likelihood ratio tests also show that the built environment variables have great effects on rural household vehicle ownership. All these findings can help rural policy makers and planners create effective policies and design potential interventions by considering personal driving skills and rural built environment. For the reduction of travel energy consumption and carbon emissions, we suggest the following. (1) Important destinations, such as schools, hospitals/health centers, and bazaars should be planned, such that they are within walking and cycling distance. (2) In the process of urbanization, a reasonable scale of centralized residential areas should be established for urbanized rural households. (3) Rural residents should be encouraged to learn to ride bicycles, and electric bicycles. (4) Rural residents also should be encouraged to learn to ride motorcycles for rural residents riding motorcycles can effectively reduce household car ownership. Some possible future research opportunities could include: (1) exploring the effects of the rural built environment on rural individual’s mode choice (including long and short distances); (2) testing the influence of fast changes in the rural built environment on rural residents’ travel behavior and/or travel-activities; (3) considering the interaction among rural individual self-selection, rural built environment, vehicle ownership and travel-activities; and (4) incorporating a greater range of rural built environments to test the spatial heterogeneity of rural China contexts.

References Anowar, S., Yasmin, S., Eluru, N., & Miranda-Moreno, L. F. 2014. Analyzing car ownership in Quebec City: a comparison of traditional and latent class ordered and unordered models. Transportation, 41(5): 1013–1039. Bhat, C. R., & Guo, J. Y. 2007. A comprehensive analysis of built environment characteristics on household residential choice and auto ownership levels. Transportation Research Part B: Methodological, 41(5): 506–526. Broberg, A., & Sarjala, S. 2015. School travel mode choice and the characteristics of the urban built environment: The case of Helsinki, Finland. Transport Policy, 37: 1–10. Cameron, I., Kenworthy, J. R., & Lyons, T. J. 2003. Understanding and predicting private motorised urban mobility. Transportation Research Part D-Transport And Environment, 8(4): 267–283. Cao, J. S., & Cao, X. S. 2014. The Impacts of LRT, Neighbourhood Characteristics, and Selfselection on Auto Ownership: Evidence from Minneapolis-St. Paul. Urban Studies, 51(10): 2068– 2087. Cao, X., Mokhtarian, P. L., & Handy, S. L. 2009. The relationship between the built environment and nonwork travel: A case study of Northern California. Transportation Research Part a-Policy And Practice, 43(5): 548–559. Cervero, R. 1996. Mixed land-uses and commuting: Evidence from the American housing survey. Transportation Research Part a-Policy And Practice, 30(5): 361–377. Chatman, D. G. 2008. Deconstructing development density: Quality, quantity and price effects on household non-work travel. Transportation Research Part a-Policy And Practice, 42(7): 1008–1030.

162

6 The Relationship Between the Rural Built Environment …

Chatman, D. G. 2013. Does TOD need the T?: On the importance of factors other than rail access. Journal of the American Planning Association, 79(1): 17–31. Cheng, T. J., & Selden, M. 1994. The origins and social-consequences of China Hukou system. China Quarterly (139): 644–668. Choudhary, R., & Vasudevan, V. 2017. Study of vehicle ownership for urban and rural households in India. Journal of Transport Geography, 58: 52–58. Cullinane, S. 2003. Hong Kong’s low car dependence: Lessons and prospects. Journal of Transport Geography, 11: 25–35. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. 2017. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100: 65–80. Ewing, R., & Cervero, R. 2001. Travel and the built environment—A synthesis. Land Development and Public Involvement in Transportation (1780): 87–114. Ewing, R., & Cervero, R. 2010. Travel and the built environment. Journal of the American Planning Association, 76(3): 265–294. Ewing, R., & Handy, S. 2009. Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban Design, 14(1): 65–84. Ewing, R., Tian, G., Goates, J. P., Zhang, M., Greenwald, M. J., Joyce, A., Kircher, J., & Greene, W. 2015. Varying influences of the built environment on household travel in 15 diverse regions of the United States. Urban Studies, 52(13): 2330–2348. Feng, J. X. 2017. The influence of built environment on travel behavior of the elderly in urban China. Transportation Research Part D-Transport And Environment, 52: 619–633. Feng, J. X., Dijst, M., Wissink, B., & Prillwitz, J. 2014. Understanding mode choice in the Chinese context: The case of Nanjing metropolitan area. Tijdschrift Voor Economische En Sociale Geografie, 105(3): 315–330. Fu, L., & Farber, S. 2017. Bicycling frequency: A study of preferences and travel behavior in Salt Lake City, Utah. Transportation Research Part A: Policy and Practice, 101: 30–50. Guo, Z. 2013. Does residential parking supply affect household car ownership? The case of New York City. Journal of Transport Geography, 26: 18–28. Handy, S., Cao, X., & Mokhtarian, P. 2005. Correlation or causality between the built environment and travel behavior? Evidence from Northern California. Transportation Research Part D: Transport and Environment, 10(6): 427–444. He, S. Y., & Thøgersen, J. 2017. The impact of attitudes and perceptions on travel mode choice and car ownership in a Chinese megacity: The case of Guangzhou. Research in Transportation Economics, 62: 57–67. Hess, D. B., & Ong, P. M. 2002. Traditional neighborhoods and automobile ownership. Travel Demand and Land Use 2002 (1805): 35–44. Hirota, K. 2010. Comparative studies on vehicle related policies for air pollution reduction in ten Asian countries. Sustainability, 2(1): 145–162. Holtzclaw, J., Clear, R., Dittmar, H., Goldstein, D., & Haas, P. 2002. Location efficiency: Neighborhood and socioeconomic characteristics determine auto ownership and use—Studies in Chicago, Los Angeles and San Francisco. Transportation Planning and Technology, 25(1): 1–27. Hong, J. H., Shen, Q., & Zhang, L. 2014. How do built-environment factors affect travel behavior? A spatial analysis at different geographic scales. Transportation, 41(3): 419–440. Jiang, Y., Gu, P., Chen, Y., He, D., & Mao, Q. 2017. Influence of land use and street characteristics on car ownership and use: Evidence from Jinan, China. Transportation Research Part D: Transport and Environment, 52: 518–534. Keller, R., & Vance, C. 2013. Landscape pattern and car use: Linking household data with satellite imagery. Journal of Transport Geography, 33: 250–257. Krizek, K. J. 2003. Residential relocation and changes in urban travel—Does neighborhood-scale urban form matter? Journal of the American Planning Association, 69(3): 265–281. Lee, C.-K., & Shaw, M.-S. 1996. Constrained Diffusion Models for the Prediction of Multi-Class Motor Vehicle Ownership. Volume 1: Travel Behavior. Paper presented at the World Transport

References

163

Research. Proceedings of the 7th World Conference on Transport ResearchWorld Conference on Transport Research Society. Li, S., & Zhao, P. 2017. Exploring car ownership and car use in neighborhoods near metro stations in Beijing: Does the neighborhood built environment matter? Transportation Research Part D: Transport and Environment, 56: 1–17. Lin, C.-Y., & Liao, Y. 2017. Perceptions of activity-supportive environment and motorcycle use among urban Taiwanese adults. BMC public health, 17(1): 665. Ma, L., & Dill, J. 2015. Associations between the objective and perceived built environment and bicycling for transportation. Journal of Transport & Health, 2(2): 248–255. McCormack, E., Rutherford, G. S., & Wilkinson, M. G. 2001. Travel impacts of mixed land use neighborhoods in Seattle, Washington. Land Development and Public Involvement in Transportation (1780): 25–32. Miller, E. J., & Ibrahim, A. 1998. Urban form and vehicular travel—Some empirical findings. Land Use and Transportation Planning and Programming Applications (1617): 18–27. Moshe, B.-A., & Lerman, S. 1985. Discrete choice analysis: Theory and application to travel demand. Massachusetts: Massachusetts Institute of Technology. National Bureau of statistics of the People’s Republic of China 1986–2017. China Statistical Yearbook. Oyedepo, O. J., & Etu, J. 2015. Binomial logistic regression model of household motorcycle ownership in Akure, Ondo State, Nigeria. Journal of Transport Literature, 9(4): 40–44. Potoglou, D., & Kanaroglou, P. S. 2008. Modelling car ownership in urban areas: A case study of Hamilton, Canada. Journal of Transport Geography, 16(1): 42–54. Protection, M. o. E. 2016. 2015 Annual report of automobile air pollution in China. In: Protection, M. O. E(Ed.). Robert Cervero,G. B., & Arrington. 2008. Vehicle trip reduction impacts of transit-oriented housing. Journal of Public Transportation, 11 (3): 1–17. Ryan, J., & Han, G. 1999. Vehicle-ownership model using family structure and accessibility application to Honolulu, Hawaii. Transportation Research Record: Journal of the Transportation Research Board (1676): 1–10. Stewart, O. T., & Moudon, A. V. 2014. Using the built environment to oversample walk, transit, and bicycle travel. Transportation Research Part D-Transport and Environment, 32: 15–23. Train, K. E. 2009. Discrete choice methods with simulation: Cambridge university press. Tyrinopoulos, Y., & Antoniou, C. 2013. Factors affecting modal choice in urban mobility. European Transport Research Review, 5(1): 27–39. Van Acker, V., & Witlox, F. 2010. Car ownership as a mediating variable in car travel behaviour research using a structural equation modelling approach to identify its dual relationship. Journal Of Transport Geography, 18(1): 65–74. Vance, C., & Hedel, R. 2007. The impact of urban form on automobile travel: Disentangling causation from correlation. Transportation, 34(5): 575–588. Wang, D., Chai, Y., & Li, F. 2011. Built environment diversities and activity–travel behaviour variations in Beijing, China. Journal of Transport Geography, 19(6): 1173–1186. Wang, Y., Peng, Z., Wang, K., Song, X., Yao, B., & Feng, T. 2015. Research on urban road congestion pricing strategy considering carbon dioxide emissions. Sustainability, 7(8): 10534–10553. Yao, B., Hu, P., Lu, X., Gao, J., & Zhang, M. 2014. Transit network design based on travel time reliability. Transportation Research Part C: Emerging Technologies, 43: 233–248. Yang, Y. 2013. A research on neighborhood built environment and household travel energy in Jinan, Tsinghua University. Zegras, C. 2010. The built environment and motor vehicle ownership and use: Evidence from Santiago de Chile. Urban Studies, 47(8): 1793–1817. Zhang, Y., Wu, W., Li, Y., Liu, Q. X., & Li, C. Y. 2014. Does the built environment make a difference? An investigation of household vehicle use in Zhongshan metropolitan area, China. Sustainability, 6(8): 4910–4930.

164

6 The Relationship Between the Rural Built Environment …

Zhang, Y., Yang, X., Li, Y., Liu, Q., & Li, C. 2014. Household, personal and environmental correlates of rural Elderly’s cycling activity: Evidence from Zhongshan Metropolitan area, China. Sustainability, 6(6): 3599–3614. Zhao, P. J. 2011. Car use, commuting and urban form in a rapidly growing city: evidence from Beijing. Transportation Planning and Technology, 34(6): 509–527. Zhao, P. J. 2014. The impact of the built environment on bicycle commuting: Evidence from Beijing. Urban Studies, 51(5): 1019–1037. Zhao, Y., & Chai, Y. 2013. Residents’ activity-travel behavior variation by communities in Beijing, China. Chinese Geographical Science, 23(4): 492–505.

Chapter 7

The Impact of the Rural Built Environment on Household Car Ownership, Adjusted for Preference Bias

Abstract This chapter collected data from 374 households from rural areas in Sichuan, China, to examine the effects of the built environment on the number of cars in a household. It considered family structure, socioeconomic characteristics, and individual’s perceptions of the built environment, preferences towards the built environment and attitudes towards car ownership (shortened to perceptions, preferences and attitudes from now on). Geographic information system (GIS) technology, combined with on-site measurement, was used for data collection. The multinomial logit model was applied for estimation. Household structure and the built environment (including the perceived built environment and the objective built environment) significantly influence the number of cars in a household. By contrast, preference and attitude attributes have less influence on car ownership. Most of the findings are in line with the literature that considers Chinese urban centers. Nevertheless, new results are also found. For example, rural hukou, and building density have significant positive impacts on household car ownership in China’s rural areas, which is in contrast with their effects on cities. As the first study on rural areas in China, this research provides some insights for rural planners and policymakers to understand better the relationship between built environment and household car ownership. Keywords Car ownership · Rural built environment · Preference and attitude · Multinomial logit model · Exploratory factor analysis · Rural China

7.1 Introduction China, as the most populous country in the world, is undergoing rapid economic and infrastructure development. With the accelerated road construction and economic growth, the need for cars is increasing in China (Belgiawan et al. 2014; Jiang et al. 2017; Pucher et al. 2007). In the past decade, the number of cars per household and the kilometers traveled by cars have exhibited a rapid growth trend. The total number of owned private vehicles reached 163,302,200 by the end of 2016, which was 26.15 times that of 2000 (National Bureau of statistics of the People’s Republic of China 2017). Moreover, the average distance traveled has increased dramatically. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_7

165

166

7 The Impact of the Rural Built Environment …

For example, the average distance traveled by private vehicles in Beijing exceeded 15,000 km/year in 2010 (Li and Zhao 2017). The rapid increase in the number of cars and traveled kilometers leads to high energy consumption, air pollution, traffic congestion, and obesity (Handy et al. 2005; Li and Zhao 2017; Potoglou and Kanaroglou 2008). It has caused various problems, such as photochemical smog pollution, in China (Li and Zhao 2017; Protection 2016)_ENREF_11. The annual growth rate of fossil fuel demand in the transportation sector has reached 10.56% (Wang et al. 2011). A slight increase in car ownership per capita will cause a huge increase in energy consumption and carbon dioxide emissions considering the large population of China. Therefore, the importance of research on car ownership has become increasingly prominent (Aditjandra et al. 2016; Buehler 2017; Li and Zhao 2017). A growing number of researchers have begun to study car ownership in the context of China or Asia (Belgiawan et al. 2014; Ding et al. 2017b; Jiang et al. 2017; Pan et al. 2009; Pucher et al. 2007; Yang 2010). However, they have mainly focused on metropolises, such as Beijing, Shanghai, Shenzhen, Nanjing, and Guangzhou (Wang and Zhou 2017). A massive infrastructure investment has recently occurred in rural areas. By the end of 2016, the mileage of China’s highway routes reached 4,696,263 km, which was 3.35 times that in 2000. In 2016, the fixed-asset investment of rural households was 3.43 times that in 2000, while the disposable income of rural residents was 5.49 times that in 2000 (National Bureau of statistics of the People’s Republic of China 1986–2017). Moreover, the number of cars per 100 households was 17.4, which was 13.18 times that in 2000 (National Bureau of statistics of the People’s Republic of China 1986–2017) (Fig. 7.1). In particular, the number of cars per 100 rural households has increased rapidly in the past 5 years, considering that

Highway Mileage (100,000 kilometers) China's Urbanization Rate (%)

60 50 40

20

Highway Mileage (100,000 kilometers) China's Urbanization Rate (%) Number of Cars Per 100 Rural Households

17.4 56.1 57.4 54.8 52.6 53.7 50.0 51.3 13.3 48.3 45.9 47.0 45.8 47.0 43.9 43.6 44.6 43.0 42.4 40.5 41.8 40.1 41.1 39.1 11.0 37.7 37.3 38.6 36.2 35.8 9.9 34.6

18 16 14 12 10

30 20 10

8 18.7 19.3 17.0 17.7 18.1

6

14.0 1.3

3.8 1.2

1.3

1.4

1.4

1.8

1.8

1.9

2.0

2.3

4.1

2.4

0

4

Number of Cars Per 100 Rural Households

70

2 0

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Fig. 7.1 Urbanization rate, highway mileage and car ownership of rural China (Sources China Statistical Yearbook [2001−2017])

7.1 Introduction

167

China has a rural population of nearly 50%. Therefore, the study on car ownership and its influencing factors in rural China is of utmost significance for achieving energy conservation and emission reduction in transportation. To the best of our knowledge, however, few studies have focused on car ownership in China’s rural areas. This study aims to fill in the aforementioned research gap by exploring the relationship between the built environment and household car ownership using geographic information and survey data from 374 households in rural Sichuan, China. The multinomial logit (MNL) model was applied. It also takes account for socio-demographic factors and individual’s preference and attitude in the context of rapid urbanization and new construction in rural Sichuan. Moreover, a comparison of effects between the perceived built environment, and the objective built environment on the car ownership of rural households was conducted. The remainder of this paper is organized as follows. Section 7.2 discusses the existing literature about household car ownership. Section 7.3 presents the methodology used in the study. The data collection method, variable specification, and model specification are described in Sect. 7.4. The model results are discussed in Sect. 7.5. Finally, Sect. 6 concludes the paper by providing the key findings of the study and policy recommendations.

7.2 Literature Review Car ownership is an important intermediate factor in the study of the built environment and travel behavior _ENREF_17 (Cao et al. 2007). In the past few decades, a large number of studies in Western developed countries have used empirical data and various research methods to explore the relationships among the built environment, car ownership, and travel behavior (Jiang et al. 2017). The built environment, individual’s preference and attitude, and socio-demographic factors are the three main factors that influence household car ownership based on the existing literature (Cao et al. 2009; Handy et al. 2005; Salon 2015).

7.2.1 Built Environment The built environment has been widely accepted as 6D, including density, diversity, design, destination accessibility, distance to transit, and demand management (Ewing and Cervero 2001, 2010; Ewing and Handy 2009; Ewing et al. 2015; Vance and Hedel 2007). Numerous studies have demonstrated that there is a significant relationship between the built environment and car ownership. The reasonable design could reduce travel energy consumption. In particular, there is a significant negative effect of the built environment density on household car ownership. People are less likely to have a car or have fewer cars if they are living in a high density built environment (Holtzclaw et al. 2002; McCormack et al. 2001; Krizek 2003; Ewing et al. 2015).

168

7 The Impact of the Rural Built Environment …

However, road density acts as a proxy for accessibility by cars to external locations in a rural area (Headicar et al. 1994). It may have a positive correlation with household car ownership in the rural area. (Holtzclaw et al. 2002) also found that the impact of design is weaker than those of density and diversity on household car ownership. Destination accessibility is a built environment index at a regional level and generally includes distance to a central business district (CBD) and job accessibility (Krizek 2003; McCormack et al. 2001). Empirical studies have confirmed that households are less likely to own cars as the distance from their residence to a CBD decreases (Miller and Ibrahim 1998; Van Acker and Witlox 2010). Similarly, driving mileage and the number of family vehicles are fewer if a household’s residence is close to the job or business center (Ewing and Cervero 2010; Potoglou and Kanaroglou 2008). By contrast, the household car number increases as the distance of residence to a CBD increases (Robert Cerver and Arrington 2008). A significant relationship has been found also between the distance to transit and household car ownership level. For example, Potoglou and Kanaroglou (2008) found that the number of household cars is fewer if public transportation stations are set up within walking distance from residential areas. Moreover, “the number of the parking lots”, “parking price”, and “distance to parking lot” near residential locations also have impacts on the level of car ownership. Car ownership level is high if low-cost parking lots are available in a community (Guo 2013; Tyrinopoulos and Antoniou 2013). Therefore, Chatman (2008) suggested reducing the number of parking lots to impede the increase in household car ownership level.

7.2.2 Socio-Demographic Factors Socio-demographic factors have been proven to have considerable effects on household car ownership (Li and Zhao 2017). The main socio-demographic factors include household size, income, number of children, number of workers, parking space, and educational level (Bhat and Guo 2007; Cao and Cao 2014; Holtzclaw et al. 2002; Keller and Vance 2013). Many empirical studies have found that income is an important determinant of car ownership. High-income households may have more cars than low-income households (Potoglou and Kanaroglou 2008). Moreover, a large household size, numerous children, and a high number of licensed drivers and workers are positively related to high car ownership level (Li and Zhao 2017; Potoglou and Kanaroglou 2008; Tyrinopoulos and Antoniou 2013). Educational level is closely related to income; thus, education and income have the same effect on household car ownership (Ding et al. 2017a; Fu and Farber 2017). In addition, with the changes in the age structure of the Chinese society, the household structure is changing (Zhao 2014). Together with the urbanization process, China’s household hukou is also changing, which influence on Chinese households

7.2 Literature Review

169

significant (Cheng and Selden 1994). Hukou is a legal document made by the household administration organization of the public security organization to record and retain the basic information of the household population (Baidubaike 2017a). There are two types of hukou, which are urban hukou and rural hukou in China. Urban hukou is the hukou held by people who do not have the right to distribute rural land, while the rural hukou is the opposite (Baidubaike 2017b). Rural urbanization refers to the process of transforming agricultural population into non-agricultural population and gathering into residential areas of urban nature, and transforming rural areas into urban areas or gradually increasing urban elements in rural areas (Baidubaike 2018). In the current situation of population mobility and rapid rural urbanization, there are three types of hukou situations in the rural areas of China. (1) For most rural residents, they hold the rural hukou and live in rural areas for most of the year. (2) There are also some people who hold the rural hukou, but living in urban areas for more than half a year in order to work in cities and towns. They normally go back to the rural area for family visiting or other social events. They become the typically separated population in urban areas, i.e. urban residents who hold rural hukou. (3) In addition, along with China’s rapid urbanization in rural areas, the land of the rural residents with rural hukou was expropriated and these people were turned into urban hukou residents without rural land distribution rights in the process of rural urbanization, but their living area still does not have the city characteristics, or has not yet moved to the urban area. Therefore, they are the people who are living in rural areas with urban hukou in the course of urbanization in China. Many studies have found that residents who hold the rural hukou in urban areas have fewer cars comparing with urban hukou residents (Jiang et al. 2017; Li and Zhao 2017).

7.2.3 Perception, Preferences and Attitudes Relatively few studies have considered psychological determinants, such as personal attitudes and preferences (Ajzen 1991). Steg (2005) stated that people drive occasionally not because they need to, but because they want to. Her research demonstrated that the symbol of a car is an important factor of travel mode choice among people. Van and Fuji (2006) found that attitude variables significantly impact driving and commuting intentions in Japan, China, and Vietnam, but not in Indonesia, Thailand, and the Philippines by conducting a research on the six Asian countries. Belgiawan et al. (2014) studied the car purchase motivation of undergraduates from China, Indonesia, Japan, Lebanon, the Netherlands, Taiwan, and the USA. Their results showed that significant differences exist between developed and developing countries. Students in developed countries are unwilling to buy cars. The expectation of others seems to be an important determinant factor of purchase intention, whereas income and the symbolic significance of a car are less related to purchase intention. Another important factor influencing household car ownership is the environmental attitude. People with a concept of sustainable consumption are likely to live in a neo-traditional or transit-oriented community (Cao and Cao 2014; Cao et al. 2009;

170

7 The Impact of the Rural Built Environment …

Cervero 2007; Robert Cervero and Arrington 2008). Flamm (2009) found that households with high environmental awareness are willing to own few private cars and buy energy-efficient cars. Researchers also found that household car ownership is related to their residential self-selection. People will select residential locations based on their individual psychological determinants, which will affect their car purchase decisions and travel behavior (Handy et al. 2005). In summary, one of the most significant characteristics of the study of the built environment and car ownership in China is that research areas are mainly concentrated in large cities, such as Beijing, Guangzhou, Nanjing, and Jinan (Feng et al. 2014; Jiang et al. 2017; Li and Zhao 2017; Zhao 2011). No existing literature has focused on rural areas.

7.3 Research Methodology 7.3.1 Model Specification 7.3.1.1

Exploratory Factor Analysis (EFA)

First, EFA is used to identify important broad perceptions, preferences, and attitudes to evaluate the built environment and car ownership. EFA is a multivariate statistical analysis method that can convert measured variables to a small number of non-related comprehensive factors. These comprehensive factors reflect the main information of the original measured variables and explain the relationship between measured variables. In particular, EFA studies the condensation of a large number of measured variables into a few factors with the least information loss. The general form of the EFA model is X i = μ + ai1 Fi1 + ai2 Fi2 + . . . + ain Fin + δi (i = 1, 2, . . . , p)

(7.1)

where X i is a random observed variable, Fi is a common factor, ai j (i = 1, 2,…, p) is a factor load, and δi is a special factor part that is excluded from the common factors. The EFA result will be used in the MNL model to investigate the effects of perception, preference and attitude on household car ownership.

7.3.1.2

MNL Model

Models of car ownership are classified based on the underlying choice response mechanism, namely, ordered, e.g., ordered probit and ordered logit (ORL) models, and unordered response, e.g., logit and probit models (Potoglou and Kanaroglou 2008). The ordered response mechanism assumes that the choice for the number of cars to own arises from a 1D latent index, which reflects the propensity of a household to own cars (Bunch 2000). By contrast, the unordered response mechanism is based on

7.3 Research Methodology

171

the assumption that households associate a utility value with each car ownership level and choose the level in which utility is maximized (Potoglou and Kanaroglou 2008). Thus, the global utility maximization framework in the unordered response mechanism adds theoretical and behavioral contexts to the study of household car ownership, whereas no clear theoretical framework exists when estimating models using the ordered response mechanism (Potoglou and Kanaroglou 2008). Bhat and Guo (2007) compared the unordered with the ordered response mechanism and concluded that MNL provides a better representation of the car ownership decision-making process of a household when several datasets are used. Thus, we mainly adopt the MNL model to explore the effects of the rural built environment on household car ownership. Household structure attributes, individual skills, and individual’s preference and attitude factors (the result of EFA) are also inputted into the model. This study classifies car ownership into three levels: 0, 1, and 2 + . Therefore, the utility function for the car ownership of each household can be expressed as 

U (0) = β0 xi0 + εi0 , 

U (1) = β1 xi1 + εi1 , 

U (2) = β2 xi2 + εi2 , yi = ji f Ui ( j) > Ui (q)∀q = j

(7.2)

The MNL component is based on the assumption that a household i, when faced with a finite set {0,1,2} of alternative j, chooses option j, which provides the maximum utility Ui ( j). β denotes the parameters to be estimated for the observed explanatory variables x. ε is assumed to be independently and identically distributed (i.i.d) with identical extreme value distribution. The distribution function is    Fε j = exp − exp −ε j .

(7.3)

On the basis of this specification, the choice probabilities are 

exp(β j xi j ) Pr ob j = Pr obUi ( j) > Ui (q) =  J ∀q = j, j = 1, 2,  q=0 exp(βq x iq ) 2 

Pr ob(yi = j) = 1, 0 ≤ Pr ob(yi = j) ≤ 1, j = 0, 1, 2

(7.4)

j=0

We use the option “0 car” as the reference option, setting the coefficient β0 = 0. This equation is rewritten as Pr obyi = j = Pr obUi ( j) > Ui (q) =

1 , j = 0,   exp(β1 xi1 ) + exp(β2 xi2 ) + 1

172

7 The Impact of the Rural Built Environment … 

exp(β j xi j )

Pr obyi = j = Pr obUi ( j) > Ui (q) =





exp(β1 xi1 ) + exp(β2 xi2 ) + 1

, j = 1, 2 (7.5)

The relative probability of Pr obyi = j to Pr obyi = 0 can be expressed as Pr obyi = j = exp(β j ’xi j ), j = 1, 2 Pr obyi = 0

(7.6)

Thus, the calculation formula for the predicted number of cars is N = Pr ob(yi = 1) ∗ 1 + Pr ob(yi = 2) ∗ 2.

(7.7)

7.3.2 Data Collection and Descriptive Analysis Data were collected from Sichuan Province, China. The total rural population in Sichuan Province reached 41.96 million by the end of 2016, and the urbanization rate was 49.12%. With the increasing of a mileage of the highway, the number of cars per 100 households in rural areas increased 12.38 times compared to the number in 2010. The household car ownership, urbanization rate, and mileage of highway lines of Sichuan rural areas from 2000 to 2016 are shown in Fig. 7.2.

14

Highway Mileage of Sichuan(10,000 kilometers) Urbanization Rate of Sichuan (%)

50

Number of Cars Per 100 Rural Households

40 33.0

34.3

35.6

37.4

38.7

40.2

41.8

43.5

44.9

46.3

49.2 47.7 12.5

10

31.6 32.4 30.2 31.0 8.6 28.3 29.3 26.6 7.5 24.9

30

8 6

22.4 18.9

20

10

12

16.5 9.1

5.3

4

10.9 11.2 11.3 11.3 11.5 2.7

3.0

2

Number of Cars Per 100 Rural Households

Urbanization Rate of Sichuan (%) Highway Mileage of Sichuan(10,000 kilometers)

60

1.7

0

1.5 1.4 1.2 1.2 1.0 1.0 1.1 0.9 0.9 1.2 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

0

Fig. 7.2 Urbanization rate, Highway mileage and car ownership per 100 rural households of sichuan (Sources China Statistical Yearbook [2001−2017])

7.3 Research Methodology

173

To collect the data, a rural household survey had been conducted from October 1st, 2017 to January 31st, 2018. Meanwhile, geographic information system (GIS) data were collected. Sample villages were chosen based on the following two conditions. First, local residents should be willing to participate in the questionnaire survey. Second, at least one road should be directly accessible to the sample villages, and all the roads should be paved. Finally, 7 sample villages were identified, including four traditional scattered-living villages (Da Zhuang, Shuang Yan, Xin Long, and Wu Gang) and three new concentrated-living villages (Yan Jing, Dong Xing, and Shang Teng). Based on the literature, an initial survey was developed, and was sent to the 117 rural students who were coming from Sichuan rural areas (Fu and Farber 2017; Handy et al. 2005; Ma and Dill 2015). The survey includes two parts, which are the village questionnaire and household questionnaire. The village questionnaire requests the general information about the village, which could be filled in by on-site measurement and interviewing with the village chief. The household questionnaire includes questions regarding the household socio-demographic information, individual activity-travel dairy, and built environment preferences, perceptions and attitudes. Later, meetings were organized to get feedbacks. Based on these feedbacks, the questionnaire was revised and improved in order to capture the actual situation in rural Sichuan. 30 students were recruited as surveyors from the 117 rural students. They were trained uniformly before conducting the survey. To guarantee the quality of the data, pre-investigation was organized to investigate the reaction of rural residents to the survey, the length of time required to complete a questionnaire, and the type of token that can entice rural residents to participate in the survey. 10 households from two sample villages (5 each) were randomly selected to participate in the investigation. After the investigation, we found that although rural residents lack the patience to complete the questionnaire, incentives have a significant effect on encouraging rural residents to complete the questionnaire. Each household questionnaire took around 1–1.5 hours to be finished. Finally, 374 completed household questionnaires and 7 completed village questionnaires were collected. The sample villages are shown in Fig. 7.3 with the number of valid questionnaires. Comparing to the China Statistical Yearbook, the sampled household structure and vehicle ownership are more or less the same. However, the value of rural household car ownership was significantly higher than the average value in China. The significant difference was attributed to several reasons. The two main reasons are: (1) In this survey, the household number was counted as the person who was holding a rural hukou, even though the person left the village to work in a city. The cars owned by rural hukou residents who worked in cities have been included. (2) The infrastructure construction and economic development levels in these sample villages were higher than the average levels in China and Sichuan. Therefore, the randomly selected households adequately represent the sample villages. The detailed comparison is shown in Table 7.1. The rural built environment differs from the urban built environment in China. Thus, we regulated the measurement scope by considering the living style of the residents from the sample villages when measuring the built environment indicators. (1)

174

7 The Impact of the Rural Built Environment …

Fig. 7.3 Built environment, regional location and valid questionnaires

7.3 Research Methodology

175

Table 7.1 Sample vs. population characteristics Total population

Householda

Rural Sichuanb

Rural Chinab

1758

419.6 2016:billion

5897.3 2016:billion

–

–

Total permanent residents

1388

Total number of households

374

Average household size

4.70

3.03 (2015)

3.88 (2012)

Average household permanent residents

3.71

–

–

Per capita income(10 k yuan)

1.36

1.13 (2016)

1.24 (2016)

Average household income(10 k yuan)

4.44

–

–

Average number of household cars

0.54

0.13 (2016)

0.17 (2016)

Average number of household autobikes

0.58

0.50 (2016)

0.65 (2016)

Average number of household ebikes

0.71

0.27 (2016)

0.58 (2016)

Average number of household bicycles

0.59

0.31 (2012)

0.79 (2012)

a Data

from face to face household survey between 29th December 2017 to 5th January 2018 China Statistical Yearbook (2013, 2016 and 2017)

b Source

The calculation scope for the built environment indicators was a circle with a 1 km radius from the village center (village committee or neighborhood committee office) for Huojing and Dongxing. Although the respondents from Shangteng also exhibit a centralized living style, Shangteng is a new rural village under construction. Thus, this village demonstrates many scattered-living characteristics. (2) The administrative village boundary is the scope for calculating the built environment indicators in traditional scattered-living villages. This provision is mainly explained by the considerably varying administrative areas of traditional scattered-living villages in Sichuan Province and the inconsistent decentralization degree. The surveyed household samples were not completed within the scope of the circle with a 1 km radius from the center of the administrative village. Two approaches were mainly applied to obtain the basic data of the objective built environment considering the limited geographical information of rural areas in China. First, the researchers conducted on-site measurement using the Baidu navigation app to search and measure the distance to the nearest bus station, train station, bus stop, main road, market, school, health center (hospital), and the city (county) center from the village center. Table 7.2 presents the basic data that were measured on-site. Second, the basic data of buildings and roads were coded from Tencent Street View imagery (map.qq.com) using ArcGIS 10.2. The road and building land information from ArcGIS 10.2 is shown in Fig. 7.4.

176

7 The Impact of the Rural Built Environment …

Table 7.2 The basic data measured on-site Bus Station (KM)

Train Station (KM)

18.20

19.90

0.20

70.00

Shuangyan

16.30

13.40

Xinlong

Dazhuang Wugang

Public transport station (KM)

Main road (KM)

Market (KM)

School (KM)

Health Center (hospital) (KM)

City Center (KM)

2.50

2.50

3.00

0.50

0.05

19.60

16.00

0.00

3.50

2.50

0.20

16.00

0.50

0.50

1.60

1.60

0.60

13.50

13.40

13.40

1.20

0.80

0.80

3.00

4.90

4.90

Dongxing

3.90

16.40

3.90

0.50

0.00

2.10

0.00

10.00

Shangteng

22.40

24.80

0.69

0.69

1.50

1.50

1.60

14.00

0.50

125.00

34.00

0.50

1.50

0.50

1.70

35.00

Yanjing

Fig. 7.4 Road and building land information

7.3.3 Variable Specification We directly asked the respondents about the number of cars that their household owned during the face-to-face interview. Among the 374 sample rural households, the maximum number of cars is 3. The dependent variable was truncated at 2 because the proportion of households with 3 cars was only 1.3%. In addition, the proportions of households with 0, 1, and 2 cars were 54.3%, 38.8%, and 5.6%, respectively. The independent variables can be mainly categorized into three, namely, socio-demographic factors, preference, attitude and the built environment perception, and the objective built environment, based on the literature.

7.3 Research Methodology

7.3.3.1

177

Socio-Demographic Variables

Socio-demographic factors will influence household car ownership decision. Therefore, this study mainly selects the variables of household structure and respondents’ driving skill based on the literature, which has proven that these variables significantly impact household car ownership. The variables included in this study are as follows: (1) household size; (2) number of permanent residents, workers, members under 18, and driver’s license holders; (3) household income; (4) household highest educational level; (5) number of houses and household parking spaces; (6) rural hukou; (7) number of motorcycles, e-bikes, and bicycles; and (8) whether the respondents can drive a motorcycle, e-bike, and bicycle and if they have a driver’s license. Finally, we categorize socio-demographic variables into two categories, which are household structure attributes and personal skills.

7.3.3.2

Perception, Preference and Attitude Variables

To explore the effects of perception, preference and attitude on household car ownership level, 20 indicators of built environment perception and preference and 12 indicators of car ownership attitude, combined with the characteristics of rural areas, are included in the model (Cao et al. 2007; He and Thøgersen 2017). The perception of the built environment is measured based on the respondents’ judgment of the 20 built environment questions. The respondents were asked to judge the truthfulness of the built environment questions based on the current living environment using a five-point Likert-scale. 1 point indicates not true at all, whereas 5 points indicate completely true. The respondents were also asked to answer how important the built environment indicators are under the hypothetical situation of buying a house to express the preference for the built environment using a five-point Likert-scale. 1 point indicates not important at all, and 5 points indicate very important. To measure the attitudes toward car ownership, they were asked to assess 12 statements using a semantic differential scale ranging from 1 (completely disagree with the statement) to 5 (completely agree with the statement). EFA was conducted to identify the important broad perceptions, preferences, and attitudes that rural residents in Sichuan used to evaluate the built environment and car ownership using SPSS 23.0. To test the suitability for factor analysis, we used the Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s test to explore the factorability of individual perception, preference, and attitude variables. All the KMO values are considerably above 0.6, particularly 0.838, 0.864, and 0.679, whereas all the P values are 0.000. The results indicate high correlations among multiple variables, thereby suggesting that the data are suitable for EFA. The items are listed in Tables 7.3, 7.4, and 7.5. The variables with a factor load below 0.5 are considered insignificant and are eliminated. The factor analysis identifies five built environment perception factors, four built environment preference factors, and four car ownership attitude factors. The identified factors respectively explain 63.69, 68.81, and 66.81% of the variance, whereas

178

7 The Impact of the Rural Built Environment …

Table 7.3 Built environment perception component analysis summary: Rotated Component Matrix Component Accessibility

Public space and services

Good neighborhood environment

Physical activity options

Few bad accidents

Convenient to go 0.902 to a market Convenient to go 0.861 to school Convenient to go 0.853 to city Convenient to go 0.780 to transit Convenient to go 0.602 to health center/hospital Spacious public courtyard

0.724

Good road lighting facilities

0.719

Enough parking lots

0.709

Park or public open space

0.702

Good maintenance service

0.625

Quiet living environment

0.736

It’s safe to walk

0.686

Good neighborhood relationship

0.666

A child safe environment

0.609

Good bike path

0.807

Good sidewalk

0.670

No traffic incident

0.812

No criminal incident

0.752 (continued)

7.3 Research Methodology

179

Table 7.3 (continued) Component Accessibility

Public space and services

Good neighborhood environment

Physical activity options

Few bad accidents

Eigen value

3.417

3.067

2.075

1.593

1.312

Proportion of variance explained

18.986

17.040

11.526

8.848

7.290

Cumulative variance explained

18.986

36.026

47.552

56.4

63.69

Summary statistics

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation converged in 5 iterations. N = 374. Factor loading below 0.5 are considered insignificant and not shown in the table. The rest of the factor analysis output can be acquired from the authors

the number of factors is correspondingly reduced to 25.0, 20.0, and 33.33%. That is, only 36.31, 31.18, and 33.18% loss in information were incurred by the 75.0, 80.0, and 66.67% reduction in the number of variables, respectively. Accordingly, the obscure concepts of “built environment perception”, “built environment preference,” and “car ownership attitude” can be better interpreted and represented. Then, the extracted components can be used in the MNL model and can effectively represent built environment perception, built environment preference, and car ownership attitude of the rural household.

7.3.3.3

Objective Built Environment Variables

The objective built environments in rural areas of China are simpler than those in cities. Therefore, this study mainly focuses on the “4Ds + 1S” built environment variables based on the on-site measurement of basic data and GIS extraction data. The “4Ds + 1S” built environment variables include design, diversity, distance to transit, destination accessibility, and living style. Due to the difficulty of obtaining accurate data on the population and households within the circle of 1 km radius from the central residential area, the density variables such as the population density and dwelling unit density were neglected in this study. To be more specific, these built environment variables are described as follows. Here, mu is a unit of land area used in China. In particular, 1 acre = 4046.86 m2 . The design used in this study denotes road density and is calculated as Design index = Totallengthof roads (m)/Total surveyed area (mu)

(7.8)

180

7 The Impact of the Rural Built Environment …

Table 7.4 Built environment preference component analysis summary: Rotated Component Matrix Component Safety It’s safe to walk

0.800

No criminal incident

0.776

No traffic incident

0.770

A child safe environment

0.763

Good village appearance

0.559

Accessibility

Public space

0.504

Convenient to go to a market

0.883

Convenient to go to school

0.876

Convenient to go to city

0.778

Convenient to go to transit

0.665

Enough parking lots

0.782

Enough parking lots

0.757

Park or public open space

0.716

Good road lighting facilities

Good neighborhood and service

0.504

0.570

No economic difference

0.810

Good maintenance service

0.631

Good maintenance service

0.521

Summary statistics Eigen value

3.379

2.848

2.791

1.991

Proportion of variance explained

21.117

17.803

17.442

12.443

Cumulative variance explained

21.117

38.92

56.362

68.805

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation converged in 6 iterations. N = 374. Factor loading below 0.5 are considered insignificant and not shown in the table. The rest of the factor analysis output can be acquired from the authors

7.3 Research Methodology

181

Table 7.5 Car ownership attitude component analysis summary: Rotated Component Matrix Component Independent on car I am fine without a car

0.739

Good public transportation, no need for cars

0.734

Inconvenient without car

−0.677

I need a car to do what I like to do

−0.505

Economy, status symbol

If I have (more) a car, others will pay more attention to me

0.829

If I have (more) a car I will be very happy

0.763

Car is a symbol of economic development

0.550

Cost

Not recommended to buy a car for pollution

0.707

Cars should be taxed on the amount of pollution

0.687

High cost of car purchase and maintenance

0.575

Fuel efficiency and road

Fuel efficiency is an important factor to buy a car

0.834

Need to build more roads to reduce traffic congestion

0.769

Summary statistics Eigen value

2.562

1.944

1.225

1.086

Proportion of variance explained

26.224

14.841

12.987

12.753

Cumulative variance explained

26.224

41.065

54.052

66.805

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation converged in 5 iterations. N = 374. Factor loading below 0.5 are considered insignificant and not shown in the table. The rest of the factor analysis output can be acquired from the authors

Land use mix is consistently used to calculate the diversity index in most related studies. However, land use in rural areas is relatively singular, and we can only read building lands using GIS technology. Thus, building density was used to calculate the diversity of rural land use in this study.     Diversity index = Building land area m2 /Total surveyed area m2

(7.9)

182

7 The Impact of the Rural Built Environment …

Anowar et al. (2014) used a mixed index to calculate the distance to transit. We simplified his formula to calculate the distance-to-transit mix index and the destination accessibility mix index. Distance − to − transitmix index =



[1/(dk + 1)]

(7.10)

k

where k = 1, 2, 3, 4, and dk represents the distance from the village center to the nearest bus station, train station, public transportation station, and main road. Destination accessibility mix index =



[1/(dk + 1)]

(7.11)

k

where k = 1, 2, 3, 4, and dk represents the distance from the village center to the nearest market, school, health center (hospital), and city (county) center. As a consequence of urbanization, the living style of rural residents is gradually shifting from traditional scattered living to urbanized centralized living, and such change in living style directly influences a household’s decision with regard to car ownership. Accordingly, in addition to the influences of the D variables, the influence of living style on rural household car ownership was investigated in this study.

7.3.3.4

Variables Multicollinearity

Multicollinearity problems may cause low significance levels of various spatial variables (Ding et al. 2017a). Therefore, the multicollinearity of the independent variables must be examined. The variable expansion factor (VIF) was used to test for multicollinearity in this study. When VIF value is high, a particular explanatory variable is likely to be represented by a linear function model for other explanatory variables, and thus, multicollinearity problems may occur in the model (Yao et al. 2014). The VIF values of the explanatory variables are considerably below 10 (Table 7.6) according to our analysis, thereby indicating that no multicollinearity problem occurs.

7.4 Results and Discussion The car ownership MNL model of rural household comprises all the explanatory variables described in the previous section. The explanatory variables were individually inputted into the model following the categories of socio -demographic factors, preference and attitude, and the perceived and objective built environment with NLOGIT 5.0. This process demonstrates that each variable category contributes significantly to the model, and the log-likelihood value increases from −330.76 to −164.49. The model result with all the variables is presented in Table 7.7. A total of 35 explanatory variables, excluding the specific constant variable, are adopted, and 2 parameters are

7.4 Results and Discussion

183

Table 7.6 Descriptive statistics summary of variables used in this study Variables

Description

Dependent variable

Mean

Std. dev

Min

Max

0.54

0.67

0.00

3.00

Car ownership Number of cars

Type

VIF

Continuous

Explanatory variables Household structure Household permanent residents

Number of people

4.70

1.58

1.00

12.00

Continuous

3.32

Resident population

Number of people

3.71

1.68

0.00

12.00

Continuous

2.13

Population under 18

Number of people

0.88

0.85

0.00

4.00

Continuous

2.28

4.16

1.25

2.00

8.00

Nominal (7 levels)

1.38

Number of people

2.01

1.23

0.00

11.00

Continuous

1.55

Number of Number of people license holders

1.16

0.95

0.00

4.00

Continuous

2.05

Household income

Yearly, in 10000 RMB

4.44

3.68

0.00

40.00

Continuous

1.48

Number of housing units

Number

1.24

0.70

0.00

7.00

Continuous

1.39

Household parking space

Number

0.55

0.53

0.00

4.00

Continuous

1.39

0.83

0.37

0.00

1.00

Binary: 0-no/1-yes

2.43

Household highest education Household number of workers

Rural hukou Motorcycle ownership

Number of auto bike owned

0.58

0.68

0.00

6.00

Continuous

1.29

E-bike ownership

Number of e-bike owned

0.72

0.65

0.00

4.00

Continuous

1.47

Bicycle ownership

Number of bicycle 0.60 owned

0.74

0.00

4.00

Continuous

1.46

Holing a Whether they have 0.25 driver’s license a drive license

0.44

0.00

1.00

Binary: 0-no/1-yes

1.65

Can ride motorcycle

0.49

0.00

1.00

Binary: 0-no/1-yes

1.69

Personal skills

Whether they can ride motorcycle

0.39

(continued)

184

7 The Impact of the Rural Built Environment …

Table 7.6 (continued) Variables

Description

Mean

Std. dev

Min

Max

Type

VIF

Can ride e-bike

Whether they can ride e-bike

0.64

0.48

0.00

1.00

Binary: 0-no/1-yes

2.43

Can ride bicycle

Whether they can ride bicycle

0.71

0.45

0.00

1.00

Binary: 0-no/1-yes

2.25

Independent on car

See Table 7.4

0.00

1.00

−2.90

3.18

Common factor

1.30

Economy, status symbol

See Table 7.4

0.00

1.00

−2.89

2.53

Common factor

1.22

Cost

See Table 7.4

0.00

1.00

−3.19

3.36

Common factor

1.11

Fuel efficiency See Table 7.4 and road

0.00

1.00

−3.07

2.56

Common factor

1.24

Car ownership attitude

Built environment preference Safety

See Table 7.3

0.00

1.00

−4.08

2.38

Common factor

1.21

Accessibility

See Table 7.3

0.00

1.00

−4.10

2.37

Common factor

1.27

Public space

See Table 7.3

0.00

1.00

−3.59

2.64

Common factor

1.14

Good neighborhood and service

See Table 7.3

0.00

1.00

−3.51

2.07

Common factor

1.16

Accessibility

See Table 7.2

0.00

1.00

−3.33

2.52

Common factor

1.46

Public space and services

See Table 7.2

0.00

1.00

−2.91

2.27

Common factor

1.60

Good neighborhood and service

See Table 7.2

0.00

1.00

−3.09

2.34

Common factor

1.25

Physical activity options

See Table 7.2

0.00

1.00

−2.56

3.93

Common factor

1.19

Few bad accidents

See Table 7.2

0.00

1.00

−2.15

3.23

Common factor

1.15

Perceived Built environment

Objective built environment (continued)

7.4 Results and Discussion

185

Table 7.6 (continued) Variables

Description

Mean

Std. dev

Min

Max

Type

VIF

Building density

See Eq. (7.8)

11.81

5.60

4.76

19.52

Continuous

6.04

Road density

See Eq. (7.9)

2.91

0.69

1.74

3.51

Continuous

2.26

Distance to transit mix index

See Eq. (7.10)

1.70

1.18

0.67

4.46

Continuous

6.06

Destination accessibility

See Eq. (7.11)

1.59

0.42

1.14

2.41

Continuous

3.57

Living style

Whether to live in concentration

0.39

0.49

0.00

1.00

Binary: 0-no/1-yes

2.91

estimated for each explanatory variable. Thus, the variables are χ 2 distributed with 70 degrees of freedom, and the critical χ 2 at the 0.01 significant level for K = 70 is 100.43. The likelihood ratio value of this model is considerably above the critical value, thereby indicating that the null hypothesis (i.e., all parameters are zero) is rejected. In addition, the relatively high rho-squared (R2 ) value also indicates that the model fits well. We specify a set of specific constants for car ownership. Specific constants for one car and two or more cars are statistically significant. In general, specific constants mainly capture unobserved information beyond all explanatory variables (Moshe and Lerman 1985). In the car ownership model, specific constant items refer to the costs associated with car ownership, such as purchase, lease, usage, and maintenance costs (Ryan and Han 1999). This model does not include specific cost variables because the cost information of households with cars is difficult to collect accurately. Moreover, the relationship between unobserved information, such as cost and car ownership, is reflected by specific constants. The utility of rural households with cars decreases as the number of owned cars increases if other conditions are equal. This result is consistent with the findings of other scholars (Potoglou and Kanaroglou 2008; Ryan and Han 1999). Among the family structure variables, the number of household members under the age of 18, the number of household members with driver’s license, total household income, and the number of household-owned parking spaces are continuous variables. The beta values of these variables are all positive. The beta values of these variables are higher for 2 + cars than 1 car (Table 7.7). This finding indicates that the households which have a higher number of household members under 18, household members with driver’s license, total household income, and household-owned parking spaces increase, they are more likely to own a car and have more than 1 cars. This finding agrees with the conclusion of existing research (Choudhary and Vasudevan 2017; He and Thøgersen 2017; Potoglou and Kanaroglou 2008). Interestingly, the rural hukou variable has a positive relationship with car ownership, and the probability of a household with rural hukou owning more than 1 cars is considerably high and statistically significant. This result contradicts the conclusion

186

7 The Impact of the Rural Built Environment …

Table 7.7 MNL, ORP and ORL estimated parameters of household car ownership Variables

Specific constants

MNL

ORP

ORL

0

1

2+

–

Beta

–

– 7.825 0.000 – 27.080 0.000 – 4.170 0.000 – 7.002 0.000

P

Beta

P

Beta

P

Beta

P

Household structure Household size

–

– 0.085 0.655 0.862

0.064 – 0.069 0.444 – 0.124 0.448

Resident population

–

– 0.071 0.616 – 0.573

0.151 – 0.038 0.571 – 0.047 0.694

Population under 18

–

0.338

0.024 0.415

Household highest education

–

– 0.267 0.084 – 0.670

0.129 – 0.087 0.246 – 0.171 0.199

Household number of workers

–

0.118

0.468 – 0.034

0.950 0.054

0.482 0.112

0.419

Number of driver license holders

–

1.173

0.000 2.744

0.000 0.650

0.000 1.145

0.000

Household income

–

0.359

0.000 0.715

0.000 0.170

0.000 0.316

0.000

Number of housing units

–

– 0.017 0.955 0.045

0.947 0.086

0.488 0.162

0.445

Household – parking space

1.109

0.002 3.119

0.013 0.607

0.000 1.018

0.001

Ruralhukou

–

0.496

0.435 6.761

0.037 0.622

0.040 1.076

0.043

Motorcycle ownership

–

– 0.573 0.071 0.351

0.675 – 0.001 0.996 0.010

0.966

Ebike ownership

–

0.122

0.712 0.139

0.373

Bicycle ownership

–

– 0.047 0.851 – 1.227

0.161 – 0.129 0.283 – 0.227 0.285

Holing a driver’s license

–

0.937

0.352 0.438

Can ride motorcycle

–

– 0.299 0.474 – 3.133

0.020 – 0.381 0.036 – 0.628 0.060

Can ride bicycle

–

0.875

0.018 0.770

0.211 1.777

0.690 – 0.384

0.002 0.705

0.324 0.230

0.003

Driving skills 0.037 1.085

0.093 4.852

0.025 0.690

0.001 1.326

0.050

0.002 (continued)

7.4 Results and Discussion

187

Table 7.7 (continued) Variables

MNL

ORP

ORL

0

1

–

Beta

–

– 0.525 0.281 – 0.166

0.915 – 0.390 0.100 – 0.699 0.100

– 0.203 0.274 – 0.948

0.062 – 0.174 0.047 – 0.342 0.031

Economy, – status symbol

0.011

0.953 – 0.328

0.592 – 0.064 0.462 – 0.108 0.473

Cost

– 0.241 0.162 – 0.673

0.255 – 0.154 0.074 – 0.249 0.104

0.083

0.782 0.055

Can ride ebike

2+ P

Beta

P

Beta

P

Beta

P

Car ownership attitude Independent on car

–

–

Fuel – efficiency and road

0.638 0.137

0.511 0.143

0.340

Built environment preference Safety

–

0.139

0.434 – 0.413

0.492 0.014

0.881 0.010

0.952

Accessibility

–

0.071

0.679 0.813

0.236 0.094

0.279 0.145

0.348

Public space

–

0.031

0.849 – 0.271

0.575 0.014

0.866 0.039

0.788

Good – neighborhood and service

– 0.275 0.100 – 0.881

0.091 – 0.130 0.112 – 0.242 0.095

0.005 0.217

Perceived built environment Accessibility

–

0.053

Public space and services

–

0.788 1.889

0.019 0.381

0.020

– 0.088 0.669 – 0.185

0.775 – 0.003 0.980 0.035

0.845

Good – neighborhood environment

– 0.080 0.648 0.732

0.220 0.004

Physical activity options

–

– 0.011 0.949 0.229

0.672 – 0.060 0.472 – 0.100 0.499

Few bad accidents

–

0.472

0.042 0.209

0.019 1.231

0.963 – 0.001 0.995

0.010 0.376

0.008 (continued)

188

7 The Impact of the Rural Built Environment …

Table 7.7 (continued) Variables

MNL

ORP

ORL

0

1

2+

–

Beta

P

Building density

–

0.117

0.097 0.579

0.023 0.081

0.018 0.152

2.470

Road density

–

0.926

0.013 1.339

0.191 0.355

0.038 0.560

0.071

Distance to transit mix index

–

0.149

0.646 – 2.766

0.068 – 0.139 0.375 – 0.291 0.301

Destination accessibility

–

0.137

0.849 – 1.239

0.571 – 0.088 0.792 – 0.262 0.663

Living style

–

0.228

0.671 – 1.984

0.285 – 0.050 0.850 – 0.102 0.828

Beta

P

Beta

P

Beta

P

Objective built environment

Number of observations

374.000

374.000

374.000

Log-likelihood with alternate specific constants (L(C))

– 330.754

– 330.754

– 330.754

Log-likelihood model (L(β))

– 164.493

– 187.461

– 187.637

Likelihood ratio = –2[L(c)–L(β)]

332.522

286.586

286.233

0.503

0.433

0.433

0.450

0.323

0.327

Rho-squared(R2

= 1–[L(β)/L(c)])

Adjusted rho-squared (Adj-R2 = 1–[(L(β)–M)/L(c)])

Note ORL represents ordered logit model, ORP represents ordered probit model

of Yang et al.’s (Jiang et al. 2017) study conducted in Jinan City, China. However, the study area of Yang et al. is different from this study which focuses on the city area. Their findings are consistent with the situation in urban areas, due to rural residents who work in urban areas tend to be indeterminate. An in-depth analysis indicates that our finding is in line with the actual conditions of rural areas in China. In the process of China’s urbanization, some rural residents are urbanized. The lands of these rural residents are expropriated, and they have to rely on government subsidies to live without stable work. In addition, these rural residents face a series of problems, such as choosing a relocation area and determining the best time to transfer. Therefore, urbanized rural households in rural areas have a low utility for household-owned cars and are less likely to own more cars. Vice versa, un-urbanized rural households are more likely to own cars. With regard to household size, this variable agrees with the conclusion of Ding et al. (Ding et al. 2016) that households which have more people are likely to own more cars. Finally, a negative correlation exists between the number of motorcycles and cars owned by a rural household. Under the existing rural built environment, rural households with motorcycles can easily reach surrounding areas. Thus, a motorcycle is a good alternative to a car.

7.4 Results and Discussion

189

Apart from household structure variables, driving skills of the respondents were also collected. The driving skills of the respondents involve four variables: whether they have a driver’s license and whether they can ride a motorcycle, a bicycle, and an e-bike. The first three variables exhibit statistical significance. The beta values of having a driver’s license and being able to ride a bicycle are positive and increase with an increase in the number of household cars. They indicate that in rural Sichuan, a complementary relationship exists between bicycling and driving. Moreover, as the number of household cars increases, the beta values increase which indicates that there is a stronger relationship between whether they can ride/drive, and the households owning more than one cars. By contrast, the beta value of being able to ride a motorcycle is negative and decreases significantly with an increase in car ownership. It indicates that households that use motorcycles are less likely to own cars. As mentioned above that the motorcycle is a good alternative to a car for the rural area of Sichuan. For this type of rural residents or families, either riding a motorcycle or owning a car may satisfy their basic needs for travelling. This finding is consistent with the finding for the number of vehicles in the previous section. For car ownership attitudes, only one common factor variable has a significant relationship with household car ownership. The significant finding is that the less dependent rural residents are on cars, the less likely they are to own cars. This finding is consistent with that of (Cao et al. 2007), i.e., the more people rely on cars, the more likely they are to own cars. Although the variable of a private car as a status and economic symbol is insignificant, the beta sign is extremely interesting. People who own a car as a status and economic symbol are more likely to own only one car than two or more cars. This finding disagrees with the result of Yang et al. Jiang et al. (2017) in their study conducted on a Chinese city. These contrasting findings exhibit the difference in the influence of car ownership attitudes on car ownership between rural and urban households in China. Rural households own one car to satisfy the psychological demand of cars as a status and economic symbol. Owning more cars will have a negative impact because of the increasing costs. However, Chinese urban residents may have travel, work, and other needs. Such households tend to own two or more cars. For the relationship between the preferences for the built environment of rural residents and the car ownership, only one of the four common factors has a significant negative relationship with the household car ownership, i.e., a good neighborhood and service. Here, such relationship indicates that no difference exists in terms of economic conditions among neighbors. Neighbors frequently chat with one another, and the maintenance service for public facilities is excellent. This finding complies with our expectation. People who are living in a good neighborhood service area tend to have a lower possibility to own a car or more than one cars. Their scope of activities might be limited within the village area. To compare relationships between the objective built environments and household car ownership with the perceived built environment and household car ownership, we discuss these two categories together. The beta values of the objective built environment for accessibility to public institutions and public transport is completely opposite to that of the perceived built environment for rural households who have

190

7 The Impact of the Rural Built Environment …

more than one cars. The result shows that the objective access to public transport and public institutions has a negative relationship with the household car ownership of 2+ , which indicates that the more accessible the built environment, the less likely are rural households to own more than one cars. This finding is consistent with the findings of Ding et al. (2017a). The positive relationship between the perceived built environment of rural residents and household car ownership is logical as well. This finding is due to the rapid urbanization of rural Sichuan and the rapid development of rural infrastructure (Figs. 7.1 and 7.2). In the past, traveling was inconvenient for rural residents. In the perceived built environment, however, rural residents experience a fundamental change brought about by being able to drive a car. Therefore, perceived accessibility positively related with the option of owning two or more cars by rural households. The perceived built environment of few bad accidents such as traffic accident and crime incident near homes has a positive relationship with household car ownership. In addition, the relationship is stronger for the households which have more than one cars, which is consistent with our expectation. The actual building density positively influences the car ownership of rural households. This finding is contrary to the results of the existing literature on cities. The general belief is as follows: the higher the building density in a city, the more conducive to physical activities for residents, and the lower the level of household car ownership (Bhat and Guo 2007; Ding et al. 2017a; Li and Zhao 2017; Zegras 2010). However, for the rural areas of Sichuan, the building density is much lower than an urban area. The higher building density in rural areas at this moment reflects the better living conditions with more activity locations. Therefore, it is logical to have a positive relationship between building density and household car ownership after controlling the household income variable. Road density has a significantly positive relationship with owning of one car. It means that rural residents are more likely to hold one car in a higher road density area. The result is logical since, in rural areas of Sichuan, the density of the roads is acting as a proxy for accessibility by cars to external locations. For a higher road density area, people are more likely to connect their cars with external locations and expand their activities. This finding is consistent with the conclusion of (Headicar et al. 1994). Overall, socio-demographic factors have the most significant relationship with the car ownership of rural households based on the estimation of the standardized parameters of various variables, which agree with the finding of the existing literature (Cullinane 2003). Compared with perception, preference and attitude, the characteristics of the built environment (including perceived and objective built environments) have more significant beta values. This result is contrary to the findings of (Cao et al. 2007). Such contradiction could be explained by the evident changes in the built environment during the rapid development of rural areas. Finally, we use the same variables to estimate the ORP and ORL models. The results are presented in Table 7.7 as well. Most parameters have the same signs. MNL is the best model according to the likelihood ratio and Rho-squared. Therefore, in this study, we mainly focused on the results interpretation of the MNL model.

7.5 Conclusion

191

7.5 Conclusion Rapid rural urbanization and new rural construction have brought about rapid changes in rural China, along with a dramatic increase in the energy consumption of rural residents. One of the key factors that affect the travel behavior of rural residents, and which also impacts air pollution, is the rapid increase in car ownership of rural households. This study uses survey data from Sichuan rural households and GIS data for the first time to analyze the relationship between the built environment and rural household car ownership, after controlling for individual’s preference and attitude, and socio-demographic factors using the MNL model. Household characteristics, such as the number of members under the age of 18, the number of members with driver’s license, total household income, the number of household-owned parking spaces, and household size significantly and positively affect household car ownership. This finding is consistent with the conclusions in existing research. Rural hukou, however, has a significant positive impact on household car ownership, which is contrary to the research findings in the urban context but is in line with the actual situation in rural China. The individual driving skills of the respondents significantly affect household car ownership. Driving and cycling skills positively influence household car ownership, whereas knowing how to ride a motorcycle has negative impacts on household car ownership. In addition, individual’s preference and attitude also impact the car ownership of rural households to a certain extent. Individuals who do not rely on private cars and who live in a well serviced neighborhood are less likely to own a car. This finding is consistent with existing research. The built environment (including perceived and objective built environment) has a significant impact on car ownership of rural households. The perceived build environment where there are few or no traffic accidents and where crime incidents near homes are minimal, positively affects household car ownership. There is a significant positive relationship between perceived accessibility and household car ownership. It might indicate that being able to travel by car considerably impacts the convenience perception of rural residents. The relationship between objective accessibility and household car ownership of 2+ is consistent with the literature, which confirms that the more accessible the built environment, the less likely are rural households to own more than one car. In addition, there is a positive relationship between objective building density and household car ownership. A similar relationship has been found regarding road density and one car ownership as well. In Sichuan’s rural areas, building density and road density can, to a certain extent, represent the development level of infrastructure. Although rural areas are urbanizing rapidly, the rural built environment indicators still lag behind their counterparts in cities. For example, the average building density of the 20 districts in Jinan City is 0.407 (Yang 2013), whereas the average building density of the sample villages selected in this study is only 0.119. Infrastructure will be continuously improved with an increase in road density and building density in rural areas. Incentivizing rural residents to own as few cars as possible will greatly impact the future quality of rural development.

192

7 The Impact of the Rural Built Environment …

References Aditjandra, P. T., Cao, X. Y., & Mulley, C. 2016. Exploring changes in public transport use and walking following residential relocation: A British case study. Journal Of Transport And Land Use, 9(3): 77–95. Ajzen, I. 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2): 179–211. Anowar, S., Yasmin, S., Eluru, N., Miranda-Moreno, L. F., 2014. Analyzing car ownership in Quebec City: a comparison of traditional and latent class ordered and unordered models. Transportation, 41(5): 1013–1039. Baidubaike. 2017a. Hukou. Baidubaike. 2017b. Uuban Hukou. Baidubaike. 2018. Rural urbanization. Belgiawan, P. F., Schmocker, J. D., Abou-Zeid, M., Walker, J., Lee, T. C., Ettema, D. F., & Fujii, S. 2014. Car ownership motivations among undergraduate students in China, Indonesia, Japan, Lebanon, Netherlands, Taiwan, and USA. Transportation, 41(6): 1227–1244. Bhat, C. R., & Guo, J. Y. 2007. A comprehensive analysis of built environment characteristics on household residential choice and auto ownership levels. Transportation Research Part B: Methodological, 41(5): 506–526. Buehler, R., Pucher, J., Gerike, R., & Gotschi, T. 2017. Reducing car dependence in the heart of Europe: Lessons from Germany, Austria, and Switzerland. Transport Reviews, 37(1): 4–28. Bunch, D. S. 2000. Automobile demand and type choice. Handbook of transport modelling, 463: 479. Cao, J. S., & Cao, X. S. 2014. The impacts of LRT, neighbourhood characteristics, and self-selection on auto ownership: Evidence from Minneapolis-St. Paul. Urban Studies, 51(10): 2068–2087. Cao, X., Mokhtarian, P. I., & Handy, S. I. 2007. Cross-sectional and quasi-panel explorations of connection between the built environment and auto ownership. Environment and Planning A, 39: 830–847. Cao, X. Y., Mokhtarian, P. L., & Handy, S. L. 2009. The relationship between the built environment and nonwork travel: A case study of Northern California. Transportation Research Part a-Policy and Practice, 43(5): 548–559. Cervero, R. 2007. Transit-oriented development’s ridership bonus: a product of self-selection and public policies. Environment and Planning A, 39(9): 2068–2085. Chatman, D. G. 2008. Deconstructing development density: Quality, quantity and price effects on household non-work travel. Transportation Research Part a-Policy and Practice, 42(7): 1008–1030. Cheng, T. J., & Selden, M. 1994. The origins and social-consequences of China Hukou system. China Quarterly (139): 644–668. Choudhary, R., & Vasudevan, V. 2017. Study of vehicle ownership for urban and rural households in India. Journal of Transport Geography, 58: 52–58. Cullinane, S. 2003. Hong Kong’s low car dependence: Lessons and prospects. Journal of Transport Geography, 11: 25–35. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. 2017a. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100: 65–80. Ding, C., Wang, D. G., Liu, C., Zhang, Y., & Yang, J. W. 2017b. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part a-Policy And Practice, 100: 65–80. Ding, C., Wang, Y., Tang, T., Mishra, S., & Liu, C. 2016. Joint analysis of the spatial impacts of built environment on car ownership and travel mode choice. Transportation Research Part D: Transport and Environment. Ewing, R., & Cervero, R. 2001. Travel and the built environment—A synthesis. Land Development And Public Involvement In Transportation (1780): 87–114.

References

193

Ewing, R., & Cervero, R. 2010. Travel and the built environment. Journal of the American Planning Association, 76(3): 265–294. Ewing, R., & Handy, S. 2009. Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban Design, 14(1): 65–84. Ewing, R., Tian, G., Goates, J. P., Zhang, M., Greenwald, M. J., Joyce, A., Kircher, J., & Greene, W. 2015. Varying influences of the built environment on household travel in 15 diverse regions of the United States. Urban Studies, 52(13): 2330–2348. Feng, J. X., Dijst, M., Wissink, B., & Prillwitz, J. 2014. Understanding mode choice in the Chinese context: The case of Nanjing metropolitan area. Tijdschrift Voor Economische En Sociale Geografie, 105(3): 315–330. Flamm, B. 2009. The impacts of environmental knowledge and attitudes on vehicle ownership and use. Transportation Research Part D-Transport and Environment, 14(4): 272–279. Fu, L., & Farber, S. 2017. Bicycling frequency: A study of preferences and travel behavior in Salt Lake City, Utah. Transportation Research Part A: Policy and Practice, 101: 30–50. Guo, Z. 2013. Does residential parking supply affect household car ownership? The case of New York City. Journal of Transport Geography, 26: 18–28. Handy, S., Cao, X., & Mokhtarian, P. 2005. Correlation or causality between the built environment and travel behavior? Evidence from Northern California. Transportation Research Part D: Transport and Environment, 10(6): 427–444. He, S. Y., & Thøgersen, J. 2017. The impact of attitudes and perceptions on travel mode choice and car ownership in a Chinese megacity: The case of Guangzhou. Research in Transportation Economics, 62: 57–67. Headicar, P., Curtis, C., Education, P., & Research Services, L. 1994. Residential Development and Car-Based Travel: Does Location Make A Difference? Holtzclaw, J., Clear, R., Dittmar, H., Goldstein, D., & Haas, P. 2002. Location efficiency: Neighborhood and socioeconomic characteristics determine auto ownership and use—Studies in Chicago, Los Angeles and San Francisco. Transportation Planning and Technology, 25(1): 1–27. Jiang, Y., Gu, P., Chen, Y., He, D., & Mao, Q. 2017. Influence of land use and street characteristics on car ownership and use: Evidence from Jinan, China. Transportation Research Part D: Transport and Environment, 52: 518–534. Keller, R., & Vance, C. 2013. Landscape pattern and car use: Linking household data with satellite imagery. Journal of Transport Geography, 33: 250–257. Krizek, K. J. 2003. Residential relocation and changes in urban travel - Does neighborhood-scale urban form matter? Journal of the American Planning Association, 69(3): 265–281. Li, S., & Zhao, P. 2017. Exploring car ownership and car use in neighborhoods near metro stations in Beijing: Does the neighborhood built environment matter? Transportation Research Part D: Transport and Environment, 56: 1–17. Ma, L., & Dill, J. 2015. Associations between the objective and perceived built environment and bicycling for transportation. Journal of Transport & Health, 2(2): 248–255. McCormack, E., Rutherford, G. S., & Wilkinson, M. G. 2001. Travel impacts of mixed land use neighborhoods in Seattle, Washington. Land Development and Public Involvement in Transportation (1780): 25–32. Miller, E. J., & Ibrahim, A. 1998. Urban form and vehicular travel—Some empirical findings. Land Use And Transportation Planning and Programming Applications (1617): 18–27. Moshe, B.-A., & Lerman, S. 1985. Discrete choice analysis: Theory and application to travel demand. Massachusetts: Massachusetts Institute of Technology. National Bureau of statistics of the People’s Republic of China 1986–2017. China Statistical Yearbook. National Bureau of statistics of the People’s Republic of China 2017. China Statistical Yearbook. Pan, H. X., Shen, Q., & Zhang, M. 2009. Influence of urban form on travel behaviour in four neighbourhoods of Shanghai. Urban Studies, 46(2): 275–294. Potoglou, D., & Kanaroglou, P. S. 2008. Modelling car ownership in urban areas: a case study of Hamilton, Canada. Journal of Transport Geography, 16(1): 42–54.

194

7 The Impact of the Rural Built Environment …

Protection, M. o. E. 2016. 2015 Annual Report of Automobile Air Pollution in China. In: Protection, M. O. E(Ed.). Pucher, J., Peng, Z. R., Mittal, N., Zhu, Y., & Korattyswaroopam, N. 2007. Urban transport trends and policies in China and India: Impacts of rapid economic growth. Transport Reviews, 27(4): 379–410. Robert Cervero,G. B., & Arrington. 2008. Vehicle trip reduction impacts of transit-oriented housing. Journal of Public Transportation, 11 (3): 1–17. Ryan, J., & Han, G. 1999. Vehicle-ownership model using family structure and accessibility application to Honolulu, Hawaii. Transportation Research Record: Journal of the Transportation Research Board (1676): 1–10. Salon, D. 2015. Heterogeneity in the relationship between the built environment and driving: Focus on neighborhood type and travel purpose. Research In Transportation Economics, 52: 34–45. Steg, L. 2005. Car use: Lust and must. Instrumental, symbolic and affective motives for car use. Transportation Research Part a-Policy and Practice, 39(2–3): 147–162. Tyrinopoulos, Y., & Antoniou, C. 2013. Factors affecting modal choice in urban mobility. European Transport Research Review, 5(1): 27–39. Van Acker, V., & Witlox, F. 2010. Car ownership as a mediating variable in car travel behaviour research using a structural equation modelling approach to identify its dual relationship. Journal of Transport Geography, 18(1): 65–74. Van, H. T., & Fuji, S. 2006. A cross asian country analysis in attitudes toward car and public transport. Transportation Studies: Sustainable Transportation, Proceedings Of the 11th International Conference of Hong Kong Society for Transportation Studies: 803–803. Vance, C., & Hedel, R. 2007. The impact of urban form on automobile travel: Disentangling causation from correlation. Transportation, 34(5): 575–588. Wang, D., Chai, Y., & Li, F. 2011. Built environment diversities and activity–travel behaviour variations in Beijing, China. Journal of Transport Geography, 19(6): 1173–1186. Wang, D., & Zhou, M. 2017. The built environment and travel behavior in urban China: A literature review. Transportation Research Part D: Transport and Environment, 52: 574–585. Yang, J. W. 2010. Spatial and social characteristics of urban transportation in Beijing. Transportation Research Record (2193): 59–67. Yao, B., Hu, P., Lu, X., Gao, J., & Zhang, M. 2014. Transit network design based on travel time reliability. Transportation Research Part C: Emerging Technologies, 43: 233–248. Yang, Y. 2013. A Research on Neighborhood Built Environment and Household Travel Energy in Jinan, Tsinghua University. Zegras, C. 2010. The built environment and motor vehicle ownership and use: Evidence from Santiago de Chile. Urban Studies, 47(8): 1793–1817. Zhao, P. 2011. Car use, commuting and urban form in a rapidly growing city: evidence from Beijing. Transportation Planning and Technology, 34(6): 509–527. Zhao, P. J. 2014. Private motorised urban mobility in China’s large cities: the social causes of change and an agenda for future research. Journal of Transport Geography, 40: 53–63.

Chapter 8

The Impact of the Rural Built Environment on the Travel Mode Preferences of Rural Residents

Abstract This chapter combines on-site measurement methods, geographic information system (GIS) technology, and activity diary survey to obtain basic data regarding the built environment and the daily activities of rural residents. The multinomial logit (MNL) model is used to explore the relationship between the rural built environment and the travel mode choice of rural residents. Results show that building density significantly positively affects private car trips. This finding challenges earlier urban built environment research conducted in the urban context. An increase in road density increases the travel frequency of electric bicycles and motorcycles. Accessibility perception and preferences positively affect the likelihood of choosing to walk. Safety and neighborhood harmony perception positively affect the travel frequency of motorcycles and private cars. Rural residents who prefer a safe living environment are likely to choose walking as their means of daily travel. Despite the considerable achievements in the construction of rural roads, the frequency of public transportation remains low for rural residents. Therefore, additional attention should be given to the investment and construction of public transport facilities commensurate with the level of rural urbanization. Keywords Urbanization · Rural built environment · Travel mode choice · MNL model · Rural Sichuan

8.1 Introduction As a large agricultural country, China has nearly 600 million people living in rural areas. The rapid transition from a rural society to an urban society has become a development characteristic of China in recent years (Qian et al. 2013). Since 2005, the Party Central Committee has proposed that “building a new socialist countryside is a major historical task in the process of China’s modernization” (Gao 2016). China has entered the stage of rapid development in new rural construction and urbanization. In

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_8

195

196

8 The Impact of the Rural Built Environment on … 1000

10000

894.9

8000

7434.5

10546.7 10755.8 10409.8 900 9840.6 9964.9 10041.67 9554.4 9089.1 800 7886.0

700

5951.8

6000

5123.3

4436.2 3940.6 3362.7 4000 3123.2 3201.0 2904.32976.6 294.1 325.2 246.6 2000 216.1 176.7 159.3 147.4 141.2 151.6

678.86 666.2 600

656.4 566.9

534.5

570.8 509.7

560.3

582.2

500 400 300 200 100 0

0

Sichuan' rural household fixed assets investment

Chinese rural household fixed assets investment

12000

Chinese rural household fixed assets investment Sichuan' rural household fixed assets investment

Fig. 8.1 Fixed asset investment of rural households

2020, China will build a well-off society in all aspects.1 Rural construction is the key point and is driven by the upsurge of building a moderately prosperous society that covers all aspects. By the end of 2018, the fixed asset investment of rural households in China was 1004.167 billion yuan, which was 2.55 times that in 2005. The growth of the fixed asset investment of rural households in Sichuan has become increasingly significant since 2000 (Fig. 8.1). In the context of rural reform, China is experiencing rapid rural urbanization. Although the urbanization rate in Sichuan is lower than that in China, their growth rates are coordinated (Fig. 8.2). China’s urbanization rate is estimated to reach approximately 65% by 2030 (Meng et al. 2018). Rapid rural urbanization has brought tremendous changes in the rural construction environment of China. The total number of rural roads has reached 4.05 million km at present. The access rate of hardened roads to towns and villages was 99.64%. The access rate of hardened roads to incorporated villages was 99.47% (Ministry of transport 2018). With the changes in the rural built environment, the number of household vehicles of rural residents has also changed (Fig. 8.3). The number of bicycles per 100 households has been declining annually in the past decade. By contrast, the number of motorcycles and cars has been increasing. By the end of 2018, the number of cars per 100 households was 14.83 times that in 2005. Studies have shown that the demand for private cars in rural areas will continue to increase rapidly and steadily in the next 5 years (Zhang and Zhang 2016). People use different travel modes to participate in activities in various locations. The built environment creates a difference in the spatial distribution of activities, and such difference causes variations in travel mode choices. The built environment directly affects people’s participation in activities, and consequently, 1 According to China’s economic development goals, By 2020, China’s per capita GDP will exceed

us $6,000, reaching the level of a middle-income country. To build a well-off society. in all respects is the goal of the party and the country by 2020.

8.1 Introduction

197

70.00% 65.00% 58.52%

60.00%

59.58%

57.35% 56.10% 53.73%

55.00%

54.77%

52.57%

52.29%

51.27%

50.79% 49.21%

50.00% 43.90%

45.00%

44.94%

45.68%

46.59%

47.69%

47.50% 46.30% 44.90% 43.53% 41.83% 40.18%

38.70%

40.00%

37.40% 35.60% 34.30%

35.00% 30.00%

Urbanization rate of China Urbanization rate of sichuan

60 50

52.06 42.98

44.93

46.31 41.88 42.18

40 30.73

30

24.85

33.90

40.85

51.50

49.50 49.50

45.17

40.15 44.23

17.80

36.53 37.88

13.40

30.87

27.85

12.50

26.18 16.45 8.56

20

7.54 5.27

10 0

1.20

1.38

1.18

1.23

1.73

2.70

49.00

3.00

1.53

20 18 16 14 12 10 8 6 4 2 0

The number of Private cars per 100 rural households

The number of bicycles or motorcycles per 100 rural households

Fig. 8.2 Urbanization rates of Sichuan and China

The number of bicycles per 100 rural households The number of motorcycles per 100 rural households The number of Private cars per 100 rural households

Fig. 8.3 Number of vehicles per 100 rural households in Sichuan

their travel behavior (Hägerstrand 1986). During rural urbanization, a percentage of rural population will be transformed into urban population. Studies have indicated that the original travel modes of rural residents will change accordingly with the

198

8 The Impact of the Rural Built Environment on …

changes in the built environment. For example, existing nonmotorized and motorcycle trips will be replaced by public transportation and private cars for long-distance travel (Luo and Zhang 2017). Moreover, an increase in the daily activities of rural residents will promote the use of private cars (Zhang and Zhang 2016). In the context of rapid rural urbanization, this study explores the influence of the built environment on the travel mode choice of rural residents, providing a theoretical basis for further rural planning and construction practice. For this study, 10 sample villages in Sichuan Province were selected to explore the influences of the rural built environment, built environment perception and preferences, social demographic factors, and daily travel-related variables on the travel mode choice of rural residents using the multinomial logit (MNL) model. The remainder of this paper is structured as follows. Section 8.2 reviews the relevant literature. Section 8.3 discusses the research method, and the data collection and sample selection processes. Section 8.4 presents the empirical results and discussion. The final section concludes the study.

8.2 Literature Review China’s rapid urbanization has brought major changes in the built environment (Zhang and Lin 2012), generating growing research interest (Cai et al. 2020; Cao et al. 2014). In the process of urbanization, traditional villages on the urban perimeter have been absorbed and transformed within expanding cities, with rural populations and the rural labor force flowing into those cities (Deng et al. 2019). The result is that the relentless construction of rural housing, infrastructure and roads dramatically and irreversibly impact the rural built environment. The built environment is different from the natural environment; it represents a large number of artificial human environments, including urban landscapes. Handy et al. (2002) divided the built environment into three parts: land use patterns, urban design, and transportation systems. Cervero and Kockelman (1997) believed that the built environment is composed of three dimensions: density, diversity, and design. This idea is called “3D” theory. Ewing et al. (2008) added the dimensions of destination accessibility and distance to transit, expanding 3D theory into 5D theory. In the 1960s, researchers began exploring the influence of the built environment on travel behavior. A series of social problems emerges as a city expands. The majority of researchers have attempted to explore the relationship between the built environment and travel behavior to avoid environmental pollution and traffic congestion. Studies have found that physical activity-oriented urban planning is beneficial for the health of residents (Sun et al. 2017). Travel behavior includes travel mode, frequency, distance, time, purpose, and travel chain (Chen et al. 2016). Three primary travel modes are available: cars, public transportation, and nonmotorized travel (Cao et al. 2009). Numerous studies have shown that different built environment measurement indicators exert varying effects on travel behavior. Existing research on the urban built environment shows that land

8.2 Literature Review

199

use indicators influence the frequency of motorized and nonmotorized travel behavior (Boarnet and Crane 2008; Chatman 2008). Sun and Dan (2015) used an MNL model and found that an increase in the population density of a residential area, the land use mix, the proportion of crossroads, and accessibility, can reduce the probability of residents choosing to travel by private car. This is consist with the findings of Cervero and Duncan (2003), where residents prefer to use public transport. Since public transport hubs in such areas are usually concentrated, taking public transport proves to be convenient (Cervero and Kockelman 1997; Ding et al. 2014; Zhao 2013). Similarly, the distance to transit has a significant impact on residents’ choice of transportation. Cervero (2007) and Zhao (2013), using a Nested Logit Model and an MNL model respectively, found the closer the residential area is to public transport points, the more likely people are to choose public transport when commuting. During weekdays, the probability of choosing to self-drive is negatively correlated with business density near a workplace (Cervero 2002), whereas the relationship is the opposite during weekends (Ta et al. 2015). In Bogotá, small blocks with high road density and high connectivity are likely to promote walking and biking trips (Cervero et al. 2016). To accurately study travel behavior, the specific characteristics of the built environment that are closely related to travel behavior must be analyzed. For example, people are likely to choose walking in a built environment that is close to a school, has a high population density, and has a good walking route to the school with safety infrastructure and traffic (Saelens and Handy 2008). In addition to objective built environment factors, travel attitudes and socioeconomic factors may also determine the travel mode choice of residents (Ohnmacht et al. 2009; Scheiner 2010). For example, an individual’s preference for social activities in a neighborhood will caused him/her to select nonmotorized over motorized transport (Cao et al. 2009). Liu et al. (2015) studied the relationship between built environment perception and walking; they found that improving the walkable built environment may help increase adult walking times. Etminani-Ghasrodashti and Ardeshiri (2015) used structural equation modeling (SEM) to evaluate the influence of built environment preferences on travel mode choice. They determined that built environment preference affects the travel frequency of individuals during nonworking days. With regard to research on the built environment in rural China, Wan and Ng (2016) proposed a “Mountainous Rural Built Environmental Sustainability Assessment Framework” based on sustainable rural development models. The design and planning processes for rural built environments should be flexible, and methods should be diverse. A design should consider changes in the long-term lifestyles and needs of different groups within the village (Gao 2016). In a study of travel behavior, the rural residents of Haining, Zhejiang prefer electric bicycle trips, followed by private car trips, and then walking (Kong and Yao 2015). Meanwhile, the rural residents of Northern Jiangsu prefer electric bicycle trips, followed by walking and then motorcycle trips (Chen and Zhu 2013). Given that the majority of young people travel to work, only children and the elderly have limited mobility, resulting in relatively few trips in rural areas (Yang et al. 2014). Yang et al. (2014) found that rural residents in economically developed areas make more trips, with shorter distance and time, than those in economically underdeveloped areas. Feng et al. (2010) used MNL

200

8 The Impact of the Rural Built Environment on …

model predicted the choice of transport mode of rural residents in terms of social demographic attributes, travel distance, and travel time. They found that personal attributes and total family income are less likely to affect travel mode choice; by contrast, travel distance exerts greater impact on travel mode choice. Zheng and Chen (2017) determined that the major daily travel modes for relocated rural residents in Shanghai were bus, electric bus, and walking. They proposed that bus routes and stops should be added to improve bus service. An increase in rural residents’ daily travel may increase their demand for urban and rural public transport. Government planning can be expected to have a significant impact on public transport (Wang and Cao 2017). Zhuo et al. (2019) and Huang et al. (2019) assert urban and rural public transport coordination to be an emerging problem requiring the optimization of rural public transport routes and facilities. The effects of rural built environment on rural household vehicle and private car ownership were investigated by Ao et al. (2018) and Ao et al. (2019b). These studies found that the rural built environment significantly affects household vehicle and private car ownership. Subsequently, Ao et al. (2019a) further determined that the rural built environment exerts considerable impact on travel carbon emissions. However, these authors did not find any literature on the relationship between the built environment and travel mode choice in rural China.

8.3 Research Methodology 8.3.1 Model Specification Scholars studying the built environment and travel modes have used a variety of methods and models. These included the MNL model (Feng et al. 2010; Kamruzzaman et al. 2013; Sun and Dan 2015; Thrane 2015; Zhao 2013), the SEM model (Etminani-Ghasrodashti and Ardeshiri 2015; Sun et al. 2017), the Nested logit model (Cervero 2007), the linear regression model (Ta et al. 2015), and the ordered probit regression model (Feng 2017). With the maturation of travel behavior research, scholars have demonstrated a preference for the MNL model in examining travel mode choice (Bouscasse et al. 2019; Hasnine et al. 2018; Kamruzzaman et al. 2013; Thrane 2015). However, a search of the relevant literature on rural travel in China reveals minimal research on the relationship between the rural built environment and travel mode choice of residents. This limitation restricts the further planning of countryside transition to urbanization. Thus, a disaggregate model was used to explore the relationship between the rural built environment and the travel mode choice of rural residents.

8.3 Research Methodology

8.3.1.1

201

Model Introduction

A disaggregate model studies the relationship between travel behavior and built environment in terms of individuals. The MNL model and the multinominal probit (MNP) model are the most common disaggregate models (Cao 2015). The former can obtain the probability of an individual choosing different travel modes through the calculation of the utility function; this model has been applied to many travel mode studies (Feng et al. 2010; Sun and Dan 2015). The current work selects individual rural residents as the research object, and thus, the non-lumped model is more suitable for this study. The general model for travel mode as a discrete variable will result in a large deviation, and this study aims to examine the factors that affect the different travel mode choice of rural residents, compare the probability of residents’ choice of each travel mode, and apply the MNL model to establish the relationship between the built environment and the travel mode choice of rural residents.

8.3.1.2

MNL Model

Assume that the effect of the nth traveler choosing the ith travel mode is U ni , and J n is the scheme set. Then, i ∈ Jn , Uni = Vni + εni , and Vni = β  X nk , where εni is the random error term, and X nk is the Kth factor that influences the selection behavior of the X nk traveler. β  is the parameter to be estimated. Subsequently, the probability that the nth traveler chooses the ith travel mode is calculated as follows: Pn (i) = Pr ob(Uni ≥ Un j , j ∈ Jn , i = j) Pr ob(Vni + εni , j ∈ Jn , i = j)    = Pr ob Vni + εni ≥ max Vn j +εn j

(8.1)

j∈Jn

If each random term εni is assumed to follow an independent homology distribution, then f (ε1 , ε2 , . . . , εn ) = Πn g(εn )

(8.2)

where g(εn ) is the distribution function that corresponds to the nth traveler. If g(εn ) is assumed to follow the double exponential distribution, then the probability of selecting the ith mode of travel in jn is calculated as follows: exp(Vin ) 1  =    − exp V exp V jn jn j∈J n j∈J n j∈J n exp(Vin )    exp β X nk =  j∈J n exp(β X nk )

ρin = 

(8.3)

202

8 The Impact of the Rural Built Environment on …

8.3.2 Factor Analysis Factor analysis is an important branch of modern statistics; it is a statistical method for simplifying and analyzing high-dimensional data. The principle behind factor analysis is to reduce dimensionality and consolidate complex variables into a few factors (Zhang et al. 2019). This method provides the least missing information, guaranteeing the integrity of the original information to the utmost extent (Zeng et al. 2019). The theoretical model for factor analysis is generally expressed as follows: Suppose N samples and P indicators are available X = (X1 , X2 , . . . . . . , Xn )T is a random vector. The common factor that should be determined is F = (F1 , F2 , . . . . . . , Fm )T . X 1 = a11 F1 + a12 F2 + · · · + a1m Fm + ε1 , X 2 = a21 F1 + a22 F2 + · · · + a2m Fm + ε2 , ... X p = a p1 F1 + a p2 F2 + · · · + apm Fm + ε p .

(8.4)

  The previous equations compose the factor model. Matrix A = aij is the factor load matrix, and aij is the factor loading. The essence of A is the correlation coefficient between common factor Fi and variable X i . ε is a special factor that represents factors other than the common factor.

8.3.3 Data and Sample Collection Existing research on the travel behavior of rural residents is based on national data or focused on the areas of Jiangsu, Zhejiang, and Shanghai (Kong and Yao 2015; Zheng and Chen 2017); by contrast, nearly no research has been conducted on the travel behavior of rural residents in Southwest China. Located in Southwest China, Sichuan has a population of 83.41 million, with the rural population accounting for 47.71% of the provincial population, or 39.79 million (China Statistical Yearbook 2018). The rapid change in the rural built environment and the change in the household vehicle ownership of residents (Fig. 8.3) exert an unknown impact on the travel behavior of rural residents. Considering this condition, the current study selects 10 typical villages in Sichuan Province as examples to explore the impact of the rural built environment on the travel patterns of rural residents. This work provides theoretical basis for sustainable rural construction and reference suggestions for rural transport organizations. Prior to this study, the research group conducted two field surveys on a rural infrastructure construction in Sichuan Province and the satisfaction of rural residents with such construction (Ao et al. 2017). The surveys were performed in October

8.3 Research Methodology

203

2016 and April 2017. On basis of the aforementioned research, the present study determines the following basic principles for sample village selection. 1. The rural transport facility is in good condition. In particular, the roads in the sample village should reach every rural household, enabling private cars to travel conveniently. 2. The sample village is a representative for a certain type of Sichuan countryside. On the basis of the previous research, the authors classify the rural areas in Sichuan into three major and six minor categories (Ao et al. 2018; Ao et al. 2019a; Ao et al. 2019b) Therefore, rural representativeness is the second basic principle of a sample village. 3. The research group has good communication channels with the rural residents in the sample village, and villagers exhibit a welcoming attitude or are willing to participate in the questionnaire survey. The current study conducted two questionnaire surveys, and the selected sample villages met the three basic principles. In the first questionnaire survey, which was conducted in January 2018, seven sample villages were identified, namely, five scattered-living and two concentrated-living villages. In the second questionnaire survey, conducted in August 2018, three concentrated-living villages were identified. Figure 8.4 shows the geographical location and distribution of the sample villages. Questionnaire survey, on-site measurement, and geographic information system (GIS) data were collected. In the questionnaire survey, several researchers used the activity center as the center of the sample village and conducted divergent household visits to residents to ensure the uniform distribution of respondents. Rural residents were randomly selected to complete the collection of data regarding personal, family, and socioeconomic conditions. Built environment perceptions and preferences were obtained through face-to-face interviews. For the on-site measurement and GIS data survey, the researchers measured travel distance from the activity center to major destinations and public transportation facilities using Baidu navigation. Building and road densities were extracted using GIS technology at a later stage. Finally, 428 valid questionnaires and 1,242 travel records of rural residents were collected. Table 8.1 provides additional information about demographic data.

8.3.4 Variables Description This study focuses on the effects of the built environment on the travel mode choice in rural Sichuan. The dependent variable represents the travel mode choice of rural residents, namely, private car, public transportation(in the context of rural Sichuan, public transportation refers to the buses connecting villages, towns and cities), motorcycle, electric bicycle, bicycle, walking, and other modes (primarily refer to tractors and small trucks). The questionnaire survey data show that rural residents are likely to choose walking for their daily travel, with a rate of up to 48.60%, followed by electric bicycle (18.80%), private car (10.10%), motorcycle (8.00%), bicycle (6.00%),

204

8 The Impact of the Rural Built Environment on …

Fig. 8.4 The geographical location and distribution of sample villages

8.3 Research Methodology

205

Table 8.1 Basic data of samples

Basic data of respondents and their families

Rural Sichuan

Total population

2010

408.5 billion (2017)a

Total permanent residents

1605

397.95 billion (2018)b

Total number of households

428

–

Travel records

1242

–

a Source b Source

China statistical yearbook Sichuan statistics bureau

and public transportation (5.40%). Only 3.10% of rural residents chose other travel modes. The built environment, travel distance/time, and sociodemographic variables are considered key factors that influence travel mode choice in a large number of studies (Stead 2001; Van Acker et al. 2010). Psychological factors, such as environmental perception and preference, also affect the travel mode choice of residents (Cao et al. 2009; Liu et al. 2015).

8.3.4.1

Sociodemographic Variables

Sociodemographic variables, including gender, age, education level, household type, highest education level of family members, household size, and total household income, are considered in this study. The number of cars, motorcycles, electric bicycles, and bicycles owned per household is also considered. The survey data show that the number of female respondents is more than that of male participants, female makes up as much as 60.50% of the respondents which is shown in Table 8.2. It can be explained by a common phenomenon in rural China wherein the economy is less developed than that in cities; hence, the rural male labor force goes to work, leaving the women to take care of their families. In addition, the phenomenon of “left-behind women” exists in China’s rural areas. Studies show that the number of left-behind women exceeded 50 million in 2011 (Hao et al. 2018). Moreover, 69.9% of the respondents are over 40 years old because young and middle-aged residents are also likely to leave for work. Although the sample villages are selected from rural Sichuan, some rural residents have become urban residents during rural urbanization. Therefore, 14.00% of the respondents are holders of urban hukou. For education level, 35.3 and 30.80% of the respondents have attained junior high school and primary school levels, respectively. Moreover, the highest education level of family members is higher than that of the respondents. Most of the sample households have one or two members who are working, with a total household income between 1,000 and 10,000 yuan. Most of the sample households have one vehicle, with electric bicycles

206

8 The Impact of the Rural Built Environment on …

Table 8.2 Sociodemographic variables Personal attributes

Frequency (%) Personal attributes

Gender

Male

39.50

Female

60.50

Age

Under 20 years old

Education level

Frequency (%)

Rural hukou

86.00

Town hukou

14.00

Primary school and below

36.60

20–29 years old

14.70

Junior high school

35.30

30–39 years old

10.30

High school

17.10

40–49 years old

26.20

Skill-related training

6.10

50–59 years old

24.10

Bachelor’s degree

4.70

60 years old or older

19.60

Master’s degree and above

0.20

Family attributes Family member highest education

5.10

Hukou type

Frequency

Primary school and below

5.10

Junior high school High school

Family attributes 0

44.80

28.10

1

44.20

32.90

2

9.00

Skill-related training

10.60

3 or more

2.00

Bachelor’s degree

21.90

Master’s degree and above Number of 1 or less people 2 working 3 4 or more

Number of cars

Frequency

1.20

Number of 0 motorcycles 1

48.55

0.20

2

5.31

29.31 43.48 18.11 9.1

3 or more Number of electric bicycles

3.1

45.17

0.97

0

32.30

1

56.30

2

10.50

Total Less than ¥10,000 household ¥10,000–50,000 income ¥50,000–100,000

54.7

¥100,000–150,000

5.5

2

9.26

¥150,000–200,000

1.2

3 or more

1.77

More than ¥200,000

1.2

34.3

3 or more Number of bicycles

0.90

0

39.53

1

49.44

8.3 Research Methodology

207

accounting for the highest proportion, followed by bicycles, cars, and motorcycles. Table 8.2 provides additional detail regarding sociodemographic information.

8.3.4.2

The Objective Built Environment

In the extant literature, scholars have studied the most extensive built environment indicators: density, diversity (Ao et al. 2019b; Cervero et al. 2016; EtminaniGhasrodashti and Ardeshiri 2015; Wang et al. 2019; Zhao and Li 2019), and accessibility (Ao et al. 2019b; Cao et al. 2009; Etminani-Ghasrodashti and Ardeshiri 2015; Feng 2017). On the basis of previous studies, the objective built environment indicators used in the current work include design (road density), diversity (building density and number of nearby markets), destination accessibility mix, and distance to transit mix indexes. All the built environment indicators are calculated from the basic data collected via the questionnaire survey, on-site measurement, and GIS extraction. The research scope varies in accordance with different types of sample villages. The research scope of scattered-living villages (Da Zhuang Village, Wu Gang Village, Xin Long Village, Shuang Yan Village, and Shang Teng New Village) is the village administrative boundary. By contrast, the research scope of concentrated-living villages (Dong Xing Community, Huo Jing Village, Jin Ning Village, Tian Du Village, and Wu Xi Community) is a circle with the travel activity center as the core and 1 km as the radius (Ao et al. 2019b), as shown in Fig. 8.5. Table 8.3 provides the basic data used to calculate the built environment indicators. The Objective Built Environment Indicators Are Calculated from Table 3. the Formulas for Rural Road and Building Densities Are as Follows (Ao et al. 2019b) Road density = total length of road/total area of research land, Building density = area of construction land/total area of research land. Distance to transit and destination accessibility are simplified from the research of Anowar et al. (2014), Ao et al. (2019a) and Ao et al. (2019b). The two formulas are expressed as follows: Distance to transit = k {1/(dk + 1)}, k = 1, 2, 3, 4. “k” = 1, 2, 3, 4, “d”, and “k” represent the distances from the village center to the nearest bus station, train station, public transportation station, and main rural road, respectively.  Destination accessibility = k {1/(dk + 1)}, k = 1, 2, 3, 4. “k” = 1, 2, 3, 4, “d”, and “k” represent the distances from the village center to the nearest market, school, health center/hospital, and city/county center, respectively.

8.3.4.3

Built Environment Perceptions and Preferences

Built environment perception and preference significantly influence travel mode choice. This study explores such influences on the travel mode choice of rural residents. A total of 20 questions about built environment perception were formulated. Table 8.4 provides additional details regarding the 20 perception items and the

208

8 The Impact of the Rural Built Environment on …

Da

Shang

Xin Long

Wu Gang

Huo Jing

Jin Ning

Shuang

Dong Xing

Tian Du

Wu Xi Fig. 8.5 Boundary range of each sample village

exploratory factor analysis (EFA) results. The respondents were asked to judge the realization degree of 20 built environment items. A five-point Likert scale was used to measure the degree of agreement with the actual situation of built environment items. Points from 1 to 5 indicate “not at all” to “completely” agree. EFA was used to analyze built environment perception using SPSS 23.0, and a Kaiser–Meyer–Olkin (KMO) value of 0.839 illustrated the applicability of EFA. Factor rotation was performed

8.3 Research Methodology

209

Table 8.3 Basic data for the destination accessibility mix and distance to transit mix indexes Distance to the nearest transit/destination from the village activity center (km)

Da Zhuang Wu Gang

Bus station

Train station

18.20

19.90

Public transport station

Main road

Market

School

Health center (hospital)

City center

2.50

2.50

3.00

0.50

0.05

19.60

0.20

70.00

16.00

0.00

3.50

2.50

0.20

16.00

Shuang Yan

16.30

13.40

0.50

0.50

1.60

1.60

0.60

13.50

Xin Long

13.40

13.40

1.20

0.80

0.80

3.00

4.90

4.90

Dong Xing

3.90

16.40

3.90

0.50

0.00

2.10

0.00

10.00

Shang Teng

22.40

24.80

0.69

0.69

1.50

1.50

1.60

14.00

Yan Jing

0.50

125.00

34.00

0.50

1.50

0.50

1.70

35.00

Jin Ning

3.00

60.00

3.00

2.00

3.00

3.00

3.00

27.00

Tian Du

6.00

40.00

0.10

0.10

0.60

1.00

1.00

6.00

Wu Xi

4.00

4.00

4.00

0.10

0.10

0.10

0.20

20.00

using the maximum variance method, and variables with a factor load below 0.4 were deleted. The cumulative interpretation variance ratio of common factors is 62.964%. Lastly, 20 variables were grouped into 5 common factors. The common factors of built environment perception are accessibility, public space and services, safety and neighborhood harmony, less accidents, and neighborhood differences (Table 8.4). The respondents were asked to state the importance of the 20 questions in selecting a new residence to reflect their built environment preference. Table 8.5 presents additional information regarding the preference items and the EFA results. Five-point Likert scales and EFA were used to analyze the built environment preference of rural residents. The KMO test result (0.878) showed that EFA is suitable for analyzing built environment preference data. Factor rotation was also performed using the maximum variance method, and variables with a factor load below 0.4 were deleted. The cumulative interpretation variance ratio of common factors is 70.176%. Lastly, the 20 variables were reduced to 5 common factors. The built environment preference factors are accessibility, good roads and public spaces, safety, neighborhood harmony and maintenance services, and less accidents.

8.3.4.4

Daily Travel-Related Variables

Daily activity, travel distance, and travel time influence travel mode choice (Feng et al. 2010; Pan et al. 2007). Daily activities were divided into three categories (Yang and

210

8 The Impact of the Rural Built Environment on …

Table 8.4 Component matrix of built environment perception Does the Composition environment in Accessibility which you are currently living in meet the following conditions? 1

Public space and services

Safety and neighborhood harmony

Less accidents

2

3

4

Neighborhood differences

5

It is convenient to go to school from my current home.

0.863

0.103

0.090

0.038

−0.034

It is convenient to go to the market from my current home.

0.905

0.128

0.109

−0.018

0.060

It is convenient to go to the city from my current home.

0.885

0.089

0.093

0.021

0.083

It is convenient to go to public transportation stations (bus, subway, train stations) from my current home.

0.768

0.181

0.021

−0.057

0.272

It is convenient to go to a health center (hospital, clinic) from my current home.

0.533

0.306

0.113

−0.046

−0.131

The bike paths are good near my current home.

0.027

0.770

0.074

0.091

−0.021

The walkway is good near my current home.

0.112

0.763

0.154

0.082

−0.068

(continued)

8.3 Research Methodology

211

Table 8.4 (continued) Does the Composition environment in Accessibility which you are currently living in meet the following conditions? 1

Public space and services

Safety and neighborhood harmony

2

3

Less accidents

Neighborhood differences

4

5

Parks or other public open spaces are available near my current home.

0.193

0.665

0.042

−0.055

0.184

A wide public courtyard is available near my current home.

0.219

0.650

0.115

0.053

0.204

Village appearance is good in my current home.

0.233

0.476

0.432

0.037

−0.008

Public facility maintenance service is good in my current home.

0.197

0.579

0.421

0.072

0.212

A wide public courtyard is available near my current home

0.175

0.505

0.016

−0.045

0.639

Sufficient parking space is available near my current home.

0.138

0.487

0.249

−0.168

0.517

(continued)

Timmermans 2015): mandatory (e.g., working, including field work; going to school; and picking up and dropping off children), maintenance (e.g., going to markets and shopping), and leisure (e.g., walking and participating in recreational activities) activities. The statistics of daily activities shows that mandatory activities account for the largest proportion (40.80%), followed by maintenance (32.30%) and leisure (26.90%) activities.

212

8 The Impact of the Rural Built Environment on …

Table 8.4 (continued) Does the Composition environment in Accessibility which you are currently living in meet the following conditions? 1

Public space and services

Safety and neighborhood harmony

Less accidents

Neighborhood differences

2

3

4

5

It is safe to walk near my current home.

0.063

0.335

0.752

0.004

0.108

It is safe for children to play outside my current home.

0.145

0.186

0.683

0.127

0.321

The living −0.018 environment is quiet in my current home.

0.259

0.700

0.146

−0.129

Neighborhood relationship is good in my current home.

0.108

−0.130

0.614

0.022

0.110

No crime incident occurs near my current home.

0.051

−0.016

0.057

0.867

0.020

No traffic −0.091 incident occurs near my current home.

0.151

0.155

0.835

0.118

No economic −0.019 difference near my current home.

−0.044

0.151

0.209

0.672

Summary statistics Eigenvalues Percentage of variance

3.511

3.510

2.463

1.596

1.514

17.553

17.550

12.313

7.980

7.568 (continued)

8.3 Research Methodology

213

Table 8.4 (continued) Does the Composition environment in Accessibility which you are currently living in meet the following conditions?

Cumulative variance percentage

Public space and services

Safety and neighborhood harmony

Less accidents

Neighborhood differences

1

2

3

4

5

17.553

35.104

47.416

55.396

62.964

Extraction method: Principal component analysis Rotation method: Varimax with Kaiser normalization Rotation converged in six iterations

The travel distances of rural residents are short, and the proportion of travel distance within less than 1 km is up to 54.7%, and the proportion of travel time within 20 min is 86.40%. The reason for these results is that most of the respondents are farming or working nearby. The peak period of rural residents’ daily travel in the morning is 6:00–10:00, which is similar to existing rural research (Liu et al. 2016). The evening peak period is 16:00–19:00.

8.3.5 Multicollinearity of Variables When numerous independent variables exist in a research, a correlation may occur between two or more independent variables. Such correlation is called multicollinearity. When the collinearity trend of an independent variable is evident, it will seriously affect model fitting (Ma, 2008). In the current study, the variance inflation factor (VIF) is used to test multicollinearity. The larger the variance inflation factor, the stronger the multicollinearity. When VIF is greater than 10, a strong multicollinearity exists between variables; such situation is unacceptable. The VIF values of the independent variables in this study are less than 4 (Table 8.6); thus, no multicollinearity exists between independent variables.

8.4 Results and Discussion SPSS 23.0 was used to estimate the MNL model. Walking was set as the reference alternative because it represents the largest proportion among all travel mode alternatives. The results show that the data fits the model well [P = 0.000 (< 0.050)],

214

8 The Impact of the Rural Built Environment on …

Table 8.5 Component matrix of built environment preference When you select Composition a new home, will Accessibility the following conditions be important? 1

Good roads and public spaces

Safety

2

3

Neighborhood harmony and maintenance services

Less accidents

4

5

−0.04

−0.032

It should be convenient to go to school from the new home.

0.797

0.085

0.268

It should be convenient to go to the market from the new home.

0.860

0.114

0.183

0.171

−0.012

It should be convenient to go to the city from the new home.

0.854

0.206

0.058

0.152

0.108

It should be convenient to go to public transportation stations (bus, subway, train stations)from the new home. transportation stations (bus, subway, train station)

0.704

0.228

−0.028

0.293

0.173

It should be convenient to go to a health center (hospital, clinic) from the new home.

0.610

0.264

0.076

0.214

0.309

The bike paths should be good near the new home.

0.395

0.671

−0.171

0.177

0.234

The walkway should be good near the new home.

0.323

0.658

−0.014

−0.023

0.444

(continued)

8.4 Results and Discussion

215

Table 8.5 (continued) When you select Composition a new home, will Accessibility the following conditions be important? 1

Good roads and public spaces

Safety

Neighborhood harmony and maintenance services

Less accidents

2

3

4

5

Parks or other public open spaces should be available near the new home.

0.238

0.723

0.163

0.211

0.101

A wide public courtyard should be available near the new home.

0.180

0.622

0.225

0.431

−0.167

Sufficient parking area should be available near the new home.

0.070

0.588

0.463

0.237

−0.023

Road lighting should be complete near the new home.

0.031

0.505

0.595

0.210

0.106

Village appearance should be good in the new home.

0.126

0.468

0.501

0.247

0.347

Walking should be safe near the new home.

0.194

0.090

0.809

0.056

0.198

Children should be able to play safely outside the new home.

0.124

−0.033

0.795

0.079

0.291

(continued)

and most of the exploratory variables exerted significant influences on travel mode choice; exceptions being, the number of nearby markets, accessibility in perception, public space and services, less accidents, and departure time. Table 8.7 provides additional details. Table 8.8 presents the travel mode choice results of the MNL model.

216

8 The Impact of the Rural Built Environment on …

Table 8.5 (continued) When you select Composition a new home, will Accessibility the following conditions be important? 1

Good roads and public spaces

Safety

Neighborhood harmony and maintenance services

Less accidents

2

3

4

5

The living environment should be quiet in the new home.

0.153

0.182

0.442

0.515

0.190

Neighborhood relationship should be good in the new home.

0.175

0.176

0.190

0.784

0.156

No economic difference should exist near the new home.

0.197

0.167

−0.041

0.799

0.085

Public facility maintenance service should be good near the new home.

0.125

0.379

0.386

0.521

0.160

No crime incident should occur near the new home.

0.101

0.070

0.348

0.190

0.796

No traffic incident should occur near the new home.

0.116

0.169

0.329

0.145

0.785

Summary statistics Eigenvalues Percentage of variance

3.528

3.076

2.940

2.465

2.027

17.641

15.378

14.699

12.324

10.134 (continued)

8.4.1 Sociodemographic Variables All of the sociodemographic variables exert significant effects on travel patterns, and all of the variables’ significance levels are below 0.05 (Table 8.7). For the sociodemographic variables (Table 8.7), the probability of choosing a private car (−0.595), public transportation (−0.948), a motorcycle (−0.332), an electric bicycle (−0.341), and a bicycle (−0.376) is significantly negatively correlated with age compared with

8.4 Results and Discussion

217

Table 8.5 (continued) When you select Composition a new home, will Accessibility the following conditions be important?

Cumulative variance percentage

Good roads and public spaces

Safety

Neighborhood harmony and maintenance services

Less accidents

1

2

3

4

5

17.641

33.019

47.718

60.042

70.176

Extraction method: Principal component analysis Rotation method: Varimax with Kaiser normalization Rotation converged in 19 iterations Table 8.6 Multicollinearity test of independent variables Explanatory variables

Explanatory variables

Sociodemographic variables Gender

Objective built environment 1.162

Building density

1.477

Age

2.110

Road density

2.052

Hukou type

1.645

Distance to transit mix index

2.167

Education level

2.281

Destination accessibility mix index

1.752

Highest education level of family members

1.292

Number of nearby markets

3.59

Number of household members

1.259

Built environment perception

Total household income

1.628

Accessibility

1.314

Number of cars

1.425

Public space and services

1.257

Number of motorcycles

1.233

Safety and neighborhood harmony

1.242

Number of electric bicycles

1.426

Less accidents

1.126

Number of bicycles

1.279

Neighborhood differences

1.275

Travel-related variables

Built environment preferences

Activity

1.482

Accessibility

1.274

Travel distance

1.413

Good roads and public spaces

1.234

Departure time

1.334

Safety

1.166

Travel time

1.386

Neighborhood harmony and maintenance services

1.222

Less accidents

1.098

218

8 The Impact of the Rural Built Environment on …

Table 8.7 Likelihood ratio test of independent variables Effect

Constant

Model fitting condition

Likelihood ratio test

Simplified model −2 log likelihood

Bangla

Degree of freedom

Significant

2132.871a

0

0

.

2176.071

43.2

6

0.000

Sociodemographic variables Age Gender

2179.383

46.512

6

0.000

Hukou type

2154.622

21.751

6

0.001

Education level

2159.087

26.216

6

0.000

Highest education level of family members

2154.836

21.965

6

0.001

Number of people working

2153.425

20.555

6

0.002

Total household income

2157.848

24.977

6

0.000

Number of cars

2198.81

65.939

6

0.000

Number of motorcycles

2212.948

80.077

6

0.000

Number of electric bicycles

2189.364

56.493

6

0.000

Number of bicycles

2165.496

32.625

6

0.000

Building density

2155.973

23.102

6

0.001

Road density

2172.271

Objective built environment 39.4

6

0.000

Distance to transit mix index 2147.43

14.559

6

0.024

Destination accessibility mix 2146.039 index

13.168

6

0.040

2141.361

8.49

6

0.204

Accessibility

2141.642

8.771

6

0.187

Public space and services

2136.146

3.275

6

0.774

Safety and neighborhood harmony

2177.494

44.624

6

0.000

Number of nearby markets Built environment perception

Less accidents

2135.277

2.406

6

0.879

Neighborhood differences

2152.335

19.464

6

0.003

Accessibility

2145.345

12.475

6

0.052

Good roads and public spaces

2144.435

11.564

6

0.072

Built environment perception

(continued)

8.4 Results and Discussion

219

Table 8.7 (continued) Effect

Model fitting condition

Likelihood ratio test

Simplified model −2 log likelihood

Bangla

Degree of freedom

Significant

Safety

2147.943

15.072

6

0.020

Neighborhood harmony and maintenance services

2164.488

31.617

6

0.000

Less accidents

2153.687

20.816

6

0.002

Travel-related variables Daily activity

2312.809

179.938

12

0.000

Travel distance

2516.428

383.557

6

0.000

Departure time

2137.608

4.738

6

0.578

Travel time

2201.286

68.415

6

0.000

The chi-square statistic presents the difference in −2 log likelihood between the final and reduced models. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0 This reduced model is equivalent to the final model because omitting the effect does not increase the degree of freedom

walking. Thus, the older the traveler, the greater the willingness to choose walking for daily travel, which is consistent with the result of (Feng 2017), i.e., that older individuals are likely to walk. Men are more willing to choose private cars (1.281), motorcycles (1.367), and other travel modes (1.250) over walking than women. This finding is consistent with the research result in urban and rural areas (Cao et al. 2009; Kong and Yao 2015). Individual education level is significantly negatively correlated with the probability of choosing public transportation (−0.780), motorcycle (−0.294), electric bicycle (−0.415), and bicycle (−0.215), but insignificantly correlated with the driving a private car probability. Cao et al. (2009) found that education level exerts a significant negative influence on driving a private car probability in urban areas. The proportion of bachelor’s degree and above is only 11.0% in this study, which is considerably lower than the education level of residents in urban areas. Respondents who hold a rural hukou are more willing to choose a private car (0.149) for daily travel than walking. The reason for this finding is that the living standards of rural residents and the construction of rural infrastructure have improved with the development of rural urbanization and construction. In Ao et al. (2019b), residents who hold a rural hukou are willing to choose vehicles for daily travel, which is consistent with our research result. The number of people working is significantly negatively correlated with the probability of choosing a motorcycle (−0.184), an electric bicycle (−0.349), or other travel modes (−1.024). The higher the total family income, the greater the possibility of choosing a private car (0.697) for daily travel. This finding is consistent with existing research results (Etminani-Ghasrodashti and Ardeshiri 2015). The number of private cars owned by a rural household exerts a positive influence on the probability of choosing a private car (1.538). Thus, the

0.012

0.246

0.000

0.003

0.978

0.486

Highest education level −0.412 of family members

−0.184

0.697

1.538

−0.165

Total household income

Number of private cars

Number of motorcycles −0.820

0.008

Number of household members

Number of electric bicycles

Number of bicycles

9.376

0.126

Building density

Road density

Objective built environment

0.452

0.149

Education level

0.614

0.005

0.002

0.001

2.203

Rural hukou (Urban hukou = ref.)

0.000

0.000

1.281

−0.595

0.400

−0.876 0.192

0.823

0.472

0.431

−0.259 0.183

0.000

0.699

0.209

0.268

−1.397

0.118

0.311

−0.195

0.095

0.000

−0.780 0.272

0.087

0.436

0.000

0.064

1.087

0.285

−0.948

−4.139

0.000

−7.518

Male (Female = ref.)

Age

Sociodemographic variables

Intercept

B

B

P

Public transport

P

Car

0.796

5.436

−0.285

−1.192

0.002

0.138

0.261

0.000

0.000

0.190

1.650

0.685

0.089

0.028

0.457

0.111

0.932

0.000

0.010

0.000

P

−0.358

−0.343

0.091

−0.294

0.052

1.367

−0.332

−7.764

B

Motorcycle

Table 8.8 Estimation result of MNL model parameters for travel mode choice

0.995

9.719

−0.613

0.784

−0.411

0.493

0.424

−0.349

0.037

−0.415

0.409

−0.193

−0.341

−6.266

B

0.000

0.000

0.001

0.000

0.028

0.006

0.010

0.002

0.700

0.002

0.221

0.391

0.001

0.000

P

Electric bicycle

0.117

6.054

0.631

0.065

−0.176

0.652

0.066

0.002

0.809

0.505

0.203

0.167

−0.367 0.320

0.831

0.462

0.244

0.713

0.711

0.015

0.094

P

0.031

0.096

−0.215

0.156

−0.119

−0.376

−3.226

B

Bicycle

0.765

6.022

−0.471

−0.410

−0.546

1.947

−1.447

−1.024

−0.698

0.080

(continued)

0.154

0.391

0.337

0.506

0.298

0.001

0.021

0.008

0.012

0.827

0.020

−1.744

0.149

0.804

0.014

0.424

0.958

P

1.250

B

Others

220 8 The Impact of the Rural Built Environment on …

0.474

0.016

0.226

Safety and neighborhood harmony

Less accidents

Neighborhood differences

0.029

0.273

−0.359

Safety

0.116

0.017

Good roads and public spaces

0.924

0.199

0.927

0.015

0.395

Accessibility

Built environment preference

0.164

0.640

0.081

−0.435

−0.081

0.890

0.222

−0.669

0.062

B

B

0.970

−0.007

−0.050 0.781

0.438

0.262

−0.214

0.138

0.004

−0.585

−0.135

0.223

−0.248

−0.28

0.501

−0.126

0.271

0.082

0.271

0.599

0.315 −0.926

P

Public transport

P

Car

Public spaces and services

Accessibility

Built environment perception

Number of nearby markets

Destination accessibility

Distance to transit

Table 8.8 (continued)

0.587

−0.208

−0.107

−0.112

−0.303

−0.073

0.835

−0.086

−0.081

0.206

−0.169

B

Motorcycle

0.136

0.507

0.526

0.062

0.621

0.000

0.633

0.638

0.378

0.688

0.284

P

0.205

−0.181

−0.357

0.192

−0.082

0.188

−0.014

−0.006

−0.077

−0.776

−0.329

B

0.078

0.140

0.003

0.112

0.464

0.114

0.912

0.959

0.622

0.011

0.394

P

Electric bicycle

0.005

0.171

−0.118

0.268

−0.156

0.036

0.012

0.418

−0.278

0.171

−1.427

B

Bicycle

0.975

0.323

0.491

0.100

0.317

0.817

0.945

0.016

0.259

0.707

0.021

P

0.068

−0.416

−0.376

0.773

−0.064

−0.206

0.035

0.017

−0.083

1.102

−2.291

B

Others

(continued)

0.794

0.201

0.184

0.008

0.810

0.415

0.906

0.957

0.868

0.190

0.048

P

8.4 Results and Discussion 221

1.616

2.892

Mandatory activities (leisure activities = ref.)

Daily activities (leisure activities = ref.)

0.000

0.003

0.122

0.000

0.140

−1.223

Time spent on the road

Departure time

0.000

0.637

−0.077

2.271

0.501

0.111

0.747

2.574

0.984

0.000

0.155

0.882

−0.037 0.037

0.000

0.355

0.856

1.932

0.171

0.035

B

P

Public transport

B

P

Car

Travel distance

Travel-related variables

Less accidents

Neighborhood harmony and maintenance services

Table 8.8 (continued)

2.885

2.819

−1.295

−0.056

1.635

0.000

0.000

0.000

0.551

0.000

0.003

0.000

−0.677

0.456

P

B

Motorcycle

0.072

1.581

0.258

0.111

3.422

3.085

−0.84

B

0.000

0.000

0.000

0.257

0.000

0.023

0.328

P

Electric bicycle

1.869

1.177

−0.14

0.075

1.604

0.448

−0.239

B

Bicycle

0.000

0.011

0.483

0.388

0.000

0.006

0.138

P 0.622

1.674

4.024

−1.694

0.021

−0.737

−0.150

B

Others

0.093

0.000

0.003

0.885

0.290

0.466

0.036

P

222 8 The Impact of the Rural Built Environment on …

8.4 Results and Discussion

223

higher the number of available private cars, the higher the possibility of choosing them for daily travel. The number of motorcycles owned by a household is significantly positively correlated with choosing a motorcycle (1.650) for daily travel and negatively correlated with choosing other travel modes. The relationship between the number of bicycles owned by a household and choosing a bicycle (0.631) for travel is similar to that of motorcycles. These results are consistent with the research finding that people with bicycles are likely to ride them.

8.4.2 The Objective Built Environment Four out of five built environment indicators exert significant effects on travel mode choice, except for number of nearby markets (Table 8.7) and building density (Table 8.8), which has the greatest influence. In particular, building density is significantly positively correlated with choosing a private car and an electric bicycle for daily travel (Table 8.8). The most likely travel mode choice is an electric bicycle (9.719), followed by a private car (9.376) and then a bicycle (6.054). An electric bicycle is flexible and convenient, and it causes less congestion. The higher the building density, the more the residents are willing to choose electric bicycles for daily travel. This result is consistent with that of (Sun and Dan 2015). In many studies based on the urban built environment, the greater the building density, the less likely the urban residents will choose private cars for daily travel (Cervero and Kockelman 1997; Etminani-Ghasrodashti and Ardeshiri 2015). This finding is contrary to the conclusion of our study. The rural built environment is relatively different from the urban built environment, and building density in rural areas is considerably lower than that in cities (Ao et al. 2019b). In rural Sichuan, areas with high building density have good infrastructure for driving, and when the income level of rural residents is sufficient, they are willing to buy cars (Ao et al. 2018). Therefore, the results are consistent with the situation in rural Sichuan. The effects of road density on travel mode choice are similar to those of building density, but to a less extent. Road density is positively correlated with the probability of choosing other travel modes apart from walking. The reason is that good rural road connectivity improves the possibility and convenience of riding electric bicycles and motorcycles. The probability of choosing an electric bicycle (0.995) is slightly greater than that of choosing a motorcycle (0.796), as shown in Table 8.8. The reason for this result is that the cost of electric bicycles is lower than that of motorcycles. This result is consistent with those of Kong and Yao (2015) and Chen and Zhu (2013), i.e., that the electric bicycle is the first choice of rural residents for daily travel in Haining, Zhejiang, and northern Jiangsu. Public transport accessibility is significantly negatively correlated with the probability of choosing a bicycle and other travel modes, indicating that farmers are willing to walk for their daily travel. Public transport accessibility actively encourages rural residents to choose public transportation for daily travel, and walking is the best complementary travel mode with public transportation. Destination accessibility is significantly negatively correlated with the probability of choosing electric

224

8 The Impact of the Rural Built Environment on …

bicycles (−0.776) for daily travel, with an insignificant influence on the probability of choosing other travel modes. Therefore, when destination accessibility is good, rural residents are willing to choose walking as their daily transport mode.

8.4.3 Built Environment Perceptions and Preferences Built environment perception and preference are subjective built environment indicators, and relevant studies have been conducted on the basis of the psychological feelings of residents (Badland et al. 2012; Feng 2017). Perception aspects (Table 8.8): When daily destinations and transport facilities are easily accessible, the probability of choosing a bicycle (0.418) is the highest among rural residents, followed by walking, which is consistent with the result of Cao et al. (2006) on walking behavior. When rural residents feel that their environment is safe and their neighborhood is harmonious, they are willing to choose motorcycles (0.835) for their daily travel, followed by private cars (0.474), with public transport as the last priority (−0.585). When rural residents feel no economic difference in the neighborhood, they are likely to choose other travel modes (0.773) (which mostly refers to tractors and small trucks), followed by walking. Villages without economic difference among rural residents, such as Wugang and Shangteng, frequently have distinctive industries. The rural households in these villages often have tractors or small trucks associated with specialty agriculture instead of other travel modes. Therefore, this result is consistent with the current situation in rural Sichuan. Built environment preferences (Table 8.8): Rural residents who prefer an accessible built environment are more likely to choose walking as their daily travel mode over electric bicycles (−0.357). This result is consistent with that of Cao et al. (2009). The influences of the objective, preferred, and perceived built environments in terms of accessibility by walking are consistent. Thus, the accessibility of destinations and transport facilities can prompt rural residents to choose walking as their daily travel mode. Rural residents who prefer a harmonious neighborhood and good facility service are likely to choose other travel modes (e.g., tractors and small trucks). This result is the same as the influence result of built environment perception on travel mode choice. This finding is also consistent with the research result of EtminaniGhasrodashti and Ardeshiri (2015); i.e., that neighborhood harmony is likely to promote walking choice probability. Residents who prefer a safe living environment have a low probability of choosing a private car (−0.359) and are likely to choose walking as their daily travel mode. Rural residents who prefer a living environment with less accidents are likely to choose motorcycles (0.456), electric bicycles (0.258), and bicycles (0.448) as their daily travel modes. Therefore, a safe and harmonious built environment can prompt rural residents to choose convenient and fast modes of daily travel.

8.4 Results and Discussion

225

8.4.4 Travel-Related Variables Daily activity, travel distance, and departure time are significant influencing factors for travel mode choice (Table 8.7). Among travel-related variables (Table 8.8), distance is significantly positively correlated to the probability of choosing private cars (2.271), public transport (1.932), motorcycles (1.635), bicycles (1.604), and electric bicycles (1.581) as travel modes. When traveling for a long distance, the probability of using motorized transport is high (Pan et al. 2007). Travel distance exerts the greatest effect on the probability of choosing a private car because driving is fast and efficient, and people are willing to drive. The next choice is public transport. Residents who choose public transport can freely arrange this period (e.g., for chat and entertainment), which is consistent with the existing research results (Feng et al., 2010). Rural residents who choose walking spend more time on the road, and those who choose driving a private car (−1.223), motorcycle (−1.295), electric bicycle (−0.840), and bicycle (−0.140) spend less time. The effect of departure time on the travel mode choice of rural residents is insignificant. Activities have a positive impact on travel mode choice. Leisure activities are used as the reference group. In mandatory activities, the probability of choosing other travel modes (4.024) is the highest, followed by electric bicycles (3.085), motorcycles (2.819), private cars (1.616), bicycles (1.177), and public transportation (0.984). The field work of rural residents may directly use farm vehicles. Older parents will choose electric tricycles when picking up and dropping off their children. An electric tricycle has peripheral maintenance, and it is safe for children to ride. This case results in a high probability in other travel modes. In daily activities, the probability of choosing electric bicycles (3.422) is the highest, followed by private cars (2.892). The car purchases of rural residents will continue to rise steadily (Zhang et al. 2019). Therefore, families with high financial capability will choose to drive for their daily activities.

8.5 Conclusion and Policy Implications China has entered an important stage in the goal of building a moderately prosperous society. The construction of rural areas is the key to a comprehensive developed society. China’s rapid urbanization has changed the rural built environment and the lives and travel behavior of rural residents. This study collects research data through a household survey. On-site measurement methods using GIS technology and the MNL model are used to analyze the effect of the rural built environment on the travel mode choice of rural residents. The perception and preference of rural residents are also considered. In sociodemographic variables, age is significantly negatively correlated with the probability of choosing a private car, public transport, motorcycle, electric bicycle, and bicycle compared with walking. Men are inclined to choose a private car, motorcycle, and other travel modes. Education level is significantly positively correlated

226

8 The Impact of the Rural Built Environment on …

with the probability of choosing public transport and electric bicycles. Residents who hold a rural hukou have a higher probability of choosing a private car than those with an urban hukou. The number of household members is significantly negatively correlated with the probability of choosing a motorcycle, an electric bicycle, and a bicycle. The ownership of each type of transportation is significantly positively correlated with its choice as the corresponding travel mode. Objective built environment indicators, namely, building density, road density, distance to transit, and destination accessibility, are statistically significant. Moreover, building density exerts the greatest influence on the travel mode choice of rural residents. The higher the building density, the greater the probability that rural residents will choose electric bicycles, followed by private cars. The higher the road density, the more willing people will be in choosing motorcycles and electric bicycles. When accessibility is good, residents are inclined to walk. In terms of built environment perception, rural residents living in a safe neighborhood with a harmonious environment favor choosing private cars and motorcycles. In terms of built environment preference, rural residents who value good accessibility, a safe and harmonious neighborhood, and well serviced environments, tend to prefer walking. Those residents who prioritize an environment with less crime and fewer traffic accidents will tend to use motorcycles. Among travel-related variables, travel distance is significantly positively correlated with the probability of choosing a private car, public transportation, motorcycle, electric bicycle, and bicycle. The longer the trip distance, the more the people tend to use public transportation. Rural residents are likely to choose other travel modes when conducting mandatory activities and electric bicycles when performing their daily activities, compared with leisure activities. On the basis of the characteristics of the influences of various variables on travel mode choice, suggestions are proposed as follows. a. The investment construction and maintenance of rural public transportation infrastructure should be given more attention in the rural urbanization process with increasing private car use. The connection between rural and urban roads should also be improved to increase urban accessibility. b. The government should place importance on developing public transportation in rural areas and increase the number of bus routes and bus stops, along with greater frequency of public transportation scheduling. c. Nonmotor vehicle lanes should be added to meet the travel needs of electric bicycles and bicycles. This is despite the current declining trend in the number and use of bicycles in rural areas. Connectivity between villages will promote further usage of electric bicycles, and bicycling should be encouraged because it is a healthy and zero-carbon mode of travel. d. Daily destinations, such as markets, schools, and health centers/hospitals, should be located in villages or within walking distance in future rural planning. In this manner, rural residents can easily meet their daily needs by walking.

References

227

References Anowar S, Yasmin S, Eluru N, Miranda-Moreno LF. 2014. Analyzing car ownership in Quebec City: a comparison of traditional and latent class ordered and unordered models. Transportation 41: 1013–1039. Ao Y, Chen C, Yang D, Wang Y. 2018. Relationship between Rural Built Environment and Household Vehicle Ownership: An Empirical Analysis in Rural Sichuan, China. Sustainability 10: 1–18. Ao Y, Li J, Wang Y, Liu C, Xu S. 2017. Farmers’ Satisfaction of Rural Facilities and Its Influencing Indicators: A Case Study of Sichuan, China. Mathematical Problems in Engineering 2017: 12. Ao Y, Yang D, Chen C, Wang Y. 2019a. Effects of rural built environment on travel-related CO2 emissions considering travel attitudes. Transportation Research Part D: Transport and Environment 73: 187–204. Ao Y, Yang D, Chen C, Wang Y. 2019b. Exploring the effects of the rural built environment on household car ownership after controlling for preference and attitude: Evidence from Sichuan, China. Journal of Transport Geography 74: 24–36. Badland HM, Oliver M, Kearns RA, Mavoa S, Witten K, Duncan MJ, Batty GD. 2012. Association of neighbourhood residence and preferences with the built environment, work-related travel behaviours, and health implications for employed adults: Findings from the URBAN study. Soc Sci Med 75: 1469–1476. Boarnet M, Crane R. 2008. The influence of land use on travel behavior: specification and estimation strategies. Transportation Research Part A 35: 823–845. Bouscasse H, Joly I, Peyhardi J. 2019. A new family of qualitative choice models: An application of reference models to travel mode choice. Transportation Research Part B: Methodological 121: 74–91. Cai Z, Liu Q, Cao S. 2020. Real estate supports rapid development of China’s urbanization. Land Use Policy 95: 104582. Cao J. 2015. Examining the relationships between neighborhood built environment and travel behavior: A review from US perspective. Urban Planning International 04: 50–56. Cao S, Lv Y, Zheng H, Wang X. 2014. Challenges facing China’s unbalanced urbanization strategy. Land Use Policy 39: 412–415. Cao X, Handy SL, Mokhtarian PL. 2006. The Influences of the Built Environment and Residential Self-Selection on Pedestrian Behavior: Evidence from Austin, TX. Transportation 33: 1–20. Cao X, Mokhtarian PL, Handy SL. 2009. The relationship between the built environment and nonwork travel: A case study of Northern California. Transportation Research Part A 43: 548–559. Cervero R. 2002. Built environments and mode choice: toward a normative framework. Transportation Research Part D Transport & Environment 7: 265–284. Cervero R. 2007. Transit-oriented development’s ridership bonus: a product of self-selection and public policies. Environment and Planning part A 39: 2068–2085. Cervero R, Duncan M. 2003. Walking, bicycling, and urban landscapes: evidence from the San Francisco Bay Area. Am J Public Health 93: 1478–1483. Cervero R, Kockelman K. 1997. Travel demand and the 3Ds: Density, diversity, and design. Transportation Research Part D Transport & Environment 2: 199–219. Cervero R, Sarmiento OL, Jacoby E, Gomez LF, Neiman A. 2016. Influences of Built Environments on Walking and Cycling: Lessons from Bogotá. Urban Transport of China 3: 203–226. Chatman DG. 2008. Deconstructing development density: Quality, quantity and price effects on household non-work travel. Transportation Research Part A 42: 1008–1030. Chen b, Zhou s, Jiang c. 2016. The impact of built environment on the commuting efficiency of guangzhou residents from the perspective of activity. Journal of urban planning 67–74. Chen y, Zhu y. 2013. Rural residents in northern jiangsu travel characteristics analysis and traffic countermeasures. Traffic standardization: 17–19. Deng Y, Zeng Y, Li Z. 2019. Real Estate Prices and Systemic Banking Crises. Economic Modelling 80: 111–120.

228

8 The Impact of the Rural Built Environment on …

Ding C, Lin Y, Liu C. 2014. Exploring the influence of built environment on tour-based commuter mode choice: A cross-classified multilevel modeling approach. Transportation Research Part D: Transport and Environment 32: 230–238. Etminani-Ghasrodashti R, Ardeshiri M. 2015. Modeling travel behavior by the structural relationships between lifestyle, built environment and non-working trips. Transportation Research Part A Policy & Practice 78: 506–518. Ewing R, Bartholomew K, Winkelman S, Walters J, Chen D. 2008. Growing cooler: the evidence on urban development and climate change. Journal of the American Planning Association 75: 6–13. Feng J. 2017. The influence of built environment on travel behavior of the elderly in urban China. Transportation Research Part D Transport & Environment 52: 619–633. Feng z, Liu h, Zhang j. 2010. Rural population travel mode selection model. J Transp Eng 10: 77–83. Gao Y. 2016. Top-Down and Bottom-Up Processes for Rural Development and the Role of Architects in Yunnan, China. Buildings 6: 47–52. Hägerstrand T. 1986. What about People in Regional Science? *. Transport Sociology 24: 143–158. Handy SL, Boarnet MG, Ewing R, Killingsworth RE. 2002. How the built environment affects physical activity Views from urban planning. Am J Prev Med 2002;23(2S): 64–73. Hao H, Siqi L, Aditi S, Futing Z, Fangqiong T, Hui W. 2018. Factors associated with life satisfaction among married women in rural China: a cross-sectional study based on large-scale samples. Psychology research and behavior management 11. Hasnine MS, Lin T, Weiss A, Habib KN. 2018. Determinants of travel mode choices of postsecondary students in a large metropolitan area: The case of the city of Toronto(Article). Journal of Transport Geography 70: 161–171. Huang z, HUang j, Liu x. 2019. Connection Model and Planning Strategies for Urban Rural Public Transportation Integration. Planners 35: 26–31. Kamruzzaman M, Baker D, Washington S, Turrell G. 2013. Residential dissonance and mode choice. Journal of Transport Geography 33: 12–28. Kong r, Yao z. 2015. Comparison of travel characteristics between urban and rural residents in haining city. Transportation enterprise management 30: 17–19. Liu Q, Ren Y, Cao C, Su M, Lyu J, Li L. 2015. [Association between walking time and perception of built environment among urban adults in Hangzhou]. Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi 36: 1085. Liu y, Zhang j, Liu y. 2016. Discussion on characteristics of rural residents’ daily trips. Journal of wuhan university of technology (traffic science and engineering edition) 40: 390–393. Luo x, Zhang n. 2017. Research on key issues of rural road network planning under the background of new-type urbanization. 2017 China urban transportation planning annual conference, Shanghai, China 2017: 7. Meng h, Li y, Zhao y. 2018. Explore the sustainable development path of China’s new urbanization under the new normal. Journal of hebei university of geosciences 41: 69–75. Ministry of transport. 2018. The total length of rural roads reached 4.05 million kilometers at the end of last year. https://newssinacn/2019-02-28/detailihsxncvf8546686dhtml?oid=3811167501454995&vt=4&pos=3. National Bureau of Statistics. China Statistical Yearbook. 2018. Available online: http://tongjicnkinet/kns55/Navi/HomePageaspx?id=N2017100312&name=YINFN&floor=1 (accessed on 28 April 2019). Ohnmacht T, Götz K, Schad H. 2009. Leisure mobility styles in Swiss conurbations: construction and empirical analysis. Transportation 36: 243–265. Pan H, Shen Q, Zhang M. 2007. Influence of Urban Form on Travel Behavior in Four Neighborhoods of Shanghai. In Transportation Research Board Meeting. Qian J, He S, Liu L. 2013. Aestheticisation, rent-seeking, and rural gentrification amidst China’s rapid urbanisation: The case of Xiaozhou village, Guangzhou. Journal of Rural Studies 32: 331–345.

References

229

Saelens BE, Handy SL. 2008. Built environment correlates of walking: a review. Medicine and science in sports and exercise 40: 550–566. Scheiner J. 2010. Social inequalities in travel behaviour: trip distances in the context of residential self-selection and lifestyles. Journal of Transport Geography 18: 679–690. Stead D. 2001. Relationships between land use, socioeconomic factors, and travel patterns in Britain. Environment and Planning B Planning and Design 28: 499–528. Sun b, Dan b. 2015. The influence of urban built environment on residents’ commuting mode choice in Shanghai. Journal of geographical 70: 1664–1674. Sun B, Yan H, Zhang T. 2017. Built environmental impacts on individual mode choice and BMI: Evidence from China. Journal of Transport Geography 63: 11–21. Ta n, Chai y, Guan m. 2015. The influence of built environment on weekday car travel of Beijing suburb residents. Journal of geographical 70: 1675–1685. Thrane C. 2015. Examining tourists’ long-distance transportation mode choices using a Multinomial Logit regression model. Tourism Management Perspectives 15: 115–121. Van Acker V, Van Wee B, Witlox F. 2010. When Transport Geography Meets Social Psychology: Toward a Conceptual Model of Travel Behaviour. Transport Reviews 30: 219–240. Wan L, Ng E. 2016. Assessing the Sustainability of the Built Environment in Mountainous Rural Villages in Southwest China. Mountain Research and Development 36: 4–14. Wang D, Cao X. 2017. Impacts of the built environment on activity-travel behavior: Are there differences between public and private housing residents in Hong Kong? Transportation Research Part A Policy & Practice 103: 25–35. Wang Y, Ao Y, Zhang Y, Liu Y, Zhao L, Chen Y. 2019. Impact of the Built Environment and Bicycling Psychological Factors on the Acceptable Bicycling Distance of Rural Residents. Sustainability 11: 4404. Yang D, Timmermans H. 2015. Analysis of consumer response to fuel price fluctuations applying sample selection model to GPS panel data: Dynamics in individuals’ car use. Transportation Research Part D: Transport and Environment 38: 67–79. Yang q, Yuan h, Feng s. 2014. Travel characteristics of rural residents under different economic conditions. Journal of chang’an university (natural science edition) 34: 76–83. Zeng w, Yan s, Tian j. 2019. Quality evaluation of new urbanization in wuhan city based on factor analysis. Statistics and decision making 35: 114–117. Zhang C, Lin Y. 2012. Panel estimation for urbanization, energy consumption and CO 2 emissions: A regional analysis in China. Energy Policy 49: 488–498. Zhang w, Zhang h. 2016. From the rural travel mode and the experience of developed countries to see China’s rural automobile market prospects. Automobile vertical and horizontal 02: 106–109. Zhang y, Yang h, Ma k. 2019. Study on residents’ happiness in tonghua city based on factor analysis. The wind of science and technology 03: 232–241. Zhao P. 2013. The Impact of the Built Environment on Individual Workers’ Commuting Behavior in Beijing. International Journal of Sustainable Transportation 7: 389–415. Zhao P, Li P. 2019. Travel satisfaction inequality and the role of the urban metro system. Transport Policy 79: 66–81. Zheng r, Chen c. 2017. Study on the travel characteristics of rural residents in Shanghai—a case study of zhoukang daju. Integrated transport 39: 79–85 + 93. Zhuo J, Wei w, Zhuo J. 2019. Dilemma and Strategies of Urban-rural Public Transportation Integration. Planners 35: 19–25.

Chapter 9

The Effects of the Rural Built Environment on Travel-Related CO2 Emissions, Adjusted for Travel Preferences

Abstract This chapter contributes to the understanding of the impacts of the rural built environment on travel-related CO2 emissions by considering the mediating effects of household car ownership, travel frequency, travel distance, and individual travel attitudes through structural equation modeling. The travel data were collected from an activity diary survey in rural Sichuan. Geographic information system technology, combined with on-site measurement, was used to obtain data on the built environment. After controlling for socio-demographic factors, the model results confirm that all built environment variables have a significant total impact on car ownership, travel distance, travel frequency, and travel emissions. Specifically, residents living in villages with more accessible markets, better roads, and higher building density travel shorter distances and consequently generate less CO2 . Meanwhile, residents living in villages with a centralized living style and higher transit and destination accessibility, travel less frequently but emit more CO2 . Individual travel attitudes have a limited effect on travel behavior and CO2 emissions. This study suggests that planners and policymakers should consider shortening the distance between destination/transit and residential areas and increasing road and building densities. Moreover, promoting the construction of cycling facilities and separate bicycle lanes to encourage rural residents to ride electric bicycles, bicycles, and motorcycles, will reduce transport CO2 emission in Chinese rural areas. Keywords Travel-related CO2 emission · Rural built environment · Travel attitudes · Structure equation modeling · Exploratory factor analysis · Rural China

9.1 Introduction Accumulated scientific evidence shows that climate change is a real and daunting threat to global human development (Stocker et al. 2013). Global climate change caused by energy consumption from human activities and related CO2 emissions has attracted widespread attention from the international community (Ou et al. 2013; Solomon et al. 2007). Transportation is the fastest growing sector in terms of global energy consumption and CO2 emissions (Agency 2009; Yan and Crookes © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_9

231

232

9 The Effects of the Rural Built Environment …

2009). According to the International Energy Agency (IEA), the global transportation sector produced 7001.1 Mt of CO2 in 2011, accounting for 22.3% of all emissions and making it the second largest source of CO2 emissions. Road traffic accounts for around three-quarters of the total CO2 emissions from transportation (73.9%). China’s transportation has a relatively low proportion of CO2 emissions but ranks second only to the United States (Statistics 2011). This situation means that China faces enormous challenges in reducing carbon emissions from transport (Yang et al. 2015). By the end of 2015, China’s energy production and energy consumption were 2.93 and 2.61 times that in 2000, respectively. Meanwhile, the number of car ownership per 100 rural households in China in 2016 was 13.18 times the number in 2000. Reducing CO2 emissions from transport is the primary way to achieve climate change mitigation goals (Ma et al. 2015), and transportation is supposed to be the most challenging sector in terms of reducing CO2 emissions (Brand et al. 2012; Marsden and Rye 2010). Numerous studies have investigated the relationship between transport planning and individual travel behavior (Cao et al. 2009; Cao and Yang 2017; Ding et al. 2017b; Handy et al. 2005; Li and Zhao 2017; Liu et al. 2016; Sun et al. 2017). Several investigations related to mobility have demonstrated the validity and importance of these relationships through empirical research (Bamberg et al. 2003; Haustein and Hunecke 2007; Heath and Gifford 2002; Schoenau and Müller 2017; Thorhauge et al. 2016). Scholars have affirmed that high population density, mixed land use, and pedestrian-friendly street designs are positively related to small numbers of vehicles, short travel distances, and reduced motor vehicle travel (Ewing and Cervero 2010; Ewing et al. 2015; Khattak and Rodriguez 2005; Krizek 2003). China is in a process of rapid urbanization and new rural construction. However, all studies in relation to the relationship among the built environment, travel behavior, and travel-related CO2 emissions focused on large cities in China, such as Beijing, Guangzhou, Shanghai, and Nanjing (Cao and Yang 2017; Liu et al. 2016; Ma et al. 2015; Yang et al. 2015). The scale of China’s rural urbanization and new rural construction is unprecedented. The interrelationship among rural space reorganization, rural resident travel attitudes, travel behavior, and travel-related CO2 has undergone profound changes. Exploring the relationship among them is crucial for the further establishment of a new ecological and low-carbon countryside and fills the abovementioned research gap. Therefore, this study focuses on rural areas in Sichuan, China, and explores the direct and indirect impacts of China’s rural built environment on travel CO2 emissions. The residents’ psychological factors are considered in this work. The structure of this paper is as follows. Section 9.3 explains the methodology used in this research, data collection, and variable specification. The results and discussion of the structural equation model (SEM) are presented in Sect. 9.4. The conclusions and policy implications are summarized in Sect. 9.5.

9.2 Literature Review

233

9.2 Literature Review Built environment exerts a significant influence on travel behavior and transport carbon emissions (Hankey and Marshall 2010). The built environment is measured by the D variable. With the accumulation of relevant research, the built environment measurement indicator has developed from 2D to 4D and is now widely accepted as 6D (Ewing and Cervero 2001, 2010; Ewing and Handy 2009; Ewing et al. 2015; Vance and Hedel 2007). The “6Ds” of the built environment, namely, density, diversity, design, destination accessibility, distance to transit, and demand management, have been widely utilized (Ewing and Cervero 2001, 2010; Ewing and Handy 2009; Ewing et al. 2015; Vance and Hedel 2007). Travel behaviors are measured in many ways, including travel mode, distance, frequency, purpose, and time. (Boarnet 2011; Ewing and Cervero 2001, 2010; Handy et al. 2005). Overall, scholars have found that high population density, mixed land use, and pedestrian-friendly street designs are positively related to small numbers of vehicles (Brownstone and Golob 2009; Ewing and Cervero 2010), short distances (Ewing et al. 2015; Khattak and Rodriguez 2005), and reduced motor vehicle travel (Krizek 2003; Modarres 2013) because compact and high-density urban forms promote public transport development and reduce the use of private cars (Ewing 1997; Kenworthy and Laube 1996). For example, Ding et al. (2014) discovered that job density in urban centers is important in reducing travel CO2 emissions compared with the situation in household dwelling areas (Ding et al. 2014). Hong (2017) found a nonlinear relationship between density and transport CO2 emissions. However, the relationship between CO2 emission and population density is not significant to some extent. In other studies, the correlation between residential density and transport CO2 emissions is not significant (Barla et al. 2011; Jiang et al. 2011; Xiao et al. 2011). Moreover, increasing road capacity is a viable means to increase energy efficiency in transportation and reduce related emissions. However, Shim et al. (2006) revealed an inverse relationship between transport energy consumption and road density in their study of 61 small and medium-sized cities in South Korea. Improvements in road capacity may encourage rampant driving, which may increase CO2 emissions. Ma et al. (2015) examined commuting travel data in Beijing and found that subway accessibility is negatively correlated with CO2 emissions. Another study in China showed that the proportion of bus travel has a significant negative impact on CO2 in transportation (Su et al. 2011). Ribeiro and Balassiano (1997) reported that CO2 emissions from private cars used for daily commute are nearly eight times higher than those from public transport. Yang et al.’s (2015) study indicated that a significant negative relationship exists between urban public transportation and per capita CO2 emissions from transportation. Therefore, public transport plays a key role in reducing carbon emissions. Zahabi et al. (2012) discovered that if density, transit accessibility, and land use mix index are increased by 10% separately, travel-related greenhouse gas emissions will decrease by 0.5, 5.8, and 2.5%, respectively. Zhao (2010) found that the urban sprawl in Beijing’s urban borders increases travel distance and car use, leading to increased emissions. Moreover, parking service as the 6th D variable (demand management) has an impact

234

9 The Effects of the Rural Built Environment …

on CO2 emissions as well. Researchers have found that low-cost parking lots are correlated with high CO2 emission due to car ownership (Guo 2013; Tyrinopoulos and Antoniou 2013). However, most of the studies above focused on the urban built environment. Only a few researchers have investigated the rural household travel behavior associated with CO2 emission. Dargay (2002) reported that car ownership by rural households is much less sensitive to car costs than car ownership by urban households. Therefore, measures to control travel-related CO2 emissions in rural areas through car use cost are not necessarily appropriate for rural areas. Moreover, a study showed that Chinese rural residents have a strong desire to own a car due to the lower rate of household car ownership compared with urban dwellers, and this will lead to a rapid increase in the number of cars in rural areas (Zhu et al. 2012). Christie and Fone (2003) used data from Wales and found that although car ownership is related to household income level, no evidence indicates that low-income households in rural areas own fewer cars than those in urban areas. This result indicates that car ownership is minimally correlated with household income. Once rural residents have cars, they become increasingly dependent on their cars because of the few alternative transport modes available, which will increase travel-related CO2 emissions in rural areas (Wang et al. 2011). Meanwhile, a few studies have focused on travel mode analysis in rural China. As for the choice of travel mode, rural residents in different regions have slightly different choices. Rural residents in Haining, Zhejiang, prefer electric bicycles to cars and walking (Kong and Yao 2015), whereas rural residents in northern Jiangsu prefer electric bicycles, followed by walking and motorcycles (Chen and Zhu 2013). Children and elderly people with limited mobility needs are the major residents in rural areas because most of the young population are working outside the rural area. Therefore, rural residents have fewer trips and lower travel CO2 emissions compared with urban residents (Yang et al. 2014). For rural households, literature has found that socio-demographic characteristics influence travel-related CO2 emission, similar to the situation for urban households. Specifically, men, middle-aged individuals, and elderly people who live in rural areas and own bicycles have a significant but weak association with CO2 emissions (Brand et al. 2013). The studies above investigated the effects of the built environment on CO2 emissions but did not consider psychological determinants, such as preference and personal attitudes. Only a few studies have considered these perspectives (Ao et al. 2019; Belgiawan et al. 2014). These studies have found that significant differences exist between developed and developing countries in terms of car purchase motivation. The expectation of others exerts substantial impacts on purchase intention in developing countries. Attitude is an essential determinant factor for driving and commuting intentions in developed countries. People view the car as a symbol of wealth, which may decide their travel mode. Environment attitudes may influence private car purchase decisions. In summary, the studies above did not reach a consistent conclusion. This scenario indicates that the impacts vary from country to country, and this variation might be

9.2 Literature Review

235

related to attitudes and preferences. Compared with Western countries, urban residents in China have particular travel-related attitudes and preferences (Wang and Lin 2014). In addition, a massive difference in the built environment exists between urban and rural areas in China. For example, rural households cannot select residential locations according to their preferences because of the fixed homestead location, which is contrary to urban households in China. With the rapid development of new rural construction and urbanization, great changes have taken place in China’s rural built environment. However, in China, all related studies on the relationship among the built environment, travel behavior, and travel CO2 emissions were conducted in China’s first-tier or second-tier cities, such as Shanghai, Nanjing, Guangzhou, and Beijing (Cao and Yang 2017; Liu et al. 2016; Ma et al. 2015; Yang et al. 2015). Research on rural areas is lacking. The scale of China’s rural urbanization and new rural construction is unprecedented. The interrelationship among rural space reorganization, rural residents’ travel attitudes, travel behavior, and travel-related CO2 emissions has undergone profound changes. Exploring the relationship among them has significant impacts on the further establishment of a new ecological and lowcarbon countryside. In addition, most existing studies only considered the direct impact of the built environment on CO2 emission from daily traveling; they ignored the indirect effects of the built environment, which may affect other variables and ultimately influence travel-related CO2 emissions (Cao and Yang 2017).

9.3 Research Methodology 9.3.1 Model Specification Two models were applied in this study. First, exploratory factor analysis (EFA) was used to reduce data to a small number of non-related comprehensive variables. EFA identified the structure of the relationship between the variables and obtained the important and common travel attitudes. Second, the EFA result was used in the structural equation model (SEM) to investigate the influence of travel attitudes on travel behavior and travel-related CO2 emissions. SEM is a research technique that has been used in its present form since the 1970s. This technique is widely utilized in the majority of qualitative research in psychology, sociology, biological sciences, education research, political science, and marketing (Van Acker et al. 2007). Recently, SEM was used to explore the complex effects of the built environment on travel behavior (Cao and Yang 2017; Liu et al. 2016; Ma et al. 2015; Van Acker et al. 2007). SEM can solve endogeneity problems between variables, and it can analyze the indirect, direct, and total effects between exogenous and endogenous variables (Glaser 2001; Jahanshahi and Jin 2016; Jahanshahi et al. 2015; Kline and Santor 1999). The variables used in this study were all observational. SEM without latent variables can be defined as follows:

236

9 The Effects of the Rural Built Environment …

Fig. 9.1 Conceptual framework for SEM analysis

y = By + Γ x + ξ

(9.1)

The definitions of the letters in the formula above are as follows: y—vector of endogenous variables; x—vector of exogenous variables; B—matrix of coefficients representing the effects of endogenous variables on each other; Γ —matrix of coefficients representing the effects of exogenous on endogenous variables; ξ —vector of errors. The conceptual framework based on SEM is shown in Fig. 9.1. The sociodemographic attributes and built environment affect household car ownership, travel behavior (e.g., travel frequency and distance), and CO2 emission from daily traveling. Several existing studies used travel frequency, travel distance, and car ownership as dependent variables to explore the effects of socio-demographic attributes and the built environment. In addition, studies on travel-related CO2 emissions typically defined travel frequency and distance, household car ownership, socio-demographic information, and built environment as exogenous variables but ignored their endogeneity. However, based on literature, we found that travel behavior and car ownership have a significant impact on CO2 emissions from daily traveling. CO2 emissions are affected by socio-demographic variables and the built environment as well (Van Acker and Witlox 2010; Yang and Cao 2018). Therefore, this study adopted travel frequency, travel distance, and car ownership as intermediary variables to explore the mediating effects of the built environment and socio-demographic characteristics on travel-related CO2 emissions. Car ownership also affects travel frequency and travel distance (Van Acker and Witlox 2010). People with different socio-demographic attributes choose various built environments due to residential self-selection. Many studies have considered the impact of socio-demographic attributes on the built environment (Ding et al. 2017a; Ma et al. 2015; Yang and Cao 2018). However, Chinese rural residents have limited freedom to select their residential location because of

9.3 Research Methodology

237

the fixed homestead location in China. Therefore, this study did not consider the influence of socio-demographic attributes on the built environment but considered that travel attitudes directly affect travel behavior and travel emissions. Then, we assumed that different travel conditions and rural residents have unique travel preferences and attitudes in rural China. We also assumed that the built environment and socio-demographic attributes directly affect travel attitudes.

9.3.2 Sample Selection and Data Collection The data collection was implemented in Sichuan rural areas. Rural areas can be divided into three categories based on living places. The first category is scattered living places, which is the traditional way of living in Sichuan rural areas. The infrastructure has been dramatically improved in the last decades; however, the living places are still the same. The second category is new-style living spaces. The traditional residential patterns in the countryside have been changed by moving rural residents to concentrated living spaces in rural areas. They often occupy agricultural land as well. The third category is the mixture of traditional—and new-style living spaces. It is normally a transition stage from traditional living to new-style living spaces.

9.3.2.1

Sample Selection

Based on our experiences in data collection in Sichuan rural areas and the purpose of this research, we selected sample village areas that satisfy the following criteria. (1) The area should have the necessary road infrastructure that can be used by vehicles, including buses. A road should be connected to at least one highway/freeway/motorway, which can be used by personal vehicles. (2) The residents in the rural area should support the research and are willing to cooperate for a survey or interview. We found that if a person in our research team came from the village, then obtaining support from the residents would be easy. Based on the two criteria, we organized the sample village selection in four steps. First, we recruited volunteer students who are from Sichuan rural areas and interested in this research (1st Oct. 2017 to 31st Oct. 2017). To minimize knowledge barriers, we recruited students studying in the Environmental and Civil Engineering Department of Chengdu University of Technology. The purpose was to have at least one person in each sample village to set up communication with residents in the village. In total, 117 students submitted their applications. With the criteria mentioned above, a pre-selection was carried out, and 37 rural village areas were selected. Second, intensive training was organized for the recruited students (1st Nov. 2017 to 10th Nov. 2017). We held discussions with the students to determine if the pre-selected rural village areas are suitable for this research. After the training and examination, 14 preselected rural village areas were eliminated. Third, we established a connection with the village communities from the pre-selected rural village areas (10th Nov. 2017 to

238

9 The Effects of the Rural Built Environment …

120th Nov. 2017). We asked if the residents are willing to cooperate and whether the research team can approach them or not. After the discussion, only 10 pre-selected village communities provided consent. Lastly, we organized pre-interview groups for 1–2 residents living in each approved rural village area to understand the residents’ willingness to cooperate (21st Nov. 2017 to 10th Dec. 2017). We found that residents were unwilling to assist and showed precaution from Helin Village (Chongren Town, Dongpo District, and Meishan City), Nanliu Community (Huangshui Town, Shuangliu District, and Chengdu City), and Shiguan Village (Sanshui Town and Guanghan City). In the end, seven rural village areas were selected for the data collection. The seven rural village areas covered 16,953 individuals and 5,888 households.

9.3.2.2

Data Collection

Three types of data were collected in this study, and these are GIS, field measurement, and individual data. GIS and field measurement data were collected for built environment measurement. A huge difference exists between the built environment in rural and urban areas, especially traditional rural living areas. Therefore, we pre-defined a few rules to measure the built environment. For the new-style living spaces in rural areas, we measured the built environment by using the village committee office as the center point and with 1 km as the radius. For the traditional-style living areas, the village committee office was set at the center location. The built environment indicators were calculated based on the data on the natural boundary of the village. The main reason is that the scattered rural areas in Sichuan vary greatly. Using a 1 km radius to measure the area is impossible. Based on the definitions, we used Arcgis 10.2 to obtain building and road information. However, the GIS data for rural areas in China are very limited. Field measurement is necessary for collecting objective built environment information. All interviewees were equipped with the same Baidu navigation system. They used the navigation system to measure the driving distance between the center of the village and the nearest public transport stations (bus, coach, and train), main subway/freeway/motorway, open market/supermarket, school, hospital, and administration center of the city/town. Individual information was collected via a face-to-face interview. The household survey contained a list of socio-demographic variables, including individual and household information that may help explain travel behavior decisions. We executed the data collection via a household questionnaire survey. A total of 560 questionnaires were distributed, and 413 were collected back. Owing to the missing data, 39 out of the 413 questionnaires could not be used for the analysis. In the end, 374 valid questionnaires were used from the seven rural village areas. The survey covered 1,758 individuals. The survey sample distribution matched the rural population in Sichuan and China well, as shown in Table 9.1. The socio-demographic distribution of the sample is listed in Table 9.2. The data of the on-site measurement are shown in Table 9.3. The sample area location is shown in Fig. 9.2, and the GIS data are presented in Fig. 9.3 (Fig. 9.4).

9.3 Research Methodology

239

Table 9.1 Sample vs. population characteristics Total population Total number of households

Householda

Villagea

Rural Sichuanb

Rural Chinab

1758

16953

419.6 2016: Billion

5897.3 2016: Billion

374

5888

–

–

3.03 (2015)

3.88 (2012)

Average household size

3.71

Per capita income (10 k yuan)

1.36

–

1.13 (2016)

1.24 (2016)

Average household income (10 k yuan)

4.44

–

–

–

a Data

2.88

from face-to-face household survey between 16th December 2017 and 5th January 2018 China Statistical Yearbook (2013, 2016, and 2017)

b Source

Table 9.2 Distribution of socio-demographic information Variables

Level

Number of sample

Percent

Male

0 for female

226

60.43

1 for male

148

39.57

47

12.57

2 represent age 25–40

65

17.38

3 represent age 41–50

112

29.95

4 represent age 51–60

80

21.39

Age

1 represent age 16–25

5 represent age 61–70 Driving certificate

70

18.72

279

74.60

95

25.40

0 for cannot ride a motorcycle 229

61.23

o for no driving certificate 1 for have driving certificate

Ride a motorcycle

1 for can ride a motorcycle

145

38.77

Ride a bicycle

0 for cannot ride a bicycle

108

28.88

1 for can ride a bicycle

266

71.12

Ride electric bicycle

0 for cannot ride an electric bicycle

137

36.63

1 for can ride an electric bicycle

237

63.37

1 represents 0

Income (ten thousand yuan)

116

31.02

2 represents 0–0.5

38

10.16

3 represents 0.5–1

80

21.39

4 represents 1–2

60

16.04

5 represents 2–4

59

15.78

6 represents>4

21

5.61

240

9 The Effects of the Rural Built Environment …

Table 9.3 Data measured on-site Name of villages

Distance to the nearest bus station (KM)

Distance to the nearest train station (KM)

Distance to the nearest public transportation station (KM)

Distance to the nearest main road (KM)

Dazhuang (DZ)

18.20

19.90

2.50

2.50

Wugang (WG)

0.20

70.00

16.00

0.00

Shuangyan(SY)

16.30

13.40

0.50

0.50

Xinlong (XL)

13.40

13.40

1.20

0.80

Dongxing (DX)

3.90

16.40

3.90

0.50

Shangten g(ST)

22.40

24.80

0.69

0.69

0.50

125.00

34.00

0.50

Yanjing (YJ) Name of villages

Distance to the nearest market (KM)

Distance to the nearest school (KM)

Distance to the nearest hospital (KM)

Distance to the nearest city centre (KM)

Dazhuang (DZ)

3.00

0.50

0.05

19.60

Wugang (WG)

3.50

2.50

0.20

16.00

Shuangyan(SY)

1.60

1.60

0.60

13.50

Xinlong (XL)

0.80

3.00

4.90

4.90

Dongxing (DX)

0.00

2.10

0.00

10.00

Shangteng (ST)

1.50

1.50

1.60

14.00

Yanjing (YJ)

1.50

0.50

1.70

35.00

Note The map is from the National Bureau of Surveying, Mapping, and Geographic Information

9.3.2.3

Calculation of CO2 Emissions

This study used Cao and Yang’s (2017) formula to calculate travel-related CO2 emissions, as shown below. C E i j = Distancei j × E_ f actori j , where C E i j represents CO2 emissions from a trip using mode j for respondent i, Distancei j is the distance with mode j for respondent i, and E_ f actori j is the emission factor of travel mode j for respondent i. The emission factor data of different transport modes are unavailable for rural areas. By referring to various studies and reports (more details can be seen in Table 9.4) (Entwicklungsbank 2008; Proost et al. 2006; Yang et al. 2018; Cai and Xie 2010; Xiao et al. 2011) and in accordance with the relative intensity of energy consumption and carbon emission for each transport mode, we used the following emission factors in this study, as shown in Table 9.4.

9.3 Research Methodology Preparation stage Sample villages solicitation and selection

Investigators recruitment and screening

Formal training of investigators

241 Three new concentrated-living villages and four traditional scattered-living villages were selected. 30 surveyors were recruited , Including 17 undergraduate students from Construction Management Major and 13 graduate students from Civil Engineering Major. One new concentrated-living and one traditional scattered-living villages were selected randomly from the seven sample villages, and the five households were selected randomly from the two villages respectively.

Pilot test survey Implementation stage Village map and gift preparation for interviewees Formal investigation preparation Finally, 374 valid questionnaires were obtained and the information of socio-demographic was shown in table 1. Household questionnaire survey The information measured on-site was shown in table 3. On site measurement investigation

GIS data extraction

The information extracted from GIS was shown in figure 4

Fig. 9.2 Data and sample collection flowchart

9.3.3 Variable Specification 9.3.3.1

Socio-Demographic Variables

Travel behavior and travel-related CO2 emissions are influenced by sociodemographic variables, as proven by literature. In this study, based on literature, eight variables were included in the final model. These variables are gender, age, income, driving license ownership, ability to drive a motorcycle, ability to drive an electric bicycle, and ability to ride a bike.

9.3.3.2

Travel Attitude Variables

To explore the effects of travel attitudes on travel behavior and travel-related CO2 emissions, 30 statements on travel attitudes were provided in the questionnaire (Cao et al. 2007; He and Thøgersen 2017). A Likert scale ranging from 1 to 5 was used, wherein 1 signifies “completely disagree” and 5 means “completely agree.” The respondents were asked to assess the 30 statements based on their attitudes. To identify the important broad attitudes, EFA was applied using SPSS 23.0. The Kaiser– Meyer–Olkin (KMO) measure was used to test the suitability for EFA, and Bartlett’s test was applied to examine the factorability of individual attitude variables. The final result of KMO was 0.806, and the P value of 0 confirms that high correlations

242

9 The Effects of the Rural Built Environment …

Fig. 9.3 Map of the study area’s location

exist among the attitude variables. EFA must be used to identify the main factors. The results of EFA are shown in Table 9.5. We eliminated the variables with a factor loading below 0.5. Finally, six travel attitude factors were identified, and they accounted for 63.814% of the variance. That is, only 36.186% loss in information was incurred by the 80.0% reduction in the number of variables. Accordingly, the obscure concepts of “travel attitude” can be interpreted and represented well

9.3 Research Methodology

243

Fig. 9.4 GIS information on road and building land

Table 9.4 Emission factors of different travel modes (kg CO2 /person•km) Walk, bike Electric bike Bus

Motorcycle Car

Coach

–

–

0.026

–

0.0203 Entwicklungsbank (2008)

0

–

0.0738 0.01136

0.0606

0.01786 –

References Proost et al. (2006)

0

0.008

0.035

–

0.126

–

Yang et al. (2018)

0

0.008

0.035

–

0.135

–

Xiao et al. (2011)

–

–

–

0.0472

–

–

Cai and Xie (2010)

0

0.008

0.035

0.0472

0.126

–

Selected in this research

9.3.3.3

Built Environment Variables

Six built environment variables were calculated and selected in this study according to the actual situation in rural Sichuan; these variables are road density, building density, transit accessibility, destination accessibility, living style, and number of accessible markets. Road and building density were calculated from the GIS extraction data, which are shown in Fig. 9.4. Transit and destination accessibility were calculated from the on-site measurement data used the Baidu navigation application, which is shown in Table 9.3. The calculation methods of road density, building density, transit accessibility, and destination accessibility are presented in Table 9.6 and can also be found in our previously published papers (Ao et al. 2018; Ao et al. 2019). The living style of rural residents is defined as the spatial form of different villages in this study according to the actual situation in rural Sichuan. The living style of rural residents is categorized into two types, which are scattered and centralized living styles. The shift from traditional scattered living style to centralized living style occurs gradually due to urbanization in China. It directly influences rural residents’ decisions on travel behavior and travel-related CO2 emissions. Therefore, based on

244

9 The Effects of the Rural Built Environment …

Table 9.5 Travel attitude component analysis Statements

Component Pro_wb Pro_Eb Pro_Ab Less_out Use_cost Buy_cost

Cycling exercises your body

0.762

Cycling is a low-carbon, environmentally friendly travel mode

0.767

Bicycle parking is convenient

0.738

The low cost of bicycle purchase and use poses no economic burden at all

0.588

Quick and easy to walk

0.602

Walking exercises your body

0.742

Walking is a low-carbon and environmentally friendly travel mode

0.724

It is quick and easy to ride electric bicycles

0.769

It is safe and environmentally friendly to ride electric bicycles

0.828

The low cost of electric bicycle purchase and use poses no economic burden at all

0.591

Electric bicycle parking is convenient

0.586

It is safe and environmentally friendly to ride motorcycles

0.523

The low cost of motorcycle purchase and use poses no economic burden at all

0.771

Motorcycle parking is convenient

0.696

I often make reasonable arrangements to minimize the number of outings

0.726

For problems that can be resolved by the telephone or the Internet, they will not be resolved on site.

0.793

The price of gasoline affects my choice of travel mode

0.739

Parking costs are high everywhere, and driving is not worthwhile

0.754

There is no economic pressure to buy a car

0.811 (continued)

9.3 Research Methodology

245

Table 9.5 (continued) Statements

Component Pro_wb Pro_Eb Pro_Ab Less_out Use_cost Buy_cost

Eigen value

5.181

1.930

1.474

1.285

1.176

1.079

Proportion of variance explained

20.311

11.500

10.185

8.623

7.274

5.922

Cumulative variance explained

20.311

31.811

41.996

50.618

57.892

63.814

Table 9.6 Built environment variables used in this study Variable

Calculation method

Road density

Total length of roads (m)/total surveyed area (hectares)

Building density Transit accessibility

Building land area (m2 )/total surveyed area (m2 )  [1/(dk + 1)], where k = 1, 2, 3, 4 and dk represents the

Destination accessibility

distance from the village center to the nearest bus station, train station, public transportation station, and main road  1  dk +1 , where k = 1, 2, 3, 4 and dk represents the distance

k

k

from the village center to the nearest market, school, health center (hospital), and city (county) center Living style

Respondents living in traditional scattered areas were measured at 0, whereas those in centralized areas were measured at 1 (only two types of living style existed in the sample villages)

Number of accessible markets The number of accessible markets was obtained from the face-to-face questionnaire survey according to actual statistical data; this variable is expressed in ordinal numbers

the current living situation, we set the living style for the seven sample villages as 0 for the traditional scattered living style and 1 for the centralized living style. Markets are the center of transactions in rural areas. Therefore, it is important to determine if the number of accessible markets has an impact on rural travel behavior and travelrelated CO2 emissions. The number of accessible markets was obtained from the face-to-face questionnaire survey. The actual built environment of the seven villages photographed by the researchers is shown in Fig. 9.5.

9.3.3.4

Travel-Related Variables

All travel-related data were collected from an activity diary survey, in which respondents were asked to record two entire days of activity from 30th December, 2017 to 5th January, 2018. We considered three travel-related variables, namely, frequency, distance, and travel-related CO2 emissions. Then, we obtained 1,042 trips with average frequency, distance, and CO2 emissions of 1.393 times, 6.359 km, and 0.343 kg per person per day, respectively. On the average, each household has

246

9 The Effects of the Rural Built Environment …

Fig. 9.5 Actual built environment of the seven villages

9.3 Research Methodology

247

Table 9.7 Information on travel behavior variables and CO2 emission for the sample villages Name of villages

Mean values Number of trips (Number)

Distance traveled (KM)

Car ownership (Number)

CO2 emission (kgce)

Dazhuang

1.482

6.648

0.333

0.285

Wugang

1.277

6.364

0.702

0.419

Shuangyan

0.858

7.525

0.447

0.425

Xinlong

2.148

5.577

0.328

0.106

Doxing

0.957

3.713

0.862

0.122

Shangteng

1.092

3.640

0.490

0.140

Yanjing

1.988

12.751

0.675

1.142

All samples

1.393

6.359

0.540

0.343

Table 9.8 Information on the 1042 trips Car

Motorcycle

Electric bike

Bicycle

Walking

The others

Travel frequency

103

85

179

61

519

95

Frequency ratio

9.88%

8.16%

17.18%

5.85%

49.81%

9.12%

0.540 cars (Table 9.7). Compared with the average travel-related CO2 emissions of 1.91 kg/person day in Chinese urban areas (Liu et al. 2016), the average CO2 emissions of rural residents was lower. Out of the 1,042 trips collected in this study, 49.81% were conducted by walking. The electric bike was the second most commonly used transport mode (17.18%). Car trips accounted for 9.88% of the total trips. Motorcycles and bicycles accounted for 8.16% and 5.85% of all trips, respectively (Table 9.8). The current sample is similar to that of Kong and Yoa (2015), who reported that electric bikes are more preferred than cars.

9.4 Results and Discussion 9.4.1 Goodness-of-Fit for SEM Amos 21.0 was used to estimate the conceptual SEM. This study adopted the Bollen– Stine bootstrap estimation method, and the number of bootstraps was set to 1,000 (Yang and Cao 2018). Links with no statistical significance (P > 0.1) were removed, and the model was re-estimated. The model was modified and improved according to the modification indices (MI). Table 9.9 presents the goodness-of-fit statistics of the final model and the corresponding reference values. The degree of freedom in the

248

9 The Effects of the Rural Built Environment …

Table 9.9 Goodness-of-fit statistics of the SEM model Model fit indices

Values of our model

Chi-square

167.555

Degress of freedom (df)

120

Reference value

Probability level

0.003

0.9

Adjusted Goodness of Fit Index (AGFI)

0.92

>0.9

Comparative Fit Index (CFI)

0.987

>0.9

Normed Fit Index (NFI)

0.958

>0.9

Non-Normed Fit Index (NNFI)

0.976

>0.9

Root Mean Square Error of Approximation (RMSEA)

0.033

40 thousand yuan Acceptable bicycling distance (km)

70

21

5.61

[0,1]

65

17.38

(1,2]

108

28.88

(2,3]

88

23.53

(3,4]

44

11.76

(4,5]

56

14.97

(5,8]

13

3.48

to ride bicycles) to 5 (i.e., strongly positive to ride bicycles) for the following question: Are you willing to ride bicycles when the following conditions are satisfied? 2. Bicycling motivation. The respondents were asked to evaluate eight statements using a 5-point Likert scale that ranges from 1 (i.e., strongly disagree) to 5 (i.e., strongly agree) for bicycling motivation. 3. EFA was performed using SPSS 23.0 to identify the latent structures underlying the aforementioned attitude and motivation response. Three common factors of attitude on bicycling infrastructure conditions and two common factors of bicycling motivation were eventually obtained. The common factors will enter the multivariate models. 4. Bicycling purpose. The respondents were asked to select their bicycling purpose if they want to ride bicycles from the given seven options. The seven bicycling purposes were transferred to seven binary variables (1 = yes, 0 = no).

10.3 Materials and Research Method

271

Fig. 10.3 Map of the study area’s location and the sample village location (Note The map is from the National Bureau of Surveying, Mapping, and Geographic Information)

5. Riding preference. To accurately collect riding preference data, the respondents were asked to rank car, public transportation, motorcycle, bicycle, and electric bicycle on the basis of their preferences. The scale ranges from 5 (i.e., most favorite travel mode) to 1 (i.e., least favorite travel mode). Lastly, bicycle, motorcycle, and electric bicycle were selected as the riding preference variables using the given numbers (1–5).

272

10 The Impact of the Built Environment …

Table 10.2 Basic data of the actual built environment Name of villages

Valid sample number

Distance to nearest hospital (km)

Distance to nearest market (km)

Distance to nearest school (km)

Distance to nearest public transportation station (km)

Dazhuang

57

0.05

3.00

0.50

2.50

Wugang

56

0.20

3.50

2.50

0.2

Shuangyan

53

0.60

1.60

1.60

0.50

Xinlong

61

4.90

0.80

3.00

1.20

Doxing

58

0.00

0.00

2.10

3.90

Shangteng

49

1.60

1.50

1.50

0.69

Yanjing

40

1.70

1.50

0.50

0.50

10.3.3.3

Built Environment Variables

This study focuses on the objective and perceived built environment information on the bases of the on-site measurement of the built environment and perceived data of the respondents. Moreover, this research mainly considers four destinations to explore the connection between the objective and perceived built environment. All destinations are in the most acceptable bicycling distance scope (i.e., 5 km) (see Fig. 10.4). First, objective built environment indicators were measured on-site using the Baidu navigation app (see Table 10.3). Second, the respondents were asked to assess the four statements using a 5-point Likert scale that ranges from 1 (strongly disagree) to 5 (strongly agree) to measure the perceived built environment indicators (see Table 10.3).

Fig. 10.4 Acceptable bicycling distances of the rural residents of Sichuan

10.3 Materials and Research Method

273

Table 10.3 Objective and perceived built environment variables Objective built environment indicators

Perceived built environment indicators

1

Distance to nearest hospital from village center. (O_H)

It is very convenient to hospital. (P_H)

2

Distance to nearest market from village center. (O_M)

It is very convenient to market. (P_M)

3

Distance to nearest school from village center. (O_S)

It is very convenient to school. (P_S)

4

Distance to nearest public transportation station from village center. (O_P)

It is very convenient to public transportation station. (P_P)

10.3.3.4

Dependent Variable

The acceptable bicycling distance of rural residents is the dependent variable in this study. The respondents were asked to indicate the bicycling distance that is acceptable to them. The preliminary statistical analysis indicated that the maximum, minimum, and average acceptable bicycling distances are 8, 0, and 2.873 km, respectively (see Table 10.3). The dependent variable was dealt with in two ways to accurately analyze the influence of the independent variables on the dependent variable. First, the acceptable bicycling distance data filled out by the respondents is retained because it is a continuous variable. In particular, the data were entered into the multiple linear regression model as a continuous dependent variable. Second, an ordered variable was processed using serial numbers to indicate the acceptable bicycling distance of the respondents. The data processing is as follows: 0 represents 0 km, 1 represents an acceptable bicycling distance of over 0 km and below or equal to 1 km, 2 represents the distance above 1 km and below or equal to 2 km, 3 represents the distance above 2 km and below or equal to 3 km, and so on. Lastly, 6 represents distance above 5 km and below or equal to 8 km. Figure 10.4 shows the frequency statistics of the ordered variable. The most number of people who can accept a bicycling distance above 1 km and below 2 km reached 28.88%, followed by over 2 km and below or equal to 3 km. Over 50% of the respondents believe that bicycling distance above 1 km and below or equal to 3 km is acceptable. Table 10.4 shows all the variables. The variance inflation factor (VIF) test indicates that this study has no multicollinearity problem.

10.4 Results and Discussion This section presents the results of the analysis. First, we present the descriptive statistics for bicycling psychology, including the stated bicycling purpose, motivation, attitudes on bicycling infrastructure conditions, preferences for bicycling, riding motorcycles, and riding electric bicycles. Thereafter, the current section provides the

274

10 The Impact of the Built Environment …

Table 10.4 Descriptive statistical summary of the variables used in this study Variables

Mean

S.D

Minimum

Acceptable bicycling distance

2.874

1.433

0.000

Acceptable bicycling distance

2.873

1.542

Male

0.396

Age

Maximum

Type

VIF

6.000

Ordinal (0,1,2,3,4,5,6)

–

0.000

8.000

Continuous

–

0.489

0.000

1.000

Binary: 0-female/1-male

1.257

3.163

1.270

1.000

5.000

Nominal (5 levels)

1.273

Income

1.407

1.730

0.000

15.000

Continuous

1.374

Cycling ancillary facilities

0.000

0.999

(4.018)

4.177

Common factor

1.193

Bicycle lane conditions

0.000

0.999

(2.752)

3.383

Common factor

1.172

Safety

0.000

0.999

(3.554)

3.885

Common factor

1.200

Other motivations

0.000

0.999

(2.922)

2.233

Common factor

1.405

Convenient

0.000

0.999

(3.058)

2.104

Common factor

1.216

Physical activity

0.489

0.500

0.000

1.000

Binary: 0-no/1-yes

1.390

Go to work/school

0.160

0.367

0.000

1.000

Binary: 0-no/1-yes

1.155

Bike with children

0.147

0.354

0.000

1.000

Binary: 0-no/1-yes

1.209

Go shopping

0.340

0.474

0.000

1.000

Binary: 0-no/1-yes

1.537

Visit friends

0.267

0.443

0.000

1.000

Binary: 0-no/1-yes

1.308

To entertainment

0.112

0.316

0.000

1.000

Binary: 0-no/1-yes

1.174

Others

0.134

0.340

0.000

1.000

Binary: 0-no/1-yes

1.168

Liking riding motorcycles

2.631

1.258

1.000

5.000

Ordinal

1.441

Liking riding electric bicycles

3.393

1.123

1.000

5.000

Ordinal

1.164

Liking bicycling

2.693

1.350

1.000

5.000

Ordinal

1.395

O_P (Table 10.3)

1.426

1.282

0.200

3.900

Continuous

2.445

O_M (Table 10.3)

1.695

1.155

0.000

3.500

Continuous

2.967

O_S (Table 10.3)

1.742

0.880

0.500

3.000

Continuous

Dependent variables

Independent variables

1.688 (continued)

10.4 Results and Discussion

275

Table 10.4 (continued) Variables

Mean

S.D

Type

VIF

O_H (Table 10.3)

1.313

1.702

Minimum 0.000

Maximum 4.900

Continuous

2.706

P_S (Table 10.3)

3.439

0.971

1.000

5.000

Ordinal

2.906

P_M (Table 10.3)

3.508

0.939

1.000

5.000

Ordinal

3.810

P_P (Table 10.3)

3.179

1.118

1.000

5.000

Ordinal

2.204

P_H (Table 10.3)

3.634

0.839

1.000

5.000

Ordinal

1.498

EFA results of the latent attitude and motivation and the multiple linear regression of the acceptable bicycling distance.

10.4.1 Rural Residents’ Attitudes Regarding Bicycling Infrastructure Conditions, in Sichuan Figure 10.5 shows the results of the questionnaire survey on attitude. Over half of the respondents believe that every bicycle infrastructure condition (except shower facilities at the destination) encourages them to ride a bicycle. Over 70% of the respondents agree that these bicycling infrastructures, such as good-quality route surface, bicycle lanes separated from motor vehicle lanes, safety of bicycle lanes, and sufficiently wide bicycle path, encourage them to ride bicycles. Therefore, the

Fig. 10.5 Attitude on bicycling infrastructure conditions for rural residents

276

10 The Impact of the Built Environment …

good conditions of bicycle lanes have the most evident positive impact on bicycling for the current rural residents of Sichuan, followed by other ancillary facilities, such as bicycle parking, traffic lights, shade, and shower facilities. These statistical results are consistent with the actual situation in Sichuan without efficient specialized bicycle infrastructure.

10.4.2 Rural Residents’ Bicycling Use Motivation Figure 10.6 shows the preliminary statistics. A total of 83.42% of the respondents believe that bicycling is beneficial for the physical and mental health, although a only few of them selected physical activity as their bicycling purpose. Over 75% of the respondents believe that bicycling is beneficial to the environment, bicycle parking, and money saving. Over 60% of the respondents believe that bicycling can ease traffic congestion and is also interesting. A total of 34.22% of the respondents disagree with the idea that “bicycling is fast,” 34.22% of them agree with such statement. The respondents believe that bicycling does not have a fast advantage in speed compared with other bicycling motivations (Fu and Farber 2017).

Fig. 10.6 Bicycling motivations of the rural residents of Sichuan

10.4 Results and Discussion

277

Fig. 10.7 Bicycling purposes of the rural residents of Sichuan

10.4.3 Rural Residents’ Purposes for Bicycling Figure 10.7 shows the distribution of the bicycling purpose of the rural residents based on the preliminary statistics. A total of 46.26, 33.96, and 27.27% of the respondents selected going shopping, visiting friends, and going to work and school, respectively. Only 10.70 and 13.64% of the respondents selected physical activity and recreation options, respectively, as bicycling purposes. These two options have the lowest selection rates. This finding is inconsistent with that of Fu and Farber (2017), who conducted an investigation in Salt Lake City and found that bicycling is mainly a physical and recreational means of traveling for residents of this city. This finding shows a certain gap in the living conditions of residents between China and developed Western countries between urban and rural areas in China. These developments are consistent with the current situation in rural China.

10.4.4 Preferences for Riding Bicycles, Motorcycles, and Electric Bicycles Figure 10.8 shows the statistical results of the respondents’ travel model preference. Their preference for public transportation and private cars is relatively more dispersed compared with their riding preferences. A total of 27.27% of the respondents expressed that they prefer bicycling, whereas 48.93% do not like bicycling. However, preference for electric bicycles is evidently higher than that of bicycles. A total of 46.79% of the respondents said that they prefer electric bicycles, whereas only 22.46% did not prefer this mode of transportation. The respondents’ preference for motorcycles is more consistent than that for bicycles. This study mainly considers the influence of the riding preferences of rural residents on the acceptable riding distance. Therefore, the current research eventually chooses the preference data of bicycles, motorcycles, and electric bicycles to enter the multivariate models.

278

10 The Impact of the Built Environment …

Fig. 10.8 Travel mode preferences of the rural residents of Sichuan

10.4.5 Analysis of the EFA Results The Kaiser–Meyer–Olkin (KMO) and Bartlett’s tests were used to investigate the factorability of individual attitude to bicycling infrastructure conditions and bicycling motivation variables. The values for the KMO and Bartlett’s tests are 0.956 and 0.809, respectively, thereby indicating that the data are suitable for EFA. Tables 10.5 and 10.6 list the items. EFA identifies three common factors for the attitude to bicycling conditions. The identified factors explain 79.196% of the variance, while the number of factors is reduced from 13 to 3. EFA also identifies two bicycling motivation factors, which explain 59.445% of the variance, while the number of factors is reduced from 8 to 2. Therefore, the EFA results indicate that the number of variables reduced is considerably less than the information loss. Accordingly, the common factors of “bicycling condition attitude” and “bicycling motivation” can be interpreted and represented. Thereafter, the extracted components can be used in the linear regression and can effectively represent the bicycling condition attitude and bicycling motivation of the respondents.

10.4.6 Multivariate Models of Acceptable Bicycling Distances The multivariate models of the acceptable bicycling distance of rural residents comprise all the previously described independent variables. Multiple linear regression was estimated using NLOGIT 5.0. Table 10.7 shows the results. Table 10.7 shows that the socio-demographic attributes of rural residents do not significantly affect their acceptable bicycling distance, except for age, which negatively affects their acceptable bicycling distance. This result is inconsistent with the conclusions of the majority of the existing relevant empirical studies. Moreover, the

10.4 Results and Discussion

279

Table 10.5 EFA result of attitude on bicycling condition The items of attitude on bicycling condition

Component Cycling ancillary facilities

Bicycle lane conditions

The route has traffic lights for cyclists

0.793

Shower facilities available at destination

0.704

Vehicular speeds are limited

0.633

The bike lane can be safer

0.632

The route has enough lighting

0.630

The route is flat

0.608

0.600

Bikeway with trees on both sides

0.592

0.556

0.625 0.591

The width of the bikeway is adequate

0.828

The route surface is of good quality

0.790

The route is sufficiently direct

0.780

Small motor vehicle traffic Secure bicycle parking at destination

Safety

0.766 0.582

0.623

There is a bikeway separated from traffic

0.618

% of Variance

30.125%

29.686%

19.385%

Cumulative %

30.125

59.811

79.196

3.916

3.859

2.520

Eigenvalues

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization

majority of the related studies have shown that socio-demographic attributes often have more significant effects on bicycling behavior than other variables (Fu and Farber 2017; Msambichaka et al. 2018). This finding is significantly related to the fact that there is no fixed income for the left-behind population in rural areas in Sichuan and that the left-behind population members are generally older, young, or female-dominated. The following subsections mainly analyzes the impact of the individual bicycling psychology and rural built environment on the acceptable bicycling distance of rural residents.

280

10 The Impact of the Built Environment …

Table 10.6 EFA result of bicycling motivation

The items of bicycling motivation

Component Other motivations

Bicycling can improve health

0.817

Bicycling can protect the environment

0.809

Bicycling can save money

0.666

It is easier to park a bicycle

0.653

Bicycling can avoid traffic jams

0.639

I cycle for fun

0.614

Convenient

Bicycling is faster

0.893

Bicycling is convenient

0.795

% of Variance

38.862

20.584

Cumulative %

38.862

59.445

3.109

1.647

Eigenvalues

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization

10.4.6.1

Influence of the Psychological Factors of Bicycling on the Acceptable Bicycling Distance

Table 10.7 shows that the individual psychological factors of bicycling of each respondent significantly affect the acceptable bicycling distance of rural residents. Safety attitude (i.e., third common factor) on bicycling infrastructure conditions has a significant and positive influence on the acceptable bicycling distance of rural residents. This result notes that small motor vehicle traffic, dedicated separate bicycle lanes, and secure bicycle parking lot at the destination can significantly encourage rural residents to accept long bicycling distances. Note that 75.67% of the respondents believe that the good quality of the route surface can encourage them to ride a bicycle. However, this infrastructure does not significantly affect the acceptable bicycling distance. That is, the bicycling infrastructure conditions that meet the requirement to choose bicycles and acceptable longer bicycling distance are not precisely the same. The special separate bicycle lane can encourage rural residents to choose bicycles but also satisfy their requirement for long bicycling distances. For bicycling motivation, the second common factor (i.e., convenience) significantly influences the acceptable bicycling distance of rural residents. In particular, rural residents who believe that bicycling is convenient and fast are reluctant to ride long distances. Instead, they believe that short-distance travel can better reflect the convenience brought by bicycles. Thus, people who think that bicycles are convenient and fast may ride bicycles frequently. This conclusion indirectly coincides with the conclusion of Fu and Farber (2017) on bicycling frequency (Fu and Farber 2017).

10.4 Results and Discussion

281

Table 10.7 Multiple linear regression results of the acceptable bicycling distance

Variables Constant Male Age

Original linear regression

The result of linear regression without insignificant variables

Coef.

Coef.

2.494***

P-value 0.000

0.071

0.362

−0.033

0.279

Income

−0.007

0.751

Cycling ancillary facilities

−0.018

0.636

0.034

0.363

Bicycle lane conditions Safety

0.170***

0.000

Other motivations

0.043

0.287

−0.084**

0.026

Convenient

P-value

2.615***

0.000

0.171***

0.000

−0.080**

0.049

Physical activate

0.395***

0.000

0.398***

0.000

Go to work/school

0.293***

0.003

0.281***

0.009

−0.542***

0.000

Bike with children

−0.083

0.433

Go shopping

−0.528***

0.000

Visit friends

0.192**

0.029

To entertainment

0.005

0.963

Others Liking riding motorcycles Liking riding electric bicycles

0.113

0.233

−0.318***

0.003

−0.340***

0.004

0.084***

0.010

0.116***

0.010

−0.083**

0.011

-0.073

0.186

Liking bicycling

0.167***

0.000

0.054

0.322

O_P (Table 10.3)

0.098**

0.019

0.144***

0.000

0.125**

0.011

−0.117***

0.001

O_M (Table 10.3)

−0.063

O_S (Table 10.3)

−0.056

0.210 0.261

O_H (Table 10.3)

0.139***

0.000

P_S (Table 10.3)

0.056

0.352

P_M (Table 10.3) P_P (Table 10.3) P_H (Table 10.3) Rho-squared (R2 = 1−[L(β)/L(c)]) Number of observations *Significant at the 10% level **Significant at the 5% level ***Significant at the 1% level

0.029 −0.139*** 0.015 0.274 374

0.677 0.002 0.770 0.243 374

282

10 The Impact of the Built Environment …

Although 83.43% of the rural residents believe that bicycling can enhance physical fitness, this bicycling motivation does not significantly affect their acceptable bicycling distance. Similar bicycling motivations include protecting the environment and saving money. Therefore, motivation and specific behavior are not constantly consistent (Xing et al. 2018). For bicycling purposes, five of the seven indicators significantly affect the acceptable bicycling distance. Among the indicators of bicycling purpose, physical activity had a significant and positive influence on the acceptable bicycling distance of rural residents, with the largest impact coefficient among all the influencing factors. However, only 10.70% of all the respondents selected physical activity as their bicycling purpose. That is, only a few residents in the rural areas of Sichuan are bicycling for physical activity. However, if the respondents selected physical activity as their bicycling purpose, then they would accept long bicycling distance. Fu and Farber (2017) studied urban residents and showed that 79.55% of the respondents ride bicycles for physical activity, although this endeavor did not significantly affect their bicycling frequency. Thus, the two similar research results indicate that the influence of bicycling purpose on bicycling behavior is indirectly related to the number of respondents who select this bicycling purpose option. Moreover, bicycling to work and school and visiting friends has significant and positive effects on acceptable bicycling distances compared with daily shopping and other purposes of transportation. This aspect reflects the current status of the daily destination distances of rural residents. Typically, workplaces, schools, and relatives and friends are not close to home. For the purpose of these trips, the acceptable bicycling distance is relatively distant (subconsciously). Other daily trips, such as shopping, are often selected at a nearby market or the nearest convenience store. Hence, the acceptable bicycling distance is considerably short. For riding preferences, this study mainly considers the influence of the likeness of bicycles, motorcycles, and electric bicycles on the acceptable bicycling distance of rural residents. The research results (see Table 10.7) show that all indicators of the riding preference have a significant impact on the acceptable bicycling distance of rural residents. This result indicates that liking bicycling is the most important factor in explaining bicycle ownership and use (Handy et al. 2010). Moreover, liking bicycling is closely related to acceptable bicycling distance (Handy et al. 2010; Xing et al. 2010a). The indicators of liking riding a bicycle and motorcycle positively affect the acceptable bicycling distance of rural residents, with liking riding a bicycle having a considerable impact. Thus, rural residents who like bicycling can accept long bicycling distances. Furthermore, those who like to ride motorcycles can accept relatively long bicycling distances. By contrast, rural residents who like riding electric bicycles would accept short bicycling distances. This research conclusion completely illustrates the complementarity of bicycles and motorcycles for travel and the mutual replacement of bicycles and electric bicycles (Ao et al. 2018b).

10.4 Results and Discussion

10.4.6.2

283

Influence of the Rural Built Environment on Acceptable Bicycling Distance

Table 10.7 shows that two of the four objective built environment indicators significantly influence the acceptable bicycling distance of rural residents. The distance from the village center to the nearest health center (O_H) and public transportation station (O_P) has a significant and positive influence on the acceptable bicycling distance of rural residents. That is, the distance between the two locations has a substantial influence. Moreover, the distance from the village center to the nearest school has a significant negative impact on the acceptable bicycling distance. Note that the authors of this study perceive that the market is a trading center in the rural areas of Sichuan and should significantly influence the bicycling behavior of rural residents. However, the market is not significant in this study. The indicators of the perceived built environment of the rural residents have limited impact on their acceptable bicycling distance. Only the perceived convenience to public transportation station significantly and negatively affects the acceptable bicycling distance. This result indicates that rural residents who believe in the convenience of going to public transport stations will accept short bicycling distances. This result is consistent with our expectations. This study compares the influences of the objective and perceived built environment on acceptable bicycling distance of rural residents and determines that only the perceived and objective health center built environment indicator significantly influences the acceptable bicycling distance of rural residents. However, the effects are opposite. This result shows that the objective and perceived built environment have relatively independent effects on the acceptable bicycling distance of rural residents. Moreover, this result is consistent with that of (Ma and Dill 2015) on bicycle frequency. The multiple regression model was used to explore the socio-demographic attributes, individual bicycling psychological factors, and rural built environments on the acceptable bicycling distance of rural residents. The R-squared of the linear regression is 0.274 and the significance of the F-test of the model is sig = 0.000. Therefore, the model rejects the null hypothesis (all parameters are zero). Moreover, this study re-estimated the linear regression model by dropping out the insignificant variables and the results are similar to the initial model (see Table 10.7). Therefore, the results are consistent with our assumptions and this study adopt the original model to interpret the relationship between dependent and independent variables for more information. The influence of the individual bicycling psychological factors of rural residents on their acceptable bicycling distance is significant, followed by the built environment indicators, whereas the impact of social demographic characteristics is limited. However, the socio-demographic characteristics in the relevant research literature significantly affect travel behavior (Fu and Farber 2017; Ma and Dill 2015; Msambichaka et al., 2018). This finding may be related to the characteristics of the left-behind population in rural Sichuan.

284

10 The Impact of the Built Environment …

10.5 Strengths and Limitations Given that this study is the first on the bicycling behavior of rural residents in Sichuan, the current research completely considers the impact of the individual bicycling psychological factors on the acceptable bicycling distances of rural residents. This research also compares and analyzes the influences of the objective and perceived built environment of rural residents on their acceptable bicycling distances. Thus, the current study substantially contributes in explaining the bicycling behavior (i.e., acceptable bicycling distance) in rural Sichuan (i.e., undeveloped areas). Moreover, this study deploys two multivariate models to accurately fit the relationships between variables. Accordingly, the results are consistent. This study has two limitations. First, cross-sectional data were used without considering the impact of changes on the rural built environment and the psychological determinants of rural residents on the acceptable bicycling distance. Second, the aggregate data of the objective built environment in rural areas used in this study (i.e., distance from the village center to various destinations) do not specify the distance from each sample family to various destinations. Therefore, accurately obtaining the inner link between the objective and perceived built environment is virtually impossible. This finding is mainly the result of lack of rural geographic information data. Thus, additional research in the future is recommended.

10.6 Conclusion and Policy Implications An increasing number of studies have focused on bicycling frequency and the choices made in this regard by residents in large cities. The number of rural household bicycles has continued to decline owing to rapid urbanization and new rural construction in China. In all modes of travel, the proportion of bicycling is the lowest. The current study uses face-to-face questionnaire surveys and on-site measurement data to analyze the impact of individual cycling psychology and built environment on the acceptable cycling distance of rural residents. The purpose of this research is to provide a theoretical basis for the ecological construction of new rural areas in China and encourage low-carbon travel for rural residents. The results suggest that further investment in the construction of special bicycle lanes is needed to provide efficient road infrastructure for rural residents, thereby encouraging them to choose bicycles as a mode of transport. Moreover, separating bike lanes from motor vehicle lanes reduces motor vehicle traffic. The perceived convenience and speed of cycling will encourage rural residents to travel short distances by bicycle. Therefore, daily destinations, such as grocery stores, shops, and markets, should be planned to be within the range of the acceptable bicycling travel distances (e.g., the average acceptable bicycling distance is 2.9 km; see Fig. 10.4). The riding preference of rural residents significantly affects their acceptable cycling

10.6 Conclusion and Policy Implications

285

distance. Therefore, local governments should improve the provision of information on the advantages of cycling, such as benefits to physical and mental health, zero carbon emissions, and environmental protection. In addressing the bicycling requirements of rural residents, local governments and provincial agencies that are involved in planning the construction of new rural areas should consider the impact of rural infrastructure and perceived built environment on the travel behavior of rural residents.

References Akar, G., Fischer, N., & Namgung, M. 2013. Bicycling choice and gender case study: The Ohio State University. International Journal of Sustainable Transportation, 7(5): 347–365. Ao, Y., Chen, C., Yang, D., & Wang, Y. 2018a. Relationship between rural built environment and household vehicle ownership: An empirical analysis in rural Sichuan, China. Sustainability, 10(5): 1–18. Ao, Y., Chen, C., Yang, D., & Wang, Y. 2018b. Relationship between rural built environment and household vehicle ownership: An empirical analysis in rural Sichuan, China. Sustainability, 10(5): 1566. Ao, Y., Yang, D., Chen, C., & Wang, Y. 2019. Exploring the effects of the rural built environment on household car ownership after controlling for preference and attitude: Evidence from Sichuan, China. Journal of Transport Geography, 74: 24–36. Beenackers, M. A., Foster, S., Kamphuis, C. B. M., Titze, S., Divitini, M., Knuiman, M., van Lenthe, F. J., & Giles-Corti, B. 2012. Taking up cycling after residential relocation built environment factors. American Journal of Preventive Medicine, 42(6): 610–615. Berrigan, D., Troiano, R. P., McNeel, T., DiSogra, C., & Ballard-Barbash, R. 2006. Active transportation increases adherence to activity recommendations. American Journal of Preventive Medicine, 31(3): 210–216. Bonham, J., & Koth, B. 2010. Universities and the cycling culture. Transportation Research Part D-Transport and Environment, 15(2): 94–102. Broach, J., Dill, J., & Gliebe, J. 2012. Where do cyclists ride? A route choice model developed with revealed preference GPS data. Transportation Research Part a-Policy and Practice, 46(10): 1730–1740. Brownson, R. C., Hoehner, C. M., Day, K., Forsyth, A., & Sallis, J. F. 2009. Measuring the built environment for physical activity state of the science. American Journal of Preventive Medicine, 36(4): S99–S123. Cao, X. Y. 2015a. Examining the impacts of neighborhood design and residential self-selection on active travel: a methodological assessment. Urban Geography, 36(2): 236–255. Cao, X. Y. 2015b. Examining the relationship between neighborhood built environment and travel behavior: a review from the US perspective. Urban Geography, 30(4): 46–52. Chataway, E. S., Kaplan, S., Nielsen, T. A. S., & Prato, C. G. 2014. Safety perceptions and reported behavior related to cycling in mixed traffic: A comparison between Brisbane and Copenhagen. Transportation Research Part F-Traffic Psychology and Behaviour, 23: 32–43. Daley, M., & Rissel, C. 2011. Perspectives and images of cycling as a barrier or facilitator of cycling. Transport Policy, 18(1): 211–216. Damant-Sirois, G., Grimsrud, M., & El-Geneidy, A. M. 2014. What’s your type: A multidimensional cyclist typology. Transportation, 41(6): 1153–1169. de Geus, B., De Smet, S., Nijs, J., & Meeusen, R. 2007. Determining the intensity and energy expenditure during commuter cycling. British Journal of Sports Medicine, 41(1): 8–12.

286

10 The Impact of the Built Environment …

Dill, J. 2009. Bicycling for transportation and health: The role of infrastructure. Journal of Public Health Policy, 30: S95–S110. Elvik, R., & Bjornskau, T. 2005. How accurately does the public perceive differences in transport risks? An exploratory analysis of scales representing perceived risk. Accident Analysis and Prevention, 37(6): 1005–1011. Ewing, R., & Handy, S. 2009. Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban design, 14(1): 65–84. Fernández-Heredia, Á., Monzón, A., & Jara-Díaz, S. 2014. Understanding cyclists’ perceptions, keys for a successful bicycle promotion. Transportation Research Part A: Policy and Practice, 63: 1–11. Fishman, E., Washington, S., & Haworth, N. 2012. Barriers and facilitators to public bicycle scheme use: A qualitative approach. Transportation Research Part F-Traffic Psychology And Behaviour, 15(6): 686–698. Foster, C. E., Panter, J. R., & Wareham, N. J. 2011. Assessing the impact of road traffic on cycling for leisure and cycling to work. International Journal of Behavioral Nutrition and Physical Activity, 8. Fu, L., & Farber, S. 2017. Bicycling frequency: A study of preferences and travel behavior in Salt Lake City, Utah. Transportation Research Part A: Policy and Practice, 101: 30–50. Garrard, J., Rose, G., & Lo, S. K. 2008. Promoting transportation cycling for women: The role of bicycle infrastructure. Preventive Medicine, 46(1): 55–59. Gatersleben, B., & Appleton, K. M. 2007a. Contemplating cycling to work: Attitudes and perceptions in different stages of change. Transportation Research Part A: Policy and Practice, 41(4): 302–312. Gatersleben, B., & Appleton, K. M. 2007b. Contemplating cycling to work: Attitudes and perceptions in different stages of change. Transportation Research Part a-Policy And Practice, 41(4): 302–312. Gebel, K., Bauman, A. E., Sugiyama, T., & Owen, N. 2011. Mismatch between perceived and objectively assessed neighborhood walkability attributes: Prospective relationships with walking and weight gain. Health & Place, 17(2): 519–524. Götschi, T., Castro, A., Deforth, M., Miranda-Moreno, L., & Zangenehpour, S. 2018. Towards a comprehensive safety evaluation of cycling infrastructure including objective and subjective measures. Journal of Transport & Health, 8: 44–54. Greene, W. H. 2003. Econometric analysis. Pearson Education India. Handy, S. 2005. Critical assessment of the literature on the relationships among transportation, land use, and physical activity. Transportation Research Board and the Institute of Medicine Committee on Physical Activity, Health, Transportation, and Land Use. Resource Paper for TRB Special Report, 282(1): 1–81. Handy, S., & McCann, B. 2011. The regional response to federal funding for bicycle and pedestrian projects. Journal of the American Planning Association, 77(1): 23–38. Handy, S. L., & Xing, Y. 2011. Factors correlated with bicycle commuting: A study in six small U.S. cities. International Journal of Sustainable Transportation, 5(2): 91–110. Handy, S. L., Xing, Y., & Buehler, T. J. 2010. Factors associated with bicycle ownership and use: a study of six small U.S. cities. Transportation, 37(6): 967–985. Haskell, W. L. 2004. General dose response issues concerning physical activity and health. Health Enhancing Physical Activity, 6: 149–168. Heinen, E., Maat, K., & van Wee, B. 2011. The role of attitudes toward characteristics of bicycle commuting on the choice to cycle to work over various distances. Transportation Research Part D-Transport and Environment, 16(2): 102–109. Heinen, E., van Wee, B., & Maat, K. 2010. Commuting by bicycle: An Overview of the literature. Transport Reviews, 30(1): 59–96. Hong, J., & Shen, Q. 2013. Residential density and transportation emissions: examining the connection by addressing spatial autocorrelation and self-selection. Transportation Research Part D: Transport and Environment, 22: 75–79.

References

287

Kerr, J., Emond, J. A., Badland, H., Reis, R., Sarmiento, O., Carlson, J., Sallis, J. F., Cerin, E., Cain, K., Conway, T., Schofield, G., Macfarlane, D. J., Christiansen, L. B., Van Dyck, D., Davey, R., Aguinaga-Ontoso, I., Salvo, D., Sugiyama, T., Owen, N., Mitas, J., & Natarajan, L. 2016. Perceived neighborhood environmental attributes associated with walking and cycling for transport among adult residents of 17 cities in 12 countries: The IPEN study. Environmental Health Perspectives, 124(3): 290–298. Krizek, K. J., & Johnson, P. J. 2006. Proximity to trails and retail: Effects on urban cycling and walking. Journal of the American Planning Association, 72(1): 33–42. Kuzmyak, J. R., Walters, J., Bradley, M., & Kockelman, K. M. 2014. Estimating bicycling and walking for planning and project development: A guidebook. Ma, L., & Dill, J. 2015. Associations between the objective and perceived built environment and bicycling for transportation. Journal of Transport & Health, 2(2): 248–255. Moore, S. C., Lee, I. M., Weiderpass, E., Campbell, P. T., Sampson, J. N., Kitahara, C. M., Keadle, S. K., Arem, H., de Gonzalez, A. B., Hartge, P., Adami, H. O., Blair, C. K., Borch, K. B., Boyd, E., Check, D. P., Fournier, A., Freedman, N. D., Gunter, M., Johannson, M., Khaw, K. T., Linet, M. S., Orsini, N., Park, Y., Riboli, E., Robien, K., Schairer, C., Sesso, H., Spriggs, M., Van Dusen, R., Wolk, A., Matthews, C. E., & Patel, A. V. 2016. Association of leisure-time physical activity with risk of 26 types of cancer in 1.44 million adults. Jama Internal Medicine, 176(6): 816–825. Msambichaka, B., Abdul, R., Abdulla, S., Klatser, P., Tanner, M., Kaushik, R., Bringolf-Isler, B., Geubbels, E., & Eze, I. C. 2018. A cross-sectional examination of physical activity levels and their socio-demographic determinants in Southern Tanzania. International Journal of Environmental Research and Public Health, 15(6). National Bureau of Statistics of China. 1986–2017. China Statistical Yearbook. Nocon, M., Hiemann, T., Mueller-Riemenschneider, F., Thalau, F., Roll, S., & Willich, S. N. 2008. Association of physical activity with all-cause and cardiovascular mortality: A systematic review and meta-analysis. European Journal of Cardiovascular Prevention & Rehabilitation, 15(3): 239–246. Parkin, J., Wardman, M., & Page, M. 2007. Models of perceived cycling risk and route acceptability. Accident Analysis and Prevention, 39(2): 364–371. Parkin, J., Wardman, M., & Page, M. 2008. Estimation of the determinants of bicycle mode share for the journey to work using census data. Transportation, 35(1): 93–109. Pucher, J., & Buehler, R. 2008. Making cycling irresistible: Lessons from the Netherlands, Denmark and Germany. Transport Reviews, 28(4): 495–528. Pucher, J., Dill, J., & Handy, S. 2010. Infrastructure, programs, and policies to increase bicycling: An international review. Preventive Medicine, 50: S106–S125. Pucher, J., & Renne, J. L. 2003. Socioeconomics of urban travel: Evidence from the 2001 NHTS. Transportation Quarterly, 57(3): 49–77. Rodriguez, D. A., Evenson, K. R., Roux, A. V. D., & Brines, S. J. 2009. Land use, residential density, and walking the multi-ethnic study of atherosclerosis. American Journal of Preventive Medicine, 37(5): 397–404. Saelens, B. E., Sallis, J. F., & Frank, L. D. 2003. Environmental correlates of walking and cycling: Findings from the transportation, urban design, and planning literatures. Annals of Behavioral Medicine, 25(2): 80–91. Sallis, J. F., Conway, T. L., Dillon, L. I., Frank, L. D., Adams, M. A., Cain, K. L., & Saelens, B. E. 2013. Environmental and demographic correlates of bicycling. Preventive Medicine, 57(5): 456–460. Sallis, J. F., Frank, L. D., Saelens, B. E., & Kraft, M. K. 2004. Active transportation and physical activity: opportunities for collaboration on transportation and public opportunities health research. Transportation Research Part a-Policy and Practice, 38(4): 249–268. Sener, I. N., Eluru, N., & Bhat, C. R. 2009. An analysis of bicycle route choice preferences in Texas, US. Transportation, 36(5): 511–539. Shaheen, S. A., Guzman, S., & Zhang, H. 2010. Bikesharing in Europe, the Americas, and Asia past, present, and future. Transportation Research Record(2143): 159–167.

288

10 The Impact of the Built Environment …

Stinson, M. A., & Bhat, C. R. 2003. Commuter bicyclist route choice - Analysis using a stated preference survey. Pedestrians and Bicycles 2003(1828): 107–115. Thakuriah, P., Metaxatos, P., Lin, J. N., & Jensen, E. 2012. An examination of factors affecting propensities to use bicycle and pedestrian facilities in suburban locations. Transportation Research Part D-Transport and Environment, 17(4): 341–348. Thomas, B., & DeRobertis, M. 2013. The safety of urban cycle tracks: A review of the literature. Accident Analysis And Prevention, 52: 219–227. Timperio, A., Ball, K., Salmon, J., Roberts, R., Giles-Corti, B., Simmons, D., Baur, L. A., & Crawford, D. 2006. Personal, family, social, and environmental correlates of active commuting to school. American Journal of Preventive Medicine, 30(1): 45–51. Titze, S., Stronegger, W. J., Janschitz, S., & Oja, P. 2008. Association of built-environment, socialenvironment and personal factors with bicycling as a mode of transportation among Austrian city dwellers. Preventive Medicine, 47(3): 252–259. Van Acker, V., Derudder, B., & Witlox, F. 2013. Why people use their cars while the built environment imposes cycling. Journal Of Transport And Land Use, 6(1): 53–62. Van Dyck, D., Veitch, J., De Bourdeaudhuij, I., Thornton, L., & Ball, K. 2013. Environmental perceptions as mediators of the relationship between the objective built environment and walking among socio-economically disadvantaged women. International Journal of Behavioral Nutrition and Physical Activity, 10. Wang, J. Y., Mirza, L., Cheung, A. K., & Moradi, S. 2012. Transforming Auckland into a bicyclefriendly city: Understanding factors influencing choices of cyclists and potential cyclists. Paper presented at the Australasian Transport Research Forum, Perth. Willis, D. P., Manaugh, K., & El-Geneidy, A. 2015. Cycling under influence: summarizing the influence of perceptions, attitudes, habits, and social environments on cycling for transportation. International Journal of Sustainable Transportation, 9(8): 565–579. Xing, Y., Handy, S. L., & Mokhtarian, P. L. 2010a. Factors associated with proportions and miles of bicycling for transportation and recreation in six small US cities. Transportation Research Part D-Transport and Environment, 15(2): 73–81. Xing, Y., Handy, S. L., & Mokhtarian, P. L. 2010b. Factors associated with proportions and miles of bicycling for transportation and recreation in six small US cities. Transportation Research Part D: Transport and Environment, 15(2): 73–81. Xing, Y., Volker, J., & Handy, S. 2018. Why do people like bicycling? Modeling affect toward bicycling. Transportation Research Part F: Traffic Psychology and Behaviour, 56: 22–32. Yang, L., Sahlqvist, S., McMinn, A., Griffin, S. J., & Ogilvie, D. 2010. Interventions to promote cycling: Systematic review. British Medical Journal, 341.

Chapter 11

The Impact of Building Features and Attitudes Regarding Water Conservation on the Water Use Behavior of Rural Residents

Abstract This chapter takes rural residents from four villages in Chengdu as the research object in order to conduct research on the impact of building features and attitudes to water conservation. A total of 165 valid questionnaires are collected after face-to-face interviews. First, descriptive analysis is used to analyze the current situation of rural residents’ water conservation behavior. Second, exploratory factor analysis and the binary logistic regression model are used to explore the relationship between building characteristics, water conservation attitudes, and water conservation behavior. Results show that: (1) rural residents’ water conservation attitude plays an important role in water conservation behavior; with “environmental values” being the most significant factors, followed by “saving money and joint participation;” (2) rural building characteristics such as layout of the kitchen and shower facilities significantly affect the water conservation behavior of rural residents. Based on the analysis, several suggestions are offered for improving water-saving in rural Chengdu. These are strengthening the publicity and education of water-saving behavior, and subsidizing water-saving facilities. This research provides a theoretical basis for local government departments to formulate relevant policies, and serves as a reference for the protection of water resources in other rural areas. Keywords Building characteristics · Water conservation behavior · Water conservation attitude

11.1 Introduction Water resource plays an important role in people’s life (Liu et al. 2017). However, the global water crisis is increasing daily with rapid urbanization and growing population (Domene and Saurí 2006; Russell and Fielding 2010). China is one of the countries with the poorest per capita water resources in the world. The per capita water resource was only 1,971.8 m3 in 2018 (Source: National Bureau of Statistics of the People’s Republic of China 2019 Table 11.8), accounting for 1/4 of the world average (Qiu 2018). Water pollution, water scarcity, and flooding are the top three challenges to water resource in China (Qian et al. 2002). Groundwater is overdrawn by 10 billion © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_11

289

290

11 The Impact of Building Features …

cubic meters annually on average, but water supply always cannot meet demand (Liu and Speed 2009). The lack of water resources has seriously affected people’s life not only in the urban area, but also in the rural area. For instance, more than 300 million residents in rural China have no access to safe drinking water (Cho 2011). Thus, research on water shortage problem in rural China is of great significance. Water conservation efficiently alleviates the challenges to water shortage (Yin et al. 2006) (Lin et al. 2015). This issue has gained much interest from practitioners and academia. There are various determinants affecting water conservation behavior. Various determinants affect water conservation behavior. Several researchers argue that socio–demographic factors help predict water conservation behavior (Willis et al. 2013a). Some studies suggest that economic incentive, such as price control, is effective to encourage saving of water (Worthington and Hoffman 2008). Several others propose that psychological characteristics, such as attitude, are critical determinants of water conservation (Aprile and Fiorillo 2017; Willis et al. 2011b). Many studies believe that degree of urbanization will affect the water conservation behavior of residents (Wang et al. 2011; Zhang et al. 2009). Most of the existing research focuses more on developed urban areas and less on rural areas (Li 2011). Past decades, the Chinese government built many infrastructures in rural areas, such as roads, hospitals, schools, and sewage treatment plants, which has greatly improved the living standards of rural residents. Moreover, the demand for water resources in rural areas is growing. China’s rural areas are facing several challenges, such as lack of water conservancy infrastructure, huge gap of the water quality between urban and rural areas, and scarce water supply (Zhu et al. 2015). Therefore, it is imperative to investigate the water conservation behavior of rural residents. Therefore, this study takes rural residents in Chengdu, China as the research object and investigates the influence of various determinants on water conservation behavior using exploratory factor analysis (EFA) and binary logistic regression. Based on the results, several policy suggestions are made on the planning, construction, and management of rural water resources in Chengdu which can also be promoted to other regional governments. The rest of the study is organized as follows. Section 11.2 reviews related studies. Section 11.3 introduces research design and data collection. Section 11.4 describes the empirical analysis results and discussion. Section 11.5 concludes the study.

11.2 Literature Review The water conservation behavior of residents is quite complex and mainly affected by socio-demographical factors, water conservation attitudes, and building characteristic, etc.

11.2 Literature Review

291

11.2.1 Social Demographic Factors Related research shows that the social demographic characteristics of family and individual can significantly affect the domestic water consumption. The influencing factors include income, age, belief, gender, family population size, etc. Many scholars have studied the relationship between household income and water consumption. The results show that the higher the income is, the higher the water consumption (Jiang et al. 2019; Otaki et al. 2017). High-income families often have more water equipment, such as dishwasher, washing machine, swimming pool, and outdoor garden watering facilities, than low-income families (Lam 2006). Their daily water demand is much greater than that of low-income cohort. Fan believes that highincome groups have a greater capacity to pay, and their water fees only account for a small part of their total income. Thus, they are not sensitive to the regulation and control of water price, and are more likely to waste water resource (Fan 2014). Social variables such as age, household component, and education affect water usage as well. Foreign studies have found that the water consumption of families with numerous teenagers is generally on the high side, which is mainly because the water consumption of teenagers to pursue more comfortable living habits is much greater than that of the elderly (Thomas and Nauges 2000). Studies in China also show that the elderly are usually willing to take more active measures to save water mainly because they have a stronger sense of saving compared with the young (Zhao et al. 2019). The water consumption of residents of different races or faiths is also quite different (Yan et al. 2018), and that of collectivist families is relatively lower than that of other families (Andersen 2008). Most of the female groups are the main users of domestic water; thus, their behavior represents that of the whole family. As the manager or maintainer of the family, women often use more water than men. They usually take more active measures to save water (Tong et al. 2017). Numerous studies show that the greater the number of permanent residents is, the lower the per capita water consumption (Höglund 1999). Chen et al. (2005) also proposed that the total household water consumption and the proportion of the family educated population have a significant correlation.

11.2.2 Water Conservation Attitudes People’s attitudes also affect people’s behavior (Conner and Armitage 2006; Cook and Berrenberg 1981). Over the years, many researchers attempt to understand the rationale of the behavior from the fundamental psychology. Theory of planned behavior (Ajzen 1991) has been a leading model. This theory describes that an “individual makes behavioral decisions based on rational considerations”, which implies that human behavior is an complexed result of behavioral intention, attitude, self-efficacy (Koop et al. 2019). Attitude which is defined as “strengths of beliefs about consequences of behaviors and evaluations of such consequences” (Ajzen and

292

11 The Impact of Building Features …

Fishbein 2000) plays an important role in determining people’s behavior (Cook and Berrenberg 1981; Lee and Tansel 2013). Different environmental consciousness or water-saving attitudes will lead to different water-use behaviors (Chaudhary et al. 2018; Kumar Chaudhary et al. 2017). Generally, the stronger the environmental awareness is, the lower the water consumption. Household water consumption is higher given an insufficient or incorrect understanding of environmental awareness (Corral-Verdugo et al. 2003). Based on 728 questionnaires collected in Bulgaria, Clark and Finley (Clark and Finley 2007) testified that an awareness of future water shortage and a positive environmental attitude have a significant influence on water conservation behavior. Willis et al. (2011a) found a significantly positive relationship between general environmental attitudes and water conservation behavior through field experiment and questionnaire survey in Australia. Previous research argues that pricing control is an efficient way to motivate water conservation behavior, that is, multi-tier pricing mode. Multi-tier pricing divides the water consumption of residents into several levels, and different water prices are set in each stage (Ma et al. 2018). Average water prices increase with the progression of the stage. Increasing water price has been widely proven effective in reducing domestic water consumption (Nieswiadomy and Molina 1991). However, more current research has found that water attitude is much more influential than price control. Hu (2006) pointed out that the effect of water-saving publicity is more evident than that of water price on water-saving behavior. Mu, Zhang (Mu, Zhang, & Ma, 2014) found that compared with price regulation, residents’ water-saving awareness plays a more important role in water-saving behavior based on survey data of 5,000 Beijing residents in 2010. Chang (2013) proposed that people could be led to save water through publicity and education, specifically using radio, banners, posters, or flyers. Related research shows that water conservation attitudes are vital to water conservation behavior.

11.2.3 Building Characteristics Characteristics of housing, such as age, density, size, location, and number of taps, have a great effect on water consumption. Generally, the larger the house is, the greater the amount of water that will be used. Studies have shown that low-density buildings tend to use more water than high-density ones because they have more landscaping (Archibald 1998). Kumar Chaudhary pointed out that water conservation is related to the residents’ perception of landscape/outdoor benefits. The more beneficial their view is, the higher the possibility that they will adopt water conservation behavior (Chaudhary et al. 2019). Huang and Huang investigated several communities in Shanghai by principal component analysis and found a significant difference between house price and house age on water consumption (Huang et al. 2017). The water consumption of ordinary or low-price communities is significantly lower than that of high-price communities.

11.2 Literature Review

293

In summary, precious, scarce water resource has attracted many scholars to focus on the research of water conservation behavior, but a few research gaps remain: (1) Most of the existing research focuses on developed areas, lacking attention to developing areas. Economic development level and residents’ awareness of water conservation are quite dissimilar, which need to be differentiated. (2) Domestic scholars’ research on water resource conservation and utilization mostly focus on urban areas. Research is difficult, and the literature on rural areas is relatively limited due to the scattered residents. (3) The existing research mainly focuses on residential water consumption and domestic water quota of urban residents (Jiang et al. 2019; Xiong 2018). Their focus is more on economic incentive or regulations, and social demographic characteristics. Therefore, this study aims to explore the influences of building characteristics and water-saving attitude on water conservation behavior after controlling the social demographic variables, and provide theoretical implications for constructing sustainable rural areas.

11.3 Research Design and Data Collection 11.3.1 Sample Village Selection The subjects of this study are rural residents in Chengdu, China. Chengdu is one of the pilot areas of urban–rural integration in China and has a wide range of rural areas, with 52 towns and 2,658 villages. After presurvey, the final selected typical sample villages should meet two conditions: 1. The villages need to be connected with tap water because this research studies the water-saving behavior of residents, which is measured using water-saving appliances (water-saving faucets, water-saving toilets, and water-saving washing machines). Such water-saving appliances need to be used with tap water. 2. The research team has a good communication with the selected typical sample villages. Carrying out household interview and comprehensive questionnaire survey in rural areas is difficult, and the attitude of village cadres and the trust of villagers are important. Therefore, a good communication channel with the local community must be established. The research team solves this problem as follows: ➀ At least one undergraduate student originally from the sample village must be willing to participate in the survey and act as the contact person. ➁ The village cadres are contacted in advance; they accept and are willing to support the research group for the on-site survey. ➂ Villagers’ attitude toward the household questionnaire survey must be determined, and the village will be selected if most households welcome the research group. Based on the above principles and preliminary selection, four villages around Chengdu were finally selected as the sample villages. These villages are Jin Ning Village in Pidu District, Tian Du Village in Xinjin County, Wu Xing community in

294

11 The Impact of Building Features …

Fig. 11.1 Location of the sample villages

Pujiang County, and Hua Guo Village in Longquanyi District. The specific geographical location is shown in Fig. 11.1. The selected villages are all new villages built in recent years. The specific investigation area of each village was limited within a circle of 1 km from the village center due to the complex topography of rural areas and the large difference in size, as shown in Fig. 11.2. The original map was downloaded from the Qiaofeng map, and ArcGIS 10.2 was used to depict the land for buildings. In Fig. 11.2, the village center is set as the center of the circle, and the research area is within a 1 km radius. The solid black line is the village road, and the pink area is the construction land.

11.3.2 Data Collection 11.3.2.1

Questionnaires

The questionnaire was revised after preliminary investigation and field visit. It was mainly divided into four parts: social demographic factors, building characteristic factors, water conservation attitudes, and water conservation behavior. Except the specific questions in the first part, all the other parts adopt multiple-choice question. 1. Social-demographic factors. These factors mainly include the number of permanent residents, gender, age, annual income, and education level (Willis et al. 2013b). The basic information of the family and individual is an important part of the questionnaire survey. On this basis, we can understand the different background of water-saving behavior of rural residents in Chengdu. Moreover, we can study the specific attitude of villagers with different levels of income and living environments to water-saving behavior.

11.3 Research Design and Data Collection

295

Fig. 11.2 Study area of sample villages

2. Building characteristic factors. These factors mainly include house size, building age, lighting and ventilation conditions, number of faucets and toilets, and the length of time for hot water to come out of the shower and kitchen. The watersaving behavior of residents is closely related to the structure, age, and size of houses (Roccaro et al. 2011). Through the study of building information, we can find out the influence of various building cha We can determine the influence of various building characteristics on the water-saving behavior of villagers through the study of building information. Accordingly, building houses reasonably can effectively promote the water-saving behavior of residents. 3. Water conservation attitudes. Specific questions involve personal attention to environmental issues, understanding of water resources, environmental control

296

11 The Impact of Building Features …

concept of water resources, and other attitudes and perception issues. Environmental attitude is initially considered a tendency of subjects to environmental behavior, which influences the prediction of environmental behavior (Ajzen 1991). Specific questions involve personal attention to environmental issues, understanding of water resources, environmental control concept of water resources, and other attitudes and perception issues. Environmental attitude is initially considered a tendency of subjects to environmental behavior, which influences the prediction of environmental behavior (Polonsky et al. 2014). Therefore, this study mainly reflects the psychological tendency of the interviewees to save water through the expression and questions of water-saving-related attitude. This part contains 25 questions, and details of the measurement are presented in Table 11.5. 4. Water conservation behavior. The water-saving behavior involves many indexes which are difficult to quantify. The use of water-saving appliances can greatly reduce the domestic water consumption of residents and play a good role in water saving; thus, it can be taken as a representative. A study in the United States, Australia, and the United Kingdom has shown that the water consumption of households equipped with water-saving appliances such as water-saving toilets, water-saving faucets, or low-flow shower heads has been reduced by 9– 12%. If all existing household water equipment are replaced with efficient watersaving equipment, water consumption can be reduced by 35–50% (Inman and Jeffrey 2006). Geller et al. (1983) found that water-saving appliances can significantly reduce water consumption. Roccaro et al. (2011) proposed that technical support has a significant effect on water saving, including installation of water meters, implementation, and popularization of water-saving appliances. Therefore, whether to use water-saving equipment is selected to represent water-saving behavior and works as the explanatory variable in this research. From the above, all the main indicators of the questionnaire and their measurement can be shown in following Table 11.1.

11.3.2.2

Data Collection

Data collection mainly went through two stages: presurvey and formal questionnaire survey. The team conducted a presurvey in January 2018. According to the presurvey data, the validity of the questionnaire was verified, and the questionnaire was improved and modified. A formal survey was conducted in August 2018. In each village, two researchers formed a team. Considering the average low education level of villagers in China’s rural areas, they may not be able to understand the questionnaire correctly. Thus, this survey adopted one-to-one and face-to-face interview. Moreover, our team contacted the village committee or the leader and sought their help to motivate the villagers before the questionnaire survey. In the sample villages, random sampling was adopted for household survey, rural household was randomly selected, if the selected rural household do not agree to participant in the survey, the

11.3 Research Design and Data Collection

297

Table 11.1 Variables and measurement Variables

Measurement

Gender

1 = Male; 2 = Female

Age

1 = 0–18 years old; 2 = 19–45 years old; 3 = 46–59 years old; 4 = 60–74 years old; 5 = 75 years old or above

Education: highest degree of the family

1 = no education; 2 = primary school; 3 = middle school; 4 = high school (or vocational and technical secondary schools); 5 = junior college (or higher vocational); 6 = Bachelor degree or above

Household size

1 = 1; 2 = 2; 3 = 3; 4 = 4; 5 = 5; 6 = 6 or above

Household composition

1 = alone; 2 = couple; 3 = couple and elderly parents; 4 = couple and their children; 5 = three generations; 6 = others

Annual income per family

1 = 0–20 K; 2 = 20–40 K; 3 = 40–60 K; 4 = 60 K or above;

Building age

1 = 0–2 years; 2 = 3–4 years; 3 = 5–6 years; 4 = 7–8 years; 5 = 8 years or above

Size of the house

1 = 0–99 m 2 ; 2 = 100–139 m 2 ; 3 = 140–179 m 2 ; 4 = 180–219 m 2 ; 5 = 220 m 2 or above

Water heating time from the shower or the kitchen

1 = 0–15 s; 2 = 15–30 s; 3 = 30–45 s; 4 = 45–60 s; 5 = 60 s or above;

Lighting and ventilation conditions

1 = Very poor; 2 = Poor; 3 = commonly; 4 = preferably; 5 = good

No. of taps

1 = 2 or less; 2 = 3; 3 = 4; 4 = 5; 5 = 6 or above

No. of toilets

1 = 1; 2 = 2; 3 = 3; 4 = 4; 5 = 5 or above

25 statements of water-saving attitudes

1 = Totally disagree; 2 = Quite disagree; 3 = Neutral; 4 = Quite agree; 5 = Totally agree

Use of water-saving appliances

0 = no; 1 = yes

researcher will Randomly transfer to the next rural household. Each villager would get a reward after completing the questionnaire. These measures helped increase the reliability of the questionnaire.

11.3.2.3

Sample Component

In this survey, 200 questionnaires were distributed, and 186 questionnaires were finally collected because the interviewees gave up answering during the survey. After double-checking, 21 invalid questionnaires were discarded, 165 valid questionnaires were obtained, and the sample effective rate was 82.5%. Details are shown in Table 11.2. The final questionnaires consisted of the following: 39 in Jin Ning Village, accounting for 23.64%; 50 in Tian Du Village, accounting for 30.30%;

298

11 The Impact of Building Features …

Table 11.2 Sample component Villages

Number of efficient questionnaires

Number of household to be sampled

Total households

Proportion (%)

Jin Ning

39

24.03

621

6.28

Tian Du

50

24.55

1376

3.63

Wu Xing

39

23.92

554

7.04

Hua Guo

37

23.49

781

4.74

39 in Wu Xing community, accounting for 23.64%; and 37 in Hua Guo Village, accounting for 22.42%. The basic data show 621 households in Jin Ning Village, 1,376 households in Tian Du Village, 554 households in Wu Xing community, and 781 households in Hua Guo Village. The proportions of households in the sample village are 6.28%, 3.63%, 7.04% and 4.74%. Thus, the distribution proportion of the number of questionnaires in each village is reasonable and representative at the sampling level with an precision of 20% (Fernandez et al. 2018).

11.3.3 Descriptive Analysis 11.3.3.1

Social-Demographic Variables

In terms of social and demographic factors, this study starts from two dimensions: individual and family. The statistical result of the original data is shown in Table 11.3. From the individual perspective, the number of women in this survey far exceeds that of men, accounting for 69.3% of the total mainly because most of the male villagers work outside or in farming, whereas most of the female villagers stay at home. In rural areas, women are often responsible for household living, laundry, and other work. Their water-saving behavior can represent the overall water-saving behavior of the family. The age distribution of the respondents is representative. Respondents who are over 75 years were strictly removed during data processing because respondents who are very much advanced in years may have problems with the questionnaire. The education level of the respondents is mainly junior high school, accounting for 43.6%; followed by primary school and senior high school (or vocational high school, technical secondary school), accounting for 22.4% and 8.2%, respectively. The education level of junior high school (or vocational high school) and above is very small. For individual annual income, distinguishing between individual and family income is not practical because most of the income of rural residents comes from the planting industry, which is run by the entire family. Thus, only the family annual income is considered. The number of family permanent residents is mostly distributed in the range of 2–5, accounting for 95%. Most of the families surveyed are three generations living together, accounting for 50.3% of the total; followed by two couples living together

11.3 Research Design and Data Collection

299

Table 11.3 Social demographic factors Factors

No. of sample

Percentage

Gender

Male

Age

0–18

6

3.6

19–45

63

38.2

46–59

59

35.8

60–74

37

22.4

0

0

Female

75 and above Household size

Household composition

Highest family education background

Education background of the interviewee

Family annual income

50

30.3

115

69.7

1

9

5.5

2

32

19.4

3

39

23.6

4

23

13.9

5

38

23.0

6 and above

24

14.5

Alone

4

2.4

Couple

14

8.6

Couple and elderly parents

8

4.8

Couple and their children

41

24.8

Three generations

83

50.4

Others

15

9.0

No education

1

0.6

Primary school

6

3.6

Junior middle school

36

21.8

High school (or vocational and technical secondary schools)

55

33.3

Junior college (or higher vocational)

25

15.2

Bachelor degree or above

42

25.5

No education

5

3.0

Primary school

37

22.4

Junior middle school

72

43.6

High school (or vocational and technical secondary schools)

30

18.2

Junior college (or higher vocational)

9

5.5

Bachelor degree or above

12

7.3

0–20 K (20 K not included)

28

17.0

20–40 K (40 K not included)

53

32.1

40–60 K (60 K not included)

43

26.1 (continued)

300

11 The Impact of Building Features …

Table 11.3 (continued) Factors

No. of sample

60 K and above

41

Percentage 24.8

with their children, accounting for 24.8%; the proportion of couples living together with their parents or living alone is very small. The highest percentage of educational background of family members is high school (or vocational high school, technical secondary school), followed by undergraduate and above, junior high school, junior college (or vocational high school), and primary school.

11.3.3.2

Building Characteristic Variables

The building characteristic has a significant influence on water consumption. Thus, the questionnaire survey specifically investigates the information of the houses where the respondents live. The related variables include housing size, housing age, lighting, ventilation status, number of faucets and toilets, and the length of time for hot water to come out of the shower and kitchen. The preliminary analysis results are shown in Table 11.4. The housing size is mainly concentrated in 100–180 m2 , accounting for 65.4%. The housing size of a few large families is much beyond the average level because many newly built houses in rural areas are distributed according to the Table 11.4 Building characteristic Parameter

Options and proportions

Housing Size (m 2 )

0-99 9.1%

29.1%

36.3%

14.0%

11.5%

Housing age (years)

0–2

3–4

5–6

7–8

8 and above

100–139

140–179

180–219

220 and above

18.8%

17.6%

20.6%

26.7%

16.4%

The time of hot water out of the shower

Less than 15 s

15–30 s

30–45 s

45–60 s

60 s and above

19.4%

18.8%

10.3%

10.3%

41.2%

The time of hot water out of the kitchen

Less than 15 s

15–30 s

30–45 s

45–60 s

60 s and above

29.7%

21.8%

10.3

15.8%

22.4%

Lighting

Very poor

Poor

Commonly

Preferably

Good

4.2%

2.4%

12.1%

46.7%

34.5%

Ventilation

Very poor

Poor

Commonly

Preferably

Good

4.2%

2.4%

8.5%

47.9%

37%

1–2

3

4

5

6 and above

19.3%

21.1%

30.7%

15.8%

13.2%

1

2

3

4

5 and above

23.0%

62.4%

12.7%

1.2%

0.6%

Number of taps Number of toilets

11.3 Research Design and Data Collection

301

number of registered permanent residences. The villages in this survey are newly built rural area in recent years, with a house age of less than 10 years. The water for shower, and washing dishes and clothes is the most important, largest proportion of the villagers’ domestic water consumption. A large amount of hot water will be used in the shower or washing dishes. Generally, a large amount of water will be wasted because the water heater is very far from the water outlet. According to the survey, the average time for hot water to flow out of the shower faucet varies: 41.2% takes more than 60 s, and only 19.4% of the users can use hot water within 15 s. The situation improves for the hot water usage in the kitchen: 22.4% of the users wait for more than 60 s when using hot water, and 29.7% of the users could use hot water within 15 s. If we can reasonably plan during the construction of the house and control the distance between the kitchen, shower, and the position of the water heater, we can greatly reduce the waiting time for hot water and save water. Most of the rural houses are built by the residents themselves, and no high-rise buildings in the countryside block the sunlight. Thus, the villagers generally think that the lighting and ventilation conditions of the houses are good: 81.2% think that the lighting conditions are good or very good, and 84.9% think that the ventilation conditions are satisfying.

11.3.3.3

Water Saving Attitudes Variables

We can understand the basic situation of respondents’ water consumption behavior and various influencing factors through descriptive analysis of water-saving attitude related factors. Table 11.5 shows four latitudes with an average score of more than “4,” namely, X6 , X8 , X14 , and X15 . The standard deviation of these four indexes is less than 0.8, which shows that the respondents have good values of respecting nature. Although the scores of the four influencing factors X1 , X2 , X7 , and X16 decreased, their scores are all between 3.5–4 points, and their reflected ecological values are almost the same as those reflected by the previous factors. By contrast, the average scores of X17 and X18 are the lowest and less than 2.5, whereas X25 only has 2.68, which shows that the local farmers have a serious lack of awareness of water-saving equipment, and lack of publicity and education on water-saving equipment from the local government. The scores of X5 , X9 , X11 , X12 , X13 , and X3 are also low, all between 2.5–3.0. These scores show that the local rural residents have certain limitations in the sense of moral responsibility for the environment. The local rural residents generally have good ecological values but often behave badly in terms of their interests. In addition, the scores of other remaining indicators such as X21 , X22 , X23 , and X24 are all around 3.5, which shows the herd mentality of rural residents; the positive water-saving behavior of others will promote their own water-saving behavior.

302

11 The Impact of Building Features …

Table 11.5 Mean value of all water-saving attitude related statements Description

References

Mean

Median

Standard deviation

I usually pay close attention to the water resources issues reported by the media (X1 )

Polonsky et al. (2014)

3.54

4

1.186

I’m worried when I hear or Polonsky et al. (2014) see water related issues (X2 )

3.73

4

1.081

I take the action of saving water because of the laws and policies made by the government (X3 )

Polonsky et al. (2014)

2.7

3

1.104

In order to save water and protect environment, I am willing to sacrifice my personal interests (X4 )

Polonsky et al. (2014)

3.26

3

1.088

In order to save water and Polonsky et al. (2014) protect environment, I’m ok to offend some people if it is needed (X5 )

2.91

3

1.198

The problem of water resources is for the whole society, and everyone is responsible for it (X6 )

Geller et al. (1983)

4.35

4

0.754

As long as I am willing to do my best, I can improve or solve certain environmental problems (X7 )

Polonsky et al. (2014)

3.93

4

0.989

If we take action, it will help to protect the environment (X8 )

Polonsky et al. (2014)

4.15

4

0.721

Ordinary people can also improve or solve water problems (X9 )

Geller et al. (1983)

2.87

3

1.265

Not only the powerful people can influence the improvement of water resources (X10 )

Geller et al. (1983)

3.1

3

1.310

2.64

2

1.163

Compared with the Geller et al. (1983) comfort and convenience of life, I pay more attention to water resources (X11 )

(continued)

11.3 Research Design and Data Collection

303

Table 11.5 (continued) Description

References

Mean

Median

Standard deviation

I think saving water is more important than pursuing own favorite lifestyle and habits (X12 )

Geller et al. (1983)

3.00

3

1.137

Taking water-saving measures will not change my personal life style and habits (X13 )

Geller et al. (1983)

2.71

3

1.153

Water resource is limited, so we must save water (X14 )

Geller et al. (1983)

4.34

4

0.790

We should respect nature Geller et al. (1983) and live in harmony with it (X15 )

4.40

4

0.653

Even for economic development, we cannot sacrifice the natural environment (X16 )

Polonsky et al. (2014)

3.61

4

1.342

The water-saving equipment I want to buy is not much more expensive than ordinary equipment (X17 )

Inman and Jeffrey (2006) and Roccaro et al. (2011)

2.49

2

1.085

The technology of Inman and Jeffrey (2006 water-saving equipment is and Roccaro et al. (2011) more mature currently, and I can skillfully use it (X18 )

2.45

2

0.886

I don’t think it’s very inconvenient for individuals to take measures to save water (X19 )

3.30

3

1.217

I’m saving water because it Author helps me save money (X20 )

3.66

4

1.166

I think the implementation Geller et al. (1983) of water-saving measures lacks relevant publicity and education (X21 )

3.87

4

0.903

Information from newspapers, TV and other media will affect my decision of taking water-saving action (X22 )

3.60

4

1.087

Inman and Jeffrey (2006) and Roccaro et al. (2011)

Inman and Jeffrey (2006 and Roccaro et al. (2011)

(continued)

304

11 The Impact of Building Features …

Table 11.5 (continued) Description

Mean

Median

Standard deviation

My family, friends, etc. Geller et al. (1983) will affect my water-saving behavior (X23 )

References

3.39

4

1.056

There are too few people Geller et al. (1983) around who pay attention to water conservation and I need more people to help and participate in it (X24 )

3.96

4

0.851

I usually pay attention to the use details of water equipment (X25 )

2.68

3

1.229

Inman and Jeffrey (2006) and Roccaro et al. (2011)

11.4 Results and Discussion 11.4.1 Exploratory Factor Analysis EFA extracts the representative common factors from multiple variables. Its basic purpose is to use a few unrelated factors to describe the relationship between multiple variables and reduce the information loss during transformation as much as possible. First, the validity of “water-saving attitude related factors” in the questionnaire was tested. The Cronbach’s alpha value, which measures the internal consistency from 0 to 1, is 0.620. The larger the value is, the better the consistency. A value larger than 0.6 indicates that the questionnaire is reliable (Lai et al. 2002). Factor analysis results show that the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy index value is 0.740, which reflects the relationship between simple correlation and partial correlation coefficient among variables. The larger the KMO value is, the more common the factors among variables are, and the data are more suitable for further factor analysis. Generally, the KMO value needs to be greater than 0.7 and significant. The KMO value shows common factors of the original matrix, and the data have a certain correlation; thus, the data are suitable for conducting EFA. In this study, the principal component method was used to extract the factor whose eigenvalue is greater than 1. The final extraction factors total seven, and the total variance rate of interpretation is 62.034%. The composition matrix after rotation is shown in Table 11.6. According to the distribution of each factor, we obtained seven common facets, thus reducing 25 statements into seven factors, which are more representative. F1 (environmental value) includes X6 , X8 , X14 , X15 , and X16 . This factor can be regarded a person’s worldview which reflects his/her belief about the relationship of humankind with the natural world (Russell and Fielding 2010). It shows how the residents care about the importance of the environment. This fundamental value determines the potential behavior he/she is going to take.

11.4 Results and Discussion

305

Table 11.6 Composition matrix after rotation Variables

Component 1

2

3

4

5

6

7

X15

0.782

−0.001

0.056

−0.129

−0.197

0.122

0.075

X6

0.744

0.025

0.153

−0.046

0.190

0.154

−0.093

X14

0.686

0.146

0.041

−0.132

0.135

0.004

0.371

X8

0.555

0.490

−0.129

−0.019

0.046

−0.057

0.084

X16

0.456

−0.083

0.238

0.303

−0.395

−0.253

−0.170

X4

−0.114

0.776

0.024

−0.041

−0.100

0.314

−0.101

X5

−0.103

0.736

0.102

−0.122

0.106

0.312

0.043

X2

0.262

0.593

−0.131

−0.174

0.099

−0.053

0.064

X1

0.115

0.490

0.000

−0.067

0.438

−0.001

0.073

X7

0.405

0.474

−0.073

−0.092

0.263

−0.212

0.039

X11

0.083

0.002

0.842

0.035

0.020

−0.064

0.053

−0.095

−0.222

0.661

0.119

−0.133

−0.035

0.143

X12

0.229

0.132

0.615

0.345

0.001

0.264

−0.224

X10

0.069

0.082

0.598

0.119

−0.244

−0.250

0.333

X19

0.145

0.070

0.464

0.383

−0.317

−0.259

−0.045

X17

−0.154

0.090

0.005

0.712

−0.259

−0.158

0.276

X13

−0.020

−0.306

0.247

0.659

0.168

0.119

−0.024

X18

−0.249

−0.247

0.132

0.641

−0.080

−0.074

−0.003

X25

−0.403

0.185

−0.158

−0.414

0.181

0.104

0.176

X23

0.003

−0.031

−0.180

−0.033

0.813

0.050

−0.074

X22

0.068

0.427

−0.038

−0.120

0.670

−0.018

−0.045

X21

0.262

0.203

0.032

−0.184

−0.021

0.694

0.076

X3

0.088

−0.112

0.442

−0.075

−0.095

−0.648

0.036

X20

−0.003

0.054

−0.259

−0.140

0.141

0.051

−0.678

X24

0.363

0.170

−0.089

−0.111

0.197

0.358

0.602

Eigenvalues

4.903

3.615

1.843

1.410

1.363

1.249

1.125

Proportion of variance explained (%)

11.908

11.101

10.558

8.378

8.142

6.582

5.366

Cumulative variance explained (%)

11.908

23.009

33.567

41.945

50.087

56.668

62.034

X9

F2 (self-responsibility) includes five indicators: X1 , X2 , X4 , X5 and X7 . This factor reveals the residents’ self-awareness and responsibility about the environment. Protecting the environment is not only paying more attention but needs real actions. This factor transfers the above-mentioned value into the individual’s duty and responsibility.

306

11 The Impact of Building Features …

F3 (life habit and self-control) contains five statements: X9 , X10 , X11 , X12 , and X19 . The view of control is divided into external control and internal control. The group with the view of internal control thinks that its behavior can affect the current situation and change problems, or its behavior is meaningful and worth doing; external control holds the opposite view (Sun 2006). This factor shows the interviewees’ perspective about lifestyle change and the intention to contribute to water conservation. F4 (water conservation appliance) includes four indicators: X13, X17 , X18 and X25 , discusses about the importance and the application of water conservation equipment. Carragher et al. (2012) recorded average day diurnal water consumption patterns of 191 households in Australia, and examined the different influence of water stock (e.g., taps, shower heads, and clothes washers) efficiency. Lee and Tansel (2013) found that the implementation of water efficiency devices provides a positive synergistic effect on actual water savings. The use of water conservation equipment is one of the most observable, practical ways to save water, but the respondents do not usually pay attention to the use details of water equipment. F5 (external incentives) includes X22 and X23 . This factor can be interpreted as follows: external information, such as media publicity or the social norms, will influence the individual’s choice of water conservation. Media campaigns have been verified effective, especially in the short run, in advocating people to save water (Kneebone et al. 2018). Related research shows that people’s behavior is semi-consciously selected to conform to social environments (Otaki et al. 2017); thus, the surrounding peer pressure will affect their own water conservation behavior. However, the negative factor loading (−0.648) indicates that there is no effective policy to pressure the energy conservation behavior of rural residents, which is consistent with item 21 (X21 ). F6 (knowledge transfer) includes X3 and X21 . Measurements of publicity, education, and policy requirement about water conservation can be summarized as knowledge transfer (Koop et al. 2019). It is verified to be an efficient way to provide information to raise environment awareness, and change attitudes and resourceconsumption behavior (Fielding et al. 2013). This factor seems more effective in the short-term water conservation campaign. For a long-term effect, more feasible actions or instructions and repeated publicity can enhance the efficiency (Syme et al. 2000). F7 (money saving and co-participation) includes X20 and X24 . This factor measures two different perspectives. One is the economic incentive in water conservation which has been proven most efficient and direct for people considering taking water conservation action. Another can be interpreted as peer pressure. Peer comparison significantly influences human behavior. Based on a field experiment covering 3,896 households in California, Bhanot (2017) found that even peer rank about water usage has a different effect on higher and lower-usage groups, but the comparison between peers helps motivate people to save water. The negative factor loading (−0.678) indicates that the purpose of saving water for rural residents is not to save money.

11.4 Results and Discussion

307

11.4.2 Multicollinearity of Variables Multicollinearity may cause low significance levels of various spatial variables(Ding et al. 2017). Therefore, the multicollinearity of the independent variables must be studied. The variable expansion factor (VIF) was used to test for multicollinearity in this study. When VIF value is high, a particular explanatory variable is likely to be represented by a linear function model for other explanatory variables; thus, multicollinearity problems may occur in the model (Yao et al. 2014). The VIF values of the explanatory variables are considerably low, as shown in Table 11.7, thereby indicating that no multicollinearity problem occurs.

11.4.3 Binary Logistics Regression In this research, the dependent variable is whether rural residents select to buy watersaving appliances. It is a binary variable, and its value is either 1 or 0. Thus, binary logistic regression analysis is used. The independent variables include social demographic factors, building characteristic factors, and water-saving attitude related factors. In this model, the chi square value for the Omnibus test is 145.771 and significant (p = 0.000). Moreover, the -2lnlikelihood value is 81.483. The Cox & Snell R2 and Nagelkerke R2 are 0.589 and 0.785 respectively, which indicate that the fitting degree of the model is good. At the same time, and the estimated results of the model are shown in Table 11.8.

11.4.4 Discussion According to the significance of the influencing factors in Table 11.8, there are several factors related to the water-saving behavior in rural China. F1 (environmental value) has the most significant effect on whether residents buy water-saving appliances. This factor passed the 1% significance test and has a positive role in promoting rural residents to adopt water-saving behavior, indicating that groups with correct understanding of environmental values are more willing to buy water-saving appliances mainly because individual behavior is affected by own consciousness. Thus, different environmental values will lead to different wateruse and water-saving behaviors. The research of Corral-Verdugo et al. (2003) also confirmed that the water consumption of a group with a strong environmental awareness is often lower than that of a group with a weak environmental awareness, and a group with the view that “water resources are inexhaustible” will consume more water than other groups. F4 (water saving appliance) has a significant positive influence on water saving behavior. The components of F4 are “Taking water-saving measures will not change

308

11 The Impact of Building Features …

Table 11.7 Multicollinearity test Variables

Type

Collinearity statistics Tolerance VIF

F1 (environmental value)

Common factor 0.677

1.478

F2 (self-responsibility)

Common factor 0.777

1.288

F3 (life habit and self-control)

Common factor 0.834

1.199

F4 (water saving appliance)

Common factor 0.776

1.289

F5 (external incentives)

Common factor 0.795

1.258

F6 (knowledge transfer)

Common factor 0.758

1.319

F7 (money saving and co-participation)

Common factor 0.689

1.452

Household size

Ordinal

0.624

1.603

Household composition alone

Dummy

0.654

1.53

Couple

Dummy

0.561

1.784

Couple and elderly parents

Dummy

0.744

1.343

Couple and their children

Dummy

0.681

1.469

Three generations

Dummy

–

–

others

Dummy

0.795

1.257

Family annual income 0–20 K (20 k not included)

Dummy

0.636

1.571

20–40 K

Dummy

–

–

40–60 K

Dummy

0.561

1.782

60 K and above

Dummy

0.544

1.837

Gender

Dummy

0.425

2.35

Age 0–18

Dummy

0.491

2.038

19-45

Dummy

–

–

46–59

Dummy

0.506

1.974

60 and above

Dummy

0.485

2.062

Highest family education background No education

Dummy

0.769

1.3

Primary school

Dummy

0.62

1.613

Junior middle school

Dummy

–

–

High school (vocational and technical secondary school Dummy

0.618

1.619

Junior college (higher vocational)

Dummy

0.743

1.346

Bachelor degree or above

Dummy

0.623

1.605

Water heating time from the shower

Ordinal

0.229

4.375

Water heating time from the kitchen

Ordinal

0.231

4.327

Lighting

Ordinal

0.383

2.613 (continued)

11.4 Results and Discussion

309

Table 11.7 (continued) Variables

Type

Collinearity statistics Tolerance VIF

Ventilation

Ordinal

0.39

2.563

Housing age

Ordinal

0.622

1.609

Number of taps

Ordinal

0.645

1.55

Housing size

Ordinal

0.615

1.625

Number of toilets

Ordinal

0.516

1.939

my personal life style and habits.”, “The water-saving equipment I want to buy is not much more expensive than ordinary equipment.”, “The technology of water-saving equipment is more mature currently, and I can skillfully use it.”, “I usually pay attention to the use details of water equipment”. The more mature the water-saving appliances are, the less expensive they are, the less they change people’s habits, and the more people are willing to use water-saving appliances. People’s attention to water-saving appliances can also lead them to use water-saving appliances. This conclusion is easy to understand and consistent with existing research results (Geller et al. 1983; Roccaro et al. 2011). F5 (external incentives) has a significant negative effect on water-saving behavior, which is extracted from the information of “Information from newspapers, TV, and other media will affect my decision of taking water-saving action,” and “My family, friends, etc. will affect my water-saving behavior.” This result shows that the external interference of water-saving behavior is not conducive to the residents’ taking watersaving measures. It also shows that the public does not have a unified view on adopting water-saving behavior, and the positive effect of adopting water-saving measures is weak. Therefore, the national water-saving also needs a more effective policy guidance, popularization, and publicity (Geller et al. 1983). F7 (money saving and co-participation), which also passed the 1% significance test, has a positive effect on residents’ purchase of water-saving appliances. Residents who hold the two views of “I take water-saving action because I can save money” and “There are too few people around who pay attention to water-saving, I need more people to help and participate together” are more willing to install watersaving appliances. The statement, “The reason why I adopt water-saving behavior is that I can save money,” shows that economic cost plays an important role in the water-saving behavior of residents. Residents who hold the view, “There are too few people around who pay attention to water saving, and I need more people to help and participate together” are more willing to buy water-saving devices, indicating that such groups have a certain sense of collectivism, which is also related to the social background of traditional human society in rural China. Household population information influences the water-saving behavior of respondents. First, the size of family population significantly negatively affects the water-saving behavior of the respondent because the water-use behavior or habits

310

11 The Impact of Building Features …

Table 11.8 Result of binary logistic regression Variables

B

Wald

Sig.

F1 (environmental value)

3.207

25.872

0

F2 (self-responsibility)

−0.015

0.002

0.961

F3 (life habit and self-control)

0.268

0.653

0.419

F4 (water saving appliance)

0.75

3.411

0.065

F5 (external incentives)

−0.607

2.774

0.096

F6 (knowledge transfer)

0.162

0.279

0.598

F7 (money saving and co-participation)

1.201

9.594

0.002

Household size

−0.619

3.911

0.048

Household composition Alone

−2.695

0.944

0.331

Couple

−1.276

0.599

0.439

Couple and elderly parents

−1.595

0.541

0.462

Couple and their children

−1.227

1.807

0.179

Three generations

–

–

–

Others

−3.12

5.704

0.017

Family annual income 0-20 K (20 k not included)

0.867

0.611

0.435

20–40 K

–

–

–

40–60 K

−0.863

0.797

0.372

60 K and above

0.722

0.572

0.449

Age 0-18

17.162

0

0.999

19–45

–

–

–

46–59

1.1

1.52

0.218

60 and above

0.194

0.037

0.848

Gender

0.007

0.0226

0.0897

Highest family education background No education

0.356

0.024

0.877

Primary school

−2.237

4.646

0.031

Junior middle school

–

–

–

High school (vocational and technical secondary school

−0.283

0.071

0.789

Junior college (higher vocational)

2.281

2.798

0.094

Bachelor degree or above

1.309

0.943

0.332

Water heating time from the shower

1.503

11.93

0.001

Water heating time from the kitchen

0.905

4.724

0.03

Lighting

−1.236

5.133

0.023

Ventilation

0.454

0.737

0.391 (continued)

11.4 Results and Discussion

311

Table 11.8 (continued) Variables

B

Wald

Sig.

Housing age

0.649

4.437

0.035

Number of taps

−0.306

1.163

0.281

Housing size

0.01

0.001

0.975

Number of toilets

0.192

0.097

0.755

Constant

1.774

0.268

0.605

of family members has a direct effect on the respondents, and national water-saving consciousness is not good at present (as discussed above). The more educated the residents are, the more willing they are to conduct water-saving behavior mainly because more educated residents are more likely to have the correct environmental outlook and awareness of water and water conservation. Considerable literature studies have shown that groups with a high education level will have a strong awareness of water and energy conservation (Russell and Fielding 2010; Willis et al. 2013b). Water heating time from the shower or the kitchen also motivates the residents to buy water-saving devices. The longer the time to produce hot water is, the greater the amount of water that will be wasted, which may promote the residents to take more active action to save water. This phenomenon also shows that the residents are subconsciously not willing to waste water. The waste of water resources will increase the water cost, which will promote the residents’ awareness of water conservation to a certain extent. When building houses in the future, we can re-organize the distance between the water heater and toilet and kitchen faucet because a reasonable arrangement can effectively reduce water consumption. Housing age also has a certain role in promoting the purchase of water-saving appliances. The older the building is, the stronger the willingness of residents to purchase water-saving appliances. This finding is due to the rather low construction quality of the buildings in rural areas compared with those in urban areas. Those built environment factors shed light on future house construction.

11.5 Conclusion and Recommendations Based on a literature review and empirical analysis, 165 questionnaires are collected from four rural areas around Chengdu by face-to-face questionnaire survey. This study further uses descriptive statistics, EFA, and binary logistic regression model to analyze the water-saving behavior of rural residents interviewed. The main influencing factors of water-saving behavior of rural residents in Chengdu are systematically analyzed. The main conclusions and suggestions are as follows. For social demographic factors, the variable “education of the interviewee” is significantly positively related to water conservation behavior, while the household size significantly negatively affects water conservation behavior. This finding is

312

11 The Impact of Building Features …

consistent with other studies, indicating that people with a high level of education are more likely to have a correct environmental perspective and awareness of water and water conservation, and national water-saving consciousness is not good at present. During the site interview, we also find that the major responders are female. They are the main forces for daily water use. Thus, targeted education to those female residents will increase the effect and have a better influence on water conservation behavior. For building characteristic factors, housing age and the installation of certain appliances will affect the way the residents adopt water conservation behavior. The older the house is, the larger the possibility that the residents will install water-saving appliances. Several appliances affect the residents’ water conservation behavior as well, especially the water heater. The longer the time to produce hot water from the kitchen or toilet, the higher the tendency that residents will take water-saving action because they can reduce water consumption and save money. Economic concern is quite important in their daily life. Thus, the local government can make a more suitable arrangement or design of the new communities in rural areas. Better design, that is, shortening the distance between the water heater and the taps would help save water. More education or publicity about water-saving appliances can be promoted in rural areas. The building age is usually larger for those who are not moving to a newly-constructed community. Adequate propaganda could increase their desire to install water-saving appliances. For water conservation attitude, EFA is used to extract seven common factors from the original 25 statements related to water conservation attitudes. Binary logistic regression shows that F1 (environmental value), F4 (water-saving appliance), F5 (external incentives), and F7 (money saving and co-participation) are the most influential factors. For F1 , different environmental values will lead to different water-use and water-saving behaviors. When respondents value the importance of the environment, they would take more active actions to save water and protect the environment. For F4 , the regularization and popularization of water-saving facilities can promote residents to adopt water-saving behaviors. Residents’ attention to water-saving facilities will encourage more water-saving behavior. For F5 , external incentives such as media information and family members’ water-saving behaviors significantly negatively affect the interviewees’ water-saving behaviors mainly because the social popularization of water-saving behavior is not sufficient. For F7 , economic concern plays an important role in determining water use. Therefore, the local government should either invest more in education or subsidize the residents to buy or install water-saving appliances. Saving water by changing water-use habits is very difficult, and the most effective way is to use water-saving instruments. Therefore, the first step is to promote the use of water-saving appliances and encourage villagers to install them. Second, the local government should teach the villagers how to use those appliances effectively. Generally, most of the rural residents demonstrate an incorrect perception of water resource, such as the water resource is endless and can be used at will. Moreover, the degree of taking water-saving behavior among residents is not high, and only a few residents will recycle the wastewater. The proportion of those who purchase

11.5 Conclusion and Recommendations

313

water-saving appliances is relatively low, and many residents are not sure whether the appliances they bought are truly water saving. Even if they have installed water conservation equipment, they may be confused about how to use it. Thus, promoting the importance of water resource is urgently demanded to raise their consciousness through media campaign (television or radio), specific talks, and leaflet distribution. However, the villagers’ trust in grass-root government departments is much higher than that in other channels. Thus, we should focus on this channel as an efficient medium to strengthen the education of water-saving awareness of villagers. Guided by other literature, knowledge transfer such as education and publicity may have a strong effect in the short run (Syme et al. 2000). Repeated information distribution is needed to enhance the effect on water conservation in the long run. Related research suggests that most households are initially motivated to save water but fail to turn it into real water conservation action. Thus, the local government can take more practical strategies to help the residents maintain the behavior, such as subsidizing the adoption of water conservation appliances, providing a door-todoor guidance of the use of the equipment, and holding regular trainings. These practical, feasible actions would increase the residents’ self-efficacy to maintain water conservation habits. Admittedly, this research suffers from two distinct shortcomings: (1) generalization or transferability of the results and (2) limited sample size. Field trips in rural China is difficult in the way of transportation and communication with villagers. The research team made great efforts to enlarge the sample pool. In future research, one possible direction is to link the survey data with census data, such as the China Household Finance Survey. Another extension is to run a set of field experiments to compare the adoption of water-saving facilities.

References Ajzen, I. 1991. The theory of planned behavior. Organizational Behavior & Human Decision Processes, 50(2): 179–211. Ajzen, I., & Fishbein, M. 2000. Attitudes and the attitude-behavior relation: Reasoned and automatic processes. European Review of Social Psychology, 11(1): 1–33. Andersen, B. J. 2008. Residential landscape water use and conversation. University of Idaho. Aprile, M. C., & Fiorillo, D. 2017. Water conservation behavior and environmental concerns: Evidence from a representative sample of Italian individuals. Journal of Cleaner Production, 159: 119–129. Archibald, R. S. O. 1998. Demand side management policies for residential water use: Who bears the conservation burden? Land Economics, 74(3): 343–359. Bhanot, S. P. 2017. Rank and response: A field experiment on peer information and water use behavior. Journal of Economic Psychology, 62. Carragher, B., Stewart, R., & Beal, C. 2012. Quantifying the influence of residential water appliance efficiency on average day diurnal demand patterns at an end use level: A precursor to optimised water service infrastructure planning. Resources Conservation & Recycling, 62(2): 81–90. Chang, G. 2013. Factors influencing water conservation behavior among urban residents in China’s arid areas. WATER POLICY, 15(5): 691–704.

314

11 The Impact of Building Features …

Chaudhary, A. K., Lamm, A. J., & Warner, L. A. 2018. Using cognitive dissonance to theoretically explain water conservation intentions. Journal of Agricultural Education, 59(4): 194–210. Chaudhary, A. K., Warner, L. A., & Ali, A. D. 2019. Using perceived benefits to segment residential landscape irrigation users. Urban Forestry & Urban Greening, 38: 318–329. Chen, X., Xu, J., & Ji, Y. 2005. Research on determinant factors in residential water demand. Journal of Economics of Water Resources, 23(6): 23–24. Cho, R. 2011. How china is dealing with its water crisis. Blogs from the Earth Institute. State of the Planet. Clark, W. A., & Finley, J. C. 2007. Determinants of water conservation intention in Blagoevgrad, Bulgaria. Society & Natural Resources, 20(7): 613–627. Conner, M., & Armitage, C. J. 2006. Extending the theory of planned behavior: A review and avenues for further research. Journal of Applied Social Psychology, 28(15): 1429–1464. Cook, S. W., & Berrenberg, J. L. 1981. Approaches to encouraging conservation behavior: A review and conceptual framework. Journal of Social Issues, 37(2): 73–107. Corral-Verdugo, V. C., Bechtel, R. B., & Fraijo-Sing, B. 2003. Environmental beliefs and water conservation: An empirical study. Journal of Environmental Psychology, 23(3): 247–257. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. 2017. Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100: 65–80. Domene, E., & Saurí, D. 2006. Urbanisation and water consumption: Influencing factors in the metropolitan region of Barcelona. Urban Studies, 43(9): 1605–1623. Fan, L. 2014. Study of domestic water use behaviors in rural households: Insights from Shannxi plain, the Wei river basin, China. Northwest A&F University. Fernandez, G., Tun, A. M., Okazaki, K., Zaw, S. H., & Kyaw, K. 2018. Factors influencing fire, earthquake, and cyclone risk perception in Yangon, Myanmar. International Journal of Disaster Risk Reduction, 28: 140–149. Fielding, K. S., Spinks, A., Russell, S., McCrea, R., Stewart, R., & Gardner, J. 2013. An experimental test of voluntary strategies to promote urban water demand management. Journal of Environmental Management, 114: 343–351. Geller, E. S., Erickson, J. B., & Buttram, B. A. 1983. Attempts to promote residential water conservation with educational, behavioral and engineering strategies. Population & Environment, 6(2): 96–112. Höglund, L. 1999. Household demand for water in Sweden with implications of a potential tax on water use. Water Resources Research, 35(12): 3853–3863. Hu, F. 2006. Study on the influencing factors of domestic water demand of urban residents: A case study of Nantong City. Zhejiang University. Huang, X., Huang, Z., Zhang, L., Zhou, X., & Liu, S. 2017. The effect of block-rate water price policy on residential water consumption time series. China Population, Resources and Environment, 27(S2): 108–111. Inman, D., & Jeffrey, P. 2006. A review of residential water conservation tool performance and influences on implementation effectiveness. Urban Water Journal, 3(3): 127–143. Jiang, W., Huang, C., Liu, Q., Liu, Y., & Tian, S. 2019. Investigation on current situation of water consumption and water quota in Chongqing. Water & Wastewater Engineering, 55(02): 103–107. Kneebone, S., Fielding, K., & Smith, L. 2018. It’s what you do and where you do it: Perceived similarity in household water saving behaviours. Journal of Environmental Psychology, 55: 1–10. Koop, S., Dorssen, A. V., & Brouwer, S. 2019. Enhancing domestic water conservation behaviour: A review of empirical studies on influencing tactics. Journal of Environmental Management. Kumar Chaudhary, A., Warner, L. A., Lamm, A. J., Israel, G. D., Rumble, J. N., & Cantrell, R. A. 2017. Using the theory of planned behavior to encourage water conservation among extension clients. Journal of Agricultural Education, 58(3): 185–202. Lai, K., Ngai, E., & Cheng, T. 2002. Measures for evaluating supply chain performance in logistics. Transportation Research Part E(38): 213.

References

315

Lam, S. P. 2006. Predicting intention to save water: Theory of planned behavior, response efficacy, vulnerability, and perceived efficiency of alternative solutions 1. Journal of Applied Social Psychology, 36(11): 2803–2824. Lee, M., & Tansel, B. 2013. Water conservation quantities vs customer opinion and satisfaction with water efficient appliances in Miami, Florida. Journal of Environmental Management, 128: 683–689. Li, M. 2011. Analysis about present situation of rural domestic water and influential factors of water consumption: based on household survey in Shuyang, Jiangsu province. Nanjing Agricultural University. Lin, Z., Gong, H., Teatini, P., Li, X., & Dai, Z. 2015. Land Subsidence due to groundwater withdrawal in the northern Beijing plain, China. Engineering Geology, 193: 243–255. Liu, B., & Speed, R. 2009. Water resources management in the People’s Republic of China. Water Resources Development, 25(2): 193–208. Liu, Y., Hu, X., Zhang, Q., & Zheng, M. 2017. Improving agricultural water use efficiency: A quantitative study of Zhangye city using the static CGE model with a CES water—Land resources account. Sustainability, 9(2): 308. Ma, X., Wu, D., & Zhang, S. 2018. Multiple goals dilemma of residential water pricing policy reform: Increasing block tariffs or a uniform tariff with rebate? Sustainability, 10(10): 3526. Mu, Q., Zhang, S., & Ma, X. 2014. Driving factors of household adoption of water conservation behaviors. Acta Scientiarum Naturalium Universitatis Pekinensis, 50(3): 587–594. National Bureau of Statistics of the People’s Republic of China. 2019. China Statistical Yearbook: http://www.stats.gov.cn/tjsj/ndsj/2019/indexch.htm. Nieswiadomy, M. L., & Molina, D. J. 1991. A note on price perception in water demand models. Land Economics, 67(3): 352–359. Otaki, Y., Ueda, K., & Sakura, O. 2017. Effects of feedback about community water consumption on residential water conservation. Journal of Cleaner Production, 143: 719–730. Polonsky, M. J., Vocino, A., Grimmer, M., & Miles, M. P. 2014. The interrelationship between temporal and environmental orientation and pro-environmental consumer behaviour. International Journal of Consumer Studies, 38(6): 612–619. Qian, Z. Y., Lin, B. N., Zhang, W. Z., & Sun, X. T. 2002. Comprehensive report of strategy on water resources for China’s sustainable development. Beijing: China Water and Hydropower Press. Qiu, J. 2018. Safeguarding China’s water resources. National Science Review, 5(1): 102–107. Roccaro, P., Falciglia, P. P., & Vagliasindi, F. G. A. 2011. Effectiveness of water saving devices and educational programs in urban buildings. Water Science & Technology A Journal of the International Association on Water Pollution Research, 63(7): 1357. Russell, S., & Fielding, K. 2010. Water demand management research: A psychological perspective. Water Resources Research, 46(5): 43–63. Sun, Y. 2006. Study on Residents’ Environmental Behavior and Its Influencing Factors. Dalian University of Technology. Syme, G. J., Nancarrow, B. E., & Seligman, C. 2000. The evaluation of information campaigns to promote voluntary household water conservation. Evaluation Review, 24(6): 539–578. Thomas, A., & Nauges, C. 2000. Privately operated water utilities, municipal price negotiation, and estimation of residential water demand: The case of France. Land Economics, 76(1): 68–85. Tong, Y., Fan, L., & Niu, H. 2017. Water conservation awareness and practices in households receiving improved water supply: A gender-based analysis. Journal of Cleaner Production, 141: 947–955. Wang, Y., Feng, L., Zhao, D., & Mao, D. 2011. Relational analysis of influence factors on residential water consumption in Jinhua City. Journal of Water Resources and Water Engineering(3): 53–56. Willis, R. M., Stewart, R. A., Giurco, D. P., Talebpour, M. R., & Mousavinejad, A. 2013a. End use water consumption in households: impact of socio-demographic factors and efficient devices. Journal of Cleaner Production, 60: 107–115.

316

11 The Impact of Building Features …

Willis, R. M., Stewart, R. A., Giurco, D. P., Talebpour, M. R., & Mousavinejad, A. 2013b. End use water consumption in households: impact of socio-demographic factors and efficient devices. Journal of Cleaner Production, 60(Complete): 107–115. Willis, R. M., Stewart, R. A., Panuwatwanich, K., Williams, P. R., & Hollingsworth, A. L. 2011a. Quantifying the influence of environmental and water conservation attitudes on household end use water consumption. Journal of Environmental Management, 92(8): 1996–2009. Willis, R. M., Stewart, R. A., Panuwatwanich, K., Williams, P. R., & Hollingsworth, A. L. 2011b. Quantifying the influence of environmental and water conservation attitudes on household end use water consumption. Journal of environmental management, 92(8): 1996–2009. Worthington, A. C., & Hoffman, M. 2008. An empirical survey of residential water demand modelling. Journal of Economic Surveys, 22(5): 842–871. Xiong, S. 2018. Empirical analysis of influencing factors of urban residents’ domestic water: Chengdu city as an example. Science Technology and Industry(1): 92–97. Yan, L., McManus, P., & Duncan, E. 2018. Understanding ethnic differences in perceptions, attitudes, and behaviours: a study of domestic water use in Sydney. Geographical Research, 56(1): 54–67. Yao, B., Hu, P., Lu, X., Gao, J., & Zhang, M. 2014. Transit network design based on travel time reliability. Transportation Research Part C: Emerging Technologies, 43: 233–248. Yin, Y., Zhang, K., & Li, X. 2006. Urbanization and land subsidence in China. International Association for Engineering Geology and the Environment: 1–6. Zhang, H., Dong, J., & Yan, J. 2009. Review on the driving force of urban water use change. Acta Agriculturae Jiangxi(11): 195–199. Zhao, H., Zhong, Q., & Ao, Y. 2019. Analysis on the influencing factors of water saving behavior of residents in rural areas of Sichuan province. Rural Economy and Science-Technology, 30(03): 33–34. Zhu, L., Gong, H., Li, X., Wang, R., Chen, B., Dai, Z., & Teatini, P. 2015. Land subsidence due to groundwater withdrawal in the northern Beijing plain, China. Engineering Geology, 193: 243–255.

Chapter 12

Seismic Evacuation Preparedness Behavior of Rural Residents

Abstract This chapter utilized a random survey in the form of a sample questionnaire in the areas most affected by the Wenchuan earthquake in 2008, combined with exploratory factor analysis and binary logistic regression analysis. Results show that residents’ BE and risk perceptions positively affected their evacuation choice behavior. Specifically, when rural residents perceived a reasonable evacuation route and good quality of village roads, they would flee their homes as soon as an earthquake struck. With regard to seismic risk perception, if the residents perceived highly negative consequences from earthquakes, they would escape immediately as soon as an earthquake occurred. This conclusion highlights the importance of strengthening the construction of BE in disaster-prone areas, and raising residents’ awareness and risk perception of earthquake disasters. This study has practical significance in further construction of earthquake-stricken areas. Keywords Built environment perception · Risk perception · Evacuation behavior · Wenchuan earthquake · Exploratory factor analysis · Binary logistic regression

12.1 Introduction The Wenchuan earthquake that occurred on May 12, 2008 caused an enormous loss to China. This phenomenon, which had a 8.0 magnitude, was the most destructive earthquake since the founding of new China in 1949 and the deadliest since the 1976 Tangshan earthquake. Besides damage to land resources and infrastructure, 7.79 million houses collapsed in 14,565 villages and 24.59 million houses were damaged. The direct economic losses are approximately 845 billion Chinese yuan (CNY) and the indirect losses cost more (Wu et al. 2012). A report issued by the Chinese Ministry of Civil Affairs in 2008 stated that a total of 69,227 people were killed; 374,643 were injured; another 17,923 were listed as missing; and approximately 4.8 million were left homeless as a result of the Wenchuan earthquake (Liang et al. 2019). During the decade after the Wenchuan earthquake in 2008, scholars studied this disaster from different perspectives: (1) assessment of damage level (Zhang 2010); (2) assessment of natural disasters (such as landslides, dammed lakes, and mudslides) caused by © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_12

317

318

12 Seismic Evacuation Preparedness …

the earthquake (Peng 2011); (3) emergency response process such as emergency rescue, rescue lifeline, resource mobilization, mass medical care, public management, decision-making behavior for emergency response, and others (Ke-Sheng et al. 2010; Mao et al. 2018); (4) earthquake resistance of buildings (Lang and Qing-Song 2010); (5) social and environmental vulnerability of the affected areas (Zhang et al. 2017); (6) models and resource allocation for post-disaster reconstruction (Liu et al. 2017; Yan et al. 2011); and (7) psychological resilience of affected residents (Li et al. 2012; Meng et al. 2018). Although scholars have explored various problems caused by the Wenchuan earthquake, they have ignored the influences of risk and built environment (BE) perceptions on the evacuation behavior of residents in disaster areas, although numerous studies have been conducted on risk perception (Paton et al. 2008; Slovic 2000; Wachinger et al. 2013). All of these references on risk perception conclude that the most effective means to create awareness of potential disasters is to enhance trust in public authorities and encourage citizens to take greater personal responsibility for protection and disaster preparedness. The term “built environment” has an urban form. In the middle to late 1990s, the connotation and extension of BE was developed to emphasize the urban physical environment to which the spatial, temporal, and sociocultural backgrounds centered on human activities are attached, including land use patterns, the BE associated with their human activities, transportation systems, positive urban design, and organization of physical elements (Handy et al. 2005; Saelens and Handy 2008). Moreover, the enormous losses resulting from the Wenchuan earthquake were largely attributed to the failures in BE (Wang and Tang 2017), such as improper land use planning, lack of strict implementation of seismic design codes, and unqualified construction. Furthermore, Lindell (2010) stated that these failures in BE severely affected local and external emergency responses. The areas affected ruinously by the Wenchuan earthquake are also located in villages, towns, and other areas at the foot of the mountain or in the mountains. The BE in these areas are generally backward compared with that in cities, and the seismic resistance is relatively weak. After the Wenchuan earthquake, the government paid special attention to the post-disaster reconstruction work in the affected areas. Compared with the past BE of the affected areas, the BE improved considerably (Peng et al. 2018a). However, we should know how to effectively help the victims escape and start reconstruction after the disaster. The importance of post-disaster response is highlighted in the existing literature on postdisaster rehabilitation or reconstruction modes (Davis et al. 2015; Liu et al. 2017). We aim to explore whether residents’ perception of the new BE after the post-disaster reconstruction affects their choice of emergency evacuation behavior in areas affected by the Wenchuan earthquake. Risk perception (RP) is a psychological process that describes a subjective (conscious and unconscious) assessment of the likelihood of the effect of an impending undesirable event (as opposed to an objective risk assessment) in a specific situation as well as an assessment of one’s own perceived vulnerability and coping resources (Kinateder et al. 2015). RP is usually regarded as an important predictor of disaster evacuation behavior (Highfield 2005). When humans face an emergency natural disaster, the perception and assessment of disaster risk is a precursor to

12.1 Introduction

319

emergency escape or protection measure. The risk assessment includes the individual’s likelihood estimation of a disaster occurrence, severity and urgency of the hazard, extent of impact (such as physical injury, property damage, and interruptions in daily life), and level of concern for the hazard (Lindell and Perry 2000). In addition, Lindell et al. (2015) found a close relationship between risk perception and earthquake response measures; that is, actions related to mitigating the potential consequences of people and properties are also related to individuals’ risk perception during earthquake events. After the Wenchuan earthquake, farmers’ risk perception of concentrated rural settlements was investigated by Peng et al. (2018b) using survey data on the hardest earthquake-hit area, and critical risk factors were also identified. Human behavior is difficult to predict at all times, especially in emergencies, which are characterized by stress and chaos (Fullerton et al. 1996). During an earthquake, most people’s first reaction is to search for furniture and chairs for safety. People have been observed holding on to walls and/or other individuals in corridors and areas with no tables or other furniture (Lambie et al. 2016). Shapira et al. (2018) found that the main danger comes from falling objects that could cause injury or even death during an earthquake. In addition, they conducted a questionnaire survey of 306 residents in the earthquake disaster area, with mean age of 35 (SD = 11.5). They found that when asked how to choose disaster emergency measures, 43% of the respondents opted to evacuate buildings during the earthquake, 19% chose to hide in apartment shelters, 13% hid under heavy furniture, 8% climbed the stairs, 5% sat against the wall, and 12% did not know what to do. Regarding evacuation behavior during other disaster events, Cahyanto et al. (2014), Bowser et al. (2015), and Koshiba et al. (2018) studied the evacuation behavior strategies used in the occurrence of hurricanes and floods. The consensus is that the safest action is to evacuate from a house or low-lying area to a higher safety zone. Therefore, compared with any other proposed refuge behavior, whether an earthquake or another disaster, evacuating from the building to the safe zone in time is the safest and most survivable emergency behavior. The Wenchuan disaster has caused people, especially those living in disaster-prone areas, to pay close attention to the emergency protection behavior during earthquakes. However, in areas with frequent earthquakes, people’s first choice to evacuate is the best self-protection behavior to ensure their safety. Developed Western countries have conducted relatively mature research on this subject, whereas in China, the research is mainly carried out from the perspective of natural science and rarely from the perspective of social science. Limited studies explore the impact of BE and risk perceptions on evacuation behavior. To fill this gap, the present study aims to explore the influence of BE and risk perceptions on evacuation behavior after control of socio-demographic information. This paper is expected to provide a perspective for the further development of rural areas with low level of economy and construction, and improves the disaster resilience of residents in rural areas.

320

12 Seismic Evacuation Preparedness …

12.2 Materials and Methodology 12.2.1 Questionnaire Design The questionnaire design involves four steps: (1) The first draft was completed based on the literature review. (2) After the questionnaire was discussed with 16 undergraduates from the Wenchuan earthquake disaster area, the questionnaire was modified so that it conforms to the actual situation of target sample villages. (3) Trial investigation was conducted in random selected sample villages to test the readability and completeness of the questionnaire. (4) The questionnaire was improved according to feedback on the trial investigation. Lastly, the final questionnaire consisted of four parts focusing on the respondents’ (1) basic information, (2) BE perception, (3) disaster risk perception, and (4) seismic evacuation behavior choice. The basic information involves demographic and socioeconomic factors based on the study by Shapira et al. (2018) and Li et al. (2016), which includes gender, age, education [21], occupation [24], physical condition, and marital status. Family structure, such as number of children under age 12, (Goltz et al. 1992) and earthquake experiences (Cohen et al. 2013; Dooley et al. 2010) are also considered. In addition to the basic information considered by existing research, this study also considers number of floors of the house and housing type factors according to the trail investigation and specific situation in rural Sichuan. In the areas affected by the Wenchuan earthquake, the main housing types are single and non-detached buildings. Further details on the basic information are shown in Table 12.1. BE refers to the man-made environment formed by human activities, which are different from the natural environment (Handy et al. 2002). Each person’s perception of BE is different, resulting in the choice of emergency behavior (Haghani and Table 12.1 Distance to epicenter and number of valid questionnaires Sample villages

Distance from village center to epicenter (km)a

Number of valid questionnaires

Wen Chuan

Guo Jiaba

17

49

Lian Shanpo

22

42

Du Jiangyan

He Ming

52

52

He Jia

49

36

Lao Chang

208

53

Qing Lin

213

45

Mian Zu

Ji Xian

132

48

Hong Ming

121

53

Qing Chuan

Dong Fang

328

55

Yin Ping

331

35

Bei Chuan

a This

distance is the driving distance measured by Baidu navigation application

12.2 Materials and Methodology

321

Sarvi, 2016). Therefore, BE perception was considered to explore BE’s influences on seismic evacuation behavior. Respondents were asked to judge the degree of recognition of various BE perception items. A five-point Likert scale (Clason and Dormody 1994) was used to represent the degree of recognition of different BE perception items from non-acceptance to full acceptance. The items of BE perception are shown in Table 12.3. In accordance with existing natural hazards, risk perception is defined as a subjective belief held by an individual residing in a natural hazard zone regarding the potential harm or possibility of loss due to an earthquake or related event assessed in terms of one’s evaluation of the characteristics, probability, and severity of such risk (Paton et al. 2008). Disaster risk perception was derived from four studies that considered Chinese as their sample. Specifically, respondents were asked to assess the threat (RP1), fear (RP2), harm (RP3), and concerns (RP4) experienced regarding earthquakes (Jianguang 1994; Zhang et al. 2013). In our questionnaire, we also introduce a three-item measure (individual’s likelihood estimation of a disaster occurrence, severity and urgency of hazard, and extent of own impact) that deals with dimensions associated with environmental hazards (Lindell and Perry 2000). In terms of earthquake evacuation, respondents were asked to select whether to escape from home during the earthquake when an earthquake occurs again; this question was also found in the study by Shapira et al. (2018).

12.2.2 Sample and Data Collection 12.2.2.1

Sample Village Selection

The purpose of this study is to explore the impact of BE and risk perception on the seismic evacuation behavior of residents from the areas affected by the Wenchuan earthquake after a decade. Wen Chuan, Qing Chuan, Bei Chuan, Mian Zhu, and Du Jiangyan are the most affected areas in 2008 (Liang et al. 2019). Historical records show that since the founding of the People’s Republic of China in 1949, more than 90% of the devastating earthquakes with magnitudes above 7 on the Richter scale occurred in rural areas (ye 2008). Therefore, the research focuses on the rural areas that were the most seriously affected by the Wenchuan earthquake in 2008. Then, among the seriously affected areas, 10 villages were randomly selected for representation (Wen Chuan: Guo Jiaba and Lian Shanpo villages; Mian Zhu: Ji Xian community and Hong Ming village; Du Jiangyan: He Ming and He Jia villages; Qing Chuan: Dong Fang and Yin Ping villages; and Bei Chuan: Lao Chang and Qing Lin villages). These villages are all located on the Longmenshan earthquake zone, where the majority of the villages are surrounded by or located in the mountains. The high frequency of earthquakes has a typical research value. Figure 12.1 shows the location and distribution of the sample villages.

12 Seismic Evacuation Preparedness …

Fig. 12.1 Location of sample villages

322

12.2 Materials and Methodology

12.2.2.2

323

Data Collection

To collect valid data, we recruited a total of 32 experienced researchers who have participated in face-to-face surveys twice in rural areas; these researchers include 1 teacher, 11 graduates, and 20 undergraduates. All the researchers are from rural areas in Sichuan; thus, they are familiar with the rural environment and can communicate effectively with rural residents. The face-to-face questionnaire survey was conducted from December 28, 2018 to January 5, 2019. The respondents were randomly selected to complete the survey in the 10 villages. A total of 502 out of 600 questionnaires were completed effectively. Then, another 34 out of 502 questionnaires were excluded according to a consistency check. Finally, a total of 468 valid questionnaires were obtained with an effective questionnaire rate of 93.23% (see Table 12.1).

12.2.3 Variable Specification According to the research design, this study mainly considers three types of explanatory variables to explore their effects on earthquake evacuation behavior. These variables are basic population information, BE perception, and earthquake risk perception. Respondents were asked to answer the following question about a dependent variable with a binary alternative of yes or no: Will you evacuate from your home during the earthquake? According to the statistics of 468 valid questionnaires, 277 (59.2%) of the respondents opted to escape from their house, whereas the remaining 191 (40.8%) prefer to stay at home during the earthquake.

12.2.3.1

Basic Information of Respondents

Gender, age, education background, physical condition, marital status of the respondent, and presence or absence of casualties and property loss in the household were considered in this study. Whether the family of the respondent had children under the age of 12, the number of residential floors, and the type of residential buildings were also considered. Table 12.2 shows the basic information of the respondents. A total of 211 males and 257 females were surveyed in this study. The number of surveyed men and women was not much disparate, and the distribution of men and women in the questionnaire was relatively uniform. The age of the surveyed population ranges from 18 to 65 years. However, according to actual research, most of the rural residents are mainly middle-aged people over the age of 40 and children under the age of 12. Young people are more likely to go out to study or work. The division on the basis of educational background is based on the general classification of Chinese academic background. Residents with income from farming account for 33.3% of

324

12 Seismic Evacuation Preparedness …

Table 12.2 Basic information on individuals and their households Basic information

Options

n

%

Gender

Male

211

45.09

Female

257

54.91

Age

Less than 20

11

2.35

20–30

52

11.11

30–40

54

11.54

40–50

108

23.08

50–60

104

22.22

More than 60

139

29.70

No education at all

67

14.32

Compulsory education

291

62.18

Senior high school

69

14.74

Bachelor’s degree or above

41

8.76

Health

430

91.88

Severe disability

11

2.35

Slight disability

27

5.77

Marital status

Single

87

18.59

Married

381

81.41

With children under 12 years old

Yes

271

57.91

No

197

42.09

Yes

226

48.2

No

242

51.6

With or without property loss

Yes

426

91

No

42

Number of housing floors

Below 2

395

84.40

2 or more

73

15.60

Single building

266

56.84

Non-detached building

202

43.16

Education background

Physical condition

With or without casualties

Housing type

9

the research population, and other residents have other jobs. Moreover, 24.8% of the research population account for residents without work. The common reason is that in Chinese rural areas, elderly people and women raise their children at home without jobs and income.

12.2.3.2

BE Perception

The questionnaire design of the BE was mainly considered from the perspective of infrastructure and combined with previous relevant studies and field conditions of

12.2 Materials and Methodology

325

Table 12.3 BE perception items and sources Item

Source

The current terrain environment has a strong effect on housing earthquake resistance

Authors

The current residential infrastructure is well planned

(Shapira et al. 2018)

The residence is convenient for emergency evacuation in the area where you currently reside

(Shiwakoti 2016)

A reasonable shelter exists in the place where you currently reside

(Shiwakoti 2016)

Reasonable spacing is conducive to escape and evacuation during earthquakes

(Shiwakoti 2016)

The current house is strong and earthquake resistant

(Shapira et al. 2018)

The interior design of the house has a reasonable emergency shelter

(Shapira et al. 2018; Shiwakoti 2016)

The interior design of the house has a reasonable emergency escape route

(Shapira et al. 2018; Shiwakoti 2016)

The quality of building materials is guaranteed

(Shapira et al. 2018)

Good roads exist from villages to other villages or towns

(Shapira et al. 2018)

Roads from villages to other villages or towns are not easily damaged or congested in the event of an earthquake

(Shapira et al. 2018)

the research area. Many of the affected areas in the 2008 Wenchuan earthquake were mountainous areas. Considering that mountainous areas are not only directly affected by the earthquake but also have other derived disasters, we also designed a question from the perspective of geographical space. The main items of BE perception and their sources are shown in Table 12.3.

12.2.3.3

Earthquake Risk Perception

We considered the effect of earthquake disasters on individuals and the entire family in the questionnaire design of earthquake risk perception. Earthquake risk perception items were combined with the relevant research literature, sorted out, and shown in Table 12.4.

326

12 Seismic Evacuation Preparedness …

Table 12.4 Earthquake risk perception items and sources Items

Source

You are very sensitive to shaking things

(Lindell and Perry 2000)

Earthquakes happen easily in this area

(Lindell and Perry 2000)

The risk of a serious earthquake in the future will be greater

(Lindell and Perry 2000)

You think you will be directly affected

(Jianguang 1994; Zhang et al. 2013)

You think that an extreme earthquake will have a long-term negative effect

(Jianguang 1994; Zhang et al. 2013)

An earthquake is catastrophic

(Jianguang 1994; Zhang et al. 2013)

After the earthquake, you will always be vigilant

(Jianguang 1994; Zhang et al. 2013)

After the earthquake, you feel that aftershocks will always occur

(Jianguang 1994; Zhang et al. 2013)

The number of earthquake disasters has decreased at present

(Li and Guo 2016; Mark et al. 2013)

The earthquake will not cause devastation to the house

(Li and Guo 2016; Mark et al. 2013)

The effect of the earthquake on you is very serious

(Jianguang 1994; Zhang et al. 2013)

The effect of the earthquake on your family is very serious

(Jianguang 1994; Zhang et al. 2013)

You are not very scared of the earthquake at present

(Li and Guo 2016; Mark et al. 2013)

A minor earthquake will cause damage to your house (Li and Guo 2016; Mark et al. 2013) The occurrence of an earthquake will cause further damage to your house

(Li and Guo 2016; Mark et al. 2013)

12.2.4 Model Specification 12.2.4.1

Exploratory Factor Analysis

Exploratory factor analysis (EFA) was first proposed by Spearman (1987) in 1904. The basic purpose of this approach is to use a few unrelated factors to describe the relationship between multiple variables, namely, how to condense the measured variables into a few factors with minimal information loss. The general form of EFA is as follows: xi = ai1 F1 + ai2 F2 + · · · + ain Fn + εi (i = 1, 2, · · · p),

(12.1)

where ai j is a factor load; its essence is the correlation coefficient between common factors Fi and special factor variables xi . εi represents factors other than the common factor (2018).

12.2 Materials and Methodology

12.2.4.2

327

Binary Logistic Regression Analysis

Logistic regression belongs to probabilistic nonlinear regression. Logistic regression analysis is divided into binary logistic regression and multiple logistic regression analyses according to the different types of dependent variables. The difference is that the dependent variable of binary logistic regression analysis can only take two values, namely, 0 and 1 (virtual dependent variable), whereas the dependent variable of multivariate logistic regression analysis can be multiple values. The probability of the event being studied in the formula is P(y = 1|xi ) = pi . The probability formulas for the occurrence pi and non-occurrence 1 − pi of the event are pi =

m

1

m

1 + e−(α+

1− pi = 1−

i=1

βι +xi )

eα+ i=1 βι +xi m = 1 + eα+ i=1 βι +xi

1

1+

m e−(α+ i=1 βι +xi )

=

1

1+

m eα+ i=1 βι +xi

(12.2)

where xi is an independent variable, α is a regression intercept, and β represents regression coefficients. They are all nonlinear functions consisting only of independent variables xi . The ratio of the probability of occurrence of an event to the probability of non-occurrence is called the ratio of occurrence of an event: pi /(1 − pi ). A linear model of the logistic regression model can be obtained by comparing the event to logarithmic transformation as follows (2001):  ln

pi 1 − pi

 =α+

m i=1

βi xi .

(12.3)

In this study, the seven common factors derived from EFA combined with the basic information of respondents were the independent variables, whereas the binary alternative of evacuating from home during the earthquake was the dependent variable, where 0 represents the choice to stay at home and 1 represents the choice to escape from home during the earthquake.

12.3 Results and Discussion 12.3.1 Reliability Test A reliability analysis for all the aggregated data was performed using SPSS 24.0 software. We conducted reliability analysis on 35 questions designed in this questionnaire (9 basic information, 11 BE perception, and 15 disaster risk perception). The Cronbach’s alpha and number are shown in Table 12.5. In general, Cronbach’s alpha values of < 0.60 are considered as unsatisfactory, whereas values of > 0.70 are

328

12 Seismic Evacuation Preparedness …

Table 12.5 Reliability statistics

BE perception

Risk perception

Total items

Cronbach’s alpha

0.843

0.768

0.791

Number

11

15

26

regarded as satisfactory (Bland and Altman 1997). Therefore, the sample data used in this study passed the reliability test.

12.3.2 Exploratory Factor Analysis The maximum variance method was used to analyze the independent variables in factor analysis. Kaiser–Meyer–Olkin (KMO) and Bartlett’s tests were performed before EFA of BE and disaster risk perception. In general, if the KMO value, which varies from 0 to 1, is ≥ 0.70 and the p value for the Bartlett’s test for homogeneity of variances is < 0.05, then the data are considered to be suitable for EFA (Shiwakoti et al. 2016). The test results are shown in Table 12.6. The KMO statistic is 0.822 and 0.797, and the two p values are 0.000. The test results are significant, which indicates a certain correlation between the items of BE and risk perception, and is suitable for EFA. The EFA of BE perception is shown in Table 12.8. Finally, the 11 BE perception variables were reduced to 3 common variables (number of variables decreased by 72.727%), whereas the 3 common factors can explain 63.543% of the overall information of the 11 variables. In addition, the 16 risk perception variables were reduced to 4 common factors (number of variables decreased by 75.000%), where the 4 common factors can explain 58.263% of the overall information of the 16 variables (see Table 12.9). Therefore, EFA has less information loss and good analysis results. Tables 12.7 and 13.8 show the component rotation matrix of EFA on BE and risk perception, respectively. The three common factors of built environment perception can be concluded from Table 12.8, namely, BEP1 (reasonable outdoor evacuation route and shelter planning), BEP2 (good quality of building and village roads), and BEP3 (reasonable indoor evacuation route). Moreover, four disaster risk perception common factors are concluded from Table 12.9, namely, DRP1 (continuous Table 12.6 KMO and Bartlett’s test results

KMO and Bartlett’s test

BE perception

KMO Bartlett’s test

Chi-square Degrees of freedom P

Risk perception

0.822

0.797

2066.480

2162.960

55 0.000

105 0.000

12.3 Results and Discussion

329

Table 12.7 EFA results of BE perception: factor component matrix Component Reasonable outdoor evacuation route and shelter planning The residence is convenient for emergency evacuation in the area where you currently reside

0.849

A reasonable shelter exists in the place where you currently reside

0.837

Reasonable spacing is conducive to escape and evacuation during earthquakes

0.801

Good quality of building and village roads

Good roads exist from villages to other villages or towns

0.761

Roads from villages to other villages or towns are not easily damaged or congested in the event of an earthquake

0.709

The quality of building materials is guaranteed

0.699

The current house is strong and earthquake resistant

0.654

The current residential infrastructure is well planned

0.491

Reasonable indoor evacuation route

The current terrain environment has a strong impact on housing earthquake resistance* The interior design of the house has a reasonable emergency shelter

0.838

The interior design of the house has a reasonable emergency escape route

0.815

Eigenvalues Proportion of variance explained (%)

2.639

2.624

1.728

23.989

23.850

15.705 (continued)

330

12 Seismic Evacuation Preparedness …

Table 12.7 (continued) Component

Cumulative variance explained (%)

Reasonable outdoor evacuation route and shelter planning

Good quality of building and village roads

Reasonable indoor evacuation route

23.989

47.839

63.543

Extraction method: principal component analysis (PCA) Rotation method: Caesar normalized maximum variance method. The rotation converged after five iterations * The results were all less than 0.4, and the factor was not classified

negative psychological effects of earthquake), DRP2 (very sensitive to earthquakes), DRP3 (housing damage caused by earthquake), and DRP4 (number and impact of earthquakes are expected to decrease).

12.3.3 Multicollinearity of Variables When the multicollinearity problem exists, it will seriously affect the fitting effect of the model (Xiongwei). In this study, the variance inflation factor (VIF) was used to test multicollinearity, and the results are shown in Table 12.9. All VIF values of the variables used in this study were less than 2. Therefore, no multicollinearity problem exists between all explanatory variables.

12.3.4 Binary Logistic Regression Analysis Binary logistic regression was used to explore the influences of BE and disaster risk perception on seismic evacuation behavior after controlling the social demographic information. SPSS 24.0 software was used for binary logistic regression analysis estimation. Stepwise method was used to explore the contribution of each type of variable to the proposed model when running binary logistic regression. Basic information, three BE perception common factors, and four disaster risk perception common factors were added into the binary logistic regression model step by step. The Cox–Snell R2 values of the three models were 0.051, 0.072, and .0126. Finally, the Cox–Snell R2 value of the final binary logistic regression model was 0.126, showing that the data used in this study fit the binary logistic regression model well. Beta value represents the degree of influence of independent variables on dependent variables; positive and negative values indicate positive or negative influences, respectively. The greater the absolute value of beta is, the greater the

12.3 Results and Discussion

331

Table 12.8 EFA results of disaster risk perception: factor component matrix Items

Component Continuous negative psychological effects of earthquake

The impact of the earthquake on you is very serious

0.810

The impact of the earthquake on your family is very serious

0.779

An earthquake is catastrophic.

0.661

You think you will be directly affected

0.653

You think that an extreme earthquake will have a long-term negative impact

0.625

After the earthquake, you feel that aftershocks will always occur

0.577

After the earthquake, you will always be vigilant

0.563

Very sensitive to earthquakes

Earthquakes happen easily in this area

0.817

The risk of a serious earthquake in the future will be greater

0.737

You are very sensitive to shaking things

0.575

A minor earthquake will cause damage to your house

Housing damage caused by earthquake

Number and impact of earthquakes are expected to decrease

0.785

(continued)

332

12 Seismic Evacuation Preparedness …

Table 12.8 (continued) Items

Component Continuous negative psychological effects of earthquake

Very sensitive to earthquakes

The occurrence of an earthquake will cause further damage to your house

Housing damage caused by earthquake

Number and impact of earthquakes are expected to decrease

0.689

The number of earthquake disasters has decreased at present

0.742

The earthquake will not cause devastation to the house

0.717

You are not so scared of the earthquake at present

0.597

Eigenvalues

3.458

2.180

1.563

1.538

Proportion of 23.052 variance explained (%)

14.534

10.422

10.255

Cumulative 23.052 variance explained (%)

37.586

48.008

58.263

Extraction method: PCA. Rotation method: Caesar normalized maximum variance method. The rotation converged after five iterations

effect on the dependent variable. The results of the binary regression model are shown in Table 12.10 and will be further explained. Cross validation was performed using SPSS 24.0. Overall, 59.7% of the respondents who will stay at home were predicted as staying at home, whereas 40.3% of the respondents who will stay at home were predicted as escaping from home during the earthquake. More information about the cross-validation results are presented in Table 12.11. According to the results of binary logistical regression (Table 12.10), BE and risk perception have a significant influence on the seismic evacuation behavior of residents in the areas affected by the Wenchuan earthquake in 2008, whereas the impact of

12.3 Results and Discussion

333

Table 12.9 Multicollinearity analysis Coefficienta Model

Collinearity statistics Tolerance

VIF

Gender

0.946

1.057

Age

0.535

1.869

Educational background

0.583

1.714

Physical condition

0.960

1.042

Marital status

0.870

1.149

With or without children under 12 years old

0.915

1.093

With or without casualties

0.924

1.083

With or without property loss

0.967

1.034

Number of building floors

0.980

1.020

Housing construction type

0.916

1.092

Reasonable outdoor evacuation route and shelter planning

0.913

1.095

Good quality of building and village roads

0.889

1.125

Reasonable indoor evacuation route

0.889

1.125

Adverse continuous psychological effects of earthquake

0.865

1.156

Very sensitive to earthquakes

0.970

1.031

Housing damage caused by earthquake

0.956

1.046

Number and effect of earthquakes are expected to decrease

0.909

1.100

a Dependent

variable: whether to escape from home during the earthquake

demographic characteristic variables is limited. The conclusions are analyzed and discussed in the following. According to the results in Table 12.10, educational background positively affects the seismic evacuation behavior (OR = 1.629, 95% CI: 1.150–2.308, p = 0.006 < 0.01). That is, people with a higher education level feel that escaping from the building is the safest way to seek refuge. Moreover, these people are more inclined to escape from their residences when an earthquake occurs. This result is consistent with the findings of Koshiba et al. [21] that people with higher academic qualifications are more rational about their choice of emergency behavior when disasters occur. Alexander and Magni (2013) also found that participants with a higher education background were more inclined to implement positive behavioral strategies during the earthquake. Although the researchers referred to a number of behavioral patterns rather than just evacuating to the outdoors, their conclusions are similar in a sense to our findings that educational level has a positive influence on seismic evacuation behavior. Building construction type has a significant positive influence on evacuation behavior of the affected residents when an earthquake occurs. Residents living in nondetached houses are more inclined to choose to escape from their residence, whereas

334

12 Seismic Evacuation Preparedness …

Table 12.10 Results of Binary Logistic Regression Analysis Variable Gender

Beta

S.E.

Wald

P

OR

95%CI Lower

Upper 1.750

0.153

0.208

0.544

0.461

1.165

0.776

−0.071

0.097

0.542

0.461

0.931

0.770

1.126

Educational background

0.488

0.178

7.544

0.006

1.629

1.150

2.308

Physical condition

0.131

0.210

0.386

0.534

1.140

0.755

1.721

Marital status

0.029

0.281

0.011

0.917

1.030

0.594

1.784

−0.168

0.210

0.637

0.425

0.846

0.560

1.276

With or without casualties

−0.480

0.364

1.742

0.187

0.619

0.303

1.262

With or without property loss

−0.303

0.207

2.134

0.144

0.739

0.492

1.109

Number of building floors

−0.278

0.279

0.992

0.319

0.757

0.438

1.309

Housing construction type

0.448

0.215

4.359

0.037

1.566

1.028

2.385

BEP1 (reasonable outdoor evacuation route and shelter planning)

0.176

0.105

2.800

0.094

1.193

0.970

1.467

BEP2 (good quality of building and village roads)

0.277

0.108

6.614

0.010

1.320

1.068

1.630

BEP3 (reasonable indoor evacuation route)

0.233

0.108

4.685

0.030

1.263

1.022

1.560

DRP1 (bad continuous psychological effects of earthquake)

0.204

0.108

3.549

0.060

1.226

0.992

1.515

DRP2 (very sensitive to earthquakes)

0.227

0.104

4.817

0.028

1.255

1.025

1.538

DRP3 (housing damage caused by earthquake)

0.471

0.108

18.858

0.000

1.601

1.295

1.980

DRP4 (number and impact of earthquakes are expected to decrease)

0.080

0.106

0.572

0.450

1.083

0.880

1.334

Constant

0.045

1.212

0.001

0.970

1.047

Age

With or without children under 12 years old

those living in single-family houses do not choose to escape out. Although this result is consistent with the research conclusion of Alexander and Magni (2013) that residents living in private single-family houses are more willing to stay at home when an earthquake occurs, the reasons why they do not escape are different. According to Alexander and Magni (2013), the residents who lived in private single-family houses have more options for emergency evacuation during a disaster, and may be safer and more effective in avoiding hazards than those who directly escape from their residence. On the contrary, we found through our field household survey that due to imperfect evacuation facilities, residents who lived in single-family buildings in Chinese villages have to stay at home when an earthquake occurs. Therefore, the

12.3 Results and Discussion

335

Table 12.11 Leave-one-out Coss Validation Resultsa,c Whether to escape from home during the earthquake

Predicted Group Membership No

Original

Count %

Cross validatedb

aA

Total

Yes

No

116

75

191

Yes

99

178

277

No

60.7

39.3

Yes

35.7

64.3

100.0

100.0

Count

No

114

77

191

Yes

103

174

277

%

No

59.7

40.3

100.0

Yes

37.2

62.8

100.0

total of 62.8% of original grouped cases was correctly classified

b Cross

validation was performed only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case

cA

total of 61.5% of cross-validated grouped cases was correctly classified

earthquake disaster evacuation facilities need to be further improved in the earthquake zones, especially in remote rural areas. BEP1 (reasonable outdoor evacuation route and shelter planning) shows a significant positive correlation with seismic evacuation behavior in the binary logistic regression (see Table 12.10) (P = –0.094). If residents believe that reasonable outdoor evacuation route and shelter are available in their area, then they will immediately escape from the residence to the safe shelter when an earthquake disaster occurs to avoid the possible danger caused by the earthquake. Anbarci et al. (2005) also found that well-planned evacuation routes and shelters are more secure, which can help residents escape when an earthquake occurs. Thus, the development level of the local economy directly affects the improvement of earthquake escape facilities. BEP2 (good quality of buildings and village roads) has a significant positive influence on seismic evacuation behavior (P = 0.010, B = 0.277). That is, if residents have knowledge on the good quality of buildings and village roads in the area where they currently reside, then they are inclined to escape from their residence and wait for external rescue. Previous studies show that the enormous losses resulting from the Wenchuan earthquake in 2008 were largely attributed to failures of BE, which also severely affected local emergency response and external aid (Wang and Tang 2017). Therefore, the quality improvement of rural roads and rural buildings is necessary to protect the lives and properties of rural residents. BEP3 (reasonable indoor evacuation route) (P = 0.03 < 0.05, B = 0.233) shows that if the interior design of the house has a good emergency escape route or a good escape environment, then the residents will also opt to evacuate the building with good escape conditions.

336

12 Seismic Evacuation Preparedness …

DRP1 (continuous negative psychological effects of earthquake, P = 0.060, B = 0.205) and DRP2 (very sensitive to earthquakes, P = 0.028 < 0.05, B = 0.227) are both positively correlated to seismic evacuation behavior. In other words, the more the negative impacts caused by the earthquake disasters on residents’ psychology and their own risk sensitivity, the greater is their possibility of choosing to escape from buildings to avoid disaster and injury. The results of these two risk perception factors are in accordance with the research findings of Ruin I and Hung HV (Hung et al. 2007; Ruin et al. 2007). Individuals with low risk perception are less likely to effectively make emergency evacuation behaviors than those with higher risk perception. Therefore, residents’ earthquake disaster risk psychology can effectively predict their evacuation behavior. A noteworthy point is that the factor of DRP3 (housing damage caused by earthquake) is positively correlated with the evacuation behavior of local residents during the earthquake disaster (p = 0.000, B = 0.471). When residents perceive greater risk and damage to their homes, they have a greater likelihood of fleeing when an earthquake strikes. The larger absolute value of the coefficient also indicates that housing damage risk perception of the residents largely determines their evacuation behavior during the earthquake disaster (Brown et al. 2018). In sum, BE and disaster risk perception have more significant effects on seismic evacuation behavior than social demographic variables. All the three common factors of BE perception significantly affect the escape behavior of residents during an earthquake disaster. Moreover, three out of four common factors of disaster risk perception also significantly affect the seismic evacuation behavior of local residents. According to Burnside et al. (2007), the residents who chose to evacuate during the previous earthquake were likely to have the same evacuation behavior when another earthquake occurred. This result is also related to the psychological impact on the affected residents during the previous earthquake. Thus, the results of the present study are consistent with the existing research conclusions.

12.4 Conclusion and Policy Recommendations The main purpose of this study is to investigate the influence of BE and disaster risk perception on the seismic evacuation behavior of residents in earthquake-prone areas. A total of 10 villages located within the worst-hit areas of the Wenchuan earthquake in 2008 were randomly selected as samples for empirical research. Exploratory factor analysis and binary logistic regression were used in this study. Therefore, the following conclusions were drawn: 1. The demographic characteristics of residents have less influence on the seismic evacuation behavior than their subjective perception factors. The more educated the residents are, the more likely they are to flee their homes when an earthquake occurs.

12.4 Conclusion and Policy Recommendations

337

2. When an earthquake occurs, residents living in non-detached buildings are more likely to decide to leave their residences than those living in single-family buildings due to the difference in infrastructures. The areas of non-detached buildings are more perfect than those of single-family buildings. Alexander and Magni (2013) also concluded this finding in the foreign rural villages, which are totally different from Chinese villages. However, a similarity exists in the notion that the building type affects the evacuation choice of residents all over the world. Therefore, to better protect the lives of residents during an earthquake disaster, some of the most effective methods are to improve the residents’ education level, regional infrastructures, and building quality, especially in remote rural areas. 3. The three common factors of BE perception from the micro-building to the macro area have a positive influence on the seismic evacuation behavior of the local residents. Compared with cities, rural areas’ social and economic fragility and inequality in the built environment may further aggravate the negative consequences of disasters (Cutter et al. 2010). Therefore, for earthquake-prone areas, planning and constructing residential and regional evacuation facilities based on the perspective of disaster prevention and mitigation is particularly important. Moreover, enabling residents to intuitively perceive the convenience of escape during the occurrence of an earthquake is also necessary. 4. Risk perception plays a major role in various effective responses to disasters, and facilitates decision-making in risk management and disaster mitigation (Lindell and Hwang 2010). A total of three out of four common factors in disaster risk perception positively affect the seismic evacuation behavior. Although the frequency of major earthquakes is lower than that of minor earthquakes, the consequences can also be fatal (Wachinger 2013). Therefore, improving residents’ disaster risk perception is crucial. Moreover, preparing for future earthquake disasters is necessary. In earthquake-prone areas, strengthening awareness and education on earthquake disaster risk prevention is essential to protect the safety of residents’ lives and properties.

12.5 Limitations and Future Work The main limitation of this study is that it only conducts binary logistic regression analysis based on the dependent variable of whether to evacuate buildings or not but does not comprehensively analyze the effect of BE perception and disaster risk perception on various emergency behaviors. Earthquakes usually kill or injure people in areas with similar structures and vulnerability, mostly in developing countries. Our findings may help reduce the effect of disasters in less-developed areas such as the mountainous countryside near China’s seismic belt. In future studies, we can analyze the diverse choices of people’s emergency behaviors under various conditions that are not limited to evacuation. For example, in the study by Koshiba et al. (2018), respondents were asked about their behavior choices in 13 scenarios, such as in cold weather and on rainy days. Although their research

338

12 Seismic Evacuation Preparedness …

focused on the behavior of returning to buildings, their research can also be used to ask questions in the manner of providing scenarios when studying whether to evacuate buildings. Furthermore, a series of analyses of emergency behavior can be conducted in the study of the BE with highly objective data, such as geographic information data mined by GIS technology.

References 2001. Logistic regression model: methods and applications. 2018. Multivariate statistical method and its application. Alexander, D., & Magni, M. 2013. Mortality in the l’aquila (central Italy) earthquake of 6 april 2009. Plos Currents, 5: e50585b50588e50586efd50581. Anbarci, N., Escaleras, M., & Register, C. A. 2005. Earthquake fatalities: the interaction of nature and political economy. Journal of Public Economics, 89(9): 1907–1933. Bland, J. M., & Altman, D. G. 1997. Statistics notes: Cronbach’s alpha. BMJ, 314(7080): 572. Bowser, G. C., & Cutter, S. L. 2015. Stay or Go? Examining Decision Making and Behavior in Hurricane Evacuations. Environment: Science and Policy for Sustainable Development, 57(6): 28–41. Brown, P., Daigneault, A., Tjernstrom, E., & Zou, W. 2018. Natural Disasters, Social Protection, and Risk Perceptions. World Dev, 104: 310–325. Burnside, R., Miller, D. M. S., & Rivera, J. D. 2007. The Impact of Information and Risk Perception on the Hurricane Evacuation Decision-Making of Greater New Orleans Residents. Sociological Spectrum, 27(6): 727–740. Cahyanto, I., Pennington-Gray, L., Thapa, B., Srinivasan, S., Villegas, J., Matyas, C., & Kiousis, S. 2014. An empirical evaluation of the determinants of tourist’s hurricane evacuation decision making. Journal of Destination Marketing & Management, 2(4): 253–265. Clason, D. L., & Dormody, T. J. 1994. Analyzing Data Measured by Individual Likert-Type Items. Journal of Agricultural Education, 35(4): 31–35. Cohen, O., Leykin, D., Lahad, M., Goldberg, A., & Aharonson-Daniel, L. 2013. The conjoint community resiliency assessment measure as a baseline for profiling and predicting community resilience for emergencies. Technological Forecasting & Social Change, 80(9): 1732–1741. Cutter, S. L., Boruff, B. J., & Shirley, W. L. 2010. Social Vulnerability to Environmental Hazards. Social Science Quarterly, 84(2): 242–261. Davis, Ian, Alexander, & David. 2015. Recovery from disaster. Dooley, D., Catalano, R., Mishra, S., & Serxner, S. 2010. Earthquake Preparedness: Predictors in a Community Survey1. Journal of Applied Social Psychology, 22(6): 451–470. Fullerton, C. S., Mccaughey, B. G., & Ursano, R. J. 1996. Individual and Community Responses to Trauma and Disaster: the Structure of Human Chaos. Bmj Clinical Research, 310(6974): 269. Goltz, J., Russell, L. A., & Bourque, L. B. 1992. Initial Behavioral Response to a Rapid Onset Disaster: a Case Study of the October 1, 1987, Whittier Narrows Earthquake. Haghani, M., & Sarvi, M. 2016. Human exit choice in crowded built environments: Investigating underlying behavioural differences between normal egress and emergency evacuations. Fire Safety Journal, 85: 1–9. Handy, S., Cao, X., & Mokhtarian, P. 2005. Correlation or causality between the built environment and travel behavior? Evidence from Northern California. University of California Transportation Center Working Papers, 10(6): 427–444. Handy, S. L., Boarnet, M. G., Ewing, R., & Killingsworth, R. E. 2002. How the built environment affects physical activity: Views from urban planning. American Journal of Preventive Medicine, 23(2): 64–73.

References

339

Highfield, W. 2005. Hurricane risk perceptions among Florida’s single family homeowners. Landscape & Urban Planning, 73(2): 120–135. Hung, H. V., Shaw, R., & Kobayashi, M. 2007. Flood risk management for the RUA of Hanoi: Importance of community perception of catastrophic flood risk in disaster risk planning. Disaster Prevention & Management, 16(2): 245–258. Jianguang, Z. 1994. Environmental hazards in the Chinese public’s eyes. Risk Analysis An Official Publication of the Society for Risk Analysis, 14(2): 509–513. Ke-Sheng, G. U., Song, D. Y., Shao, Y. C., Liu, J., En-Mao, Y. E., Wang, X. H., & Yang, Z. 2010. Mental health of China rescue cadets after Wenchuan earthquake. Bulletin of the Academy of Military Medical Sciences. Kinateder, M. T., Kuligowski, E. D., Reneke, P. A., & Peacock, R. D. 2015. Risk perception in fire evacuation behavior revisited: definitions, related concepts, and empirical evidence. Fire Sci Rev, 4(1): 1. Koshiba, Y., & Suzuki, Y. 2018. Factors affecting post-evacuation behaviors following an earthquake: A questionnaire-based survey. International Journal of Disaster Risk Reduction. Lambie, E. S., Wilson, T. M., Brogt, E., Johnston, D. M., Ardagh, M., Deely, J., Jensen, S., & Feldmann-Jensen, S. 2016. Closed Circuit Television (CCTV) Earthquake Behaviour Coding Methodology: analysis of Christchurch Public Hospital video data from the 22 February Christchurch earthquake event. Natural Hazards, 86(3): 1175–1192. Lang, Z., & Qing-Song, H. 2010. Discussions of Investigation about House Damages in Rural Areas in Shifang and Dujiangyan after the “5.12” Earthquake. Earthquake Resistant Engineering & Retrofitting. Li, M., Xu, J., He, Y., & Wu, Z. 2012. The Analysis of the Resilience of Asults One Year After the 2008 Wenchuan Earthquake. Journal of Community Psychology, 40(7): 860–870. Li, Y., & Guo, Y. 2016. Be proactive for better decisions: Predicting information seeking in the context of earthquake risk. International Journal of Disaster Risk Reduction, 19: 75–83. Liang, Y., Cheng, J., Ruzek, J. I., & Liu, Z. 2019. Posttraumatic stress disorder following the 2008 Wenchuan earthquake: A 10-year systematic review among highly exposed populations in China. J Affect Disord, 243: 327–339. Lindell, M. K. 2010. Built-in resilience. Nature Geoscience, 3(11): 739–740. Lindell, M. K., & Hwang, S. N. 2010. Households’ perceived personal risk and responses in a multihazard environment. Risk Analysis An Official Publication of the Society for Risk Analysis, 28(2): 539–556. Lindell, M. K., & Perry, R. W. 2000. Household adjustment to earthquake hazard. Environment & Behavior, 32(4): 461–501. Lindell, M. K., Prater, C. S., Wu, H. C., Huang, S. K., Johnston, D. M., Becker, J. S., & Shiroshita, H. 2015. Immediate behavioural responses to earthquakes in Christchurch, New Zealand, and Hitachi, Japan. Disasters, 40(1): 85. Liu, H., Zhang, D., Wei, Q., & Guo, Z. 2017. Comparison study on two post-earthquake rehabilitation and reconstruction modes in China. International Journal of Disaster Risk Reduction, 23: 109– 118. Mao, X., Fung, O. W. M., Hu, X., & Loke, A. Y. 2018. Psychological impacts of disaster on rescue workers: A review of the literature. International Journal of Disaster Risk Reduction, 27: 602–617. Mark, C., Gaston, G., Paschal, S., & Marc, G. 2013. Some feelings are more important: cognitive attitudes, affective attitudes, anticipated affect, and blood donation. Health Psychology, 32(3): 264–272. Meng, Z., Wu, X., & Han, L. 2018. Post-traumatic stress disorder and post-traumatic growth among the adult survivors of the Lushan earthquake: Selecting resilience as the moderator. International Journal of Disaster Risk Reduction, 27: 524–529. Paton, D., Smith, L., Daly, M., & Johnston, D. 2008. Risk perception and volcanic hazard mitigation: Individual and social perspectives. Journal of Volcanology & Geothermal Research, 172(3): 179–188.

340

12 Seismic Evacuation Preparedness …

Peng, C. 2011. Formation mechanism and risk control of mountain disaster in wenchuan earthquake. Key laboratory of mountain hazards and surface processes. Peng, Y., Zhang, F., Jiang, S., Huang, L., Wang, Z., & Xu, Y. 2018a. Analysis of farmers’ satisfaction towards concentrated rural settlement development after the Wenchuan earthquake. International Journal of Disaster Risk Reduction, 31: 160–169. Peng, Y., Zhu, X., Zhang, F., Huang, L., Xue, J., & Xu, Y. 2018b. Farmers’ risk perception of concentrated rural settlement development after the 5.12 Sichuan Earthquake. Habitat International, 71: 169–176. Ruin, I., Gaillard, J. C., & Lutoff, C. 2007. How to get there? Assessing motorists’ flash flood risk perception on daily itineraries. Global Environmental Change Part B Environmental Hazards, 7(3): 235–244. Saelens, B. E., & Handy, S. L. 2008. Built Environment Correlates of Walking: A Review. Med Sci Sports Exerc, 40(7 Suppl): S550–566. Shapira, S., Aharonson-Daniel, L., & Bar-Dayan, Y. 2018. Anticipated behavioral response patterns to an earthquake: The role of personal and household characteristics, risk perception, previous experience and preparedness. International Journal of Disaster Risk Reduction, 31: 1–8. Shiwakoti, N. 2016. Understanding differences in emergency escape and experimental pedestrian crowd egress through quantitative comparison. International Journal of Disaster Risk Reduction, 20: 129–137. Shiwakoti, N., Tay, R., Stasinopoulos, P., & Woolley, P. J. 2016. Passengers’ awareness and perceptions of way finding tools in a train station. Safety Science, 87: 179–185. Slovic, P. 2000. The Perception of risk. Risk Society & Policy, 69(3): 112–112. Spearman, C. 1987. The proof and measurement of association between two things. By C. Spearman, 1904. American Journal of Psychology, 100(3/4): 441–471. Wachinger, G. 2013. The risk perception paradox–implications for governance and communication of natural hazards. Risk Analysis, 33(6): 1049–1065. Wachinger, G., Renn, O., Begg, C., & Kuhlicke, C. 2013. The risk perception paradox–implications for governance and communication of natural hazards. Risk Analysis, 33(6): 1049–1065. Wang, S., & Tang, W. 2017. Understanding the Role of Built Environment Resilience to Natural Disasters: Lessons Learned from the Wenchuan Earthquake Journal of Performance of Constructed Facilities, 31(5). Wang, X. Sustainable development of rural energy in China_ Status quo_ Challenges and Countermeasures.pdf Wu, J., Ning, L., Shi, P., Hu, A., & Liu, X. 2012. Regional indirect economic impact evaluation of the 2008 Wenchuan Earthquake. Environmental Earth Sciences, 65(1): 161–172. Xiongwei, M. The diagnosis method of multicollinearity in linear regression equation and its empirical analysis. . Journal of huazhong agricultural university (social science edition)(2): 84–87 + 91. Yan, C., Suzanne, W., David, B., Erica, S., & Regan, P. 2011. An integrated approach: managing resources for post-disaster reconstruction. Disasters, 35(4): 739–765. ye, Y. 2008. Rural construction aseismic. Architecture Technique, 166(7): 84–93. Zhang, J. S. 2010. The Study on Assessment Index of Urban Social Vulnerability to Earthquake Disaster. Guide of Sci-tech Magazine. Zhang, L., He, G. Z., Mol, A. P. J., & Lu, Y. L. 2013. Public perceptions of environmental risk in China. Journal of Risk Research, 16(2): 195–209. Zhang, W., Xu, X., & Chen, X. 2017. Social vulnerability assessment of earthquake disaster based on the catastrophe progression method: A Sichuan Province case study. International Journal of Disaster Risk Reduction, 24: 361–372.

Chapter 13

The Experience and Attitude of Rural Residents with Regard to Flood Disaster Preparedness

Abstract This chapter takes rural residents of the flood-prone regions of Sichuan Province, China, as the research subject to ascertain the effects of flood experience and attitude on residents’ disaster preparedness behavior. Results show the following. (1) The level of disaster preparedness behavior of rural residents is relatively low. (2) Respondents who are older, more educated, or who have lived locally for longer, are more likely to be better prepared for flood disasters. (3) On the other hand, respondents with higher income, either as individuals or households, are less prepared. (4) Previous flood experience and disposition towards disaster, also impacts preparedness behavior. On the basis of these findings, the following suggestions are proposed as mitigation policies for flood-prone areas. (1) Structural flood control measures should be organically combined with daily disaster preparedness. (2) Rural areas affected by devastating floods should offer psychological counseling to residents in order to reduce states of fear and helplessness. (3) Government departments should inform residents in real-time of flood warnings and of flood control measures. (4) Lastly, relevant local departments should educate local residents in flood preparedness. Taken together, these measures can be expected to improve living conditions, people’s trust in the media, and in the government’s flood control capacity. Keywords Flooding · Flood disaster · Disaster preparedness · Disaster Trauma · Sichuan · Rural China · Ordered logit regression

13.1 Introduction Disaster events are on the rise globally. In 2012 alone, 357 disasters were reported, with 9,655 deaths and a further 124.5 million people adversely impacted. The economic losses amounted to US$157 billion. In terms of disaster frequency over the last ten years, five countries stand out as having suffered the most. These are China, India, Indonesia, and the Philippines, in Asia, as well as the United States. China alone accounted for 34.7% of world disaster victims, with 446 million people affected (Ren et al. 2018). About 70% of the world’s natural disasters occur in the Asia Pacific region. Coastal regions of the Western Pacific are subject to volcanic and

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_13

341

342

13 The Experience and Attitude of Rural …

earthquake activity. Elsewhere risks include fire, hydrological and geological disasters. For example, typhoons Pablo and Yolanda, of 2012 and 2013, cost the Philippines close to US2billion in damage, along with thousands of lives. Japan’s 2011 Tohoku Earthquake, generated a tsunami 40 meters high, travelling at 700 km/hr., that reached as far as 10 km. inland. The magnitude 8.0 Wenchuan earthquake of 2008, resulted in the destruction of 80% of the buildings in Beichuan County, and was felt as far away as Shanghai. These events and others have caused serious and long-term economic loss and hardship. In particular, China’s prospects for continued, sustainable development has been severely handicapped (Yang et al. 2013). It is floods, however, that are the primary natural disaster that affects the development of human society and economy. It is estimated that the disaster loss caused by flooding accounts for approximately 40% of natural disaster losses (Daniell et al. 2016). According to the fifth comprehensive assessment report of the Intergovernmental Panel on Climate Change, the frequency and scale of floods in the various basins of the world have been affected by climate change, resulting in increased direct or indirect losses caused by disasters (Ottmar et al. 2014). In 2005, the United Nations proposed increasing the preparedness of communities with respect to disasters, especially in rural areas. This goal was ratified in the disaster risk reduction agreement of the following decade (Deng 2019). Along with growth in the economy, the consequent economic loss and impact of floods on society have also been on the rise. Li (2013) studied flood disaster data for the Asia Pacific region over the period 1990 to 2010 and confirmed that flood frequency over the entire region was on the rise. The United States is also an at-risk country in terms of the flood frequency (Zhu 2018). It also has the largest number of papers published in the field of flood disasters, followed by China (Cao et al. 2018). In September 2008, Galveston, Texas, USA was hit by Hurricane Ike. Researchers have conducted disaster preparedness and mitigation research on local enterprises. They found that enterprises in the Galveston area that experienced a low frequency of disasters had significantly improved their disaster preparedness level as compared to previously (Xiao et al. 2017). China has one of the highest frequencies of flood disasters in the world; with as much as 10% of the land area under threat of serious disaster losses. Indeed, 70% of the total industrial and agricultural output value of the country is vulnerable to flooding (Wang et al. 2014). Qin et al. (2014) reported that precipitation frequency and intensity may increase in most areas of China’s mid-latitude, and the trend of “the drier the drier, the wetter the wetter” is observed. Du et al. (2016) established a risk assessment system for rainstorm flood disasters over the rainy season, from June to August, based on a study of the Sichuan region. They found that eastern Sichuan is located in a rainstorm flood risk area. Guo et al. (2010) established a flood risk assessment system using ArcGIS and concluded that the formation of flood disaster is affected by a maximum three-day rainfall. Zhou et al. (2000) established a regional rainstorm flood risk assessment system and took regional disaster prevention and mitigation, rainstorm hazard risk, and potential vulnerability of disaster bearing

13.1 Introduction

343

body as evaluation indices. Chen (2007) proposed a flood risk assessment model to determine the potential flood risk of each section on the basis of river network and water system conditions. Moreover, Fan et al. (2016) found that Sichuan had the largest number of emergency responses to flood disasters, over the period 2011–2015. The research on disaster behavior in foreign countries can be traced back to the end of 1940s. On the basis of structural functionalism, disaster is an event with space–time characteristics, which can destroy social structure and lead to damage of social function. After 1960s, scholars with geological background have joined in and studied hazard perspective and disaster-causing factors and society interaction research, affected by human ecology; they found that people can adapt to disasters by adjusting their behavior Since the 1980s, influenced by political economy, social conflict and other theories, the social production of disaster (Sun Lei et al. 2018) has been produced. Given the heavy impacts of disaster relief costs, the concept of disaster management has evolved, with the method of disaster reduction now shifted from passive emergency response to active disaster prevention. Scholars and administrators worldwide have enriched the family disaster preparedness system, by introducing standardized disaster preparedness scales and producing disaster preparedness lists. However, in China, few scholars have proposed such disaster preparedness scales and lists. Accordingly, the present work discusses family disaster preparedness capability, and proposes a disaster preparedness ability standard scale based on quantitative analysis (Deng 2019). In the early 1990s, research in China has begun to be concerned on disaster behavior. Guo (1993) provided a definition of disaster behavior. Behavior is the expression of potential attitude, and attitude has significant influence on behavior. In addition to attitude, other factors, such as habit and experience, can affect behavior, together with attitude (Zhang et al. 2007). At the same time, the socioeconomic characteristics of farmers also significantly affect their disaster preparedness behavior (Thieken et al. 2007; Lin et al. 2008; Soane et al. 2010; Xu et al. 2018).

13.1.1 Effects of Previous Experiences on Disaster Preparedness In addition to housing supply and early warning systems (radio is more effective than television), flood experience is regarded as a major driver of disaster preparedness (Atreya et al. 2017; Priyanti et al. 2019) and one of the key predictors of preventive action (Hoffmann and Muttarak 2017). Lawrence et al. (2014) claimed that flood experience is a direct factor that affects the protective response behavior and flood preparedness behavior. Ejeta et al. (2018) believed that flood disaster experience significantly affects disaster preparedness but not relationships; however, this notion is mediated by trust and emotion (anxiety). Terpstra (2011) and Schad et al. (2012) indicated that disaster experience may change the understanding of the necessity of preventive measures and usually affects the motivation of individuals to cope with

344

13 The Experience and Attitude of Rural …

future risks. Individuals who have experienced natural disasters at an excessively high level are likely to develop coping strategies (Lawrence et al. 2014; Box et al. 2016; Soetanto et al. 2016a). Flood experience helps raise people’s awareness of risks, increase family preparedness for disasters, and supplement existing structural protection (Lawrence et al. 2014). A survey in three regions in Germany also found that flood experience significantly affected disaster preparedness (Thieken et al. 2007; Osberghaus 2015). Flood experience can affect people’s response and preparation for disasters but may be insufficient to cause behavioral change due to people unwilling to deviate from their routine (Lawrence et al. 2014; Box et al. 2016; Soetanto et al. 2016a). Some studies have also found that people with flood experience in Australia are likely to buy flood insurance but are less likely to take additional measures to prevent floods (Box et al. 2016).

13.1.2 Effects of Attitudes on Disaster Preparedness Huang et al. (2020) believed that residents’ attitudes toward flood risk directly affect their protective response behavior. Myers et al. (2016) proposed the theory of attitude formation, which holds that attitude is composed of three dimensions, namely, cognitive, behavioral tendency, and emotional components. In terms of cognitive components, cognitive factors refer to the narratives with evaluative significance to attitude objects, including factual understanding and cognition and beliefs of the attitude objects (Taylor et al. 2000). The perceptual, cognitive, and trust (cognitive trust) factors can be regarded as the cognitive component of attitude formation theory. Chinese scholars are at the early stage in cognitive research, and the Chinese lack a systematic analysis of disaster cognition, especially on disaster risk cognition. Contrariwise, Westerns scholars have developed an extensive understanding, including in respect of safety, health, ecological environment, and other aspects (Zhang et al. 2019). Scolobig et al. (2012), Reynaud et al. (2013), Hoffmann and Muttarak (2017), and Yong and Lemyre (2019) found that improvement of risk perception can enhance disaster preparedness, which is an important predictor of the latter. For example, the perception of future flood prediction significantly affects risk averse behavior (Osberghaus 2015). However, some people have not found a meaningful relationship, which may be overlooked by the method or the existence of an unknown intermediary (Yong and Lemyre 2019). In rural Nigeria, a survey of 300 households (Ajaero et al. 2016) found that more than 75% of respondents received flood information from radio or television, and a significant difference existed in the perception of reports. Flood reports did not affect people’s attitude toward disaster preparedness. Huang et al. (2020) showed that flood risk cognition is related to protective coping behavior. In the southwest poverty-stricken areas, farmers’ cognition of disaster prevention and mitigation was weak. The farmers who said that their families have no measures for disaster prevention and mitigation accounted for 69.17% of the survey, and their disaster perception and government measures had a considerable influence on their disaster prevention and mitigation behaviors (Liu and Zhuang

13.1 Introduction

345

2012). Ejeta et al. (2018) found that high trust (trust in existing flood control dams and trust in government information sources) can diminish concern for possible disasters, and thus hinder flood prevention. Chilean scholars have studied the relationship among natural disasters, disaster preparation, trust, and risk perception. They determined that for those of older age and a lower socioeconomic level, the perceived risk of natural disasters and the trust toward institutions were high, which plays a key role in disaster risk mitigation (Lin et al. 2008; Bronfman et al. 2015). Basolo et al. (2009) and Samaddar et al. (2012) found that high trust (information source) can lead to improved acceptance of preventive measures among residents. This finding is closely related to disaster preparedness. Basolo et al. (2009) and Samaddar et al. (2012) found that a high degree of trust will lead to better acceptance of precautionary measures among residents and is closely related to disaster preparedness. Residents in the eastern lakes of China developed trust in authorities after receiving disaster warnings, which also positively shifted residents’ attitudes (Shen et al. 2012). Su et al. (2008) and Zhong et al. (2010) found that trust in flood control projects reduce the public’s vigilance with respect to flooding. Consequently, a vast majority (80%) of the public underestimate the effects of flood disaster, resulting in a dual (negative and positive) psychological tendency of disaster prevention, which may eventually lead to public neglect of disaster preparedness. In terms of behavioral tendencies, Ajzen (1991) indicated that behavior is determined by intention tendency (i.e., decision to engage in a specific behavior). Intention is determined by individual motivation factors (Sinatra et al. 2012). In terms of individual cognitive decision making, behavioral tendency is the intermediate link between perception and behavior (Soetanto et al. 2016a). For residents affected by floods, disasters make them feel unsafe; thus, they tend to take structural measures to protect themselves (Schad et al. 2012). This tendency indicates that the existence of public flood control measures can profoundly affect the intention of individuals to take protective measures (Birkholz et al. 2014). However, some studies have also found that residents are unwilling to take measures to prevent disasters, as the latter relies on public flood control measures to form an “embankment effect” (Lin et al. 2008). A lack of resources, such as “time, money, knowledge, or social support,” is also noted, which can prevent individuals who intend to take action from actually taking protective actions (Grothmann and Pat 2005). In terms of emotional components, the public’s emotions about flood will significantly affect their actions in response to the disaster (Wang et al. 2018). Residents suffering from flood are emotional; have added fear and anxiety about future disasters; and worry about another flood, especially negative emotions that will encourage residents to take disaster preparedness measures (Zaalberg et al. 2009; Reynaud et al. 2013; Zaalberg and Middlen 2013). Ejeta et al. (2018) also confirmed that residents who have experienced floods are more anxious than those who have not and thus prefer to prepare for floods. By contrast, fear of future floods has been suggested to nearly have no link to past conservation measures (Schad et al. 2012; Terpstra 2011). The change in residents’ personal emotions has a remarkable influence on their decision-making behavior. Strong negative emotions increase the rational decisions of the residents to choose high-risk behaviors (Lei and Yang 2016), that is, no

346

13 The Experience and Attitude of Rural …

preventive measures are taken. In addition, the degree of fear is especially affected by citizens’ negative emotions and flood experience. The negative emotions most often reflect fear and powerlessness (Ajzen 1991), which significantly affect residents’ disaster preparedness behavior.

13.1.3 Effects of Social Demographic Characteristics on Disaster Preparedness In terms of socioeconomic characteristics, a potential relationship exists between sociodemographic factors and protective coping behaviors (Huang et al. 2020). Generally, the victims in disaster areas have a low level of education and low economic income (Lin et al. 2008). Their age (the younger age group take more preventive measures) and family size (especially young families) can affect disaster preparedness behavior (Thieken et al. 2007). Soan et al. (2010) and Xu et al. (2018) found that young individuals, high education, and high income have a positive influence on disaster preparedness behavior. Hoffmann and Muttarak (2017) found that education increases disaster preparedness actions, and they determined the basic mechanism of education that promotes disaster preparedness. However, Kim and Kang (2010) found no correlation between education and disaster preparedness. By contrast, Atreya et al. (2017) also confirmed that people with high education are active in disaster preparedness, and socioeconomic conditions are also effective predictors of disaster preparedness. For example, poor families are unlikely to prepare for disasters. Liu and Zhuang (2012) also found that economic strength has a greater influence on farmers’ disaster prevention and mitigation behavior. However, in a study by Hoffmann and Muttarak (2017), family income does not affect disaster preparedness. A study in Scotland found that women were more affected than men after a flood, but no significant effect of gender was observed on flood preparedness behavior in three high-risk areas of Lagos and the UK (Soetanto et al. 2016b). Moreover, proximity to the disaster epicenter increases the significance of the disaster preparedness behavior of residents (Xu et al. 2018), whereas house reinforcement is also a measure in dealing with flood (Atreya et al. 2017; Priyanti et al. 2019). Therefore, this study further explores the disaster preparedness behavior of residents in flood-prone areas, looking at the parameters of age, income, years of residence, number of households, annual family income, number of family members with income, residential building area, number of residential floors, and distance between home and river.

13.1 Introduction

347

13.1.4 Rationale of the Study Chen (2019) found that an awareness of disaster preparedness in rural areas is no guarantee of disaster preparedness. In the formulation of emergency plans, the community will be organized to conduct disaster prevention and mitigation or drill activities. However, the escape plan at the family level is not considered. In terms of emergency medical treatment, farmers’ access to medical supplies is cumbersome and time-consuming. Liu (2015) believed that research on disaster management at the grassroots level has a guiding role in improving people’s confidence in developing a better life. Ye (2010) found that in developed countries, where individuals, communities, and governments collectively participate in disaster prevention and mitigation management, they gradually become a system. However, China proposed to build only 1,000 comprehensive disaster reduction demonstration communities as part of its 11th national disaster reduction Five-year Plan. In fact, the government plays a role in early warning and emergency response. Thus, there is scope for disaster prevention and preparedness to take place at the family unit level. Wang et al. (2014) found that a good disaster preparedness curriculum has a positive influence on disaster preparedness behavior. Wang (2012) analyzed the problems associated with the disaster preparedness of debris flow in Zhouqu County, Gansu Province, finding that family disaster preparedness was poor; there was no family disaster preparedness plan, local warnings were late, knowledge of local disaster prevention and mitigation was insufficient, evacuation drills were lacking, and there were no family emergency supplies. Overall, the number of studies on “flood disaster” has been increasing; however, most of them focus on the formation characteristics, disaster mechanism, and engineering control measures of flood disaster (Cao et al. 2018). After signing the framework for reducing disaster risk in Sendai, 2015–2030, the number of papers on natural disaster spiked (Raikes et al. 2019). Previously, civil engineers have dictated the scientific development and progress of flood damage analysis. They focused on the technical and financial aspects and ignored the importance of socioeconomic factors and social science methods (Messner and Meyer 2006). Structural flood control measures cannot completely eliminate the risk of flood disaster, even occasionally only transferring the risk. Therefore, taking nonstructural measures to reduce flood losses is an important part of modern flood management (Osberghaus 2015). However, the current research on disaster behavior from the perspective of social sciences in China has not formed a system (Sun et al. 2018). Research scarcely focuses on the measurement of rural residents’ natural disaster risk attitude with few corresponding reviews. Attitude plays an important role in the prevention and mitigation of natural disasters. Accurate measurement of rural households’ risk attitude is the basis of their decision-making behavior (Li et al. 2015). Studying these aspects under the unique socioeconomic and cultural background of China is particularly timely, especially for the residents of flood-prone regions in China. Therefore, the current study selects two flood-prone villages along the Fujiang River, Dazhou Qujiang River, and Nanchong Jialing River in Mianyang, Sichuan Province

348

13 The Experience and Attitude of Rural …

as research sample villages. It takes local residents as the research object to explore the effects of flood experience and attitude of rural residents in flood-prone areas on disaster preparedness behavior.

13.2 Research Methodology 13.2.1 Sample Selection and Data Collection 13.2.1.1

Study Setting and Sample Selection

On the basis of the flood statistical distribution of China from 2000 to 2015 in the emergency events database (Em-dat 2017), the Sichuan Region can be seen as a frequent flood disaster area. Sichuan Province is located in the southwest of China, with complex and diverse topography, and is one of the provinces with the most significant relief (Chen et al. 2010). The areas in Sichuan with frequent rainstorm are concentrated in Ya’an and Emei areas in the west; the areas near Mianyang and Dazhou, Nanchong; and Bazhong areas (Zhou et al. 2011). The Yangtze River system is the main river system in Sichuan Province, which is one of the reasons for the frequent occurrences of flood disasters in Sichuan. Particularly, the eastern Sichuan Basin belongs to the areas with high risk of rainstorm and flood, high sensitivity, and high vulnerability (Du and Dong 2016). From 1985 to 2009, Dazhou, Mianyang, and Nanchong ranked the top three in the annual direct economic losses caused by rainstorm and flood in Sichuan, with 485, 434, and 218 million yuan, respectively (Qing et al. 2013). The affected population of Sichuan Province caused by rainstorm and flood shows an increasing trend; Mianyang, the rainstorm center of Sichuan Province, will be further affected by rainstorm and flood (Deng et al. 2016). In addition, the distribution of annual precipitation in Jialing River Basin is quite different. The annual average precipitation in Sichuan is distributed in the northeast of Sichuan (Du et al. 2015). Mianyang, Dazhou, and Nanchong are three places with large exposures to flood disasters, all of which have experienced serious flood disasters. For example, in 2004, “9.3” Dazhou suffered a 200-year flood, with a direct economic loss of 6.1 billion yuan. In 2008, “7.11” Mianyang experienced a 70-year flood, with a direct economic loss of 1.853 billion yuan. In 1981, “7.9” Nanchong experienced an 80year rare flood, with an economic loss of 43 million yuan. Therefore, Mianyang, Dazhou, and Nanchong are the typical flood disaster areas in Sichuan Province, among which the rural areas along the river are highly affected by flood disasters with a long duration. Therefore, the selection of sample villages follows the principle of randomly selecting two villages along the river in each city. Finally, six sample villages are determined, namely, Xingxiang, Fucheng, Baoshamiao, Doupengzhai, Shizi, and Xikou in Mianyang, Nanchong, and Dazhou Cities. Figure 13.1 illustrates

13.2 Research Methodology

349

Fig. 13.1 The geographical location distribution of sample villages

the geographical location distribution of the sample villages. Particularly, Pengjiaxiang and Fucheng Villages are close to Fujiang River; Baosha Temple and Doupeng Village are close to Jialing River; and Shizi Village and Xiejiaba are close to Qujiang River.

13.2.1.2

Data Collection

Data collection is divided into two stages, namely, preparation and formal implementation. In the preparation stage, the research team recruited students from the sample village as investigators for the construction management majors of Chengdu University of Technology. Through face-to-face explanation of the research purpose, content, and travel arrangements, the students with strong willingness to participate were finally determined as formal investigators, with a total of 17, including 8 undergraduate students and 9 graduate students. The research group held a formal training meeting on January 6, 2019 to train the researchers and explain all the contents of the questionnaire. On the basis of the regional location of the sample village, the researchers were divided into three groups, with five to six people in each group, to complete the questionnaire survey of entering the village and entering the household in the corresponding area. From January 15 to 17, 2019, the research group formally conducted the questionnaire survey of entering the villages. After each group members arrived at the survey village, two people were assigned to a survey implementation team. Each team, with the village committee (village center) as the point of contact, performed a survey of local households across the four regions; north, south, east and west. That is, local households within the four regions were randomly selected for oneon-one, face-to-face questionnaire surveys. If the residents in the randomly selected households declined to participate, the implementation team moved on to the next household. It was the head of each household who was interviewed. To encourage

350

13 The Experience and Attitude of Rural …

Table 13.1 Number of households surveyed Total number of households

Effective number of households

n (Number of households to be sampled)

Sampling rate (%)

Xingxiang

726

42

42

5.79

Fucheng

1155

46

43

4.36

Baoshamiao

483

59

41

12.21

Doupengzhai

854

58

42

6.79

Shizicun

1080

60

43

5.55

Xikou

760

60

42

7.89

the participation of local residents, the research group prepared daily necessities as a gift for the residents who completed the survey. On the basis of Yamane’s (1967) formula, the lowest sampling number was determined for each village. Finally, 42, 46, 59, 58, 60, and 60 questionnaires were collected from the villages of Xingxiang, Fucheng, Baoshamiao, Doupengzhai, Shizi, and Xikou, respectively, totaling 325. A total of 360 questionnaires were distributed, with an effective recovery rate of 90.28%. Sampling information is shown in Table 13.1. n=

N 1 + N e2

(13.1)

where n: number of households in the village that should be taken; N: total number of households in the village; e: precision is set to 15% (0.15; (Ullah et al. 2015).

13.2.2 Model Specification The study measures the behavior of disaster preparedness by the number of disaster preparedness measures, which is orderly. Generally, probit is suitable for the data obtained in the planned experiments, whereas the ordinal logit regression is suitable for the direct observation data (Liyan 2010). Finally, the ordinal logit (ordered regression analysis) model was selected for data fitting. The specific description of the model is as follows. Suppose that k classes exist for the ordered multiclass variables (explained variables). If the probability of each class is expressed as π 1, π 2, π 3 … π K, then π 1 + π 2 + . . . + π k = 1. The influence of P different factors (the explanatory variable is recorded as x) on the probability of each category of the explanatory variable could be analyzed from the following perspectives. A k-1 model (Xue 2013) was established as follows:

13.2 Research Methodology

351

−ln[−ln(π 1)] = α1 + β1 x1 + . . . + β p x p ,

(13.2)

−ln[− ln(π 1 + π 2)] = α2 + β1 x1 + . . . + β p x p ,

(13.3)

… −ln{− ln[π 1 + π 2 + . . . + π(k − 1)]} = αk−1 + β1 x1 + . . . + β p x p ,

(13.4)

where α is the threshold (constant term), and β is the position parameter (regression coefficient). The connection function in the model used the negative log-log model because the low category probability of the explained variable is relatively high.

13.2.3 Variable Specification 13.2.3.1

Dependent Variable

The dependent variable in this study is the respondents’ flood preparedness behavior. A total of 14 parameters regarding flood preparedness measures, drawn from the literature, were listed for the respondents to consider. See Table 13.2. 73 respondents did not select any disaster preparedness measures, whereas the second and the fourteenth items were the most frequently selected, at 145 and 144 respondents, respectively. See Fig. 13.2. In addition, this study counted the number of disaster preparedness measures taken by each family to reflect the level of flood preparedness measures taken by respondents’ household. The frequency of the number of preparedness measures is shown in Fig. 13.3. On the basis of the results of the respondents’ choice of disaster preparedness measures, the measures taken were arranged into five levels. The fist level was defined as very poor flood disaster preparedness behavior (coded as 1), which indicates that the households only took 0–2 flood preparedness measures. The second, third, fourth, and fifth disaster preparedness levels were defined as relatively poor (coded as 2), general (coded as 3), relatively good (coded as 4), and very good (coded as 5), respectively. This indicated that the households undertook 3–5, 6–8, 9–11, and 12–14 flood preparedness measures, respectively. Most respondents (46.5%) adopted very poor preparedness measures, ranging from 0 to 2, whereas very few respondents were well prepared; only 2.42% of the respondents had 12–14 disaster preparedness measures. The proportion of the five levels of flood disaster preparedness is shown in Fig. 13.3.

352

13 The Experience and Attitude of Rural …

Table 13.2 Respondents’ selection according to the actual situation of daily flood control preparation Item

Disaster preparedness measures

Reference

Item 1

Flood insurance is purchased

(Soane et al. 2010; Thieken et al. 2007; Xu et al. 2018)

Item 2

Flashlight is placed conveniently to deal with sudden floods

(Soane et al. 2010; Xu et al. 2018)

Item 3

The radio is kept in a convenient place to listen to latest flood news at any time

(Miceli et al. 2008; Soans 2011)

Item 4

Emergency telephone numbers are remembered consciously

(Miceli et al. 2008)

Item 5

At ordinary times, relatives are taught consciously how to deal with floods

(Miceli et al. 2008)

Item 6

Participated in flood emergency training

(Soane et al. 2010; Xu et al. 2018)

Item 7

Ask others (the government) about the emergency response to flood

(Ejeta et al. 2018)

Item 8

Important items are usually stored at home (Wang 2013; Danso and Addo 2017) in a higher place in case of flood

Item 9

Emergency food and drinking water are stored for emergency use in case of flood

(Soane et al. 2010; Xu et al. 2018)

Item 10

The height of the house is increased to prevent flood

(Wang 2013)

Item 11

The plating date of crops is changed in case of flood

(Terpstra 2011; Schad et al. 2012)

Item 12

Agricultural irrigation measures are added (Terpstra 2011; Schad et al. 2012) to prevent flood

Item 13

Soil protection technology is used to prevent the adverse effect of flood on soil

(Terpstra 2011; Schad et al. 2012)

Item 14

Drainage measures are improved to prevent flood

(Wang 2013)

13.2.3.2

Control Variables

On the basis of the existing research, the control variables considered in this study included age (Thieken et al. 2007; Soane et al. 2010; Xu et al. 2018), gender (Soane et al. 2010; Xu et al. 2018), education (Thieken et al. 2007; Soane et al. 2010; Xu et al. 2018), personal income (yuan; Thieken et al. 2007), family income (wan yuan; Soane et al. 2010; Xu et al. 2018), family members with income (Atreya et al. 2017), number of families (Thieken et al. 2007; Soane et al. 2010; Xu et al. 2018), family illness (Saqib et al. 2016), living years (Adelekan and Asiyanbi 2015; Lokonon 2016), building area, number of floors of houses (Samaddar et al. 2014), family drainage ditch (Danso and Addo 2017), and perceived distance between home and river (Saqib et al. 2016; Xu et al. 2018). Table 13.3 shows the statistical information of control variables.

13.2 Research Methodology

353

160 145

144

140

121

120 100 80

134

130

129

102

96

92

80

73

60

60

55

49 42

40 20 0 0

the 1st the 2nd the 3rd the 4th the 5th the 6th the 7th the 8th the 9th the item item item item item item item item item 10th item

the 11th item

the 12th item

the 13th item

0.24%

0.24%

the 14th item

Fig. 13.2 Choose frequency for each item 20% 18%

17.68% 16.46%

16% 13.80%

14%

12.35%

12%

10.65%

10% 8%

7.02% 5.57%

6%

5.81%

4%

2.91%

2.42% 1.69%

2%

1.94% 1.21%

0% 0

1 item 2 items3 items4 items5 items6 items7 items8 items9 items 1-level

2-level

3-level

10 11 12 13 14 items items items items items 4-level

5-level

Fig. 13.3 The proportion of flood preparedness level

13.2.3.3

Explanatory Variables

The explanatory variables included the respondents’ flood experience and attitudes toward flood disaster. On the basis of the existing literature, the current study designed 11 questions about flood experience and 34 questions about flood attitude. In addition, the experience of flood days and the number of flood-related deaths were objective and need not be included in the factor analysis (Tables 13.4 and 13.5). Exploratory factor analysis was used to reduce the dimension of explanatory variables. Through the reliability analysis of the aforementioned flood experience and flood attitude, the Cronbach’s α values based on standardization terms came in at 0.766 and 0.853, respectively. The value between 0.7 and 0.8 indicated that factor analysis could be

354

13 The Experience and Attitude of Rural …

Table 13.3 Basic information of respondents Indicator

Group Option

Percentage Indicator Group Option

Percentage

Gender

1

45.80

2

Female

54.20

Education

1

uneducated

26.20

2

primary

40

3

junior

28.30

4

Senior and above

5.60

Less than 500

42.6

personal 1 income (yuan/Month) 2

family income (yuan/Year)

building area

age

Man

ditch for family felt the distance between the home and the river living years

1

Yes(1)

68.80

2

No(2)

31.20

1

Less 0.5 km

43.10

2

0.5–1 km 40.00

3

1–1.5 km 3.30

4

More 1.5 km

1

Less than 32.90 10

13.60

500–1,500

32.9

2

10–20

42.20

3

More than 1,500

24.5

3

More than 20

24.90

1

Less than 10,000

37.3

Family 1 members

2

10,000–20,000 26.1

2

4–6

59.30

3

More than 20,000

36.6

3

More than 6

22.50

1

Less than 100 m2

36.60

1

1

23.50

2

100–200 m2

48.10

2

2

62.90

3

More than 200 m2

15.30

3

More than 3

13.60

1

≤30

2.70

1

yes

47.90

2

31–40

3.60

2

No

52.10

3

41–50

23.70

0–1

25.40

4

51–60

21.10

2

39.70

5

>60

48.90

Family 1 members 2 with 3 income

More than 3

34.90

number of floors of houses

family illness

Less than 18.20 3

performed with good reliability. The attitude of factor analysis showed that the KMO value was 0.825, with originality P = 0.000, KMO value of experience = 0.704, and significance P = 0.000. It also proved its suitability for factor analysis. In the process of factor analysis, items with load factor less than 0.4 in rotation matrix were deleted, and 12 (3 experienced and 9 attitudes) common factors were finally obtained. The common factors of attitude and experience accounted for 73.01% and 67.98% of the index, respectively. Tables 13.4 and 13.5 show the results of factor analysis with improved interpretation results. Five variables about flood experience were identified, namely, experience of flood days (Adelekan and Asiyanbi 2015), experience of flood-related deaths (Adelekan

13.2 Research Methodology

355

Table 13.4 Empirical component analysis of flood disaster: Rotated component matrix Questions

References

Component 1

2

3

Have you ever been helped by the community (village) during the flood?

(Lu et al. 2016, 2017)

0.859

0.003

−0.019

Have you ever been helped by your neighbors during the flood?

(Lu et al. 2016, 2017)

0.859

0.075

−0.03

Have you ever been helped by the (Lu et al. 2016, 2017) community (village) after the flood?

0.84

0.089

0.059

Have you ever been helped by your neighbors after the flood?

(Lu et al. 2016, 2017)

0.856

0.106

−0.031

Is the residential environment naturally restored after the flood?

(Lu et al. 2016, 2017)

/

/

/

Was the water supply interrupted during the flood?

(Lu et al. 2016, 2017)

0.024

0.822

0.232

Was the power supply interrupted during the flood?

(Lu et al. 2016, 2017)

0.094

0.808

0.018

Was the supply of natural gas interrupted during floods?

(Lu et al. 2016, 2017)

0.156

0.792

0.118

Was communication interrupted during floods?

(Lu et al. 2016, 2017)

−0.001

0.807

0.172

Was the residence completely surrounded by flood?

(Lu et al. 2016, 2017)

−0.02

0.134

0.86

Was the road destroyed due to the flood?

(Lu et al. 2016, 2017)

0.03

0.214

0.607

Has your own home been seriously damaged by the flood?

(Ejeta et al. 2018)

−0.032

0.055

0.786

Characteristic value

2.950

2.699

1.828

Variance (%)

26.818

24.537

16.622

Cumulative variance percentage (%)

26.818

51.355

67.977

Extraction method: Principal component analysis. Rotation method: Caesar normal maximum variance method. The rotation converged after five iterations 1: Experience of being helped during flood; 2: Experience of infrastructure damaged by flood; 3: Experience of housing affected by flood

and Asiyanbi 2015), experience of being helped during flood, experience of infrastructure damaged by flood, and experience of housing affected by flood (Ejeta et al. 2018). The first two variables belonged to objective data and were excluded in the exploratory factor analysis; however, the last three variables belonged to subjective data and could also be included in the common factors after the exploratory factor analysis.

0.829

0.727

(Terpstra 2011)

(Lokonon 2016)

(Terpstra 2011)

(Miceli et al. 2008)

Would you like to collect some information about disaster preparedness?

Would you like to buy flood insurance?

Would you like to put your flashlight and radio in a convenient place to deal with the flood?

Would you like to provide the emergency rescue telephone number of flood disaster?

Would you like to teach or (Miceli et al. 2008) arrange your relatives how to deal with flood emergencies?

Would you like to attend the flood emergency training?

Would you like to ask (Ejeta et al. 2018; someone/government what Miceli et al. 2008) measures should be taken in case of flood emergency?

(Charalambous et al. 2018)

0.713

(Terpstra 2011)

Would you like to know about flood prevention?

0.811

0.654

0.465

0.689

0.715

1

0.033

0.09

0.089

0.148

0.019

0.135

0.098

0.05

0.029

0.011

−0.018 0.035

0.086

−0.057 0.183

0.019

0.152

−0.018 −0.104 0.039

0.061

0.054

0.086

0.053

(continued)

−0.098

−0.092

−0.045

−0.054

0.19

0.361

−0.183 −0.012 0.349

−0.067 0.045

0.046

0.074

9

−0.051 0.18

8

−0.062 0.056

0.015

0.174

0.046

0.13

0.069

−0.053 0.27 0.026

7

−0.073 −0.13

6

−0.371 −0.041 0.206

0.017

5

−0.126 0.129

0.022

4

−0.227 0.132

0.189

3

−0.057 0.108

0.117

2

Component

References

Questions

Table 13.5 Composition matrix of flood attitude composition analysis after rotation

356 13 The Experience and Attitude of Rural …

(Wang 2013; Danso and Addo 0.717 2017)

(Terpstra 2011)

(Miceli et al. 2008)

(Miceli et al. 2008)

(Miceli et al. 2008)

(Miceli et al. 2008)

(Miceli et al. 2008)

(Miceli et al. 2008)

Would you like to store your important items in a safe place?

Would you like to store emergency food and water in case of flood?

You think flooding here is possible in the next five years.

The road is assumed to be destroyed by flood in the next five years.

The water supply is assumed to be interrupted if flood occurs in the next five years

The power supply is assumed to be interrupted due to flood in the next five years

The natural gas is assumed to be interrupted due to flood in the next five years

Assume the possibility of communication interruption due to floods in the next five years

0.158

0.121

0.151

0.174

0.836

0.891

0.845

0.009

0.089

−0.044 0.079

0.12

0.145

0.038

(continued)

−0.155

−0.061 −0.025 0.121 0.041

−0.05

0

−0.022 −0.063 0.033

0.185

0.167

−0.173

−0.063

−0.064 0.412

0.006

0.015

−0.047 −0.047 0.09

0.002

9

−0.136 −0.142

8

−0.017 −0.063 −0.064 0.105

0.056

0.027

−0.018 0.063

0.017

0.04

0.888

7

−0.084 −0.081 0.074

0.071

−0.033 0.18

0.18

6

−0.125 −0.053 0.221

5

−0.008 0.625

0.124

−0.014 0.055

0.178

4

0.104

0.053

0.162

3

−0.221 0.629

0.731

1

2

Component

References

Questions

Table 13.5 (continued)

13.2 Research Methodology 357

0.191 0.106

(Ejeta et al. 2018)

(Ejeta et al. 2018)

Do you believe in some news about the flood on TV?

Do you believe in some news about the flood on the radio?

Do you believe in some news (Ejeta et al. 2018) about the flood in the newspaper?

Do you feel fear when you encounter floods?

0.21 0.015

Will you be alert in case of flood? (Terpstra 2011)

(Reynaud et al. 2013)

(Reynaud et al., 2013)

(Terpstra 2011; Osuret et al. 2016)

Is there a flood dike in the river near your home?

Can the flood control dike near your home resist floods?

Do you believe that the local flood protection facilities have been maintained?

0.219

3

0.032

0.228

0.203

0.191

0.054

0.024

0.007

0.017

0.011

0.961

0.057

−0.062 0.141

0.032

0.001 −0.042 −0.012

0

−0.227 0.304

0.088

(continued)

−0.061 −0.101 −0.069

−0.062 0.11

−0.027 0.053

0.047

−0.126 0.728

−0.048 −0.022 0.943

0.026

0.67

−0.023 −0.019 −0.108 0.26

−0.067 0.83

0.02

−0.027 0.001 −0.147 0.182

0.037

−0.009 −0.026 0.094

−0.014 0.001

−0.035 0.058

0.211

9

−0.033 −0.017

8

−0.007 −0.051 −0.018 0.04

0.024

0.118

0.091

7

0.122

0.008

6

−0.029 0.113

5

−0.053 0.007

0.045

4

−0.048 0.858

0.916

0.936

0.926

−0.027 0.894

−0.003 0.015

0.022

0.165

0.179

Do you feel helpless when facing (Terpstra 2011) floods?

(Teun et al. 2009; Terpstra 2011)

0.179

(Ejeta et al. 2018)

Do you believe in some news about flood reported on the Internet?

1

2

Component

References

Questions

Table 13.5 (continued)

358 13 The Experience and Attitude of Rural …

/ 5.512 16.213

Do you care when a flood occurs? (Terpstra 2011)

Characteristic value

Variance (%)

12.829

4.362

/

11.543

3.925

/

7.199

2.448

/

5.777

1.964

/

5.612

1.908

/

5.048

1.716

/

0.002

4.725

1.607

/

0

−0.107 0.817

(continued)

4.068

1.383

/

0.842

0.027

−0.078

−0.131 0.017

−0.017 0.056

0.014

−0.115 −0.005 −0.071 0.247

(Terpstra 2011)

0.233

Do you get angry when a flood occurs?

0.02

−0.058 0.049

0.271

0.45

0.635

−0.077 −0.149 0.794

−0.04

−0.103 0.083

0.14

0.052

0.017

−0.137 0.526

Suppose that you or some of your (Miceli et al. 2008) relatives will be injured by floods in the next five years

−0.046 0.005

−0.034 0.124

−0.188 0.022

0.183

0.167

0.054

−0.064 0.005

−0.069 0.015

0.147

9

0.176

0.156

(Levavasseur et al. 2012)

Is the ditch near your house leaking?

8

−0.121 0.085

7

−0.067 −0.119 −0.046 −0.264 −0.104 −0.041 0.753

(Miceli et al. 2008)

(Levavasseur et al. 2012)

Can the ditch near your home drain smoothly?

−0.075 −0.015 0.756

−0.152 0.351

0.081

Assume that your home will be severely damaged or destroyed by floods in the next five years

(Ejeta et al. 2018; Terpstra 2011)

Do you have confidence in the ability of the village’s flood risk manager (village head, Secretary)?

0.771

−0.041 0.047

6

−0.165 0.326

5

−0.02

4

0.119

(Terpstra 2011)

Do you believe in the flood control power of your village?

3

2

1

Component

Can the ditch in your family play (Levavasseur et al. 2012) a role in flood control?

References

Questions

Table 13.5 (continued)

13.2 Research Methodology 359

References 16.213

1 29.043

2

Component 40.586

3 47.784

4 53.561

5 59.173

6 64.221

7 68.946

8

73.014

9

Extraction method: Principal component analysis. Rotation method: Caesar normal maximum variance method. The rotation converged after seven iterations 1: Willingness to prevent floods; 2: Possibility of future flood damage to infrastructure; 3: Trust degree of disaster report in media; 4: Fear and helplessness of floods; 5: Cognition of dike function; 6: Trust in flood control capacity of village; 7: Cognition of ditch function; 8: Possibility of future flood damage to home; 9: Anger against floods

Cumulative variance percentage (%)

Questions

Table 13.5 (continued)

360 13 The Experience and Attitude of Rural …

13.2 Research Methodology

361

Nine variables about flood attitude were identified, namely, the possibility of future flood damage to home (Thieken et al. 2007; Ejeta et al. 2018); the possibility of future flood damage to infrastructure, the cognition of ditch function, the cognition of dike function, the trust degree of disaster report in media (Ejeta et al. 2018), trust in flood control capacity of village, willingness to prevent floods (Teunterpstra and Guttering 2008), fear and helplessness of floods (Terpstra 2011; Schad et al. 2012), and anger against floods (Terpstra 2011; Schad et al. 2012). Twelve common factors and two objective empirical variables from the factor analysis were placed into the ordinal logit model for further analysis.

13.3 Results and Discussion After controlling for the basic information of respondents, all the above variables are put into the ordinal logit model (Table 13.6) to explore the effects of the respondents’ flood experience and flood attitude on their disaster preparedness behavior. The log likelihood of DE-2 is 976.154, the current model is 769.699, the likelihood ratio chi square is 206.454, and the probability p value is 0.000. Thus, the linear relationship between all explanatory variables and the connection function is significant, and the model selection is correct. In addition, the values of Cox and Snell R2 , Nagelkerke R2 , and McFadden R2 are 0.434, 0.393, and 0.211 respectively. The goodness of fit indices in this study are all higher than those in associated literature (Qin et al. 2020; Ruan 2019). Therefore, the data fit the model well, and the model has sufficient explanatory power regarding disaster preparedness behavior. Table 13.6 presents the results of the ordinal logit regression. A total of 28 independent variables (14 classified variables and 14 continuous variables) are involved in the results, among which 17 variables (9 positive and 8 negative) have a significant effect on disaster preparedness.

13.3.1 Effects of Control Variables on Disaster Preparedness From the perspective of socioeconomic characteristics, age groups 3 and 4 have significant positive effects on disaster preparedness behavior, compared with age group 5 (b = 0.778, P = 0.002; b = 0.5, P = 0.021). Whereas age groups 1 and 2 have no significant effect (b = − 0.066, P = 0.928; B = 0.234, P = 0.659), indicating that residents aged 40–60 are more likely to be prepared for disasters than older residents. This result is consistent with the conclusion of Soane et al. (2010). In comparison with group 4, education level groups 1, 2, and 3 have a significant negative effect on disaster preparedness behavior (b = − 2.538, P = 0.000; b = − 3.151, P = 0.000; b = − 2.624, P = 0.000), which indicates that respondents with a low education level are less inclined to take on disaster preparedness behavior. This

362

13 The Experience and Attitude of Rural …

Table 13.6 Results of ordinal logit regression Estimate

threshold

location

Std. Error

Wald

Df

Sig.

95% Confidence interval Lower bound

Upper bound

very poor preparedness

−0.446

0.847

0.276

1

0.599

−2.107

1.215

poor preparedness

1.001

0.848

1.394

1

0.238

−.661

2.663

general disaster preparedness behavior

2.768

0.847

10.694

1

0.001

1.109

4.427

better preparedness

3.690

0.85

17.795

1

0.000

1.976

5.405

experience of flood days

0.129

0.079

2.644

1

0.104

−0.026

0.284

experience of flood-related deaths

0.855

0.339

6.372

1

0.012

0.191

1.519

experience of being helped during flood

−0.020

0.090

0.051

1

0.821

−0.198

0.157

experience of infrastructure damaged by flood

−0.055

0.100

.304

1

0.581

−0.250

0.140

experience of housing affected by flood

−0.181

0.084

4.610

1

0.032

−0.347

-0.016

possibility of future flood damage to home

−0.198

0.085

5.438

1

0.020

−0.364

−0.032

possibility of future flood damage to infrastructure

−0.239

0.096

6.187

1

0.013

−0.427

−0.051

cognition of ditch function

−0.108

0.088

1.487

1

0.223

−0.280

0.065

cognition of dike function

−0.6

0.082

12.149

1

0.000

−0.447

−0.125

trust degree of 0.620 disaster report in media

0.093

44.434

1

0.000

0.438

0.803

(continued)

13.3 Results and Discussion

363

Table 13.6 (continued) Estimate

Std. Error

Wald

Df

Sig.

95% Confidence interval Lower bound

Upper bound

trust in flood control capacity of village

0.284

0.097

8.639

1

0.003

0.095

0.473

willingness to prevent floods

0.197

0.096

4.249

1

0.039

0.010

0.385

fear and helplessness of floods

−0.259

0.095

7.431

1

0.006

−0.445

−0.073

anger against floods

0.203

0.083

6.004

1

0.014

0.041

0.365

[gender = 1]

0.020

0.170

0.013

1

0.908

−0.313

0.352

[gender = 2]

0a

[education = 1.00]

−2.538

0.502

25.563

1

0.000

−3.52−

−1.554

[education = 2.00]

−3.151

0.484

42.428

1

0.000

−4.100

−2.203

[education = 3.00]

−2.624

0.455

33.287

1

0.000

−3.515

−1.732

[education = 4.00]

0a

0

0

[occupation = −1.835 1.00]

0.686

7.160

1

0.007

−3.179

−0.491

[occupation = −1.272 3.00]

1.211

1.103

1

0.294

−3.645

1.102

[occupation = −1.061 4.00]

0.772

1.885

1

0.170

−2.575

0.453

[occupation = 0.237 5.00]

0.469

0.256

1

0.613

−0.682

1.156

[occupation = 0.364 6.00]

0.357

1.040

1

0.308

−0.336

1.064

0.034

0.028

0.737

[occupation = 0a 7.00] [family illness 0.383 = 1] [family illness 0a = 2]

0 0.181

4.484

1 0

(continued)

364

13 The Experience and Attitude of Rural …

Table 13.6 (continued) Estimate

Std. Error

0.205

Wald

2.040

Df

1

Sig.

95% Confidence interval Lower bound

Upper bound

0.153

−0.109

0.695

[ditch for family = 1]

0.293

[ditch for family = 2]

0a

[family income = 1.00]

−0.027

0.243

0.013

1

0.911

−0.503

0.44

[family income = 2.00]

−0.157

0.236

0.446

1

0.504

−0.619

0.304

[family income = 3.00]

0a

[personal income = 1.00]

0.566

0.266

4.546

1

0.033

0.046

1.087

[personal income = 2.00]

0.122

0.266

0.209

1

0.647

−0.400

0.644

[personal income = 3.00]

0a

[felt the distance between the home and the river = 1.00]

−0.162

0.255

0.402

1

0.526

−0.661

0.338

[felt the distance between the home and the river = 2.00]

−0.497

0.255

3.788

1

0.052

−0.997

0.004

[felt the distance between the home and the river = 3.00]

−0.236

0.471

0.251

1

0.616

−1.158

0.687

0

0

0

(continued)

13.3 Results and Discussion

365

Table 13.6 (continued) Estimate

Std. Error

Wald

Df

Sig.

95% Confidence interval Lower bound

Upper bound

[felt the distance between the home and the river = 4.00]

0a

[living years = 1.00]

−0.598

0.224

7.119

1

0.008

−1.037

−0.159

[living years = 2.00]

−0.291

0.205

2.019

1

0.155

−0.693

0.111

[living years = 3.00]

0a

[family members = 1.00]

−0.065

0.302

0.047

1

0.829

−0.658

0.527

[family members = 2.00]

−0.203

0.224

0.818

1

0.366

−0.642

0.237

[family members = 3.00]

0a

[building area = 1.00]

0.397

0.274

2.093

1

0.148

−0.141

0.934

[building area = 2.00]

−0.035

0.254

0.019

1

0.892

−0.533

0.464

[building area = 3.00]

0a

[number of floors of houses = 1.00]

−0.461

0.292

2.498

1

0.114

−1.032

0.111

[number of floors of houses = 2.00]

−0.024

0.249

0.009

1

0.923

−0.513

0.464

[number of floors of houses = 3.00]

0a

0

0

0

0

0

(continued)

366

13 The Experience and Attitude of Rural …

Table 13.6 (continued) Estimate

Std. Error

Wald

Df

Sig.

95% Confidence interval Lower bound

Upper bound

[age = 1.00]

−0.066

0.738

0.008

1

0.928

−1.512

1.380

[age = 2.00]

0.234

0.530

0.195

1

0.659

−0.805

1.273

[age = 3.00]

0.778

0.255

9.337

1

0.002

0.279

1.277

[age = 4.00]

0.500

0.217

5.302

1

0.021

0.074

0.926

[age = 5.00]

0a

[family members with income = 1.00]

0.58

0.240

5.219

1

0.022

0.078

1.018

[family members with income = 2.00]

0.217

0.204

1.134

1

0.287

−0.183

0.617

[family members with income = 3.00]

0a

0

0

Correlation function: Negative log-log logarithm. a This

parameter is redundant; thus, it is set to zero

is consistent with relevant research results (Hoffmann and Muttarak 2017; Sooan et al. 2010; Xu et al. 2018). In comparison with individual income group 3, personal income group 1 (b = 0.556, P = 0.033) has a significant positive effect on daily disaster preparedness behavior, indicating that residents with low personal income are more likely to take on disaster preparation. In addition, the smaller the number of family members with income in a household is, the more flood preparedness measures people undertake (b = 0.548, P = 0.022). Again, this is consistent with the research result of Atreya et al. (2017) who found that socioeconomic conditions are an effective predictor of disaster preparedness. In terms of family members’ health, those experiencing illness among family members have a significant positive tendency to be disaster prepared (b = 0.383, P = 0.034). As shown in the study by Ablah et al. (2009), families with a disabled member or someone having health problems, are more likely to be disaster prepared. In comparison with age group 3, age group 1 shows a significant negative result (b = − 0.598, P = 0.008), whereas age group 2 shows no significant impact on disaster preparedness (b = − 0.291, P = 0.155). Thus, residents in age group 1 are less inclined to be disaster prepared than those of age group 3. This is consistent with the findings of Hoffmann and Muttarak (2017). Usually, people will have attachment to the place they have lived for a long time; thus, residents will have a tendency to prepare when they encounter disasters.

13.3 Results and Discussion

367

The distance between home and river in this area shows an insignificant effect, which is inconsistent with the conclusion of Baker (2011) and Xu et al. (2018). The main reason is that the region belongs to high mountain and hilly area. Although it is close to the river, given the influence of terrain elevation difference, the close distance may not necessarily encounter flood disaster; thus, it shows an insignificant effect.

13.3.2 Effects of Experiences on Disaster Preparedness Terpstra (2011), Schad et al. (2012), Lawrence et al. (2014), Atreya et al. (2017), Hoffmann and Muttarak (2017), and Ejeta et al. (2018) regarded previous flood experience as the main driving force catalyzing disaster preparedness and the taking of preventive measures. Consistency can also be seen in the simulation results. The experience of flood-related death (b = 0.855, P = 0.012) has a significant positive effect on disaster preparedness behavior. An increase in the number of deaths resulting from flood disaster will motivate residents to take on disaster preparedness measures. Therefore, floods and their impact on safety and threat to life, result in a very significant impact on the daily flood preparedness behavior of residents. Contrariwise however, residences with direct personal experience of flood disasters showed a significant negative effect regarding their flood disaster preparedness (B = − 0.181, P = 0.032). Where respondents previously experienced damage to their homes, but without adverse personal, physical impact to their health and safety, they were more resigned and took fewer precautions. Therefore, not all flood experiences can promote people to actively prepare for disasters. Only after experiencing the flood whose life safety is threatened will people take active measures to prepare for a disaster in their daily life.

13.3.3 Effects of Attitudes on Disaster Preparedness The influence of residents’ flood attitude on their daily disaster preparedness behavior is complex. Specifically, the more prone residents are to perceive damage to their homes (b = − 0.198, P = 0.020) and infrastructure (b = − 0.239, P = 0.013), the less likely they are to take disaster preparedness measures as part of their daily routine. This finding is inconsistent with those of Scolobig et al. (2012), Reynaud et al. (2013), Hoffmann and Muttarak (2017), Yong and Lemyre (2019). On exploring this finding with interviewees, it turns out that they assessed themselves as living in a chronically flood-prone area, where, despite taking precautionary measures, flooding related losses were deemed unavoidable. Thus, even where residents anticipate that floods will occur, they remain uninclined to take disaster preparedness measures. This result is consistent with the effect of respondents’ previous experience on flood preparedness, resulting in a significant negative effect of functional cognition (b = − 0.286, P = 0.000). That is, the higher cognition of the function of dam reduces

368

13 The Experience and Attitude of Rural …

the likelihood of the residents to take the measures of disaster preparedness. This finding is consistent with the conclusion of Lin et al. (2008). In view of the existence of the “dike effect,” residents tend to rely on the structural protection and relinquish daily disaster preparedness measures. The results show that the degree of trust held by rural residents in flood-prone areas on the media (b = 0.620, P = 0.000) and their confidence in the flood control capacity of villages (b = 0.284, P = 0.003) has a significant positive effect on their daily disaster preparedness behavior. That is, the higher their trust in the reliability of flood reporting and the capacity of villages, the more likely they are to adopt disaster preparedness measures. This finding fully shows that as long as the residents think that the flood remains controllable, they will actively take disaster preparedness measures to reduce flood losses. In the study of Samaddare et al. (2012), residents have high trust in the media and government institutions and high acceptance rate of early warning and preventive measures. Terpstra (2011) reported that trust is a factor that has a significant effect on people’s potential preparedness measures. For residents’ willingness to prevent disasters, Table 13.6 shows that residents’ willingness to prevent disasters has a significant positive effect on their daily preparedness behavior (b = 0.197, P = 0.039). That is, the stronger the residents’ willingness to prevent disasters is, the more likely they are to take preparedness measures in their daily life. Terpstra (2011) claimed that the willingness of coastal residents to prepare for disasters is also positively related to the situation of disaster preparedness. The residents’ feelings of fear and helplessness (b = − 0.259, P = 0.006) have a significant negative influence on their daily disaster preparedness behavior. That is, the more residents remain uncertain regarding floods, the less inclined they are to take on disaster preparedness measures. However, where residents’ emotion is one of anger with regard to flooding, the effect on disaster preparedness is positive and significant (b = 0.203, P = 0.014). That is, with an increase in anger (not to the degree of fear and helplessness), the more residents will take disaster preparedness measures. Ejeta et al. (2018) showed that emotion can affect disaster preparedness behavior; especially in the negative state, residents’ preparation intention is lower, resulting in lower disaster preparedness measures. However, when the residents are united, more disaster prevention measures will be taken. Usually, when people feel that the disaster is out of control, they leave it to fate and do not take disaster preparedness measures.

13.4 Conclusion and Recommendations By using a field survey method, data collection on flood preparedness was performed in the following Chinese villages, Sichuan Province: Xingxiang, Fucheng, Baoshamiao, Doupengzhai, Shizi, and Xikou Villages of Dazhou. On the basis of the control of demographic information and environmental characteristics, the relationship between flood experience and attitude on residents’ daily disaster preparedness

13.4 Conclusion and Recommendations

369

behavior was explored using exploratory factor analysis and ordinal logit regression. The results show that the residents’ social demographic characteristics, living environment, flood experience, and attitude in relation to flooding, significantly affects daily disaster preparedness behavior. First, the disaster preparedness behavior of rural residents remains at a relatively low level, and the vast majority (67%) of rural residents in southwest China undertake no disaster preparation (Danso and Addo 2017). The conclusion of this study is similar to that of Ning et al.’s (2013) claim that rural residents’ disaster preparedness knowledge, behavior, and stock of emergency supplies, is unacceptably poor. Disaster preparedness measures adopted by the family unit are simple, and include having a flashlight handy, storing valuable articles at height, increasing the floor level of ground-floor residences, and improving drainage measures. Second, the number of people in a household receiving an income, as well as the amount of income respondents receive, along with any incidents of family illness, all significantly negatively affect flood preparedness. On the other hand, education level, along with years of residence within the local area, have a significant positive effect on residents’ preparedness behavior. Third, the flood experience and attitude of rural residents have a significant effect on their daily disaster preparedness behavior. Where residents have experience of flood damage to their homes, and have developed a sense of inevitability with respect to flood occurrence, or where they have developed a chronic fear or feeling of helplessness with respect to floods, they are inclined to not take disaster preparedness measures. Thus, they relinquish control and enter a passive state of resignation. However, residents who have experienced flood disaster resulting not just in property damage but in human casualties, and have confidence in the ability of their village to proactively control floods, are more likely to exhibit better flood preparedness. On the bases of the above research conclusions, the following policy recommendations are proposed for the formulation of disaster prevention and mitigation policies. (1) Structural flood control measures are important. Nonetheless, the importance of daily family disaster preparedness should also be emphasized to local residents, and the two should be organically combined to enhance the capacity of flood mitigation. (2) In areas where destructive flood has occurred, relevant government departments should provide counseling and psychological guidance to allay residents’ fear and feelings of helplessness regarding recurring disaster. (3) Relevant departments of local government should promptly and accurately inform local residents of realtime flood information and flood control measures taken by relevant departments of the government, improve local residents’ trust in the media and the government’s flood control capacity, and guide residents’ daily disaster preparedness behaviors to improve residents’ confidence. (4) The relevant departments of local government should deliver flood disaster mitigation education to local residents, enhance the residents’ objective cognition of flood knowledge, improve the willingness of rural residents to be active in flood control and disaster reduction, and then take effective disaster preparedness measures.

370

13 The Experience and Attitude of Rural …

References Ablah, E., Konda, K., & Kelley, C. L. (2009). Factors Predicting Individual Emergency Preparedness: A Multi-state Analysis of 2006 BRFSS Data. Biosecur Bioterror, 7(3), 317–330. Adelekan, I. O., & Asiyanbi, A. P. (2015). Flood risk perception in flood-affected communities in Lagos, Nigeria. Natural Hazards, 80(1), 445–469. https://doi.org/10.1007/s11069-015-1977-2. Ajaero, I. D., Okoro, N. M., & Ajaero, C. K. (2016). Perception of and Attitude Toward Mass Media Reportage of the 2012 Flood in Rural Nigeria. SAGE Open, 6(3). https://doi.org/10.1177/215824 4016666887. Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179–211. Atreya, A., Czajkowski, J., Botzen, W., Bustamante, G., Campbell, K., Collier, B., Montgomery, M. (2017). Adoption of flood preparedness actions: A household level study in rural communities in Tabasco, Mexico. International Journal of Disaster Risk Reduction, 24, 428–438. https://doi. org/10.1016/j.ijdrr.2017.05.025. Baker, E. J. (2011)Household preparedness for the Aftermath of Hurricanes in Florida. Applied Geography, 31(1), 0–52. Basolo, V., Steinberg, L. J., Burby, R. J., Levine, J., Cruz, A. M., & Huang, C. (2009). The effects of confidence in government and information on perceived and actual preparedness for disasters. Environment and behavior, 41(3), 338–364. Birkholz, S., Muro, M., Jeffrey, P., & Smith, H. M. (2014). Rethinking the relationship between flood risk perception and flood management. Science of the Total Environment, 478(8), 12–20. Box, P., Bird, D., Haynes, K., & King, D. (2016). Shared responsibility and social vulnerability in the 2011 Brisbane flood. Natural Hazards, 81(3), 1549–1568. Bronfman, N. C., Cisternas, P. C., López-Vázquez, E., & Cifuentes, L. A. (2015). Trust and risk perception of natural hazards: implications for risk preparedness in Chile. Natural Hazards, 81(1), 307–327. Cao Yongqiang, Yuan Liting, Wang Feilong. (2018). Analysis of current situation and development trend of flood disaster research based on bibliometrics. Water economics, 36 (4), 33–39. Charalambous, K., Bruggeman, A., Giannakis, E., & Zoumides, C. (2018). Improving public participation processes for the Floods directive and flood awareness: Evidence from Cyprus. Water, 10(7), 958. Chen Chao, Pang Yanmei, & Zhang Yufang. (2010). Study on the characteristics of climate change in Sichuan Basin in recent 50 years. Journal of Southwest University (Natural Science Edition), 32 (9), 115–120. Chen Xiang. (2007). Risk assessment and management of rainstorm and flood in Fujian Province. Soil and water conservation research, 14 (4), 180–185. Chen Yulin. (2019). Investigation on the status quo of disaster preparedness of rural families in D Town. Agricultural Disaster Research, 9(01), 94–95. Daniell, J., Wenzel, F., & Schaefer, A. (2016). The economic costs of natural disasters globally from 1900–2015: historical and normalised floods, storms, earthquakes, volcanoes, bushfires, drought and other disasters. Paper presented at the EGU General Assembly Conference Abstracts. Danso, S. Y., & Addo, I. Y. (2017). Coping strategies of households affected by flooding: A case study of Sekondi-Takoradi Metropolis in Ghana. Urban Water Journal, 14(5), 539–545. Deng Guowei, Ma Zhenfeng, sun Jun. (2016). Analysis on the characteristics of rainstorm and flood disaster in Sichuan Province and the circulation background in the main flood season. Paper presented at the 33rd annual meeting of the Chinese Meteorological Society S2 subtropical meteorology and meteorological disaster risk. Deng Qianwen (2019). Study on assessment system of family disaster preparedness capability [J]. Today’s fire protection, 004 (002): p. 30–31. Du Huaming, Dong Tingxu. (2016). Risk assessment of rainstorm and flood in main flood season in Sichuan Yunnan region. Soil and water conservation bulletin, 36 (4), 27–31, 35.

References

371

Du Huaming, Yan Junping, Yang Dengxing, & Yang Rong. (2015). Multi time scale analysis of precipitation change and drought and flood in Jialing River Basin. Journal of natural resources, 30 (5), 836–845. Ejeta, L. T., Ardalan, A., Paton, D., & Yaseri, M. (2018). Emotional and cognitive factors influencing flood preparedness in Dire Dawa town, Ethiopia. Natural Hazards, 93(2), 715–737. EM-DAT. (2017). Emergency disasters database. Fan Zhixin, Su Jun, Feng Tao. (2016). Spatial and temporal distribution characteristics of emergency response to natural disaster relief in China from 2011 to 2015. Disaster science, 31 (3), 217–221 https://doi.org/10.3969/j.issn.1000-811x.2016.03.037. Grothmann, T., & Patt, A. (2005). Adaptive capacity and human cognition: the process of individual adaptation to climate change. Global Environmental Change, 15(3), 199–213. Grothmann, T., & Reusswig, F.(2006) People at Risk of Flooding: Why Some Residents Take Precautionary Action While Others Do Not. 38(1–2), 101–120. Guo Yongfang, Cha Liangsong. (2010). Analysis of flood risk zoning and disaster area change trend in Anhui Province. Agrometeorology, China, 31 (1), 130–136. Guo Qiang (1993). On disaster behavior. Hoffmann, R., & Muttarak, R. (2017). Learn from the Past, Prepare for the Future: Impacts of Education and Experience on Disaster Preparedness in the Philippines and Thailand. World Development, 96, 32–51. https://doi.org/10.1016/j.worlddev.2017.02.016. Huang, J., Cao, W., Wang, H., & Wang, Z. (2020). Affect Path to Flood Protective Coping Behaviors using SEM based on A Survey in Shenzhen, China. Int J Environ Res Public Health, 17(3). https:// doi.org/10.3390/ijerph17030940. Jia Liyan (2010). SPSS statistical analysis standard course: People’s post and Telecommunications Press. Jiang Zhi, Yang., Zhang Jie, et al. (2017). Analysis of the factors influencing the efficacy of the first single drug treatment for newly diagnosed epileptic children. Zhongnan pharmacy (11), 1637–1642. Kim, Y. C., & Kang, J. (2010). Communication, neighbourhood belonging and household hurricane preparedness. Disasters, 34(2), 470–488. Lawrence, J., Quade, D., & Becker, J. (2014). Integrating the effects of flood experience on risk perception with responses to changing climate risk. Natural Hazards, 74(3), 1773–1794. Levavasseur, F., Bailly, J.-S., Lagacherie, P., Colin, F., & Rabotin, M. (2012). Simulating the effects of spatial configurations of agricultural ditch drainage networks on surface runoff from agricultural catchments. Hydrological Processes, 26(22), 3393–3404. Lei Ting, & Yang naiding (2016). A study on the multi factor influence of personal emotion and behavior decision-making in the context of sudden disaster. Quest (3), 114–118. Li Jianlan, Tian min, & Shi Xiaofang. (2015). Research progress on measurement methods of farmers’ natural disaster risk attitude at home and abroad. Anhui Agricultural Science, 43 (12), 274–276. Li Jing, Wu Yijie, & he Dongyan. (2007). Comparative analysis of the causes of disasters caused by tropical cyclones “Bilis” and “pabian”. Disaster science, 22 (3), 72–76. Li Jing, Yang Hu Yu Song. (2013). Spatial and temporal patterns of flood disasters in Asia Pacific region. Journal of Shaanxi Normal University (NATURAL SCIENCE EDITION), 41 (1), 74–81. Lin, S., Shaw, D., & Ho, M.-C. (2008). Why are flood and landslide victims less willing to take mitigation measures than the public? Natural Hazards, 44(2), 305–314. Liu Hongxiu, & Zhuang Tianhui (2012). A study on the cognition of disaster prevention and mitigation of farmers in southwest poverty-stricken areas and the impact of their behavior – Based on a survey of 665 farmers in 36 villages. Agricultural economic problems, 12, 48–54. Liu Yajuan. (2015). Research on Disaster Management at the Grassroots Level: Participation Status, Experience Characteristics and Working Mechanism: Taking Mianyang Disaster Management Research Work as a Sample. Economic and Social Development Research (2), 73–74.

372

13 The Experience and Attitude of Rural …

Lokonon, B. O. K. (2016). Urban households’ attitude towards flood risk, and waste disposal: Evidence from Cotonou. International Journal of Disaster Risk Reduction, 19, 29–35. https://doi. org/10.1016/j.ijdrr.2016.08.015. Lu, Q.-C., Zhang, J., & Rahman, A. S. (2017). The interrelationship between travel behavior and life choices in adapting to flood disasters. Natural Hazards, 85(2), 1005–1022. Lu, Q.-C., Zhang, J., Wu, L., & Rahman, A. S. (2016). Job and residential location changes responding to floods and cyclones: an analysis based on a cross-nested logit model. Climatic change, 138(3–4), 453-469. Messner, F., & Meyer, V. (2006). Flood damage, vulnerability and risk perception - challenges for flood damage research. Ufz Discussion Papers, 67, 149–167. Miceli, R., Sotgiu, I., & Settanni, M. (2008). Disaster preparedness and perception of flood risk: A study in an alpine valley in Italy. Journal of environmental psychology, 28(2), 164–173. Myers, D. G., Twenge, J. M., Hou Yubo, Le Guoan, & Zhang Zhiyong (2016). Social psychology: 11th Edition. Ning Yan, Tao Maoxuan, Hu Junfeng, Li Yubo, Cheng Yulan, Zhang Gang, walnut, Li Li, Chai Yan, Nan Haitao, Zhao Hongqi, Wang Junjie. Emergency preparedness status and influencing factors of urban and rural residents in four districts (counties) of Shaanxi Province [J]. Chinese Journal of Preventive Medicine, 2013 (04): 347–351. Osberghaus, D. (2015). The determinants of private flood mitigation measures in Germany— Evidence from a nationwide survey. Ecological Economics, 110, 36–50. Osuret, J., Atuyambe, L. M., Mayega, R. W., Ssentongo, J., Tumuhamye, N., Bua, G. M., … Bazeyo, W. (2016). Coping strategies for landslide and flood disasters: a qualitative study of Mt. Elgon Region, Uganda. PLoS currents, 8. Ottmar Edenhofer, R. P.-M., Youba Sokona, Jan Christoph Minx. (2014). Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. 1454. Peterson, B. (1990). Ordinal regression models for epidemiologic data. American Journal of Epidemiology, 131(4), 745. Priyanti, R. P., Hidayah, N., Rosmaharani, S., Nahariani, P., Asri, Mukarromah, N., & Mundakir. (2019). Community Preparedness in Flood Disaster: A Qualitative Study. Int Q Community Health Educ, 40(1), 67–68. https://doi.org/10.1177/0272684x19853169. Qiao Gugu, Wang Jin, Zhou Ruiheng. (2018). Social level and well-being of the elderly - based on CGSS survey data in 2015. Reform and opening up, no.491 (14), 102–105. Qin Dahe, Thomas stocker. (2014). Highlights of the report of the first working group of the fifth IPCC Assessment Report. Progress in climate change research, 10 (1), 1–6. Qin Shanqi, Zhang Ye, Liu Junyan. Research on Influencing Factors of farmers’ demands for production and living difficulties: An Empirical Analysis Based on binary logistic regression model [J]. Hubei agricultural science, 2020,59 (03): 175–179 + 184. Qing Qingtao, Chen Wenxiu, & Zhan Zhaoyu. (2013). Characteristics of spatial and temporal distribution of losses caused by flood damage% temporal and spatial evolution characteristics of rainstorm and flood damage in Sichuan Province. Plateau mountain meteorological research, 033 (1), 47–51. Raikes, J., Smith, T. F., Jacobson, C., & Baldwin, C. (2019). Pre-disaster planning and preparedness for floods and droughts: A systematic review. International Journal of Disaster Risk Reduction, 38. https://doi.org/10.1016/j.ijdrr.2019.101207. Ren Qiuping, Xie Zerong, Wei Kemao, Huang Min, yuan Linyan, Li Xia, Yang Xiaoyun. (2018). Development and reliability and validity of family disaster preparedness questionnaire for the elderly. Nursing research (7), 1088–1090. Reynaud, A., Aubert, C., & Nguyen, M.-H. (2013). Living with Floods: Protective Behaviours and Risk Perception of Vietnamese Households. The Geneva Papers on Risk and Insurance - Issues and Practice, 38(3), 547–579. https://doi.org/10.1057/gpp.2013.16. Ruan Haibo. Empirical analysis of migrant workers’ employment service satisfaction based on ordered multi classification logistic model [J]. Research world, 2019 (12): 40–45.

References

373

Samaddar, S., Chatterjee, R., Misra, B., & Tatano, H. (2014). Outcome-expectancy and selfefficacy: Reasons or results of flood preparedness intention? International Journal of Disaster Risk Reduction, 8, 91–99. https://doi.org/10.1016/j.ijdrr.2014.02.002. Samaddar, S., Misra, B. A., & Tatano, H. (2012). Flood risk awareness and preparedness: The role of trust in information sources. Paper presented at the IEEE International Conference on Systems. Saqib, S. E., Ahmad, M. M., Panezai, S., & Rana, I. A. (2016). An empirical assessment of farmers’ risk attitudes in flood-prone areas of Pakistan. International Journal of Disaster Risk Reduction, 18, 107–114. https://doi.org/10.1016/j.ijdrr.2016.06.007. Schad, I., Schmitter, P., Saint-Macary, C., Neef, A., Lamers, M., Nguyen, L., … Hoffmann, V. (2012). Why do people not learn from flood disasters? Evidence from Vietnam’s northwestern mountains. Natural Hazards, 62(2), 221–241. Scolobig, A., Marchi, B. D., & Borga, M. (2012). The missing link between flood risk awareness and preparedness: findings from case studies in an Alpine Region. Natural Hazards, 63(2), 499–520. Shen Hong, sun Xueping, & Su Yun (2012). Trust in science and technology, trust in management and its impact on public awareness of flood risk: a social survey based on the middle and lower reaches of the Yangtze River. Disaster science, 27 (1). Sinatra, G. M., Kardash, C. M., Taasoobshirazi, G., & Lombardi, D. (2012). Promoting attitude change and expressed willingness to take action toward climate change in college students. Instructional Science, 40(1), 1–17. Soane, E., Schubert, I., Challenor, P., Lunn, R., Narendran, S., & Pollard, S. (2010). Flood Perception and Mitigation: The Role of Severity, Agency, and Experience in the Purchase of Flood Protection, and the Communication of Flood Information. Environment and Planning A: Economy and Space, 42(12), 3023–3038. https://doi.org/10.1068/a43238. Soans, S. J. (2011). Impact of health education in raising community awareness on flood preparedness. Review of Global Medicine and Healthcare Research, 2(1), 29–57. Soetanto, R., Mullins, A., & Achour, N. (2016a). The perceptions of social responsibility for community resilience to flooding: the impact of past experience, age, gender and ethnicity. Natural Hazards, 86(3), 1105–1126. https://doi.org/10.1007/s11069-016-2732-z. Soetanto, R., Mullins, A., & Achour, N. (2016b). The perceptions of social responsibility for community resilience to flooding: the impact of past experience, age, gender and ethnicity. Natural Hazards, 86(3), 1–22. Su Yun, Zhang Meihua, Gao Lilong, & Li Na. (2008). The impact of flood control project trust on public flood risk perception. Journal of natural disasters, 7 (1). Sun Lei, Qi Wenhua, Deng Yan, & Wei Benyong. (2018). Research on disaster behavior abroad: origin, issues and discoveries. Seismology of North China (March 2018), 27–34. Taylor, S., Peplau, L., & Sears, D. (2000). Social Psychology (10th Edition). In: Chinese translation, Peking University Pres. Terpstra, T. (2011). Emotions, trust, and perceived risk: affective and cognitive routes to flood preparedness behavior. Risk Anal, 31(10), 1658–1675. https://doi.org/10.1111/j.1539-6924.2011. 01616.x. Terpstra, T., & Gutteling, J. M. (2008). Households’ perceived responsibilities in flood risk management in the Netherlands. International Journal of Water Resources Development, 24(4), 555–565. Terpstra, T., Lindell, M. K., & Gutteling, J. M. (2009). Does communicating (flood) risk affect (flood) risk perceptions? Results of a quasi-experimental study. Risk Analysis: An International Journal, 29(8), 1141–1155. Tian Xiujuan, Hou Jianlin, & Dong Zhumin. (2010). Farmers’ comprehensive evaluation of the new rural cooperative medical system: Based on the analysis of 916 farmers in 13 provinces. China’s rural economy (5), 86–96. Thieken, A. H., Kreibich, H., Müller, M., & Merz, B. (2007). Coping with floods: preparedness, response and recovery of flood-affected residents in Germany in 2002. Hydrological Sciences Journal, 52(5), 1016–1037.

374

13 The Experience and Attitude of Rural …

Ullah, R., Shivakoti, G. P., & Ali, G. (2015). Factors effecting farmers’ risk attitude and risk perceptions: the case of Khyber Pakhtunkhwa, Pakistan. International Journal of Disaster Risk Reduction, 13, 151–157. Wang, Z., Wang, H., Huang, J., Kang, J., & Han, D. (2018). Analysis of the public flood risk perception in a flood-prone city: The case of Jingdezhen city in China. Water, 10(11), 1577. Wang Heng, Hu Xiuying. (2014). Effects of disaster preparedness education on disaster preparedness awareness, behavior and knowledge of college students. Chinese Journal of practical nursing, 030 (3), 76–78. Wang Weichao (2013). Study on the factors influencing the behavior of individual flood adjustment. Wang Xingping. (2012). Research on family disaster preparedness strategies in ethnic areas from the perspective of vulnerability analysis. (Master), Lanzhou University, Available from Cnki. Wang Yanjun, Gao Gao, Wang Anqian, Wang Yuyan, Zhang Feiyue, Zhai Jianqing, Su Buda. (2014). Temporal and spatial variation characteristics of exposure and vulnerability of rainstorm and flood disasters in China. Progress in climate change research, 10 (6), 391–398. Xiao Yu, peacock, W. g. (2017). Can enterprise disaster reduction and preparedness really reduce disaster losses? ——Analysis of the importance of enterprise disaster management planning. Risk Disaster Crisis Study (3), 190–215. Xue Wei (2013). SPSS statistical analysis method and application. Xu, D., Peng, L., Liu, S., & Wang, X. (2018). Influences of risk perception and sense of place on landslide disaster preparedness in southwestern China. International Journal of Disaster Risk Science, 9(2), 167–180. Yamane, T. (1967). Statistics: An introductory analysis. Retrieved from Yang peiguo, Hu Junfeng, Yu Bohua, Song Wei, & Li Jing. (2013). Spatial and temporal patterns of flood disasters in the Asia Pacific region. Journal of Shaanxi Normal University (NATURAL SCIENCE EDITION) (1), 80–87. Ye Hong. (2010). “Community Disaster Management” is the foundation of disaster prevention and mitigation. China Disaster Reduction (7), 28–29. Yong, A. G., & Lemyre, L. (2019). Getting Canadians prepared for natural disasters: a multimethod analysis of risk perception, behaviors, and the social environment. Natural Hazards, 98(1), 319–341. https://doi.org/10.1007/s11069-019-03669-2. Zaalberg, R., Midden, C., Meijnders, A., & McCalley, T. (2009). Prevention, adaptation, and threat denial: flooding experiences in the Netherlands. Risk Anal, 29(12), 1759–1778. https://doi.org/ 10.1111/j.1539-6924.2009.01316.x. Zaalberg, R., & Midden, C. J. H. (2013). Living Behind Dikes: Mimicking Flooding Experiences. Risk Analysis, 33(5), 866–876. https://doi.org/10.1111/j.1539-6924.2012.01868.x. Zhang Fuying, Gu xinbing, Peng Yi, & Yuan Tian. (2019). Knowledge map analysis of China’s disaster risk cognition research. Safety and Environmental Engineering (February 2019), 32–37. Zhang Hongtao, & Wang Erping. (2007). The current situation and development trend of the research on the relationship between attitude and behavior. Progress in psychological science, 15 (01), 163–168. Zhang Meihua, Su Yun, & Zhong jingnai. (2008). Trust in regional disaster reduction capacity and public flood risk perception based on Social Survey and analysis. Disaster science, 23 (4), 70–75. Zhong jingnai, & Zhang Meihua. (2010). The impact of flood control project trust on public flood risk perception: a case study of Kaifeng along the Yellow River and Yueyang Dongting Lake. Zhou Chenghu, Wan Qing. (2000). Study on flood risk zoning based on GIS. Acta geographica Sinica, 55 (1), 15–24. Zhou Changyan, cen Sixian, Li Yueqing, Peng Guozhao, Yang shuqun, & Peng Jun. (2011). Variation characteristics and influence of precipitation in Sichuan Province in recent 50 years. Journal of geography, 66 (5), 619–630. Zhu Ling (2018). Overview of current situation of flood disaster research and assessment. Shandong Industrial Technology (10), 192–192.

Chapter 14

Final Reflections on Current Research Contributions, Limitations and Future Research Directions

14.1 Research Contributions The author is dedicated to study how the planning and construction of Chinese rural areas influence the daily activities of local residents. The characteristics and innovations of this research are as follows: 1. It fills in the research gap and helps with the development of rural areas. At present, research on Chinese built environment and residents’ behavior is at a preliminary level, and the relationship between built environment and the CO2 emission of residents in Chinese rural areas is largely ignored. This book not only fills in the research gap, but also contributes to the urbanization in rural areas and the sustainable development of revitalized countryside in China. 2. It helps reveal and explain the rules and mechanisms at both a macro-level and micro-level. Through the comprehensive analysis of eight consecutive years’ panel data at macro level, this research helps make sense of the trends in the development of specific built environments in rural China. Moreover, through the analysis, the questionnaire data were collected from seven representative sample villages, in order to reveal at a micro-level how different measurement indexes affect the CO2 emission of Chinese rural residents. 3. It collects micro-level data with various means and add to the measurement indexes of built environment. Given that official data is currently unavailable, this research acquires its basic data on rural built environment and rural residents’ daily activities by conducting questionnaire surveys in villages and households, measuring data on site and using GIS technology. The questionnaire design and data collecting method provides valuable lessons for later studies. What’s more, considering the peculiarity of rural built environment, this research proposes new built environment measurement indexes with characteristics of rural China, such as ways of living, number of convenient and accessible markets, which further enrich indexes of built environment measurement.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0_14

375

376

14 Final Reflections on Current Research Contributions …

4. It takes into account the influence of mental determinants. This study considers various psychological factors for all research contents, including rural residents’ attitudes toward travelling, car ownership, precautionary acts against flood, motives and preferences of travelling by bikes, and their preferences and perception of built environment. This study focuses on how these elements influence the daily behaviors of rural residents (travelling, water saving, precautionary acts against flood, emergent evacuation in the case of earthquake). Thus, it reveals the mechanism of mental determinants and how they take effect.

14.2 Research Limitations Due to high collection cost, the lack of relevant research in rural areas, and the inaccessibility of data of built environment and daily behaviors in rural areas, this study has the following limitations: 1. Insufficient data precision of daily activities of rural residents. The daily activity data is determined based on the record of these activities. It makes up for the unavailability of official data, but since the data (including satisfaction, travel activity data, precautions against disasters and earthquake emergency data) is collected from the estimation and memories of rural residents, the precision might be limited. 2. Monotonous type of travel data for rural residents. Research on the micro-level activities of rural residents adopts only cross-sectional data, which could only investigate the influence of elements statically, such as that of built environment on the CO2 emission of rural residents’ travelling activities, leaving the impact of dynamic changes of built environment unaccounted for. 3. Small sample data size. Objectively speaking, there are difficulties in collecting data on rural built environment and residents’ daily activities, resulting in a small size of sample data, which further limited this model’s scope of application. On the one hand, with the number of effective samples somewhere between 165 to 700, the sample size of this study is relatively small compared with that of similar studies conducted in urban areas and abroad, whose sample size is usually in the thousands. On the other hand, the number of sample villages in similar empirical studies is always somewhere between four and nine, and the limited number of sample villages in this study restricts the use of certain measurement indexes on rural built environment, as well as their scope of universal application. Yet the small sample size does not reduce the significance of this study. As the first of its kind conducted in rural China, this study goes deep down in rural areas to conduct questionnaire survey and collect data on site. These are great effort which are very valuable to relative studies in China. 4. The statistics of rural residents’ average CO2 emission from daily travel does not consider the varieties of private vehicles, carbon emission efficiency, or road conditions. Different private vehicles have different emission and emission efficiency, especially cars and motorcycles, and this has a direct and significant

14.2 Research Limitations

377

impact on the CO2 emission of traveling by these means. The geographical environment of Sichuan rural areas varies greatly, and the diverse road conditions also have direct influence on the carbon emission of rural residents’ travel by car. Because of the difficulty in data acquiring, these factors are not taken into account, so the actual average CO2 emission from daily travel may be different from the final calculation. 5. The influence of different landforms is not considered. Rural areas in Sichuan extends over a wide range of landforms, which has a considerable impact on how rural residents choose their way of transportation. However, due to insufficient research experience, this was not fully considered when designing questionnaire and setting sample villages. Meanwhile, the limited sample number also restrained the use of built environment indexes.

14.3 Future Research Directions In reply to the aforementioned limitations, the authors would like to offer the following suggestions to future researchers: 1. Using or developing data collecting tools: with reference to related research conducted in urban areas, scholars may consider various ways in order to obtain more accurate data on the daily activities of rural residents, such as using GPS or developing apps together with GIS data to track rural residents’ travel records. 2. Collecting samples’ panel data: later scholars may consider how to fully identify the typical sample households and conduct regular return visit surveys for five or ten consecutive years. This will help obtain dynamic change data of these sample households and villages, investigate how the dynamic change of built environment in rural areas influences the mentality and daily activities of rural residents, which will be instructive to the fast urbanization in China. 3. Designing more profound experiments: the built environment in rural China is changing rapidly, with certain infrastructures growing out of nothing. Under such circumstances, researchers may provide more specific suggestions to the development of new countryside by designing experiments to study the influence of rural residents’ activities on infrastructure construction. To give an example, researchers may carry out multi-scenario designs by fully considering variables which can be weighed against each other, such as travel time, cost constraints, and the benefits of certain types of infrastructure, and offer recommendations to new countryside development based on the result. 4. Giving full account of the sample size and landforms in the rural areas: should the above-mentioned objects be fulfilled, rural areas can be investigated more extensively, with the diversity of rural landforms fully considered, so that the study may have a more universal scope of application. 5. Giving more thought to the substitute means of transportation, such as new energy vehicles. Later studies may focus on the possibility of conventional vehicles

378

14 Final Reflections on Current Research Contributions …

being substituted by the new ones, the willingness of rural residents to accept new vehicles and the prerequisite for popularizing new means of transportation in the countryside. The collection of studies contained in this book is still at a preliminary stage. The intention is to continue to refine the methodology, more accurately identify rural residents’ concerns and needs, and to pursue future research that serves the purpose of improving the rural built environment in ways that better meet the expectations of rural Chinese people, and improve their quality of life.

Appendix 1

Peasant Households Satisfaction Questionnaire on Rural Infrastructure Construction

Dear Sir/Madam, Hello, this is a questionnaire on rural infrastructure construction conducted by the Engineering and Technical College of Chengdu University of Technology. Thanks for your precious time in completing the questionnaire. The questionnaire is only for academic purposes, and it aims to provide reasonable suggestions for improving rural infrastructure construction. Please fill in according to the facts. Your participation values a lot for the survey. Thank you again for your understanding and support.

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

379

380

Appendix 1: Peasant Households Satisfaction Questionnaire on Rural Infrastructure …

1. Village address (in detail) _______________________________ 2. Village Type ( )

A. An ordinary village B. A township government stationed village C. In the rural-urban fringe D. A township government stationed village in the rural-urban fringe 3. Distance to the county (city) is ( ).

A. Less than 5 km km

B. 5–10 km

C. 10–20 km

D. 20–30 km

E. more than 30

4. Gender of the interviewee ( )

A. Male

B. Female

5. Age of the interviewee ( )

A. Under 20

B. 20–30

C. 30–40

D. 40–50

E. 50–60

F. over 60

6. Family size ( )

A. 3 people or less

B. 4 people

C. 5 people

D. 6 people and above

7. The highest academic degree among family members is ( ).

A. Primary school graduate B. Junior high school graduate C. Senior high school graduate D. Bachelor’s degree E. Master’s degree F. Doctor’s degree 8. Type of your family ( )

A. With CPC members

B. With cadres in the family C. An ordinary family

9. Sources of household income are ( ) (multiple selections)

A. Farming Others

B. Migrant work

C. Operational work

D. Public institution

E.

10. Average annual household income in the past three years is ( ).

A. Below RMB 50,000 C. RMB 100,000–200,000

B. RMB 50,000-100,000 D. Above RMB 200,000

11. Generally speaking, are you satisfied with the infrastructure of your village? ( )

A. Dissatisfied

B. Satisfied

12. How satisfied are you with the infrastructure construction in your village? (Tick in the blank) Infrastructure

Very dissatisfied

Somewhat not satisfied

Just so-so

Partly satisfied

Highly satisfied

Road











Drinking water











Environmental protection and sanitation











Building and renovation of public toilets











Village planning and renovation









 (continued)

Appendix 1: Peasant Households Satisfaction Questionnaire on Rural Infrastructure …

381

(continued) 









Renovation of  kitchen using fuels









Farmland  irrigation facilities









Medical facilities











Cultural and recreational facilities











Educational facilities











Communication, electrification and television signal

13. How do you think about the infrastructures in the village compared with those five years ago? (Tick in the blank) Infrastructures

Worse

Barely changed

Somewhat improved

Much better than five years ago

Road









Drinking water









Environmental protection and sanitation









Building and renovation of public toilets









Village planning and renovation









Communication, electrification and television signal









Renovation of  kitchen using fuels







Farmland irrigation facilities









Medical facilities









Cultural and recreational facilities









Educational facilities









382

Appendix 1: Peasant Households Satisfaction Questionnaire on Rural Infrastructure …

14. How do you like the infrastructure of the village compared to other villages you know? (Tick in the blank) Infrastructure

One of the worst

Below average

Average

Above average

One of the best

Road











Drinking water











Environmental protection and sanitation











Building and renovation of public toilets











Village planning and renovation











Communication,  electrification and television signal



















Farmland  irrigation facilities









Medical facilities











Cultural and recreational facilities











Educational facilities











Renovation of kitchen using fuels

15. How do you think about the consumption price of rural infrastructure? (Tick in the blank) No charge

Very low

Relatively low

Appropriate

Relatively high

Very high

Bus fare













Tap water price













Charge of environmental governance













Electricity price













Communication price (telephone charge)













Broadband and TV 









 (continued)

Appendix 1: Peasant Households Satisfaction Questionnaire on Rural Infrastructure …

383

(continued) Gas fee













The price of seeing a doctor and seeking medical treatment













Cultural and entertainment consumption













Cost of education













Thank you very much for your participation!

Appendix 2

Questionnaire on the Construction of Rural latrines in Sichuan Province

Dear Sir/Madam, Hello, this is a questionnaire on rural infrastructure construction conducted by the Engineering and Technical College of Chengdu University of Technology. Thanks for your precious time in completing the questionnaire. The questionnaire is only for academic purposes, and it aims to provide reasonable suggestions for improving rural infrastructure construction. Please fill in according to the facts. Your participation values a lot for the survey. Thank you again for your understanding and support.

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

385

386

Appendix 2: Questionnaire on the Construction of Rural latrines in Sichuan Province

1. Village Type ( )

A. An ordinary village B. A township government stationed village C. In the rural-urban fringe D. A township government stationed village in the rural-urban fringe 2. Gender of the interviewee ( ) A. Male B. Female 3. Age of the interviewee ( ) A. Under 20 B. 20–30 C. 30–40 D. 40–50 E. 50–60 F. Over 60 4. The highest academic degree among family members is ( ). A. Primary school graduate B. Junior high school graduate C. Senior high school graduate D. Bachelor’s degree E. Master’s degree F. Doctor’s degree and above 5. Average annual household income in the past three years is ( ). A. Below RMB 50,000 B. RMB 50,000–100,000 C. RMB 100,000–200,000 D. Above RMB 200,000 6. Distribution of dwellings in village where they live ( )

A. Very scattered B. Relatively scattered C. Generally scattered D. Relatively concentrated E. Very concentrated 7. Is there a separate family latrine? ( )

A. No

B. Yes

8. Location of family latrine is ( ).

A. Outside the courtyard B. In the courtyard C. Indoor 9. Type of family latrine ( )

A. Pit-type toilet B. Linked circled toilet C. Urine directing toilet D. Triple-chamber biogas toilet E. Double urn and funnel-type toilet F. Triplechamber septic tank G. Double-pit alternating toilet H. Flushing toilet I. Others 10. How many years have the family toilet been in use? ( )

A. Within 2 years more.

B. 2–5 years

C. Over 10 years

D. 10 years or

11. Acreage of family latrine is ( ).

A. Below 2.25 m2

B. 2.25–5 m2

C. 5–10 m2 D. Above 10 m2

12. Is there a wall ( ), a roof ( ), and a door ( ) in family latrine?

A. No

B. Yes

13. Ventilation of family latrine ( )

A. Unventilated

B. Natural ventilation

C. Mechanical ventilation

Appendix 2: Questionnaire on the Construction of Rural latrines in Sichuan Province

387

14. Is the toilet 10 cm higher than the ground ( )?

A. No

B. Yes

15. Is there a stool in the family latrine ( )?

A. No

B. Yes

16. Is the family latrine equipped with facilities for harmless disposal of excrement (such as double-urn septic tanks, triple-chamber septic tanks, complete sewerage, or pond of manure?) ( )

A. No

B. Yes

17. Does the family latrine have the following sanitation facilities: buckets, special cleaning tools, and toilet paper containers? ( )

A. No

B. It has some of them.

C. Yes, it has all of them.

18. Is there any fly control facility in the family latrine? ( )

A. No

B. Yes

19. Sanitary situation in family latrines ( )

A. Not good

B. Not bad

C. Good

20. How often do dung dregs removed from the family latrine? ( )

A. For more than two years B. One to two years C. From half a year to one year D. Within half a year 21. Is the domestic fecal storage tank closed without leakage? ( )

A. No

B. Yes

22. How smelly is the family latrine? ( )

A. Very smelly B. A little smelly C. Basically not smelly 23. Has the family latrine been renovated? ( )

A. No

B. Yes

24. Cost of construction / renovation of family latrine is ( ).

A. Below RMB 500 Above RMB 2000

B. RMB 500–1000

C. RMB 1000–2000

D.

25. The reasons for renovating family latrine is ( )

A. Haven’t renovated yet C. Government advocacy situation

B. Neighbors have renovated theirs D. The owner is unsatisfied with the previous

26. Who are responsible for the construction / renovation of family latrines ( ) (multiple selections)

A. Villagers

B. Government

C. Professional construction units

27. Who are responsible for daily management and maintenance of family latrine (

)

388

Appendix 2: Questionnaire on the Construction of Rural latrines in Sichuan Province

(multiple selections)

A. No management maintenance B. Villagers C. Government D. Professional management unit 28. Proportion of subsidy funds for the construction / renovation of family latrines ( )

A. No subsidy B. 0% ≤ 20% than 80%

C. 20% ≤ 50%

D. 50%–80%

E. more

29. The economic burden brought by the construction / reconstruction of family latrines ( )

A. Huge burden

B. Moderate burden

C. Small burden

30. The frequency of malfunction during using family latrine is ( )

A. Every 0–3 months B. Every 3–6 months D. More than 12 months per time

C. Every 6–12 months

31. How to maintain family latrine in the case of malfunction ?( )

A. On your own B. A specific person in the village is responsible for the maintenance C. Looking for professional service men 32. How long will the family latrine be repaired after malfunction? ( )

A. Within 12 hours hours

B. 24 hours

C. 36 hours

D. More than 36

33. Is there any publicity and training on the construction / renovation, use and management of family latrines in the village? ( )

A. No

B. Yes

34. Is there a unified planning for the construction / renovation of family latrines in the village ( )

A. No

B. Yes

35. Do you think the financial and human resources spent on the construction / renovation of family latrines are worthy? ( )

A. No, it’s not worthy.

B. Yes, it’s worthy.

36. How well informed and transparent are matters related to the construction / renovation of family latrines in the village in your opinion? ( )

A. The transparency is low, and the villagers are unaware of many cases. B. Just so-so. Villagers are informed of some work and information. C. Highly transparent. And villagers are well informed of relevant arrangements and information. 37. Are you satisfied with the overall status of your family latrines? ( )

Appendix 2: Questionnaire on the Construction of Rural latrines in Sichuan Province

A. Dissatisfied

389

B. Satisfied

38. Do you think that family latrine hygiene has an impact on family hygiene and village hygiene? It has ( ).

A. No effect

B. A certain influence

C. Great impact

39. Do you think family latrine hygiene has an impact on the health of family and villagers? It has ( ). A. No effect B. A certain influence C. Great impact 40. Please choose the top three items you are desperate to improve in your family latrines. ( )

A. The type of toilet B. The innocuous treatment of toilet feces C. Toilet ventilation D. The cleaning out of feces and dregs E. Toilet construction conditions F. Sanitary conditions of latrines G. Others Thank you very much for your participation!

Appendix 3

Household Questionnaire on Rural Built Environment and Rural Residents’ Travel behavior

Dear residents, Greetings! This questionnaire studies the impact of changes in rural built environment on rural residents’ consumption and travel behaviors. We conduct the survey on the basis of project “On the Influence of the Changes of Rural Built Environment on the Travel Behavior and Energy Consumption of Rural Residents”, which is funded by the Fundamental Research Funds for the Central Universities, and “Research on the Optimization of Agricultural Production Infrastructure Construction System”, a key project of natural science discipline by Education Department of Sichuan Province. Your village is the major sample area of the survey, and we sincerely invite you and your family to participate in this survey. Against the background of “urbanization” and “new village” development in rural areas, tremendous changes have, are and will take place in the built environment of rural China. We carry out the investigation from the microcosmic perspectives such as your consumption habits and behaviors on the purpose of implementing a systematic plan of the village under the premise of meeting the needs of the local residents to the greatest extent in a more scientific and reasonable way. Your objective and true feedback information will directly influence research results and the corresponding suggestions for rural construction policies. Therefore, your cooperation is crucial for the research. The results of this survey are only used for academic research, and your personal information will be strictly confidential. We sincerely hope that you can cooperate well with our researchers to complete this questionnaire. Thank you very much for your cooperation and support!

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

391

392

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

Household Questionnaire of Rural Residents Part I: Basic information of the family and the individuals Researcher’s name: _________________________ Basic information of the family

Basic personal information

Family population

Gender

Number of permanent residents

Age

Number of people under 18 years old

Type of hukou

The highest education degree in the family

Education

Number of working people

Marital status

Number of people holding driver’s license

Occupation

Total household income

Personal annual income

Number (sets) of houses the family own

Do you have a driver’s license

Does the family have a car parking space?

Driving years

Type of registered permanent residence (hukou) Can you ride a motorcycle? (1) rural hukou, (2) urban hukou, (3) others Can you ride a bike? Type of occupations: (1) village cadre, (2) teacher, Can you ride a battery cart? (3) student, (4) enterprise or factory worker, (5) farmer (6) self-employed driver (7) handicraft worker (8) enterprise or factory leaders (9) self-employed store owner (such as shops, restaurants, etc.) (10) others

Part II. Investigation on the thoughts and preference of built environmental.













The village is clean and beautiful.

Good maintenance service of public facilities.





There is a spacious public courtyard

The road lighting facilities are  complete





There are parks or other open spaces for public activities near home

There are enough parking lots 







It is convenient to the health center (hospital, clinic) now.







It is convenient to the bus, subway or railway station.

There is a good sidewalk right  now.





It is convenient to the city now.

There is a good bike lane now. 





It is convenient to the village fair now.

No



Not at all

It is convenient to school now. 

Characteristics



























Just so-so



























Yes



























Absolutely

Thoughts and judgment of built environment: do you think the following characteristics are truthful to your current living area and to what extent?



























Very important



























Important



























Just so-so



























Not important



























(continued)

Not important at all

Preference and judgment of built environment: when you select the next house, are the following conditions important and to what extent?

Appendix 3: Household Questionnaire on Rural Built Environment and Rural … 393













There isn’t any crime committed near home.

There isn’t any traffic incident  near home.





The living environment is very quiet now.

There is no difference in family economic conditions among neighborhoods.





It is safe for children to play outside.

Neighbors often chat together. 





It is safe to go for a walk and travel on foot.

(continued)

















































































































394 Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

395

Part III: Survey on the purchase, possessing, use and preference of vehicles 1. Vehicles your family possess: Vehicles

Car

Motorbike

Battery cart

Bicycle

Others

Number of vehicles you now possess

2. Survey on the travel last week (use of vehicles) Number of travels More than 7 times

5–6 times

3–4 times

1–2 times

0 times

Travel distance (km)

Type of oil: 92#95# or others

Average fuel consumption L/100 km

Motorbike Private car Public transport

No fuel consumption

Bicycle Battery cart On foot Others:

3. Do you think private cars are necessary? ___________ 1) Absolutely necessary (2) Necessary (3) No idea (4) Unnecessary (5) Completely unnecessary 4. Are you going to buy a private car in the next three years? 1) Yes

2) No

3) Not sure

5. For 1. Driving, 2. Cycling, 3. Riding a battery cart, 4. Riding a motorcycle, and 5. Taking public transport, please rank according to your preference from the most favorite to the least favorite: 6. When cycling, your purposes are _______ (multiple selections) (1) For exercises (2) Cycling to work / school (3) Cycling with the child (4) Going shopping (grocery stores, shopping malls, bazaars) (5) Visiting relatives and friends (6) Cycling to play (KTV, fitness, etc.) (7) Others 7. You would like to ride a bike when the destination is within ____ kilometers. 8. The basic information of the family population in the year when the car was purchased:

396

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

Year

The age of the head of household in that year

Number of family members

Number of the underage

When purchasing the first car When purchasing the second car When purchasing the third car

9. The life events taking place in the family in the year the car was bought: ➀

Life-events

➁

➂

The neighbor bought a car / a house / married, etc. Health condition

Some family member was in poor health / hospitalization / surgery

Education

The kids started school.

Some family member had an accident (injury). Some family member went to higher education. Some family member completed higher education. Some family member attended training for driver’s license or obtained the license. Employment

Some family member changed jobs / started business Some family member planned to change jobs. Some family member looked for a job (hard up for money) Some family member lost his/her job (on the verge or did lose a job). Some family member retired and received a pension.

Life and death

Some family member was pregnant / had a child.

Marriage

Some family member started dating.

Some family member died. Some family member ended the relationship with boyfriend or girlfriend. Some family member was engaged / married / held the wedding ceremony. Some family member ended relationship / divorced / ended cohabitation. Revenue

Household income increased a lot this year.

Moving house

Some adult child moved out.

Household income fell a lot this year. Some family member bought a house in the city (county) (continued)

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

397

(continued) The family faced land requisition and demolition. Some family member moved to the new houses they bought. Some family member moved to house given due to the land requisition and demolition. Some family member moved to self-built new houses. Crime

Some family member was victims of crime. Some family member committed a crime.

Other Events

10. Do you agree with the following point of view, and to what extent? (01–12 is about owning a car and 10–42 is about travel) Statements

Highly agree

Agree

Neutral

Disagree

Highly disagree

01. If I had (another) car, I’d be very happy.











02. If I had (another) car, people would pay more attention to me.











03. Public transport is good. I don’t need a private car.











04. It doesn’t matter if the  family doesn’t own a car.









05. Buying a car is a sign  of economic development and should be encouraged.









06. More roads need to be  built to reduce traffic jams.









07. Fuel efficiency is an  important factor in my selection of cars.









08. At present, the  environmental pollution is serious and it is not recommended to buy a car.









09. It costs a lot to repair  and maintain a car, and it is a pure expenditure to buy a car.



















10. Vehicles should be taxed according to the amount of pollution they produce.

(continued)

398

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

(continued) 11. I need a car to do what  I like to do.









12. It would be inconvenient to travel without a private car.











13. It is expensive to park now, so it is not cost-effective to drive.











14. At present, the parking  is quite under strain. It’s hard to park a car, and driving is even more troublesome.









15. Car fuel costs a lot, so  does driving a car. It is not cost-effective to drive.









16. At present, a car is  inexpensive, and there is no financial pressure to buy one.









17. The car for a test-drive  could pollute the environment, and is not conducive to environmental protection.









18. The public transport system is perfect, and it is convenient to travel anywhere by public transport.











19. The public transport costs a little, and is economic and cost-effective.











20. It takes a long time to  wait for and take public transport. Just a waste of time.









21. It’s convenient to take  public transport, and you can relax with your eyes closed on it.



















22. It is fast, convenient and easy to ride a battery cart.

(continued)

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

399

(continued) 23. Riding a battery cart is  environmentally friendly and safe.



















25. It is convenient to park  a battery cart.



















27. Riding a motorcycle is  environmentally friendly and safe.









28. Riding motorcycles pollutes the environment and affects health











29. It costs little to purchase and drive a motorcycle. There is no financial burden at all.











30. It is convenient to park  a motorcycle.









31. It is fast, convenient  and easy to ride a bike.









32. Cycling is a kind of exercise.











33. Cycling is low carbon and environmentally friendly, which is conducive to environmental protection.











34. It’s convenient to park a bicycle.











35. It costs little to  purchase and ride a bike. There is no financial burden at all.









36. It is fast, convenient and easy to walk.











37. Walking is a kind of exercise.











24. It costs little to purchase and drive a battery cart. There is no financial burden at all.

26. It is fast and convenient to ride a motorcycle.

(continued)

400

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

(continued) 38. Walking is low carbon  and environmentally friendly, which is conducive to environmental protection.









39. The price of gasoline affects my choice of transportation means.











40. I go out as little as possible through reasonable arrangements.











41. I don’t go to spot if problems can be solved by telephone and the Internet.











42. When I need to buy  something, I usually go to the nearest store.









11. With regard to the motivation of riding, do you agree or disagree with it and to what extent? Highly agree

Agree

Neutral

Disagree

Highly disagree

To avoid traffic jams











For fitness and strong body











To save money











It’s fun to ride a bike.











To protect the  environment / air quality









It is more convenient to cycle than other transportation means.











Cycling is faster than any other means of transportation.











It is easy to park a bike.











9. With the following conditions in place, would you prefer to ride a bike? Definitely I would I think so No idea I don’t think so No, I won’t There are special  bike lanes separate from the motorway.









(continued)

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

401

(continued) There is a safe bike  parking lot at the destination.









There is small traffic volume of motor vehicles.











Motor vehicles are limited in speed.











There are special bicycle streetlights on the bike lanes.











The bike lane is flat.











There are big trees  on both sides of the bike lanes.



















The road surface of  bike lanes is of good quality.









The bike lanes have no curves.











There are traffic lights on the bike lanes.











The bike lanes are safe.











There is equipment  for cleaning bicycles at the destination.









The bike lanes are wide enough.

Part IV: Records of travel activities 1. Record of travel activities on the first day

02

07

Example1

Example2

10

9

8

7

6

5

4

3

2

1

Activities

SN.

8:20

8:00

Set out time

5

15

Time on travel

0.8

2.5

Distance (km)

4

4

Transportation means

60

5

Activity duration (min)

9:25

8:20

End time

6

2

Fellow travelers

Yes

Yes

Is it a regular activity?

04. Working (including farming). 05. Studying. 06. Going to the fair to purchase. 07. Going to the fair to sell goods. 08. Going shopping in the city center. 09. Going to a hospital or a clinic to see a doctor. 10. Going to beauty salons. 11. Deposits, withdrawals, remittances and others. 12. Visiting relevant government departments. 13. Go to cultural places such as libraries and bookstores 14. Picking up packages (including all kinds of pick-up) 15. Dining out. 16. Participating in village and town activities and meetings. 17. Visiting relatives and friends. 18. Attending a banquet. 19. Receiving relatives and friends at home. 20. Indoor entertainment (Internet, TV, reading, etc.). 21. Table entertainment (mahjong, poker, billiards, etc.). 22. Going to entertainment venues (bars, cinemas, etc.). 23. Other recreational activities such as walking, square dance, etc. 24. Cycling 25. Other activities:

01. Doing housework. 02. Sending children to school (including interest classes). 03. Picking up children from school (including interest classes).

Activities

402 Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

403

Transportation means: 1. Driving, 2. Public transport (including trains, buses, etc.), 3. Motorcycle, 4. Battery cart, 5. Bike, 6. On foot, 7. Others. Fellow travelers: 1. Old family members, 2. Children, 3. Family members, 4. Friends, 5. Colleagues, 6. Alone, 7. Others. Part IV: Records of travel activities 1. Record of travel activities on the second day

02

07

Example1

Example2

10

9

8

7

6

5

4

3

2

1

Activities

SN.

8:20

8:00

Set out time

5

15

Time on travel (min)

0.8

2.5

Distance (km)

4

4

Transportation means

60

5

Activity duration (min)

9:25

8:20

End time

6

2

Fellow travelers

Yes

Yes

Is it a regular activity?

04. Working (including farming). 05. Studying. 06. Going to the fair to purchase. 07. Going to the fair to sell goods. 08. Going shopping in the city center. 09. Going to a hospital or a clinic to see a doctor. 10. Going to beauty salons. 11. Deposits, withdrawals, remittances and others. 12. Visiting relevant government departments. 13. Go to cultural places such as libraries and bookstores 14. Picking up packages (including all kinds of pick-up) 15. Dining out. 16. Participating in village and town activities and meetings. 17. Visiting relatives and friends. 18. Attending a banquet. 19. Receiving relatives and friends at home. 20. Indoor entertainment (Internet, TV, reading, etc.). 21. Table entertainment (mahjong, poker, billiards, etc.). 22. Going to entertainment venues (bars, cinemas, etc.). 23. Other recreational activities such as walking, square dance, etc. 24. Cycling 25. Other activities:

01. Doing housework. 02. Sending children to school (including interest classes). 03. Picking up children from school (including interest classes).

Activities

404 Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

Appendix 3: Household Questionnaire on Rural Built Environment and Rural …

405

Transportation means: 1. Driving, 2. Public transport (including trains, buses, etc.), 3. Motorcycle, 4. Battery cart, 5. Bike, 6. On foot, 7. Others. Fellow travelers: 1. Old family members, 2. Children, 3. Family members, 4. Friends, 5. Colleagues, 6. Alone, 7. Others.

Village-level Data Collection Questionnaire for Rural Built Environment Researcher’s Name: 1. Name of the village: 2. Address (please specify to group/team, village): 3. Acreage of the village: 4. General data Population size

Number of permanent residents

Number of households

Number of grocery stores

Number of crossings

Length of road

Road hardening rate

Road accessibility

Number of public parking lots

Number of parking spaces

Note: Statistics are based on the area under the jurisdiction of the village in survey. 5. Population distribution 0–6 years 7–17 years 18–40 years 41–65 years 65+ years Total population Number of permanent residents

6. (Walking/Biking/Driving) distance from the village center to the nearest bus station is ______________________ km. 7. (Walking/Biking/Driving) distance from the village center to the nearest railway station is ______________________ km. 8. (Walking/Biking/Driving) distance from village center to the nearest bus stop is ______ km. 9. (Walking/Biking/Driving) distance from the village center to the nearest main road is______________ km. 10. (Walking/Biking/Driving) distance from the village center to the nearest village fair is _______ km. 11. (Walking/Biking/Driving) distance from the village center to the nearest school is _____ km. 12. (Walking/Biking/Driving) distance from the village center to the nearest health center (hospital) is ________ km. 13. (Walking/Biking/Driving) distance from the village center to the nearest city (county) center is ________ km. 14. The number of village fairs within easy reach of the village is ________.

Appendix 4

Travel Satisfaction of Rural Residents

Thank you for completing the travel satisfaction questionnaire. This survey will not reveal any personal information about you, so please feel free to fill in the form. B.1 Socio-demographic variables 1. Gender: ◎Male ◎Female

2. Age: _______

3. Education level: ◎Primary school and below ◎Junior high school graduate ◎High school (technical school) graduate ◎Junior college ◎Bachelor’s degree ◎Postgraduate and above 4. Occupation: ◎Government staff ◎Farmer ◎Businessman ◎Migrant worker ◎No job ◎Others 5. Type of registered permanent residence (hukou): ◎ Rural household registration ◎ Urban household registration 6. Annual personal income: _________

7. Annual household income: __________

8. Family type: ◎Living alone ◎Living as a couple ◎Children living with their parents ◎Three generations living together ◎Four generations living together ◎Others 9. Health condition: ◎ very bad ◎ bad ◎ average ◎ healthy ◎ very healthy 10. Please fill in the number of the following

means of transportation in your household.

Family-owned cars

Family-owned delivery vans

Fuel motorcycles

Electric motorcycles

Electric battery cars

Electric counterbalance cars

Bicycles

Elderly mobility scooters

Passenger electric tricycles

Manpower tricycles

Cargo electric tricycles

Others

B.2 Satisfaction with modes of trip

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

407

408

Appendix 4: Travel Satisfaction of Rural Residents

11. Please tick the box in front of the corresponding option on your most commonly used mode of transport (the most frequently used one).  Family-owned cars

 Family-owned delivery vans

 Electric motorcycles  Electric battery cars

 Fuel motorcycles  Electric counterbalance cars

 Bicycles

 Elderly mobility scooters

 Passenger electric tricycles

 Manpower tricycles

 Cargo electric tricycles

 Large Trucks

 Agricultural vehicles  Minibuses in villages and towns  Bus  Subway

 Bike sharing

 Hitching a ride with colleagues

 Taking a taxi

 Hitching a ride with friends

 Company Shuttle buses

 Walking

 Others

B.3 Travel-related variables 12. Please rate your satisfaction with your daily mode of transport: ◎Very dissatisfied ◎Quite dissatisfied ◎Partly dissatisfied ◎Partly satisfied ◎Quite Satisfied ◎Very satisfied 13. Please choose your daily way of travel: ◎ Alone ◎ With husband/wife ◎ With parents ◎ With children ◎ With co-workers ◎ With friends ◎ Others 14. The distance from your place of residence to your workplace is: __________ km. The time of your daily trip is usually: ________ minutes. The purpose of your daily trip is _____________.

B.4 Built environment perception and travel preference 15. Please check the box according to your actual feeling

Agree

Neutral

Disagree

Strongly disagree

(1) There are good  sidewalks for travel.



















(3) There is good access  to motorized roads.









(4) There is easy access  to public transportation stations.









(5) There is convenient access to travel. destinations











(6) Private cars are highly mobile, fast and convenient.











(2) There are good bike lanes for travel.

Strongly agree

(continued)

Appendix 4: Travel Satisfaction of Rural Residents

409

(continued) (7) There is quick and  easy access to public transportation.



















(9) Riding a motorcycle  is quick and convenient.









(10) Riding a  motorcycle is quick and convenient.



















(12) It’s convenient and  easy to travel by tricycle.









(13) There are other  ways to get around easily and quickly.



















(15) I like taking public  transportation.









(16) I like riding electric bikes.











(17) I like riding motorcycles.











(18) I like riding my bike.











(19) I like walking.











(20) I like traveling by tricycles.











(21) I like other modes of travel.











(8) Riding an electric bicycle is quick and easy.

(11) Walking for exercise saves money and is easy and convenient.

(14) I like driving.

B.5 Travel Satisfaction Scale 16. Please check the following box according to your daily travel experience. (1) In terms of my daily travel, I feel time is: ◎Very urgent

◎Quite urgent

◎Somewhat urgent

◎Somewhat enough

◎Quite enough

◎Quite a lot

◎No feeling

(2) In terms of getting to a destination in the planned time for a normal trip, I feel: ◎Very worried

◎Quite worried

◎Somewhat worried

◎No feeling (continued)

410

Appendix 4: Travel Satisfaction of Rural Residents

(continued) ◎Somewhat confident

◎Quite confident

◎Very confident

(3) My emotions during my daily travels: ◎Very uneasy

◎Quite uneasy

◎Somewhat uneasy

◎Somewhat calm

◎Quite calm

◎Very calm

◎No feeling

(4) During my daily travels, I feel: ◎Very tired

◎Quite tired

◎Somewhat tired

◎Somewhat energetic

◎Quite energetic

◎Very energetic

◎No feeling

(5) During my daily travels, I feel it is: ◎Very boring

◎Quite boring

◎Somewhat boring

◎Somewhat interesting

◎Quite interesting

◎Very interesting

◎No feeling

(6) I usually feel I am _______ my daily trips: ◎Very fed up with

◎Quite fed up with

◎Somewhat fed up with

◎Somewhat attracted to

◎Quite attracted to

◎Very attracted to

◎No feeling

(7) I think my daily travel experience is: ◎Very bad

◎Quite bad

◎Somewhat bad

◎Somewhat good

◎Quite good

◎Very good

◎No feeling

(8) I think my skill of using these ways of transport is: ◎Very bad

◎Quite bad

◎Somewhat bad

◎Just so-so

◎Somewhat good

◎Quite good

◎Very good

◎Not well at all

◎Quite not well

◎Somewhat not well

◎Somewhat well

◎Quite well

◎Very well

(9) My daily trips go: ◎Just so-so

17. Are you generally satisfied with your daily travels? ◎Very dissatisfied ◎Dissatisfied ◎Somewhat dissatisfied ◎Neutral ◎Somewhat satisfied ◎Satisfied ◎Very satisfied Thank you for your participation and wish you a happy life!

B.6 Built environment data 1. Field measurements (calculated per household, using Ovi Maps) The distance from your residence to the village fair is _____ km. Village fair 1

Village fair 2

Village fair 3

By car By bus On foot By bike

The distance from your residence to the school is ____ km.

Appendix 4: Travel Satisfaction of Rural Residents

School 1

School 2

411

School 3

By car By bus On foot By bike

The distance from your residence to the health center (hospital) is ___ km. Health center 1

Health center 2

Health center 3

By car By bus On foot By bike

The distance from your residence to the city (county) center is ______ km. City center

County center

By car By bus On foot By bike

The distance from your residence to the township minibus pick-up and dropoff location is _______ km. Pick-up location

Drop-off location

By car By bus On foot By bike

The distance from your residence to the bus stop is _______ km. Bus stop 1

Bus stop 2

Bus stop 3

By car By bus On foot By bike

The distance from your residence to the nearest metro station is __________ km. The distance from your residence to the main road is ___ km.

412

Appendix 4: Travel Satisfaction of Rural Residents

Main road 1

Main road 2

Main road 3

By car By bus On foot By bike

2. GIS data collection (or aerial photogrammetry data) Acreage of the area under the jurisdiction of the village is: ______KM. Acreage of the building area under the jurisdiction of the village is: ______KM Total length of roads in the area under the jurisdiction of the village is: ______meters. The number of crossroads in the area under the jurisdiction of the village is: ______ Agricultural land of the area under the jurisdiction of the village is: ______KM.

Acreage of the afforested area under the jurisdiction of the village is _______KM. 3. Interview data from village cadres Total village population: _______________________ Total number of households in the village: _______________________ Main industries in the village: _______________________ Annual per capita income of the village: _______________________

Appendix 5

Questionnaire on the Influencing Factors of Water Saving Behavior of Rural Residents

Dear residents, Greetings! This questionnaire studies the influencing factors of water saving behavior of rural residents. We conduct the survey on the basis of project “On the Influence of the Changes of Rural Built Environment on the Travel Behavior and Energy Consumption of Rural Residents”, which is an annual social science fund project of Chengdu University of Technology, and “Research on the Optimization of Agricultural Production Infrastructure Construction System”, a key project of natural science discipline by Education Department of Sichuan Province. Your residential quarter is the major sample area of the survey, and we sincerely invite you and your family to participate in this survey. We carry out the investigation from the microcosmic perspectives such as your consumption habits and behaviors on the purpose of implementing a systematic plan of the village under the premise of meeting the needs of the local residents to the greatest extent in a more scientific and reasonable way. Your objective and true feedback information will directly influence research results and the corresponding suggestions for rural construction policies. Therefore, your cooperation is crucial for the research. The results of this survey are only used for academic research, and your personal information will be strictly confidential. We sincerely hope that you can cooperate well with our researchers to complete this questionnaire. Thank you very much for your cooperation and support! Part I: Basic information of the family and the individuals 1. Your gender is _________. 2. Your age is _________. 3. Your educational background is: _________. 4. Your total annual income is: _________. 5. The number of permanent residents in your family is: _________. 6. The highest academic degree among family members is _________. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

413

414

Appendix 5: Questionnaire on the Influencing Factors of Water Saving Behavior …

7. Total annual household income is: _________. 8. Your family type is: .

1

Living alone

2

Living as a couple

3 Couple living with

parents 4 Couple living with children

5 Three generations living together

6

Others Part II. Information on building characteristics 1. Total acreage of your home is: _________m2 . 2. The year your house was built is: _________. 3. The time it takes to produce hot water from your residential toilet shower is:

1

Within 15s 2

15–30s 3) 30–45s 4) 45–60s 5) Above 60s

4. The time it takes for your residential kitchen faucet to produce hot water is: 1

Within 15s 2

15–30s 3) 30–45s 4) 45–60s 5) Above 60s

5. How about the natural lighting of your house? 1

Very bad 2

Relatively bad 3 Just so-so 4

Relatively good 5

Very good 6. How about the natural ventilation of your house? 1

Very bad 2

Relatively bad 3

Just so-so 4

Relatively good 5

Very good 7. The number of taps in your house is _________. 8. The number of toilets in your house is _________. Part III Water-saving behavior Have you ever purchased water saving appliances (including water saving taps, water-saving water tank, water-saving washing machine, etc.)? 1. Yes 2. No Habitual water saving behavior

Never Seldom Sometimes Usually Always

I don’t wash clothes until I have accumulated enough laundry











When washing dishes, I’d like to run water into  the sink instead of leaving the faucet running.









When washing vegetables, I’d like to run water  into the basin instead of leaving the faucet running.









(continued)

Appendix 5: Questionnaire on the Influencing Factors of Water Saving Behavior …

415

(continued) 









I won’t leave the tap on all the time in the  shower (e.g.: turning off the tap when washing hair).



















When washing my face, I’d like to run water into the basin instead of leaving the faucet running.

I recycle waste water (e.g.: saving laundry water to flush toilets, etc.).

Part IV Psychological factors in water saving Strongly disagree

Disagree

Partly agree

Agree

Strongly agree











I feel worried when  hearing or seeing water-related issues.









I adopt water  conservation practices because of government laws and policies.









I’m willing to make personal sacrifices to save water and the environment.























Water issues are society-wide issues and everyone is responsible.











As long as I’m willing to do my best, I can improve or solve certain environmental problems.











If we take action, it  will help to improve environmental issues .









Ordinary people can  also improve or solve water problems.









I usually follow the media coverage of the water resources issues.

(continued)

416

Appendix 5: Questionnaire on the Influencing Factors of Water Saving Behavior …

(continued) It’s not just a few  scientists or powerful people who can have an impact on improving water issues.



















I believe that saving  water is more important than individuals pursuing their own preferred lifestyle and habits.









Adopting water conservation measures will not change my personal lifestyle and habits.











Water is a limited resource and we must conserve it.











We should respect nature and live in harmony with it.











Even for the sake of  economic development, we can’t sacrifice the natural environment.









The water-saving  equipment I want to buy isn’t much more expensive compared to regular equipment.



















I am more concerned about water issues than comfort and convenience of life.

Water-saving equipment technology is now more mature, and can be used proficiently itself.

(continued)

Appendix 5: Questionnaire on the Influencing Factors of Water Saving Behavior … (continued) I don’t think it would be inconvenient for me to take water conservation measures.











I’m saving water because that can save money.











I feel that there is a  lack of education on the implementation of water conservation measures.









What I learn from the media, such as newspapers or television, can influence whether I adopt water-saving behaviors or not.











My family, friends and so on will influence me to adopt water-saving behaviors.











There are so few people around who care about saving water, I need more people to help and participate.











I usually pay attention to the details of the use of water supply facilities.











417

Appendix 6

Questionnaire for the Study of the Impact of Built Environment or Risk Perception on Evacuation Behavior

Dear residents, Greetings! We are the research team from the Engineering and Technical College of Chengdu University of Technology. In order to explore more scientifically the emergency behavior selection of local residents in response to earthquakes and the factors influencing their emergency behavior selection, we are conducting a survey in the form of an anonymous research questionnaire, and your truthful and objective answers are crucial to our research. This study is for scientific research only, and will not involve or disclose any of your personal information. We sincerely hope that you can cooperate with the researchers to complete this questionnaire. Thank you very much for your cooperation! Basic personal and family information: Gender: Age: Education background:

◎Not educated

◎Compulsory education

◎High school (junior college)

Health condition:

◎Healthy

◎Mildly disabled

◎Severely disabled

Marital status:

◎Unmarried

◎Married

Children under 12 years old:

◎Yes

◎No

Are there any casualties:

◎Yes

◎No

Is there property damage:

◎Yes

◎No

Number of storeys: ◎1–2 floors

◎Bachelor’s (junior college) degree or above

◎More than 2 floors

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

(continued) 419

420

Appendix 6: Questionnaire for the Study of the Impact of Built Environment or Risk …

(continued) Property type:

◎Detached

◎Non-detached

Built Environment (1. Totally disagree 2. Disagree 3. Neutral 4. Agree 5. Totally agree): The terrain of the current residence has a great influence on the resistance of the 1 2 3 4 5 house to earthquake disasters. The infrastructure of the existing residential area is well planned.

1 2 3 4 5

The area where I live is convenient for emergency evacuation.

1 2 3 4 5

There are reasonable evacuation shelters where I live now.

1 2 3 4 5

Reasonable spacing is conducive to escape and evacuation during earthquakes.

1 2 3 4 5

The house I live in now is earthquake resistant.

1 2 3 4 5

The house is designed with reasonable emergency shelters.

1 2 3 4 5

The interior of the house is designed with reasonable emergency escape ways.

1 2 3 4 5

Quality of construction materials is guaranteed.

1 2 3 4 5

There are good roads connecting villages and towns.

1 2 3 4 5

Roads connecting villages and towns are not prone to break down in the event of 1 2 3 4 5 an earthquake and traffic jams

Would you choose to evacuate the building at the first time when an earthquake occurs? Yes No Risk perception (1. Not at all 2. No 3. Just so-so 4. Yes 5. Absolutely right): I’m sensitive to something shaking.

1

2

3

4

5

I think this area is prone to earthquakes.

1

2

3

4

5

The risk of serious earthquakes in the future will be greater.

1

2

3

4

5

I believe I will be directly affected.

1

2

3

4

5

I think an extreme earthquake will have long-term negative effects.

1

2

3

4

5

Earthquakes are catastrophic.

1

2

3

4

5

After an earthquake, I will always be alert.

1

2

3

4

5

After an earthquake, I feel aftershocks all the time.

1

2

3

4

5

The number of earthquake disasters has been reduced.

1

2

3

4

5

Now, earthquakes don’t have a devastating effect on houses.

1

2

3

4

5

The earthquake has a relatively serious impact on me.

1

2

3

4

5

The earthquake has a relatively serious impact on my family.

1

2

3

4

5

I’m not so scared of earthquakes now.

1

2

3

4

5

Minor earthquakes can cause major damage to my home.

1

2

3

4

5

When it shakes violently it can cause significant damage to my home.

1

2

3

4

5

Appendix 7

Disaster Preparedness Behavior in Flood-prone Areas

Dear residents, Greetings! We are the research team from the Engineering and Technical College of Chengdu University of Technology. In order to explore more scientifically the factors influencing farmers’ experiences and attitudes on disaster preparedness behavior in flood-prone areas, we are conducting a survey in the form of an anonymous research questionnaire, and your truthful and objective answers are crucial to our research. This study is for scientific research only, and will not involve or disclose any of your personal information. We sincerely hope that you can cooperate with the researchers to complete this questionnaire. Thank you very much for your cooperation! Q1 Personal information SN.

Questions

Options

Q1 Please complete the following questions. Q1.1

Gender

_____

Q1.2

Age

_____

Q1.3

Education background A Not educated B Primary school C Junior high school D Senior high school or technical secondary school E Junior college F Undergraduate and above

Q1.4

Types of profession A Government or public institutions staff B State-owned enterprise employee C Private enterprise employee D Self-employed businessman E Freelancer F Farmer G Unemployed

Q1.5

Monthly income

_____RMB

Q1.6

Years of residency

_____years

Q1.7

Family size

_____people

Q1.8

Number of income-earners in the family

_____people

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2020 Y. Ao and I. Martek, Rural Built Environment of Sichuan Province, China, https://doi.org/10.1007/978-981-33-4217-0

(continued) 421

422

Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas

(continued) Q1.9

Annual household income

_____ (unit: 10,000 RMB)

Q1.10

Is there anyone ill in your family?

A Yes B No

Q1.11

Building area

_____m2

Q1.12

Number of storeys

_____floors

Q1.13

Is there any home built gutter?

A Yes B No

Q1.14

The distance between your home and the river

____KM

Q2 Please select the most appropriate item according to your actual situation. Q2.1

How many floods have you experienced within a year?

A 0 B 1–2 C 3–4 D 5–6 E 7 times and above

Q2.2

What’s the longest duration of flood A 1–2 days B 3–4 days C 5–6 days D 7–8 you have experienced? days E 9 days and above

Q2.3

What’s the maximum number of casualties in a flood you have ever experienced?

A 0 B 1–2 C 3–4 D 4–5 E 6 people and above

Q2.4

Have you ever been helped by the community (village) during the flood?

Yes



No



Q2.5

Have you ever been helped by your neighbors during the flood?

Yes



No



Q2.6

Have you ever been helped by your neighbors after the flood?

Yes



No



Q2.7

Have you ever been helped by your neighbors after the flood?

Yes



No



Q2.8

After the flood, is the residential environment naturally restored?

Yes



No



Q2.9

In the past floods, has the residence been completely surrounded by flood?

Yes



No



Q2.10

In the past floods, has there been any road destroyed due to flood?

Yes



No



Q2.11

In the past floods, has water supply been cut during the flood?

Yes



No



Q2.12

In the past floods, has power supply Yes been cut during the flood?



No

 (continued)

Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas

423

(continued) Q2.13

In the past floods, has natural gas supply been cut during the flood?

Yes



No



Q2.14

In the past floods, has Yes telecommunication been interrupted during the flood?



No



Q2.15

Has your home been badly damaged Yes due to flood?



No



Q3 Please select the most appropriate item according to your actual situation: Absolutely No

Not likely

Not sure

Likely

Absolutely Yes

Q3.1

Do you think there is a  possibility of flooding in the next five years in this area?









Q3.2

Do you think there is a  possibility that the road will be destroyed by flood in the next five years?









Q3.3

Do you think there is a possibility that water supply will be cut by flood in the next five years?











Q3.4

Do you think there is a possibility that power supply will be cut by flood in the next five years?











Q3.5

Do you think there is a  possibility that natural gas supply will be cut by flood in the next five years?









Q3.6

Do you think there is a  possibility that telecommunication will be disrupted by flood in the next five years?









Q3.7

Do you think there is a possibility that you home will be badly destroyed by flood in the next five years?











(continued)

424

Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas

(continued) Q3.8

Do you think there is a possibility that you or your relatives will be injured by flood in the next five years?











Q3.9

Does the river near  your home have a flood control dam?









Q3.10

Does the flood control dam near your home work?











Q3.11

Can the ditch near your  home drain smoothly?









Q3.12

Are the gutters near your house leaking?











Q3.13

Can the ditch near your  home play its role in flood discharge?









Q3.14

Would you like to learn  about flood preparedness?









Q3.15

Would you like to gather some information on disaster preparedness?











Q3.16

Would you like to purchase flood insurance?











Q3.17

Would you like to put your flashlight and radio in a convenient place for a flood disaster?











Q3.18

Would you like to provide the emergency rescue telephone number of flood disaster?











Q3.19

Would you like to teach  or arrange for relatives to deal with flood emergencies?









Q3.20

Would you like to participate in flood emergency training?











(continued)

Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas

425

(continued) Q3.21

Would you like to ask someone/government what measures should be taken in case of flood emergency?











Q3.22

Would you like to keep important items in a safe place?











Q3.23

Would you like to store  emergency food and water for flood?









Q3.24

Are you confident that the local flood protection facilities have been maintained?











Q3.25

Are you confident about the flood control capacity of your village?











Q3.26

Are you confident  about the ability of people responsible for flood risk (village head or secretary of the Party committee) in your village?









Q3.27

Do you believe some of  the news reported on the internet about the floods is true?









Q3.28

Do you believe some  news about the flood on TV is true?









Q3.29

Do you believe some  news about the flood on the radio is true?









Q3.30

Do you believe some news about the flood in the newspaper is true?











Q3.31

Do you feel fear when there’s a flood?











Q3.32

Do you feel helpless when there’s a flood?











Q3.33

Do you feel indifferent when there’s a flood?











Q3.34

Do you get angry when  there’s a flood?







 (continued)

426

Appendix 7: Disaster Preparedness Behavior in Flood-prone Areas

(continued) Q3.35

Do you stay alert in case of a flood?











Q4 Which of the following flood preparedness measures do you usually do? √ (Tick the blank if it’s “Yes” ) Q4.1

Buy flood insurance.

Yes  No 

Q4.2

Put your flashlight and radio in a convenient place for a flood disaster.

Yes  No 

Q4.3

Put the radio in a convenient place so that you can listen to the latest Yes  No  news of the flood at any time.

Q4.4

Remember the emergency phone numbers.

Yes  No 

Q4.5

Train your relatives to deal with floods in normal times.

Yes  No 

Q4.6

Attend flood emergency training.

Yes  No 

Q4.7

Ask others (the government) for emergency measures for flood disaster.

Yes  No 

Q4.8

Keep your important items at home in a high place to protect against Yes  No  floods.

Q4.9

Store emergency food and drinking water to protect against floods in Yes  No  normal times.

Q4.10 Increase the height of your house to protect against floods.

Yes  No 

Q4.11 Change the planting date of crops to protect against floods.

Yes  No 

Q4.12 Increase agricultural irrigation measures to protect against floods.

Yes  No 

Q5 Which of the following flood preparedness measures do you usually do? √ (Tick the blank if it’s “Yes” ) Q4.13 Apply soil protection technology to prevent the adverse effect of flood Yes  No  on soil. Q4.14 Improve drainage measures to prevent flood.

Yes  No