Ethnicity and Fertility in Indonesia 9789814376068

Table of contents :
Contents
List Of Tables
List Of Figures
Acknowledgements
Preface
I. Introduction
II. Methodology
III. The Nature Of Ethnic Differences
IV. Ethnicity Differences In Fertility
V. Explaining Differences In Fertility
VI. Conclusions
Appendix
Bibliography

Citation preview

Ethnicity and Fertility in Southeast Asia Series General Editors: Aline K. Wong and Ng Shui Meng

ETHNICITY AND FERTILITY IN INDONESIA

by

Mely G. Tan and Budi Soeradji

Research Notes and Discussions Paper No. 53 INSTITUTE OF SOUTHEAST ASIAN STUDIES 1986

I5EA5 Institute of Southeast Asian Studies The Institute of Southeast Asian Studies was established as an autonomous organization in May 1968. It is a regional research centre for scholars and other specialists concerned with modern Southeast Asia. The Institute's research interest is focused on the many-faceted problems of development and modernization, and political and social change in Southeast Asia. The Institute is governed by a twenty-two-member Board of Trustees on which are representatives from the National University of Singapore, appointees from the government, as well as representatives from a broad range of professional and civic organizations and groups. A ten-man Executive Committee oversees day-to-day operations; it is chaired by the Director, the Institute's chief academic and administrative officer. The responsibility for facts and opinions expressed in this publication rests exclusively with the author and his interpretations do not necessarily reflect the views or the policy of the Institute or its supporters.

Published by Institute of Southeast Asian Studies Hang Mul Keng Terrace Pas I r Panjang SIngapore 0511 All rights reserved, No part of this publication may be reproduced, stored In a retrieval system, or transmitted In any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the Institute of Southeast Asian Studies, ~

1985 Institute of Southeast Asian Studies

ISSN 0129-8828 ISBN 9971-988-24-0

CONTENTS

Page LIST OF TABLES

vi

LIST OF FIGURES

ix

ACKNOWLEDGEMENTS

X

PREFACE

II

xi

INTRODUCTION

1

Purpose of the Study

1

Theoretical Background

5

Research Organization

10

Review of the Literature

11

Organization of the Report

13

METHODOLOGY

15

The Setting

15

Research Areas

18

iii

Page

III

IV

Sampling Procedure

24

Eligibility Criteria

25

Strati fi cation

26

Rural Sampling Design

26

Urban Sampling Design

26

The Questionnaire

27

Translation Problems

28

Questionnaire Checking

29

Code Book

29

Coding

29

Data Cleaning

30

THE NATURE OF ETHNIC DIFFERENCES

31

Contrasts among Ethnic Groups on Ethnic Characteristics

31

Contrasts on Demographic and Socio-economic Characteristics

52

ETHNICITY DIFFERENCES IN FERTILITY

73

Fertility and Fertility Preferences

73

On Family Planning

75

Accessibility of Birth Control Methods by Ethnic Group

83

Attitude towards Family Planning

88

Natural Fertility Factors

89

Values Attached to Children

92

iv

Page

V

VI

Disvalues Attached to Children

93

Marriage and Marriage Attitudes

102

Attitude towards Divorce and Remarriage

109

Interaction with Spouse and Sex Role Attitudes

112

EXPLAINING DIFFERENCES IN FERTILITY

119

Some Notes on the Techniques

120

Analysis of Children Ever Born

120

Analysis of Recent Fertility

126

Analysis of Use of Effective Methods of Contraception

130

CONCLUSIONS

135

APPENDIX

137

BIBLIOGRAPHY

142

V

LIST OF TABLES

Page 1

Characteristics of Ethnic Groups Studied

8

2.1

Population of the Five Provinces in 1980 and Growth Rate, 1961-71 and 1971-80

16

Population of the Four Cities in 1980 and Percentage Increase 1971-80

16

Number and Increase of Family Planning Clinics in the Five Provinces, 1971-72, 1974-75 and 1981-82

17

Number and Percentage of Current Contraceptive Users in the Five Provinces March 1982

17

Ethnic Identification, Parentage and Relative Importance of Ethnic versus National Identification, by Ethnic Group

34

3.2

Language Usage by Ethnic Group

38

3.3

Religion and Religiosity by Ethnic Group

42

3.4

Ethnic Affiliations by Ethnic Group

44

3.5

Ethnic Attitudes of Respondents from Each Ethnic Group

48

Demographic Characteristics of Respondents from Each Ethnic Group

54

2.2 2.3

2.4

3.1

3.6

vi

Page 3.7

Education, Media Exposure, and Efficacy among Respondents from Each Ethnic Group

56

Income, Assets, and Transfers of Economic Resources among Respondents from Each Ethnic Group

60

3.9 Employment Characteristics of Respondents from Each ,Ethnic Group

66

3.8

4.1 4.2

Fertility and Fertility Preferences among Respondents from Each Ethnic Group (Means)

76

Family Planning (Birth Control) by Ethnic Group

78

4.3 Accessibility of Birth Control Methods among Respondents from Each Ethnic Group

84

4.4 Attitudes towards Birth Control among Respondents from Each Ethnic Group

86

4.5 4.6 4.7

Natural Fertility Factors among Respondents from Each Ethnic Group

90

Values Attached to Children among Respondents from Each Ethnic Group

94

Disvalues Attached to Children by Ethnic Group

98

4.8 Marriage and Marriage Attitudes by Ethnic Group

104

4.9 Attitudes towards Divorce and Remarriage among Respondents from Each Ethnic Group

110

4.10 Interaction with Spouse and Sex Role Attitudes by Ethnic Group

114

5.1

Mean Number of Children Ever Born (CEB) for Selected Areas, 1980

5.2 Mean Number of Children Ever Born {CEB) by Ethnic Group Adjusting for Different Sets of Predictors vii

121

124

Page 5.3 Recent Fertility by Ethnic Group Adjusting for Different Sets of Predictors (Female respondents who have married at least five years)

128

5.4 Mean Score for Use·of Efficient Contraceptive Method by Ethnic Group Adjusting for Different Sets of Predictors (Female respondents)

132

viii

LIST OF FIGURES

Page 1

2

A Schematic Framework for Studying Differentials in Fertility

6

A Schematic Framework for Studying Differentials in Use of Contraception

7

ix

ACKNOWLEDGEMENTS

This study started out with three co-investi gators, but mid-way through the fieldwork phase Drs Amri Marzali M.A., left for Australia to continue his studies. We want to thank him for his valuable contribution as the anthropologist of the group. The field work was done with the assistance of regional co-ordinators and their team of interviewers. We are grateful for the invaluable work done by the following people: for Jakarta Drs Amri Marzali M.A. of the University of Indonesia and his team; for West Java Drs A.D. Saefullah M.A. of the Population Studies Center of Padjadjaran University, Bandung, and his team; for East Java Mr S. Soemantri M.Sc. of the Research and Development Center of the Department of Health of the Republic of Indonesia, Surabaya, and his team; for West Sumatra Drs Syahrudd in M. A. of the Faculty of Economics of And a 1as University, Padang, and his team; for North Sumatra Or Amudi Pasaribu, Rector of Nommensen University and field co-ordinators Mr O.H.S. Purba M.A., M.Sc., and Drs F.T. Rajagukguk and their team. Finally, as this study was undertaken in an agreement between the National Institute of Economic and Social Research of the Indonesian Institute of Se i ence and the Human Reproduction Programme of the World Health Organization in Geneva, Switzerland, we would like to extend our gratitude to the latter institute for its financial assistance to this study. Mel y G. Tan, Ph .D. Budi Soeradji, Ph.D. Indonesia

X

PREFACE

The Ethnicity and Fertility in Southeast Asia Project that commenced in 1980, was an outgrowth of an earlier project, Culture and Fertility in Southeast Asia, which was completed in 1979. Building upon the results of the earlier study, which established that ethnicity was a significant factor underlying the fertility differentials among the various ethnic groups in Southeast Asia, the present project aimed to explore in greater detail the extent to which ethnicity and ethnic factors like ethnic attitudes, ethnic identification and cultural practices influenced reproductive behaviour. Instead of utilizing secondary sources, the project relied on primary data collected through the survey technique. In all, twenty ethnic groups from the five ASEAN countries were surveyed in this study which spanned a total of three years. A study involving five different countries and so many ethnic groups of diverse cultural and religious backgrounds would invariably pose problems of comparability. To maximize comparability across countries, the study relied on the use of a common core questionnaire as well as a common analytical framework and data analysis procedures. While comparability was important, the incorporation of country-specific factors sa 1 i ent and relevant to the explanation of fertility behaviour was also encouraged. The final research design therefore attempted to be as comprehensive as possible in the exploration of the ethnic dimension in fertility differentials among the various ethnic groups studied. Three workshops were held during the period of the project to enable the researchers to come together to discuss and resolve problems related to the project. The first workshop was held in May 1980. At this workshop the conceptual framework and the core questionnaire were finalized. In the second workshop held in June 1981, the methods of data analysis were decided. At the xi

final workshop in September 1982, the country teams presented their preliminary findings. The final reports were completed by December 1983. A study of this scale obviously also involved many researchers. The researchers were all Southeast Asian social scientists drawn from various disciplines and backgrounds. Some were attached to universities of the region while others were from research institutes or government agencies. Dr Rodolfo A. Bulatao from the East-West Population Institute provided the initial intellectual impetus to the project by formulating the conceptual framework and research design for the study as well as the drafting of the preliminary questionnaire. In addition Dr Bul atao together with Dr Aline K. Wong from the National University of Singapore {NUS) and Dr Ng Shui Meng from !SEAS served as co-ordinators of the project. The country teams consisted of: Indonesia:

Dr Mely Tan {National Institute of Economic and Social Research of the Indonesian Institute of Sciences, LEKNAS-LIPI) Dr Budi Soeradji {Central Bureau of Statistics) Mr Amri Marzal i (Faculty of Letters, University of Indonesia)

Malaysia:

Datin Dr Noor Laily Abu Bakar Family Planning Board, NFPB)

(Malaysia

National

Dr Tan Boon Ann {NFPB) Mr Tey Nai Peng {NFPB) Mr Hew Wai Sin {NFPB) Ms Aminah Abdul Rahman {NFPB) Ms Ramlah Haji Muda (NFPB) Ms Nazi 1eh Raml i {NFPB) Mr Khal i pah Mohd Tora {NFPB) Mr Ng Tuck Seng {NFPB) Philippines:

Ms Pilar Ramos-Jimenez Council, PSSC) xii

(Philippine

Social

Science

Ms Ma. Cecilia Gastardo-Conaco (University of the Philippines) Ms Lorna Makil (PSSC) Ms Ruth N. Barniego (PSSC) Singapore:

Dr Eddie C.Y. Singapore, NUS)

Kuo

(National

University

of

Dr Chiew Seen-Kong (NUS) Thai 1and:

Dr Suchart Prasithrathsint (National Development Administration, NIDA)

Institute of

Dr Suwanlee Piampiti (NIDA) Mr Thawatchai Arthorn-thurasook (Mahidol University) Dr Laddawan Rodmanee (Mahidol University) Dr Luechai Chulasai (Chiangmai University) Ms Suranya Bunnag (Prince of Songkla University) Ms Amporn Chareoncha i ( Khon Kaen University) Funding for the project was provided by the International Development Research Centre (IDRC), Rockefeller Foundation and the Human Reproduction Programme, Task Force on Psychosocial Research in Family Planning of the World Health Organization (WHO). The results of the monographs in this series: Ethnicity Analysis

and

Fertility

study in

are

being

Southeast

Ethnicity and Fertility in Indonesia Ethnicity and Fert i1 i ty in Malaysia Ethnicity and Fertility in the Phi 1i ppi nes

xiii

published

Asia:

in

six

A Comparative

Ethnicity and Fertility in Singapore Ethnicity and Fertility in Thailand ALINE WONG and NG SHUI MENG Project Co-ordinators and General Editors of the Series

xiv

INTRODUCTION

Purpose of the Study This study is part of a larger study involving the five ASEAN countries: Indonesia, Malaysia, the Philippines, Singapore and Thailand. It forms the second of a two-phase project on the study of culture and fertility. The first phase of the project (1977-79) involved the analysis of secondary data. The findings of the study are published under the title Culture and Ferti 1 ity: The Case of Indonesia .1 The second ph a se of the prOJect, whl eh 1s the present study, is based on analysis of primary data. The project which began in mid-1980 spanned a period of more than two years. This study purports to look into the nature, causes and implications of ethnic differentials in fertility, whereby ethnicity is understood in broad terms to cover identification of individuals by race, religion, nationality, cultural traditions, and any other "primordial" characteristics, which together di st i ngui sh one ethnic group from another. The focus on ethnicity in these five countries is quite pertinent as they are all multi-ethnic and multi-religious societies. The pluralistic nature of the Indonesian society is clearly evident in the existence of about 300 ethnic groups (including groups of foreign descent, such as the ethnic Chinese, the Eurasians, and people of Arab and Indian parentage) speaking about 250 languages. They profess adherence to all the world religions Islam, Christianity, Hinduism, Buddhism, also Confucianism and a host of indigenous belief systems, which are offici a 11 y referred to as ke ercayaan ke ad a Tuhan an Maha Esa (belief in the Almighty Go en, t ere 1s t e var1at1on 1n kinship system (patrilineal among the Batak; matrilineal among the Minangkabau; and bilateral among virtually all the other ethnic groups), in marriage patterns, and in the legal systems 1

(national law).

law,

customary

adat

law,

and

religious

[Moslem]

From the beginning of the proc 1 am at ion of the re pub 1 i c on 17 August 1945, when Indonesia was established as a nation state, it carried the motto Bhinnek Tunggal Ika or Unity in Diversity. However, the position and avowed pol 1cy of the government has always been to emphasize the unity rather than the diversity aspect, as expressed for example, in the oft-repeated phrase membi na kesatuan dan ersatuan ban sa (to promote/guide the unity o t e nat10n • onet e ess, t e iversity aspect is a social fact, that has to be (and sometimes is) recognized, especially in the execution of national programmes, such as in education, health, food and nutrition, and family planning. The focus on ethnicity is also pertinent in view of the fact that the family planning programme, when it was implemented in 1970, was confined only to the six most populous provinces on the islands of Java and Bal i. Si nee 1973 it has been expanded to include ten more provinces: in DI Sumatra, Aceh, West and South Sumatra, Lampung, and North and South Sulawesi, South and West Kalimantan, and West Nusa Tenggara. In 1979 six more provinces: Riau and Jambi in Sumatra, Central and East Kalimantan, Irian Jay a and East Ti mor began to receive family planning services. 2 All these provinces are dominated by a particular ethnic group with its specific cultural characteristics and therefore may have different responses to different national programmes. The population of Indonesia (147,490,298 according to the 1980 Census) is the fifth largest in the world and the third largest in the Third World. This fact alone warrants an intensive scrutiny and understanding of Indonesia's population characteristics, in particular, its ethnic dimension. Although there exists a fair amount of population-related studies, there are few that focus directly on the ethnic factor. Today, it is generally recognized that the main population-related problems in Indonesia are related to its size and continuing rapid growth (the average rate of growth from 1930-61 is 1.5 per cent; 1961-71, 2.1 per cent; and 1971-80, 2.3 per cent), the unequal distribution of its population (two-thirds of the population live on 7 per cent of the land area), the young age structure {about 42 per cent is under fifteen years of age), and the continuing flow of people to the cities.3 It is also shown that there is regional variation in growth by island and by province. Of the major islands, Sumatra has the highest rate of growth followed by Kalimantan, Maluku, Irian Jaya, Sulawesi, Java and Nusa Tenggara {which includes Bali). The twenty-seven provinces also show a striking variation in rate 2

of growth from 1971-80, from 5. 77 per cent for Lampung (southern part of Sumatra) to 1.10 per cent for DI Yogyakarta (Central Java). Three of the provinces with more than 4 per cent growth are located in Sumatra: Lampung with 5.77 per cent, Bengkulu with 4.39 per cent and Jambi with 4.07 per cent (the second highest is East Kalimantan with 5.73 per cent), while three of the six provinces with less than 2 per cent growth are all located in Java: Central Java with 1.64 per cent, East Java with 1.49 per cent and DI Yogyakarta with 1.10 per cent. Of the five provinces included in the present study, DKI Jakarta ranks fifth in growth, West Java fourteenth, North Sumatra fifteenth, West Sumatra nineteenth and East Java twenty-fifth.4 As indicated earlier, in all provinces, except for DKI Jakarta, one can identify a dominant ethnic group. For the four provinces included in this study, the Sundanese dominate in West Java, the Javanese in East Java, the Batak in North Sumatra and the Minangkabau in West Sumatra. As none of the censuses taken after the formation of the republic provide information on ethnicity, we have no way of knowing the exact size of these various ethnic groups. The only census which has this data is the prewar census. Until today, the proportions as enumerated in that census are used as an approximation of the distribution of the major ethnic groups, as follows: Javanese 45 per cent, Sundanese 14.2 per cent, Madu re se 7. 5 per cent, Coast a 1 Ma 1 ays (Sumatra and Kalimantan) 7.5 per cent, Macassarese-Buginese 4.2 per cent, Mi nangkabau 3. 3 per cent, Ba 1 i ne se 2 per cent, and Batak 1.7 per cent.5 Another approximation which to a certain extent corroborates the 1930 figures is provided by the result to the question in the 1980 Census on 1anguage used at home. The percentages on 1anguage used are as follows: Javanese 40.1 per cent, Sundanese 15.2 per cent, Madurese 4.8 per cent, Batak 2.1 per cent, Minangkabau 2.5 per cent, Balinese 2.10 per cent, Buginese 1.9 per cent, Banjarese 1.3 per cent, Indonesian 12 per cent, Not stated 0.1 per cent and Other 17.8 per cent.6 Still another approximation that is often used is to take the provincial data for the rural area as a proxy for information on the major ethnic group of that particular province. For the Chinese it is even more difficult to get an accurate picture of their size. We have to go back to the 1930 Census 7 too, where it was recorded that there were then 1, 233,214 ethnic Chinese or 2 per cent of the total population. Based on these figures, it was calculated that approximately 3 million or 2.4 per cent of the total population in Indonesia were Chinese by 1976.8 If we assume a growth rate close to the national figure, by 1980 there should have been a 10 per cent increase, making the tot a 1 Chi ne se population at about 3 million. This i ne 1 udes both There is no direct as alien Chinese. citizens as well information on the citizens of Chinese descent, but there is

3

information on the aliens, as they are separately registered at the Immigration Office. According to the census there are about half a million alien Chinese, which makes the number of the citizens approximately 2.8 million, which is less than 2 per cent of the total population. The reasons for the selection of the five ethnic groups is rather obvious in the case of the Javanese and the Sundanese, as The Minangkabau and the Batak, they are the largest groups. though smaller in size, still rank as the fourth and fifth In addition, they both have largest groups in the population. different kinship systems from the majority of the peoples in Indonesia, that is, matrilineal among the Minangkabau and patrilineal for the Batak. Also, they are both very mobile and very dynamic, and they seem more resistant than the other groups The inclusion of the ethnic towards family planning efforts. in all the countries included in Chinese has several reasons: the project (except for Singapore}, there is a so-called "Chinese problem", that is often manifested in acts of violence towards the group whenever there is an outbreak of unrest in the country or parts of the country. This is basi ea 11 y because there is a latent enmity against this group that has its source in the hi story of their re l at i onshi p with the local population and their continuing dominant role in the economy of these various In Indonesia, since the beginning of the family countries. planning programme, accusations have been levelled at the Chinese as being least receptive to the programme even when they are perceived to have too many children. Focusing on the population problem in general and the efforts to control it, there are indications that despite the continuing growth, the process of demographic transition is starting in Indonesia, especially on the island of Java. Over the past fifteen years, fertility has been falling significantly, the CBR (Crude Birth Rate}, which in the early 1960s was still between 40 and 45 per 1,000, is now estimated to be as low as 35 per 1,000. Measured in terms of total fertility it is estimated 1967-70, 5.6; 1970-75, 5.1; and to have declined as follows: 1976-79, 4.7. Thus between 1967-79 there has been a decline of total fertility by about 16 per cent over a period of twelve years.9 These changes ea 11 for an exp 1 an at ion of the factors whi eh To what extent was fertility might have brought this about. decline the contribution of the family planning programme; and to what extent was it the result of institutional changes, such as the expansion of schools, health programmes and other development programmes? As McNicoll and Singarimbun observe, there has been a tendency to attribute fertility decline with "simplistic" explanations, when in fact one should look for more refined 4

explanations in the socio-economic and socio-cultural area that are certain to affect fertility behaviour.10 The present study, with its special focus on the ethnic factor, is an endeavour to come up with exp 1an at ions that will give insight into the variations in fertility behaviour by ethnic group that is already shown in the demographic variables by province. The questions that are to be examined include the nature and dynamics of these variations, as well as the determinants and implications of these variations within the context of the total family planning effort in Indonesia. An understanding of these variations should help in determining policies that are demographically most effective and socially and culturally acceptable by the various groups involved.

Theoretical Background The five participating countries in this study have used the same theoret i ea 1 framework as agreed upon in the preparatory meetings of the project.ll Basically, this theoretical framework or model attempts to integrate several major perspectives which serves to guide the present research. These perspectives are: ( 1) the assimilationist perspective, which explains ethnic differentials in fertility in terms of individual background characteristics and which predicts the gradual disappearance of ethnic differentials as ethnic groups become assimilated into the majority culture; (2) the structural perspective, which explains ethnic differentials in fertility as varying with the relative positions of ethnic groups in a societal stratification system, depending on the minority or majority status of the ethnic groups; (3) the cultural perspective, which attributes ethnic fertility differentials to differences in cultural values and group practices affecting fertility; and (4) the psychosocial perspective, which supplements the cultural perspective by considering i ndi vi dua 1 attitudes and preferences with regard to childbearing. Initially all four perspectives were integrated in a theoretical framework, but after further deliberations, the structural perspective was de-emphasized, as the investigators decided that there was not enough basis for including the minority-majority status variable. Eventually, the following models were developed, focusing on Fertility (Figure 1) and Use of Contraception (Figure 2) as the dependent variables. As shown in Figure 1, the relationships among variables are conceptua 1 i zed by first making a di st i net ion between proximate

5

FIGURE I A Schematic Framework tor Studying Differentials In Fertility Nuptial ity

"Character I st I cs 11 Age/Sex Education Income

Occupation Female employment Urban res I dance

Natura I Fert I 11 ty Factors Lactatlonal lnfecundabll ity Star 11 I ty (perceIved fecundIty) Spontaneous abortion

Ethnic Identity Ethn le atf I I I at ion Ethnic tertii lty 1------+-------f'\--------l,------------------~

approva I

·~·

Rei lgioslty

Gender preferences Status of women

Menstrual taboos

Psvchosoc Ia I Factors Fertility preferences Coup I e decIsIon processes Contraceptive access I blllty PsychIc costs of contraceptIon

FIGURE 2 A Schematic Framework for Studying Differentials In Use of Contraception

"CharacterIstIcs"

~

Age/Sex Education

Age at marriage Number of marriages

income Occupation Female employment Urban resIdence Mad I a exposure

Natura I Fert I I ity Factors lactatlonal infecundabi I lty Ster I I I ty ( perce 1ved fecundity)

Spontaneous abort I on

Ethnic Identity Ethnic affi llation Ethnic ferti I ity

approva I

·~·

Rei igiosity Gender preferences

Status of women Remarr 1age of wIdows Menstrua I taboos

I

Values of children Coup I e decIsion processes Contraceptive accesslbl I ity

Psych le costs ot contraception

TABLE 1 Characteristics of Ethnic Groups Studied

1.

Population size of ethnic groups No direct figures available. Approxlmatatlons can be calculated from percentages of the 1930 Census and/or from the response on the question on language used at home of the 1980 Census.

2.

Percentage of national population

1930 Census % Javanese Sundanese Madurese Coastal Malays Macassar/Buglnese Mlnangkabau Ballnese Batak Banjarese Indonesian Not stated Other Chinese 3.

"Native" l;mguage Javanese: Sundanese: Mlnangkabau: Batak: Chinese:

45.0 14.2 7.5 7.5 4.2 3.3 2.0 1.7

2.0

1980 Census

( 51>

(288)

(

50)

TABLE 4.2 Family Planning (Birth Control) by Ethnic Group Javanese Rural F

Percentage Who Have Heard of Birth Control 73.4 (207) Percentage Currently Using Birth Control

Percentage Who Have Ever Used Birth Control

Percentage of Current Non-users Who Plan to Use In the Future

Percentage of Current Users Using Each Method PI 11 IUD Diaphragm Foam, jelly, cream Douche Condom Rhythm Withdrawal Abstention Female sterll lzatlon Male sterilization lnjectables Others

Sundanese Urban

M

F

Rural M

F

Urban M

F

M

68.3 60)

83.4 (301)

74.7 87)

76.8 (198)

81.4 43)

82.7 (352)

88.7 53)

29.8 ( 171)

28.6 (49)

59.9 (222)

56.3 (64)

31.4 ( 153)

31.4 (35)

47.2 (271)

43.9 (41)

48.3 (207)

35.0 60)

68.4 (301)

57.5 87)

41.4 ( 198)

39.5 43)

55.4 (352)

52.8 53)

53.6 ( 166)

53.2 (47)

64.9 ( 188)

69.5 (59)

54.2 ( 144)

58.8 (34)

62.6 (243)

52.8 (36)

56.9 23.5

21.4 7.1

21.8 37.6 o.8

33.3 13.9

45.8 20.8 2.1

54.5 27.3

32.0 28.1 0.8

27.8 16.7 5.6

3.9 3.9

14.3 7.1

8.3 10.5

11. I

9.1

3.9

14.3

1.5 3.8

5.6 13.9

0.8 15.0 (133)

11. I

7.0 7.0 0.8 0.8 3.1 0.8 2.3 17.2 ( 128)

11. I

4.2

8.3

4.2

7.1 2.0 5.9 (51)

28.6 (14)

9.1

2.8 (36)

22.9 (48)

( 11)

11.1

5.6

22.2 ( 18)

TABLE 4.2 (cont 1 d)

Batak

Mlnangkabau

F

M

F

F

M

M

91.4 (291)

92.2 51)

87.5 (288)

86.0 50)

58.6 (222)

66.0 (47)

11.7 ( 137)

13.9 (36)

62.6

84.5 (252)

94.6 (184)

15.50 (175)

11.82 (211)

13.78 ( 174)

9.70 ( 199)

99.4

1.42 1.71 (52) (291)

2.16 2.07 (51) (283)

90.9 (274)

61.7 (269)

( 170)

9.50 (247)

8.45 ( 170)

10.08 (169)

7.75 (236)

7.25 (146)

11.11

1.76 (50)

woman to recover from child birth and to get the uterus back into position. As in most agricultural societies, breast-feeding, especially in rural areas, is a universal phenomenon. This is indicated by the fact that in all groups, the majority of women breast-feed their children (ranging from 61.7 per cent among Chinese women, to 99.4 per cent among rural Minangkabau women). The duration is also quite long especially in rural areas. It varies from a mean of 8. 45 months among Chi ne se women to 18.42 months or more than one year among rural Javanese women. We should be aware, though, that with this long duration, often there is no or little milk, and suckling at the breast has the function more of a pacifier rather than feeding.

Values Attached to Children* To elicit responses to the questions on advantages of having children needed some explaining and probing. It turned out that the respondents were more able to indicate tangible advantages, than the more intangible ones such as psychological rewards. Among the Javanese and the Sundanese, the emphasis is on financial, practical help, especially among the rural respondents. Among the urban respondents the highest proportions are for financial support, practical help and for family continuity, the latter especially among the males (34.1 per cent among the Javanese males and 42.3 per cent among the Sundanese males). Among the Minangkabau too, in all groups there is almost an equal proportion indicating financial support, practical help and family continuity (for the former ranging from 22.7 per cent to 27.5 per cent and for the latter ranging from 14.9 per cent to 37.2 per cent). Among the Batak, on the other hand, the highest proportion is for family continuity (ranging from 42.3 to 62 per cent). Among the Chi ne se, there is again an a 1most equa 1 proportion for financial support, practical help and for family continuity. The high proportion for family continuity among the Batak is very striking and clearly in line with the importance placed upon kinship or the clan system in this ethnic group. A similar situation is found among the Minangkabau, though the emphasis is less. Rather surprisingly, "help in old age" is not a very important value for having children except among the rural Javanese (20.1 per cent for the female and 23.3 per cent for the males).

*

This section refers to Table 4.6.

92

In line with the above, we find that the rural Javanese have the highest proportion indicate they expect children to be a major source of economic support in their old age (32.5 per cent among females and 20 per cent among the males). In the other groups, expectation is less (10 per cent or less), except for the rural female Sundanese (19.2 per cent). But, in all these other groups the proportions indicating that they expect some minor help is about 90 per cent and more. Most groups also have a rather positive view of the future of their children: in a 11 groups, except for the Sundanese, 90 per cent and more indicate that they expect life for their children to be better. For the Sundanese, the figures are somewhat lower; ranging from 78 per cent to 89.8 per cent.

Disvalues Attached to Children* Similar problems are involved in eliciting responses to disvalues attached to children. But with probing, such disvalues as financial disadvantages, childrearing demands (physical and emotional stress), and restrictions on parents (tied down), are mentioned most frequently. There seems to be some variation by ethnic group, in that the Javanese, Sundanese and Chinese stress more the childrearing demands and restrictions on parents, while the Minangkabau and the Batak, more the financial disadvantagaes. Among the latter two groups, close to 50 per cent or more than 50 per cent of the respondents consider the financial costs of childbearing the greatest disvalues attached to children. Looking at the Javanese and the Sundanese only, it is interesting to note that the emphasis on physical and emotional stress (that is, "emotional costs"), is in line with the finding of the va 1 ue of chi 1d ren study on these two ethnic groups where it was found that emotional costs are more salient than economic costs.** Restrictions on parents was not a very important disvalue except among the urban Javanese, rura 1 Sundanese and the Chi ne se,

*

This section refers to Table 4.7.

**

Russell K. Darroch et al., Two are not enough: The Value of Children to Javanese and Sundanese Parents. Papers of the East-West Population Institute, no. 60-D, February 1981, pp. 35-37. 93

TABLE 4.6 Values Attached to Children among Respondents from Each Ethnic Group Javanese Rural F

Advantages of Having Children Economic: household help 9.5 Help In old age 20.1 Financial, practical help 48.7 Family continuity 5.0 Psycho Iog Iea I : rewarding interactions 6.5 Companionship, love 2.5 Happiness 4.0 Play, fun, distraction 1.5 Living through children Achievement, power Character, responsibll ity, incentive to succeed Fulfilment, experience Psychological appreciation Marital bond 2.0 Rei igion, social obligation Adult status, social norms Others (199)

Sundanese Urban

M

F

Rural M

F

Urban M

F

M

6.7 23.3

6.4 12.5

10.6 8.2

19.6 5.3

9.8 2.4

16.8 7.4

5.8 1.9

33.3 15.0

30.7 16.6

16.5 34.1

37.0 9.0

43.9 14.6

21.5 16.5

21.2 42.3

8.3

8.8 2.7 9.8 2.7

8.2 3.5 8.2

16.4 1.1 6.3 0.5

9.8 2.4 7.3

7.1 0.9 22.7 0.3

5.8 15.4 3.8

4.9

0.3

1.9

6.7

1.0

1.9 0.3

1. 7

7.4

9.4

2.6

2.4

5.0

5.0

1.0

1.2

2.1

2.4

0.9 0.6

(60)

(296)

(85)

( 189)

(41)

(339)

(52)

TABLE 4.6 (cont 1 dl

Batak

Mi nangkabau Rural F

Urban M

F

Chinese

Rural M

F

Urban M

Urban

F

M

F

M

17.6 9.3

9.3 7.0

11.4 8.6

7.0 3.5

3.7 12.7

8.5 12.8

1.0 4.9

2.0 2.0

5.3 8.5

6. 1 8.2

27.5 24.2

25.6 37.2

22.7 14.9

24.6 36.8

5.3 42.3

10.6 55.3

16.3 42.4

6.0 62.0

27.3 13.1

24.5 28.6

3.3 4.9 4.9 0.5

2.3 4.7 11.6

9.8 3.1 11.0 4.7

1.8 1.8 8.8 1.8

1.6 1.6 9.5

2.1 2.1 8.6

3.1 1.4 17.7 1.4

8.0 6.0 2.0

17.7 2.1 14.2 1.8

4.1 6. 1 12.2 2.0

1.0

2.0

1. 1

4.3

0.5

1.2

0.5

0.5

6.0

2.3

6.7

8.8

2.6

6.9

4.0

5.7

1.6

3.5

9.0

2.1

6.0

1. 1

1.8

0.5

1. 7

1. 1 ( 182)

(43)

(255)

(57)

( 189)

(47)

(288)

8.2

2.1 (50)

(277)

(49)

TABLE 4,6 (cont 1 dl Values Attached to ChI Idren among Respondents from Each Ethnic Group Javanese Rural F

Expected Economic Support from Chi Idren in Old Age Major Minor None

32.5 67.5 (206)

Expectations About Lite tor Children Better The same Worse

90.3 4,2 0.5 (207)

Sundanese Urban

M

20.0 80.0 (60)

93,3 6,7 (60)

F

5.6 94.4

Rural M

3.4 96.6

(301)

(87)

92.2 7.2 0.7 (293)

92,9 5.9 1.2 (85)

F

19.2 80.8 (198)

78,0 22.0 ( 186)

Urban M

9.3 90.7 (43)

81,0 19,0 (42)

F

5,4 94.6

M

100,0

(352)

(52)

88,5 10,9 0.6 (349)

89.8 10.2 (49)

TABLE 4.6 (cont 1 d)

Mi nang kabau Urban

Rural

F

11.2 88.8 ( 187)

90.9 8.6 0.5 ( 187)

Batak

M

6.7 93.3 ( 45)

91.1 8.9 (45)

F

8.7 91.3 (263)

93.8 5.9 0.4 (256)

Chinese

Rural M

3.3 96.7 (60)

96.6 3.4 (58)

F

8.4 91.6 (203)

84.9 15.1 ( 199)

Urban M

1.9 98.1 (52)

90.2 9.8 (52)

F

Urban M

s.o

5.5 94.5 (291)

97.2 2.8 (281)

F

92.0 (51)

96.0 4.0 (50)

(283)

88.6 10.7 0.7 (275)

M

2.0 98.0 (50)

90.0 10.0 (50)

TABLE 4.7 Di sva I ues Attached to Children by Ethnic Group Javanese Rural F

Disadvantages of Having Children Financial: education 11.5 other financial costs 19.2 Chi ldrearing demands: Physical stress 23.1 Emotional stress 23.1 Hea I th, pregnancy Discipline Chi Id's sickness Worry over future 11.5 Other childbearing prob I ems 1.9 Restrictions on parents: Tied down 9.6 Cannot work Costs of social relations: Marital strain Social issues Overpopulation Mi see I I aneous (52)

Sundanese Urban

M

Rural

Urban

F

M

F

M

F

19.0

1.3

4.0

1.4

3.2

4.1

28.6

21.5

36.0

26.4

29.0

38.3

34.8

9.5 9.5

17.4 33.6 0.7

10.0 20.0 2.0

23.0 23.0

19.4 25.8

17.9 16.8 0.5

4.3 17.4

14.3

4.7

4.0

0.7

1.5

4.3

2.0

2.0

16.8

20.0

1.3

2.0

14.3

M

0.5 25.0 0.7

22.6

13.3

8.7

0.5

4.8 (21)

0.7 ( 149)

(50)

( 148)

(31)

6.6 ( 196)

30.4 (23)

4.7 (cont 1 d)

TAB~E

Batak

Minangkabau Rural F

Rural

Urban M

Chinese

F

M

F

Urban M

F

Urban M

F

M

11.9

20.6

5.4

5.0

4.8

19.4

1.9

10.0

3.4

33.3

29.4

16. 1

50.0

53.6

38.9

17.4

45.0

23.9

29.4

11.9

14.7 8.8

27.7 25.9 0.9

5.0 10.0

21.6 14.4

19.4 13.9 2.8

24.2 16.8

5.0 15.0

30.4 15.5 0.7

17.6

2.8

2.5

5.0

4.7

5.9

1.4

5.9

15.6 0.7

0.8 2.2

8.8

0.9

5.0

1.6

1.9

1.8 22.2 0.7

14.7

17 .o

15.0

0.9 0.7 0.7 ( 135)

0.8 0.8

12.4

10.0

20.3

35.3

1.6

1.9

5.0

0.7

5.9

21.1 ( 161)

5.0 (20)

21.9 (34)

3.6 ( 112)

10.0 (20)

( 125)

2.8 (36)

( 145)

( 17)

TABLE 4.7 (cont 1 dl Dlsvalues Attached to Ch lldren by Ethnic Group Javanese Rural F

Restrictions on Freedom to Enjoy Other Things from Many Chi Idren A lot A little None

Embarrassment at BeIng Pregnant If Already Have Grandch 11 dren Great embarrassment Mild embarrassment Indifference Happiness

Sundanese Urban

M

F

Rural M

F

Urban M

F

M

2.4 15.5 82.1 (207)

5.0 21.7 73.3 (60)

5.0 20.3 74.7 (300)

2.3 16.1 81.6 (87)

17.7 32.8 49.5 ( 198)

4.7 32.6 62.8 (43)

6.3 22.1 71.6 (349)

1.9 15.1 83.0 (53)

66.7 5.8 16.9 10.6 (207)

51.7 18.3 18.3 11.7 (60)

78.4 13.0 5.6 3.0 (301)

69.0 11.5 17.2 2.3 (87)

47.5 17.7 24.7 10.1 (198)

30.2 39.5 18.6 11.6 (43)

61.6 17.0 15.9 5.4 (352)

32.1 35.8 18.9 13.2 (53)

TABLE 4.7 (cont 1 d)

Batak

Mlnangkabau Urban

Rural F

M

F

5.3 38.0 56.7 ( 187)

4.4 26.7 68.9 (45)

3.4 22.9 73.7 (262)

65.8 25.1 8.6 0.5 ( 187)

64.4 22.2 13.3

73.4 19.0 5.3 2.3 (263)

(45)

Chinese

Rural M

F

Urban M

F

Urban M

F

M

(60)

16.4 19.9 63.7 (201)

9.6 11.5 78.8 (52)

13.4 27.1 59.5 (291)

3.9 27.5 68.6 (51)

5.2 27.1 67.7 (283)

12.0 88.0 (50)

45.0 38.3 10.0 6.7 (60)

75.9 10.8 6.9 6.4 (203)

75.0 5.8 11.5 7.7 (52)

78.7 14.4 4.8 2.1 (291)

68.6 11.8 9.8 9.8 (51)

79.9 13.5 5.6 1.0 (283)

58.0 22.0 .. 18.0 2.0 (50)

11.7 88.3

while marital strain, and social issues like over-population was hardly mentioned by any of the respondents at all. Hence, respondents do not necessarily consider having many children as being restrictive on their freedom. The proportion who say that they feel no restrictions on their freedom on account of having children is very high, ranging from 49.5 per cent (rural female Sundanese) to 88.3 per cent (urban male Mi nangkabau). The proportion of respondents who say that there is "a lot of restrictions on their freedom" is less than 10 per cent except for the rura 1 fema 1e Sundanese ( 17.7 per cent) and the rural female Batak (16.4 per cent). Feelings of embarrassment at getting pregnant when respondents are already grandparents seem to be shared by all groups. The urban Minangkabau (92.4 and 83.3 per cent), the Chinese (93.4 per cent and 80 per cent) and the urban Batak (93.1 per cent and 80.4 per cent) all express feelings ranging from "great embarrassment" to "mild embarrassment" if such a situation occurs. In a 11 groups there is a 1 so a higher proportion among the females than among the males who express embarrassment under such circumstances. Conversely, the proportion of those who mention feeling happiness in this situation is very small, no higher than 5 or 6 per cent with the exception of rural Javanese and Sundanese where about 10 per cent of the respondents in fact express happiness in the event of this happening.

Marriage and Marriage Attitudes Given the selection criterion of our study sample to focus on people within the childbearing age groups, the duration of current marriage is therefore not very long. The average marriage duration is only between 12-15 years (ranging from twelve years among rural male Javanese to 14.6 years among rural ma 1 e Batak). This is found to be consistent with the age of the oldest child which ranges from 9.32 years for female urban Batal'< to 11.94 years for female rural Minangkabau. The mean age at (current) marriage of the wife and of the husband is older than what one would expect. The age range of the age of the wives is from 17.90 years (rural fema 1e Sundanese) to 22.94 years ( fema 1 e Chi ne se). The age range of the husbands is from 22.94 years (rural male Batak) to 28.14 years (male Chinese). Again we note a consistent pattern of age differentiation by ethnic group: the age pattern for wives and husbands is generally younger among the Sundanese and the Javanese, slightly older among the Minangkabau and the Batak and

102

oldest among the Chinese. first marriage.

A similar pattern is found with age at

As to previous marriage, except for the Javanese and the Sundanese, a very small proportion of the respondents were previously married. Also less than 10 per cent had been married more than twice. Of those who were previously married most are respondents from the rural areas ( 26.1 per cent among the Javanese females and 18.3 per cent of the rural Javanese males) were married more than once. In the urban areas the majority of the respondents {90 per cent and higher) were never married previously with the highest among the urban Batak (99 per cent) and the Chinese (98.6 per cent among the females and 94 per cent among the males). The higher incidence of previous marriage among the Javanese and the Sundanese is not unusual, and has been found in other studies. In fact, nation a 11 y, the Sundanese and the Javanese have the highest rate of divorce and remarriage. The value of children study mentioned before, found that in their sample, of the Sundanese wives 20 per cent have been married twice or more and of the husband 24 per cent; among the Javanese the figures are 16 per cent and 22 per cent respectively. The low incidence of remarriage among the Batak and the Chinese is not unusual. The sample consists of Christian Batak who are usually strict adherents to their religion. Among them and the Chinese, divorce and remarriage is still frowned upon either for religious or social reasons. As for the reason for the termination of a marriage it was due more to divorce rather than separation or death. The incidence of having more than one wife simultaneously is almost non-existent among the respondents in this study. We found only one case each among the urban male Javanese, the urban male Minangkabau and the male Chinese and two cases among the rural ma 1 e Sundanese who profess to having po 1ygamous unions. If the information is correct, it is rather encouraging for women's organizations fighting for the abolition of polygamy. On the choice of spouse at first marriage, the Javanese, the Sundanese and also the Chi ne se are most 1 i ke 1y to have chosen their spouse with their parental approval (ranging from 49.4 per cent among the urban male Javanese to 66 per cent among the urban male Sundanese and the male Chinese). Among the Minangkabau, there is a difference between the urban and rural respondents. In the rural sample spouse selection by the parents only is very common ( 47.1 per cent among the fema 1 es and 42.2 per cent among the males). Among urban Minangkabau the highest proportion chose their spouses with parental approval {35.7 per cent among the fema 1es and 41.7 per cent among the ma 1es). Among the Batak 103

TABLE 4,8 Marriage and Marriage Attitudes by EthnIc Group Javanese Rural F

Du rat I on of Current Marriage (mean years)

Age of Wife at Current Marriage (mean)

Age of Husband at Current Marriage (mean)

Previous Marriage None One Two or more

12.62 (206)

Urban M

F

11,98 13,56 (60) (301)

* **

case 2 cases

Rural M

F

12,47 13,65 (87) ( 197)

Urban M

F

13.49 13.72 (43) (351)

M

13,39 (53)

18,03 (206)

18,12 19.53 (59) (301)

19,91 17,90 (87) (198)

18,84 18.30 (43) (352)

18,87 (53}

24,54 (206)

22,80 24.83 (60) (300)

25,15 23,86 (87) ( 198)

24,54 24.16 (43) (352)

23,92 (53)

65,1 26,1 8,7 (206)

71,7 94,0 18,3 5.6 10,0 0.3 (60) (301)

90,8 8.0 1.1 (87)

Percentage Currently Married to More than One Spouse Percent with at least One Previous Marriage ended by Death Oi vorce Separation

Sundanese

78,3 17.2 4.5 ( 198)

5.9 88.2 5,9 ( 17)

5.3 89.5 5.3 ( 19)

100.0 (6)

92,3 5,1 2.6 ( 352)

94,3 5,7 (53)

4,7**

1.1*

5,6 88.7 5,6 ( 71)

67,4 25,6 7,0 (43)

4,9 82.9 12,2 ( 41)

9,1 90.9 ( 11)

88,9 11.1 ( 27)

100,0 (3)

TABLE 4.8 (cont 1 dl

Mi nang kabau Rura I F

14.26 ( 187)

Batak Urban

M

F

14.97 13.46 (45) (263)

Rura I M

F

14.09 12.52 (60) (203)

19.25 ( 187)

20.85 19.07 (45) (263)

21.23 (60)

25.59 ( 187)

25.62 27.16 (45) (263)

27.25 24.39 (60) (201)

90.0 8.6 0.5 ( 187)

84.4 15.6 (45)

97.0 2.7 0.4 (263)

Chinese

93.3 5.0 1.7 (60)

20.42 (203)

95.1 4.4 0.5 (203)

Urban M

F

Urban M

F

14.55 12.56 (52) (291)

11.49 12.62 (51) (283)

13.78 (50)

19.50 (52)

22.10 (51)

22.94 (298)

22.36 (50)

26.98 27.73 (51) (283)

28.14 (50)

21.84 (291)

22.94 26.94 (51) (291)

92.3 7.7 (52)

99.0 1.0 (291)

96.1 3.9 (51)

98.6 0.7 0.7 (283)

71.4 28.6

75.0 25.0

75.0 25.0

(7)

(8)

(4)

94.0 4.0 2.0 (50)

2.0*

1.7*

81.3 18.7 ( 16)

M

20.0 50.0 30.0 ( 10)

20.0 60.0 20.0

66.7 33.3

( 5)

(3)

100.0 100.0 (3)

( 1)

TABLE 4.8 (cont 1 d) Marriage and Marriage Attitudes by Ethnic Group Javanese Rural

Whether Single or Married Women are Better Off Single are worse off The same Single are better off

• Opinion on Polygamy Approve Neutral Disapprove Age of Respondent at First Marriage (mean)

Urban

Rural

Urban

M

F

M

F

14.5

5.0

8.1

6.9

7.7

12.2

6.8

5.7

15.9

5.0

11.7

5.7

2.2

4.9

11.6

3.8

1.3

3.4

2.6

2.8

1.9

56.4 22.5

49.4 33.3

64.3 22.4 0.5 ( 196)

55.7 23.0

66.0 22.6

F

Who DecIded on First Spouse Parents Parents with respondents's approval Parents and respondent Respondents with parent's approval Respondents only Others

Sundanese

1.0 53.1 15.0 0.5 (207)

71.7 18.3 (60)

(298)

(87)

89.4 9.2 1.4 (207)

81.7 13.3 5.0 (60)

70.8 26.6 2.7 (301)

70.1 27.6 2.3 (87)

14.5 2.7 82.6 (207)

15.3 3.4 81.4 (59)

3.4 5.4 91.2 (297)

7.0 19.8 73.3 (86)

16.60 ( 194)

17.87 19.67 (30) (297)

1. 1

M

63.4 19.5

F

M

(41)

(352)

(53)

8.6 ( 198)

81.0 11.9 7.1 (42)

78.9 18.9 2.3 (350)

67.9 30.2 1.9 (53)

6.1 6.6 87.4 ( 198)

17.5 20.0 62.5 (40)

7.7 4.8 87.5 (351)

9.6 32.7 57.7 (52)

80.3 11. 1

16.15 17.37 ( 191) (52)

16.86 18.48 (22) (346)

16.95 (22)

TABLE 4.8 (cont 1 dl

Ml nangkabau Rural F

Batak Urban

M

F

Chinese

Rural M

F

Urban

Urban M

F

M

F

M

47.1

42.2

20.9

23.3

16.3

17.3

7.6

5.9

2.8

4.0

29.4

24.4

27.8

15.0

5.4

5.8

10.7

15.7

8.7

4.0

7.0

8.9

4.6

5.0

3.4

1.9

4.1

2.8

a.o

13.4 3.2

20.0 4.4

35.7 11.0

41.7 15.0

20.2 54.2 0.5 (203)

25.0 50.0

54.0 23.7

(52)

(291)

(51)

63.2 21.5 1.0 ( 183)

66.0 16.0 2.0 (50)

50.0 48.1 1.9 (52)

65.6 29.0 5.5 (290)

60.8 35.3 3.9 (51)

62.7 35.2 2.1 (282)

52.0 46.0 2.0 (50)

0.7 2.1 97.3 (291)

13.7 86.3 (51)

2.8 9.4

6.0 24.0 70.0 (50)

(45)

(263)

(187)

76.6 22.2 2.2 (45)

84.0 14.9 1. 1 (262)

(60)

35.0 37.1 7.9 (202)

3.2 10.2 86.6 (186)

20.0 40.0 40.0 (45)

6.6 5.4 88.0 (259)

10.5 24.6 64.9 (57)

3.5 96.5 (200)

2.0 2.0 96.1 (51)

18.73 (183)

20.89 18.44 (262) (25)

20.20 (202)

18.45 21.92 (33) (290)

(187)

76.5 23.5

(60)

ao.o 20.0

16.00 (36)

52.9 25.5

a1 .a (292)

22.68 14.64 (282) (22)

19.75 (28)

there is also a difference between the urban and rural samples. But surprisingly and in contrast to the Minangkabau, the highest proportion of the Batak chose their spouses themselves (54.2 per cent of females and 50 per cent of males). As for the urban Batak they are more 1i ke the urban Mi nangkabau. It would appear that among the Javanese, Sundanese and the Chi ne se there is a 1 ready more freedom of choice among the young people, though parent's approval is still essential, while among the Minangkabau parent's role is still very important, in line with the still prevailing clan system. This should, of course, also be the case with the Batak, but we find the figures emphasizing individual choice. As the clan system is also very much alive among the Batak, we should look at these figures with caution. The two last questions deal with attitudes towards single In all groups by far the majority of women and polygamy. respondents indicate that single women are "worse off" than This is most pronounced among the Javanese, married women. Sundanese and Mi nangkabau where some 70.1 per cent among ma 1e urban Javanese and 89.4 per cent among rural female Javanese hold this opinion. Fewer among the Batak (ranging from 50 per cent among the rural male to 65.6 per cent among the urban female) and the Chinese (62.7 per cent and 52 per cent) are likely to think thus though the proportion who hold such opinions is still very substantial. Consequently, only a small proportion (less than 10 per cent and in most cases 1ess than 5 per cent), indicate that single women are "better off" than married women. This, of course, reflects the still prevailing attitude that women should be married and that their proper role is to be a good wife and a good mother. Regarding polygamy, the prevailing attitude is generally one of disapproval as indicated by the very high proportions falling into this category (ranging from 40 per cent among rural male Minangkabau to 97.3 per cent among urban female Batak). It is noteworthy though that in each group, especially among the Javanese, Sundanese and Minangkabau disapproval is For the stronger among the females than among the males. But since Moslem groups, polygamy is religiously tolerated. the new marriage law was promulgated in 1974 and implemented in 1975, the circumstances for approving polygamy are more The consent of the first wife (or other strictly regulated. wives) is often needed before a polygamous union can be legal. Interestingly enough, there is now a new regulation applicable to civil servants, re qui ring men who want to take another wife and women who want to become an additional wife to seek permission This also applies to cabinet ministers from their superior.

108

who have to Republic.*

request

permission

from

the

President

of

the

Attitude towards Divorce and Remarriage** Two questions relate to the matter of divorce, one asking under what conditions the respondents would approve a woman divorcing her husband and the other under what conditions they would approve a husband divorcing his wife. In the first case, we note a distinct difference by ethnic group and only a slight difference by sex. Among the Javanese and the Sundanese, the conditions that received the highest proportions are "he does not support the family", "he is cruel to her" and "he has a mistress". Over all the proportions are higher among the Sundanese than among the Javanese. A similar pattern is seen among the Mi nangkabau, but the proportions are even higher. Among the Chinese, these three conditions also receive the highest approval. The interesting pattern is among the Batak, where there is almost the same proportions for the three conditions mentioned above and also for "they could not have children". This variation by ethnic group reflects the attitude towards divorce. In general, divorce in Indonesia is much more accept ab 1e, and is regulated by Is 1ami c 1aws for the Javanese, Sundanese and Minangkabau. Among the Batak, on the other hand, the customary law (adat) reinforced by the Christian religion has made divorce less-aCceptable. A similar attitude prevails among the Chinese. As for the conditions a11 owing a man to divorce his wife, there is a1 so variation by ethnic group. Among the Javanese and the Sundanese, the highest approva 1 for divorce is expressed for the conditions, "she neglects the house and fami 1y", "she is cruel to him", "she has a lover", and "they could not have

*

This refers to Peraturan Pemerintah (government regulation) or PP No. 10/1983, which was signed by President Soeharto on 21 April 1983, the day referred to as Kartini Day, to commemorate the birthday of one of the most celebrated national heroines of Indonesia. In the light of persistent rumours about high officials taking a second wife, this regulation has been given wide coverage in the press, for example, in the well-known weekly Tempo of 14 May 1983.

**

This section refers to Table 4.9.

109

TABLE 4.9 Attitudes Towards Divorce and Remarriage among Respondents from Each Ethnic Group Sundanese

Javanese Rura I F

Percentage Approving of a Woman Divorcing her Husband If They could not have children He has a mistress He Is cruel to her He does not support the family She wants to marry another man

Percentage Approving of a Man Divorcing his Wife If They could not have eh 11 dren She has a lover She is cruel to him She neglects the house and family He wants to marry another woman

Approval of a Widowed Woman Remarrying Approve Depends on her age Disapprove

Age After Which She Shou Id No Longer Remarry (mean)

Rural

Urban M

F

M

F

Urban M

F

M

22.7 43.7 46.1

28.3 40.0 50.0

32.3 53.7 68.1

32.2 52.9 65.5

32.3 59.1 72.2

30.2 53.5 74.4

28.4 61.1 71.3

30.2 64.2 81.1

55.8

61.7

66.1

72.4

76.3

83.7

74.7

86.8

17.0 (207)

23.3 (60)

12.6 (300)

16.1 (87)

12.1 ( 198)

14.3 (43)

9.9 (352)

24.5 (53)

42.0 51.0 46.1

46.7 66.7 55.0

43.3 63.7 65.7

43.7 60.9 56.3

41.4 77.8 75.3

41.9 78.6 79.1

40.9 72.2 69.6

35.8 79.2 75.5

57.3

58.3

70.4

69.0

81.3

83.7

73.0

81.1

27.7 (207)

20.0 (60)

13.3 (300)

9.2 (87)

22.2 ( 198)

18.6 (43)

14.5 ( 352)

18.9 (53)

84.1 3.4 12.6 (207)

78.3 8.3 13.3 (60)

83.4 8.3 8.3 (301>

81.6 11.5 6.9 (87)

91.9 2.0 6.1 ( 198)

95.3

90.6 9.4

4.7 (43)

86.1 4.5 9.4 (352)

39.23 ( 137)

42.01 38.69 (201) (42)

41.6 (53)

41.93 ( 103)

43.04 42.89 (214) (23)

(53)

43.71 ( 31)

TABLE 4.9 (cont 1 dl

Batak

Mlnangkabau Rural F

Rura I

Urban M

F

Chinese

M

F

Urban M

F

Urban M

F

M

13.4 79.1 89.3

41.1 84.4 86.7

27.9 71.3 76.0

43.3 60.0 78.3

52.7 44.8 35.5

63.5 51.9 32.7

23.0 41.9 41.9

21.6 49.0 43.1

16.8 63.5 71.1

18.0 38.0 60.0

88.7

88.9

76.3

86.7

58.6

59.6

45.7

49.0

67.0

68.0

24.1 (187)

33.3 (45)

16.8 (262)

30.0 (60)

11.9 (203)

5.8 (52)

10.0 (291)

21.6 (51)

16.3 (283)

34.0 (50)

12.8 92.8 66.7

15.6 93.3 82.2

35.5 73.2 71.3

43.3 83.3 75.0

53.7 45.8 34.0

63.5 57.7 30.8

24.4 45.0 38.8

21.6 56.9 49.0

29.5 64.6 60.8

34.0 56.0 52.0

76.9

88.9

81.3

83.3

55.7

53.8

50.5

47.1

73.3

72.0

28.0 ( 187)

33.3 (45)

11.8 ( 262)

15.0 (60)

12.3 (203)

7.8 (52)

14.4 (291)

11.8 (51)

14.6 (283)

20.0 (50)

54.0 32.6 13.4 ( 187)

44.4 46.7 8.9 (45)

80.2 9.2 10.7 (262)

83.3 10.0 6.7 (60)

30.5 4.9 64.5 (203)

51.9

57.4 18.2 24.4 (291)

60.0 24.0 16.0 (50)

79.4 15.0 5.6 (282)

78.0 18.0 4.0 (50)

43.75 ( 138)

41.35 45. 10 (29) ( 168)

42.86 (28)

41.71 (59)

48.1 (52)

38.46 39.22 (26) ( 174)

39.83 (30)

41.75 ( 179)

44.22 (23)

children". Again, the proportions are relatively higher among the Sundanese than among the Javanese. It is interesting to note, that the proportions for the condition "they co~ld not have children" is far higher in the case of the husband divorcing his wife than the other way around. This reflects the prevailing opinion that in the case of infertility, the wife is to be blamed. Jlroong the Minangkabau, we find a similar pattern as the Javanese and Sundanese, but the highest approval for divorce is if "she has a lover" (ranging from 73.2 per cent for urban females to 93.3 per cent for rural males). Among the Batak the conditions which favour divorce are the same for the wife as for the husband. Among the Chinese, neglect of the house and family seems the most salient suggesting that for the Chinese, the wife's domestic role is still emphasized. The next two questions deal with widow remarriage and the age limit for widow remarriage. Again, there is a distinct variation by ethnic group, reflecting the influences of religion, customary law and the value system. Among the Javanese and Sundanese, by far the majority (ranging from 78.3 per cent among the rural male Javanese to 95.3 per cent among rural male Sundanese) approve of widow remarriage. Jlroong the Mi nangkabau there is also a high proportion who approve, especially among the urban respondents (80.2 per cent for females and 83.3 per cent for males), while in the rural area the proportion is lower, because a sizeable proportion indicate that it depends on her age (32.6 per cent of females and 46.7 per cent of males). Unlike the other groups, there is a high proportion disapproving of widow remarriage among the Batak, especially rural Batak (64.5 per cent of females and 48.1 per cent of ma 1 es). In the urban area there seems to be more to 1erance with 57.4 per cent among the urban female Batak and 69 per cent among the urban male Batak approving of widow remarriage. The Chinese are more lenient, resembling the Javanese pattern, with a high proportion approving. As far as age limit for widow remarriage is concerned, there is a si mil ari ty in a 11 groups. The age 1 imi t ranges between 38.46 years (rural male Batak) to 45.10 years (rural male Mi nangkabau). This is in line with the prevailing opinion that aged forty or thereabouts is an important transition period for a woman.

Interaction with Spouse and Sex Role Attitudes* Information

*

about

interaction

with

spouse

This section refers to Table 4.10.

112

is

elicited

through

four questions as follows: "ever discussed another child", "whose opinion on having children is more important", "shares worries with spouse", and "spouse chosen to spend leisure time with". Regarding communication between husband and wife about having another child, it turned out that in all groups, urban as well as rural, about half of the respondents (ranging from 34 per cent among male Chinese to 64.7 per cent among urban male Batak) do discuss this with their spouse. This is quite encouraging, of course, as one of the obstacles in participation in family planning programme, has been the 1ack of communication between spouses. It is interesting, though, that in all groups, both ma 1e and female, the majority (ranging from 54.8 per cent among urban male Minangkabau to 89.1 per cent among rural male Javanese), indicate that when it comes to the question of having children, it is the husband's opinion that is more important. This indication of male dominance in fertility behaviour is generally recognized and is taken into consideration in the implementation of the family planning programme. Women who want to participate are encouraged to inform their husbands about their plans. The next question about sharing worries with spouse, also indicate ample communication between spouses. All groups except for the rural Sundanese more than 70 per cent, say that they share worries with spouse "all the time" or "most of the time". The response to the question on whether the spouse is chosen to spend leisure time with also suggests good communication between the spouses. Significant proportion indicates that their spouse is their first choice, although the proportion is somewhat small er among the rura 1 Batak and rura 1 Mi nangkabau. If we add "second choice", then by far the majority of the respondents indicate good communication between spouses. The study finds that the statement that a girl becomes fully a woman only when she bears her own children, does not show a high agreement score. The mean agreement on a 5-point scale indicate a range of between 2.00 (rural male Batak) to 2.77 (rural female Minangkabau). The next statement that a woman should devote a lot of time to satisfying her husband, elicits an even lower score, ranging from 1.75 (urban male Sundanese) to 2.48 (male Chinese). Considering the still prevailing idea that the proper role of woman is to be a good wife and a good mother, these scores are rather surprising. The two final questions in this section have to do with attitudes about wife working outside the home after marriage, and wife working when husband's earn ngs are adequate. It turned out 113

TABLE 4.10 Interaction with Spouse and Sex Role Attitudes by Ethnic Group Sundanese

Javanese Rural F

Ever Discussed Another Child with Spouse Yes No

Whose Opinion on Having Children Is More Important Husband's Equal Wife's

Shares Worries with Spouse AI I the time Most of the time Sometimes Seldom

Spouuse Chosen to Spend Leisure Time with First choice/ only choice Second choice Not chosen

Urban M

F

Rural M

F

Urban M

F

M

41.5 58.5 (207)

45.0 55.0 (60)

38.2 61.8 (301)

44.8 55.2 (87)

50.5 49.5 (198)

55.8 44.2 (43)

50.3 49.7 (352)

58.5 41.5 (53)

58.7

89.1

66.4

62.3

66.0

80.6

70.7

65.1

41.2 (160)

10.9 (46)

33.6 (214)

37.7 (69)

34.0 ( 162)

19.4 (36)

29.3 (294)

34.9 (43)

65.0 12.1 13.6 9.2 (206)

60.0 15.0 13.3 11.7 (60)

54.0 13.0 21.0 12.0 (300)

49.4 12.6 26.4 11.5 (87)

38.9 21.2 24.2 15.7 (198)

44.2 11.6 25.6 18.6 (43)

55.8 14.5 17.7 12.0 (351)

52.8 13.2 24.5 9.4 (53)

72.9 15.5 11.6 (207)

70.0 15.0 15.0 (60)

47.5 27.2 25.2 (301)

51.7 35.6 12.8 (87)

42.9 23.2 33.8 ( 198)

51.9 27.9 20.9 (43)

38.4 31.8 29.8 (352)

66.0 22.6 11.3 (53)

TABLE 4.10 (cont 1 d)

Mlnangkabau Rural F

Batak Urban

M

F

Chinese

Rural M

F

Urban M

F

Urban M

F

M

45.5 ( 187)

44.4 55.6 (45)

40.3 59.7 (263)

41.7 58.3 (60)

53.7 46.3 (203)

46.2 53.8 (52)

52.6 47.4 (291>

35.3 (51)

41.3 58.7 (283)

66.0 (50)

87.2

78.6

72.6

54.8

59.5

71.1

70.9

85.0

63.1

81.1

12.8

21.4 (42)

27.4

40.5 ( 158)

28.9 (38)

(223)

15.0 (40)

36.9

(208)

45.2 (42)

29.1

( 172)

(217)

18.9 (37)

28.9 34.2 25.7 11.2 (187)

33.3 37.8 20.0 8.9 (45)

53.8 17.6 19.1 9.5 (262)

47.5 18.6 27.1 6.8 (59)

60.6 14.8 18.7 5.9 (203)

75.0 11.5 9.6 3.8 (52)

62.9 16.5 15.1 5.5 (291)

60.8 15.7 19.6 3.8 (51)

51.4 16.0 18.7 13.9 (283)

38.0 12.0 38.0 12.0 (50)

33.3 55.6 11.2 (187)

42.2 42.2 15.6 (45)

40.3 33.8

48.3

39.9 48.8 11.3 (203)

44.2 40.4 15.4

60.1

54.9

17.2 22.7

31.4 13.7 (51)

37.2 32.3 30.6 (283)

26.0 18.0 (50)

54.5

25.9

(263)

23.3 28.3 (60)

(52)

(291>

64.7

34.0

56.0

TABLE 4.10 (cont 1 dl Interaction with Spouse and Sex Role Attitudes by Ethnic Group Javanese Rural F

Mean Agreement (on a 5-polnt scale) with A girl becomes fully a woman only when she bears her own eh I Idren

A woman should devote a lot of time to satisfying her husband

Wife Working Outside Home after Marriage Good idea IndIfferent Bad idea

Wife Working if Husband's Earnings are Adequate Approve Disapprove

Sundanese Rural

Urban M

F

M

F

Urban M

F

M

2.33 (207)

2.27 2.60 (60) (301)

2.48 2.70 ( 198) (87)

2.37 2.53 (43) (352)

2.62 (53)

2.10 (207)

1.90 2.20 (60) (301)

2.31 2.38 ( 197) (87)

2.10 2.35 (43) (352)

1.75 (53)

87.4 3.9 8.7 (207)

81.7 5.0 13.3 (60)

74.8 14.3 11.0 (301)

67.8 17.2 14.9 (87)

76.8 12.6 10.6 ( 198)

60.5 14.0 25.6 (43)

73.9 11.9 14.2 (352)

67.9 9.4 22.6 (53)

75.1 24.9 (189)

60.4 39.6 (53)

63.9 36.1 (274)

56.0 44.0 (75)

60.7 39.3 (183)

71.9 28.1 (32)

63.1 36.9 (306)

68.3 31.7 (41)

TABLE 4.10 (cont 1 dl

Batak

MI nangkabau Rural F

F

Urban

Rural

Urban M

Chinese

M

F

M

F

Urban M

F

M

2.77 ( 1a7)

2.6a 2.27 (44) (262)

2.60 2.05 (60) (203)

2.00 2.5a (52) (291)

2.a2 2.4a (51) (2a3)

2.a4 (50)

2.51 ( 1a7)

2.09 2.02 (43) (262)

2.47 2.01 (60) (203)

1.a2 2.1a (291) (51)

2.14 2.17 (51) (2a3)

2.4a (50)

74.3 24.1 1.6 ( 1a7)

59.1 31.a 9.1 (44)

70.3 12.9 16.7 (263)

66.7 15.0 1a.3 (60)

93.1 3.9 3.0 (203)

3.a (52)

72.3

61.9 3a.1 (42)

63.6 37.4 (225)

54.0 46.0 (50)

90.4 9.6 ( 19a)

92.2 7.a (51)

27.7 (1a4)

96.2

69.a 1a.6 11.7 (291)

63.5 36.5 (260)

70.6 17.6 (51)

71.2 15.3 13.5 (2a3)

60.0 26.0 14.0 (50)

57 .a 42.2 (45)

63.9 36.1 (247)

57 .a 42.2 (45)

ll.a

that by far the majority in all groups state that it is a good idea for the wife to work outside the home (ranging from 59.1 per cent among rural male Minangkabau to 96.2 per cent among rural male Batak). The highest proportions are among the rural Batak and the rura 1 Javanese, where it is tradition a11 y accept ab 1 e for women to work in the fields and in the market-place. .The proportion of those who approve of the wife working even if husband's earnings are adequate is also quite high, but generally slightly lower than the response to the previous question (approval ranges from 54 per cent among the urban male Minangkabau to 92.2 per cent among rural male Batak). The highest proportions are again found among the rural Batak and the rural Javanese.

118

V

EXPLAINING DIFFERENCES IN FERTiliTY

The simplest way to examine ethnic differences in fertility is to employ the tabular analysis where fertility levels of the ethnic groups under investigation are compared. Such information can be further cross classified by other variables commonly termed as control variables. Using this method, however, restricts the number of controlled variables to be employed since the more variables employed, the more difficult to read the tables. In the previous chapter, for example, we only employed two control variables, namely residence and sex. Some studies, on the other hand, have reported that fertility behaviour is influenced by other characteristics. Some of the important ones are education, occupation, income, and demographic factors. Consequently, a study of ethnic differentials on ferti 1 ity needs to take into account all or some of these characteristics. This implies the need to employ the multivariate technique of analysis. In this study, the Multiple Classification Analysis (MCA) technique is used to disentangle the effects of these cor re 1 ated vari ab 1es on the three dependent variables under investigation: children ever-born, recent fertility, and use of effective methods of contraception. The MCA is a technique for determining the interrelationship between a dependent variable and several predictor variables, within the context of an additive model. Essentially, it is a variant of the multiple regression technique using dummy vari ab 1es. Unlike the regression technique, however, MCA provides coefficients expressed as deviations from the grand mean instead of deviation from a constant term. It can be utilized for any measurement of predictor variables. It also provides a summary measure, beta, for each predictor to indicate the relative importanceDrthe predictor in explaining the dependent variable after adjusting for all other predictors included in the model. The MCA also produces the summary statistic R2 (which is 119

also produced by the regression technique) to indicate the proportion of the total variance in the dependent variable accounted for by all variables included in the model. It is a1so worth noting that the study emp 1oys the SPSS (Statistical Package for the Social Sciences) version of MCA which allows a limited number of predictor variables, namely five factors and five covari ates. Under this 1imitation, the predictor variables in the equations or models have to be entered alternately.

Some Notes on the Techniques

The analysis presented in this section is confined to data on female respondents only. Some major reasons for the decision are: 1.

Some information which are collected through the female respondents cannot be matched with male respondents because of problems of identification on the data tape. The only solution to this problem is to code again.

2.

Due to high i ne i dence of remarri ages in Indonesia, some important information, particularly number of births and age at first marriage, are either i neon si stent or unreliable.

Among the variables employed in the analysis or models, "couple income" seems to be the one which has quite high non-responses. There are about a quarter of the cases that have missing values for this variable. This high rate of non-responses, in fact, has been anticipated in the designing of the survey. Consequently, the results of the model involving this variable may suffer from reliability.

Analysis of Children Ever Born

For all the ethnic groups under study as a whole, the survey gives a grand mean of number of children ever born as 3.68 children. The Chinese are reported to have the lowest mean (2.83 children) while the Batak are shown to have the highest fertility, 4.14 children per women. Looking at these figures, the survey data show ethnic differentials of fertility. If we relate the ethnic group to the major residence of provinces (see Table 5.1l, the results of this study seem to be supported by the results of the 1980 Census. 120

TABLE 5. 1 Mean Number of Children Ever Born (CEB) for Selected Areas, 1980

Province

Major Ethnic Group

Population Census*

Ethnicity and Fertility Study

East Java

Javanese

3.25

3.49

West Java

Sundanese

3.92

3.80

West Sumatra

Mi nangkabau

4.49

3.78

North Sumatra

Batak

4.64

4.14

*SOURCE:

Central Bureau of Statistics, 1980 Population Census, unpub I I shed tab Ies.

Such differentials were also reported by Sup a rl an and Si git ** employing the results of the Intercensal Population Survey conducted in 1976. Given the consistency of these findings, we may conclude that there are differences in fertility among ethnic groups where the Chinese are found to have the lowest parity while the Batak have the highest. Line 1 of Table 5. 2 shows the mean number of children ever born for each ethnic group after controlling for ethnic affiliation and fertility approval variables. Holding constant these variables, we find that ethnic differentials on fertility remain. Comparing the figures however before and after adjusting these control variables -- the Chinese seem to show the most substantial reduction. Their parity increases by about 0.2 children. The Batak fertility, on the other hand, is shown to have been reduced after accounting for those two variables, while that of the other ethnic groups are practically not affected by those variables. Taking the Chinese and the Sundanese, we find that the difference becomes narrower from 0.97 for the unadjusted to 0. 80 for the adjusted mean CEB ones whi eh indicates that the ethnic differentials are partly due to the differences in their scores on ethnic affiliation and ethnic fertility approval.

**

Parsudi Suparl an and Hananto Si git, Culture and Fertility. The Case of Indonesia. Research Notes and D1scuss1ons senes, Paper No. 18, Institute of Southeast Asian Studies, Singapore, 1980.

121

In the next model, three major demographic variables, the age at first marriage, the marriage duration, and the number of years lived in urban areas, are added. Accounting for these five control variables, we still find ethnic differences in the number of chi 1dren ever born. If we again take those two ethnic groups ( Sundanese and Chi ne se), we find the difference is narrower for the adjusted means as compared to the unadjusted ones. In fact, this difference is even smaller than the one adjusted for ethnic affiliation and fertility approval variables only. Looking at the figures in the last column of Table 5.2, we see that the addition of these three vari ab 1es to the first two increases the proportion of explained variance from 3.1 per cent to 51.7 per cent. This indicates the predominant influence of the three variables in explaining the variation in the number of children ever born. However, if we compare the beta values for the ethnic variable in those two models, we find that the beta values are relatively stable. Hence, the relative importance of ethnic differences in explaining the number of children ever born remains the same even if we account for important demographic variables such as age at first marriage and the marriage duration. The following two models presented in Table 5.2 are expansion of the second model where socio-economic status of the husband (Model 3) and socio-economic status of the wife (Model 4) are added to the model. The data show that ethnic differences in the number of children remain unchanged in those three models. Taking the differences in the mean number of children between the Sundanese and the Chi ne se we find slight changes · in the differences. The second model gives a difference of 0.87 children while the third and fourth models give a difference of 0.78 and 0.86 children respectively. Moreover, the beta values of the ethnic variable in those three models are the same. This indicates that differences in the number of children between ethnic groups remain unchanged despite inclusion of different sets of predictor variables. It is notable that the increment of R2 value indicates that the socio-economic factor> influence fertility through the intermediate variables such as age at first marriage and marriage duration. Socio-cultural predictors are introduced in the fifth and sixth models. Comparing the data for these two models with that for the second model, we see that there are changes in the differences of the mean number of children among the ethnic groups under study. This is particularly found for the fifth model where psychosocial variables are introduced as predictors. The changes however, are not substantial and the degree of variations does not effect the beta values for the ethnic variable in the different models.

122

The analysis in the second part of Table 5.2 is confined to women who have more than one child at the time of the survey. This is due to the i ncl us ion of child mort a 1 ity experience and months of breast-feeding variables. This limitation reduces the number of cases included in the analysis by about 7 per cent (see the last line of Table 5.2) and gives a different grand mean of 3.93 children instead of 3.68 children for the whole sample of women in the survey. For this sub-sample, the survey data give the unadj usted mean number of children of 2.97 for the Chinese while it is 4.15 This again indicates differentials in the for the Sundanese. number of children among the ethnic groups under study, where the Sundanese tend to be the second most fert i 1 e popu 1 at ion whi 1e Chinese have the lowest fertility. We noted earlier that the Sundanese and Chinese have a difference of 1;18 children in the mean number of children. This difference seems to become sma 11 er when other vari ab 1es, especially the age at first marriage and the marriage duration, are held constant. A further reduction is shown when perceived fecundity, mortality experience, and months of breast-feeding are added to the model while holding constant the other ethnicity and demographic variables (see Model 7 in Table 5.2). As expected, the demographic factors --the age at first marriage and the marriage duration -- substantia 11 y affect the number of children ever born. The i ne 1 usi on of the 1ast three factors -- perceived fecundity, child mortality experience and months of breast-feeding -- in the model, however, produces very small addition to the explanation of the variation of the number of children born, only by about 0.3 per cent. In sum, for the models presented in the analysis, we find that ethnic differences in the number of chi 1dren ever born, which is considered as a measure of fertility behaviour remain unchanged even though different sets of variables ranging from social, cultural, economic, and psychosocial ones are taken into account. This is indicated by the stable beta va 1 ues for ethnic variable in the different models employing an MCA technique. The figures presented in this analysis also indicate that the Chinese consistently have the lowest fertility level while the Sundanese these Si nee 1eve 1 • fertility highest second the have differentials have been found when we are not controlling for any factors (unadjusted values or model 0 in the table), we may conclude that ethnic differentials in fertility levels are still found even when other characteristics of the population are considered. The two interrelated variables, the age at first marriage

123

TABLE 5.2 Mean Number of Children Ever Born CCEBl by Ethnic Group Adjusting for Different Sets of Predictors Ethnicity

Javanese

Sundanese

Mlnangkabau

Batak

ChI ne se

(0) None

3.49

3.80

3.78

4.14

2.83

(1) Ethnic affiliation & ethnic ferti llty approval

3.47

3.83

3.77

4.03

3.03

.12

.031

(2) 1 + years of urban residence, age at first marriage & marriage duration

3.44

3.87

3.56

4.22

3.00

• 15

.517

(3} 2 + husband's education, husband's occupation & couple Income

3.42

3.86

3.58

4.36

3.08

.16

.544

(4) 2 +wife's education wife's working experience & perceived women's role

3.42

3.85

3.58

4.25

2.99

• 16

.521

(5) 2 + gender balance, son preference & religiosity

3.45

3.94

3.56

4.19

2.91

.16

.574

(6) 2 + household type, husband dominance & spouse Interaction

3.46

3.86

3.55

4.22

3.01

.15

.517

450

494

288

Predictors

Grand mean:

N

Beta*

R2

3.68

508

550

TABLE 5.2 (cont 1 d) Mean Children Ever Born (CEB) by Ethnic Group AdjustIng for Different Sets of Predictors Ethnicity

PredIctors

Javanese

Sundanese

Minangkabau

Batak

Chinese

Beta*

R2

For those with more than one eh I Id Grand mean: 3.93 (0) None

3.71

4.15

3.97

4.46

2.97

(2) 1 + years of urban residence, age at fIrst marriage and marriage duration

3.70

4.14

3.81

4.51

3.15

.18

.50C

(7) (2) + perceived fecundIty, eh I Id mortality experience, months breastfeeding

3.66

4.02

3.75

4.35

3.35

.11

.5C3

476

501

426

457

274

N

*

for ethnicity variables.

and the influences surprising the number

marriage duration, are found to have substantial on the number of children ever born. This is not a finding since those variables are directly related to of children born.

Taking the second model which constitutes ethnicity and demographic variables as the basic one and comparing its R2 value to the values of R2 for the other models {Models 3 to 6), the data indicate that household structure (Model 6) and women's status (Model 4) do not add substantial explanation to the variation of the number of children ever born. Husband's status, however, produces more variation. This is probab 1y associated with the fact that the husband's socio-economic status determines the socio-economic status of the family. It is a 1 so worth noting that many women are still employed in the agricultural sector and their work status does not raise the household socio-economic status significantly. As the economic structure of Indonesia changes over time, women's status may become a more significant variable in the future. A more substantial explanation to the variation of the number of children ever born is produced by the variable on gender preference. These variables add almost 6 per cent to the value of R2, which is considerable. This is probably due to the fact that the Batak and Chinese favour sons to continue the familia 1 1 i ne. Methodo 1 ogi ea 11 y, however, some ea uti on must be exercised in interpreting this finding since the gender preference variables are derived from the number of existing children.

Analysis of Recent Fertility One of the common measures to indicate recent fertility experience is the number of children born during five years before the date of survey. Consequently, as also adopted by most analyses of this factor, we have to 1 imit our analysis to the respondents (women) whose years of first marriages were at least five years before the i ntervi ewi ng date. In other words, the analysis is limited to those who have been married for at least five years at the time of interview. Consequently, the number of cases included in the analysis was reduced to 1,732, a reduction of about a quarter of the whole female respondents. This study shows that the average number of children born per woman is 1.36. The unadjusted mean number of children for the Chinese is 0.96 children which is lowest among the five ethnic groups under study. The Minangkabau have given birth to

126

0.30 children more than the Chinese while the differences for the Javanese, Sundanese, and the Batak are 0.40, 0.55, and 0.61 children respectively (see first line of Table 5.3). Hence, there are recent fertility differentials among the ethnic groups in our survey. These differentials persist even after controlling ethnic affiliation and ethnic fertility approval variables. It is also indicated that controlling for those two ethnicity variables tend to decrease slightly the average number of children for all ethnic groups except for the Minangkabau. Comparing these figures with the ones for the cumulative number of children, we find that there are consistent patterns. The Chinese are reported to have the lowest fertility level while the Batak have the highest. The model with these three variables, ethnicity, ethnic affiliation, and fertility approval, also produces higher contribution of ethnic variable in explaining the recent fertility differentials as indicated by the higher value of beta for the recent fertility {0.16) than the one for cumulative fertility {0.12). Similarly, these three variables account for higher explanation of recent fertility variation (the R2 is 4.2 per cent) than cummulative fertility variation (R2 equals to 3.1 per cent). Adding and adjustment for demographic controls (years of urban residence, age at first marriage, and marriage duration) to the three ethnic factors increase the proportion of explained variation of fertility to 19.9 per cent, while it reduces the number of children born among the Javanese and Sundanese but increases the number of children born for the other ethnic groups. This is probably associated with the fact that Javanese and Sundanese tend to marry at earlier ages than the other three ethnic groups. 1

Husband s education and occupation and coup 1 e s income are found to have influenced the recent fertility level of the population. Their effects are even 1arger than those of women 1 s characteristics. However, these socio-economic characteristics are not effective enough to alter the ethnic differentials on recent fertility. In other words the ethnic differentials on recent fertility remain even after controlling for socio-economic characteristics while holding constant the effects of ethnicity and demographic factors. This is also true for household structural variables (Model 6). 1

Mode 1 5 whi eh i ne 1 udes gender ba 1 a nee, son preference and religiosity variables results in an addition of about 5 per cent of the variation in recent fertility although they do not change the relative importance of the effect of ethnic variables in explaining the recent fertility. 127

TABLE 5.3 Recent Fertility by Ethnic Group Adjusting tor Different Sets of Predictors (Female respondents who have married at least five years) Ethnicity

Javanese

Sundanese

Mlnangkabau

Batak

Chinese

(0) None

1.36

1.51

1.28

1.57

0.96

(1) Ethnic affiliation & ethnic fertility approval

1.32

1.49

1.29

1.55

(2) 1 + years of urban residence, age at fIrst marriage & marriage duration

1.25

1.44

1.34

(3) 2 + husband 1 s education, husband 1 s occupation & couple Income

1.22

1.52

(4) 2 +wife's education wife's working experience & parcel ved women's role

1.23

(5) 2 + gender balance, son preference & religiosity

1.27

Predictors

Grand mean:

Beta*

R2

0.92

.16

.042

1.60

0.97

• 16

• 199

1.33

1. 72

1.05

• 18

.227

1.43

1.36

1.61

0.99

• 16

.207

1.46

1.36

1.59

0.91

.17

.247

1.36

TABLE 5.3 (cont 1 d) Ethnicity

PredIctors

Javanese

Sundanese

MI nangkabau

Batak

Chinese

Beta*

(6) 2 + household type, husband dominance & spouse Interaction

1.26

1.44

1.34

1.60

0.96

• 16

.201

401

402

359

352

218

(0) None

1.36

1.51

1.28

1.57

0.96

(2) I + years of urban residence, age at first marriage and marriage duration*

1.29

1.47

1.36

1.62

0.97

• 16

.206

(7) (2) + percalved fecundity, eh 11 d mort a 11 ty experience & months breastfeeding

1.18

1.34

1.40

1.59

1.22

.11

.268

390

396

354

347

217

N

R2

For those with more than one child Grand mean: 1.37

N

*

For ethnicity variables.

Comparing the findings presented in this section to the one for the cumulative fertility (number of children ever born), we find similarity and consistency in the pattern of ethnic differentials. This is also true for the sub-sample constituting those who have had at least one child. Hence, the conclusions presented in the previous section are also valid for the recent fertility differentials.

Analysis of Use of Effective Methods of Contraception In the study presented here, use of effective method of contraception is coded as: 0 if the respondents never used any methods, 1 if the respondents never used any effective methods, and 2 if the respondents have ever used effective methods of contraception.* Employment of this coding system assumes that the higher the mean score the more 1 i ke 1y the respondents have ever used any effective method of contraception. We may also note, in this context, that this variable is treated as an interval scale variable while strictly speaking it is a nominal variable. The overall mean score for the female respondents in the study is 1.03. This mean varies with ethnic groups. The Chinese have the highest score, 1.40, followed by Javanese, Minangkabau, Sundanese, and Batak who have the 1owest score. The high score for the Chinese is quite a surprising finding of this study. It is not because of the fact the Chinese respondents are drawn from the urban areas only since many studies in Indonesia have shown that contraception has been equally adopted in urban and rural areas. The low practice of contraception among the Sundanese indicated in this study has also been reported by many other studies. The Sundanese (West Java) have been known to have a low age at first marriage, high fertility level and low rate of contraceptive use. The low score for the Batak is probably associated with the fact that family planning has been more recently introduced in the area (North Sumatra) whi 1e the Batak who are living in Java are more likely to be the younger age groups. Controlling for ethnic affiliation and ethnic fertility approval, we find that there is a reduction in the mean score for

*

Effective contraceptive methods include the pill, IUD, steri 1 i zat ion, condom and i nj ectab 1es. Other methods 1 i ke the douche, rhythm, withdrawal, abstention and herbs are considered "traditional" methods. 130

Chinese while the scores for other ethnic groups are practically unaffected. The ethnicity variables account for 8.6 per cent of the variation in the scores of use of contraception. The percentage increases to 10.1 if demographic factors are included in the model (see Model 2 in Table 5.4). This indicates that factors such as years in urban reside nee, age at first marriage, and marriage duration do not significantly affect the explanation of the use of contraception. As for the ethnic differentials in the use of contraception after controlling for these demographic factors, the study indicates that the differentials persist. However, it is also to be noted that the differences in the mean scores between ethnic groups become smaller. The difference between the Chinese and Sundanese, for example, is 0.19 before controlling for the demographic factors while it is 0.15 after controlling for these factors. Socio-economic characteristcs seem to be more influential than demographic factors. This can be observed from the comparison between the values of R2 for Models 1 to 4 whi eh respectively has the values of 8.6 per cent, 10.1 per cent, 16.9 per cent and 13.8 per cent. As in the case of fertility, husband characteristics are more influential than the wife's characteristics in influencing contraceptive use. The husband's education and occupation and couple's income reduce the mean scores of the use of contraception among the Minangkabau, Batak, and the Chinese. Their effects are very minimal for the Javanese and Sundanese. The reversed patterns are found for the effects of women's education, women's working experience, and perceived women's ro 1e. It is interesting to note that the increase in the predicting power of the models by the inclusion of the socio-economic variables also results in the more relative importance of the predicting power of ethnicity variables. This means that the effects of ethnicity becomes more important after accounting for their socio-economic conditions. Gender balance, son preference, and religiosity variables have contributed 4.5 per cent to the explained variance of differentia 1 s in the use of effective method of contraception. Controlling for these variables, however, we found that the effects of the ethnic vari ab 1es remain unchanged. This 1eads to conclusion that these variables do have effects on the use of contraception but they do not affect the predominant influence of the ethnic variables on the use of contraception. Unlike the variables discussed above, household type, husband dominance, and spouse interaction variables have no effects on either the use of contraception or the relative importance of the influence of ethnic variable on the use of contraception. 131

TABLE 5.4 Mean Score for Use of Efficient Contraceptive Method by Ethnic Group Adjusting for Different Sets of Predictors (Female respondents) Ethnicity

Predictors

Grand mean:

Javanese

Sundanese

Minangkabau

Batak

Chinese

Beta*

R2

1.03

(0) None

1.11

0.91

1.01

0.88

1.40

(1) Ethnic aft i I iation & ethnic fert I I ity approval

1.12

0.93

1.02

0.88

1.32

• 15

.086

(2) 1 + years of urban residence, age at first marriage & marriage duration

1. 15

0.96

0.99

0.86

1.31

• 15

• 101

(3) 2 + husband's education, husband 1 s occupation & couple Income

1.14

0.98

0.94

0.70

1.27

• 19

• 169

(4) 2 +wife's education, wife's working experience & perceived women's role

1.20

1.01

0.96

0.79

1.30

• 18

• 138

TABLE 5.4 (cont 1 d) Ethnicity

Javanese

Sundanese

Mi nangkabau

Batak

Chinese

Beta*

R2

(5) 2 +gender balance, son preference & rei igiosity

1. 15

0.98

0.98

0.85

1.29

• 15

• 146

(6) 2 + househo Id type, husband dominance & spouse interaction

1.15

0.96

0.99

0.85

1.32

.16

• 104

2 + access to contraception, desired family size & psychic cost of contraception

1.10

0.95

0.91

0.99

1.37

.15

.302

N

505

548

448

493

288

Predictors

(7)

Psychic cost of contraception, access to contraception, and de si red family size are perhaps the most important predictors in explaining contraceptive use. Together, they account for an increase of 20.1 per cent in the R2 even after controlling for ethnic and demographic factors. Taking into account these variables, the mean scores on use of contraception increase for the Chinese and Batak groups but reduce for the Javanese and Minangkabau. These variables have little effect on the Sundanese. Furthermore, these variables do not change the relative importance of the ethnic variables on the use of contraception. Consequently, the ethnic differentials on the use of contraception remain. From the above discussion, it is apparent that there are ethnic differentials in the use of effective methods of contraception in Indonesia. These differentials are consistently found even after controlling for various sets of population characteristics including socio-cultural, economic, psychological, and demographic ones. The Chinese have the highest rate of use of effective methods followed by the Javanese while the Batak are found to have the lowest rate. Given the stability of the beta values for the ethnicity variables over different models we may conclude tht there are predominant influences of ethnicity on the use of contraception. Surprisingly, this persistent value of MCA beta is still found after controlling for important variables such as economic characteristics of the respondents and the accessibility of contraception.

134

VI CONCLUSIONS

Despite the number of studies on family planning, fertility behaviour and population generated in the ten years or so since the beginning of the National Family Planning Programme in Indonesia, studies examining the role of the ethnic factor, a very pertinent topic in view of the pluralistic nature of the society, are still scarce. Those that exist, are usually limited to comparing the two largest groups, the Javanese and the Sundanese. This study is an attempt to contribute to a better understanding of the role of ethnicity in fertility behaviour by the five groups the Javanese, Sundanese, cover1ng all Minangkabau, Batak (Christians) and Chinese (the peranakan, or local-born Indonesian citizens). These five groups by v1rtue of thir numbers can be con si de red the most si gni fi cant groups influencing the structure and pattern of the Indonesian population. The main conclusions that can be drawn from this study are as follows: There are distinct differences among the five groups studied, which is to be expected as they are culturally distinct groups. Nonetheless, there appears to be more similarity between the Javanese and the Sundanese, who are both located in Java. Both groups have a bil at era l kinship system and are predominant 1y Moslem. Some similarities also exist between the Minangkabau and the Batak, who are both located in Sumatra and both have a unilateral kinship system (the Batak patrilineal and the Minangkabau matrilineal). Although the the Minangkabau are Moslem and the Batak Christian, they both share the characteristic of being adherents of their religion. The Chinese, on the other hand, form a group, distinct from the other four, despite the fact that the sample is selected from Chi ne se, who

135

are peranakan. The response on ethnic identity and affiliation indicate that there is in all groups variation by rural and urban residence, the rura 1 samp 1e being more homogeneous and therefore 1ess exposed to contact with other ethnic groups. Nevertheless, all groups seem to be aware of being a member of their own group, despite the vigorous efforts on the part of the government to pursue a policy emphasizing Indonesia as one nation and as a unitary state. In terms of ethnic affiliation, urban people of all groups have ample contact with other ethnic groups. However, Javanese and Sundanese are more 1 i ked by all other groups, than the Minangkabau and the Batak, while the Chinese are the least 1 i ked group. Variation in fertility behaviour is also very apparent, with the Batak having the highest number of living children and number of children desired, the least knowledge of family planning, and the lowest participation in the family planning programme as measured by the use of effective contraceptive methods. It should be noted that contrary to public opinion (as indicated by the response of the four other groups) the Chinese in actual fact have the lowest fertility and the highest rate in contraceptive use thereby debunking the myth that the Chi ne se have the 1argest families and are least co-operative in the national family planning programme. In trying to explain ethnic differences in fertility, study found that ethnicity is an important predictor, though should be aware of the interrelationships between ethnicity other socio-cultural and economic factors. This aspect of study will require more thorough and further analysis of data.

the one and the the

Regarding policy implications, a distinction should be made between providing services in family planning and providing information, education and communication (IEC). The study suggests that for the Batak and the Minangkabau, there should be more emphasis on IEC, for the Javanese and the Sundanese more and better services, and for the Chinese also more and better services in addition to direct inclusion of private physicians in the services.

136

APPENDIX Specification of Variables Used In the MCA Analysis

Dependent Variables CEB

Total number of children ever born

UEM

Use of effective method Effective method Traditional method

RF

2

Effective methods

PII I, IUD, Sterilization, Condom, Injectab Ies

Traditional methods

Douche, Rhythm, Withdrawal, Abstention, Other (Herb, etc.)

Recent Fertll lty, that Is, number of children ever born within last five years.

Independent Variables Ethnicity Ethnic groups of the study Ethnic Affiliation Summation of the following: ( 1)

School Ever attended own ethnic/rei lglous school = 2 No school I ng = 1 Never attended own ethnic/religious school = 0

(2)

Associations Ever belong to own ethnic associations No association = 1 Belong to other ethnic association = 0

(3)

Friends All friends from own ethnic group

137

2

2

No friends at al I = 1 Friends from at least some other ethnic group

0

Ethnic Fertility Approval Summation of the foi lowing: (1)

Better socio-economic position with many or few children Many Ooes not matter Few

(2)

Whether your ethnic group's families have too many, too few or just the right number of chl ldren Too many Right number Too few

(3)

3 2

1

2 3

An ethnic group can only be strong If there are many of them True Partly true Fa I se

3

2

Husband's Education Level of education attained In years Wife's Education Level of education attained In years Husband's Occupation Professional Services/Production Farm Unclassified

1 2 3 4

Couple's Income Perceived Woman's Role Summation of the fol lowlng: (1)

A girl becomes fully a woman only when she bears her own children

138

Strongly agree Slightly agree Neutral SI Ight dIsagree Strongly disagree (2)

(3)

5 4

3 2 1

A woman should devote a lot of time satisfying her . husband. (Score as above) Is it a good idea or a bad idea tor a woman to work outside the home after mrriage? Do you approve of a woman working even it her husband's earnings are adequate? 5 4 3 2

bad idea to work outside indifferent to the first question but no to the second question Indifferent to the first question but yes to the second good idea on the first but no to the second good Idea on the first but yes to the second

Spouse Interaction Whether you can confide in or share worries with your spouse all the time, most of the time, sometimes, or seldom All the time Most of the time Sometimes Seldom

4 3 2 1

Rei igiosity Summation of the following: (1)

How often do you pray? Several times a day Once a day At least once a week At least once a month Less than once a month Never

6 5 4 3 2

(2)

How often do you attend religious services or ceremonies? (Score as above)

(3)

How Important is religion in your daily lite? Very important

=3 139

Quite Important Not Important

2

Months of breast-feeding of the most recent child Female Employment Never worked Work at home Work nearby Work tar away

0 I 2 3

Gender Ba Iance No children Ba I ance More sons More daughters

0 2 3

Son Preference If you were to have exactly three children altogether, how many would you want to be boys and how many to be girls? 3 girls 2 girls, 1 boy

I 2

Either sex 2 boys, 1 girl 3 boys

3 4

Househo Id Type Nuclear (including unrelated persons, e.g., servants/boarders) Joint with one couple Joint with more than one couple

I 2

3

Husband Dominance If you were to disagree about how many children to have, whose opinion would be more important; yours or your (husband's/wlfe's)? Husband Wl te

2

Desired Family Size Number of living children plus number of additional children des I red

140

Perceived Fecundity Relative to other couples of the same age, how difficult Is it for (you/your wife) to become pregnant? Impossible/Infecund Very d I ff IcuI t A little difficult Fairly easy Very easy

0 2 3 4

Access to Contraception How long would it take you to go from your house to the nearest place or person to get supplies or other family planning help, If you need it? (In minutes) Psychic Cost of Contraception Scores obtained by adding all six scores from six questions on the disliked characteristics of contraception: very Important (3), fairly Important (2), not Important (1). The questions are:

2 3

4 5 6

A contraceptive method that may affect (your/your wife's) health A method that is unpleasant to use If it Is Irreversible If it requires examination (of your wife) by a man If it requires surgery If it Is against your rei lgion

Child Mortality None One Two or more

0

2

141

BIBLIOGRAPHY

of

the

Darroch, Russel K. Two are Not Enough: The Value of Children to Javanese and Sundanese Parents. East-West Popul at10n Institute Paper Series No. 60-D, East-West Centre, Hawaii, 1981. The Chinese in Indonesia: A Preliminary Mackie, J.A.C., ed. Survey. Melbourne: Thomas Nelson Ltd., 1976. McNicoll, Geoffrey and Masri Singarimbun. Fertility Decline in Indonesia. The Population Council, Centre for Pollcy Stud1es, New York. Working Papers No. 92, November 1982. National Family Planning Co-ordinating Board. Basic Intormation on Population and Family Planning Programme. Jakarta, 1982.

Suparlan, Parsudi and Hananto Sigit. Culture and Fertility. The Case of Indonesia. Research Notes and D1scuss1ons Paper No. 18. Institute of Southeast Asian Studies, Singapore 1980. Volkstell ing 1980, Deel VII. Chineezen en Andere Vreemde Oosterlingen. (Census 1930, Vol. VII, Chinese and Other Fore1gn Or1entals). Batavia: Landsdrukkerij, 1935. 143

THE AUTHORS MELY G. TAN, Ph.D. is Senior Researcher at the National Institute of Economic and Social Research, Indonesian Institute of Sciences, and Lecturer at the Faculty of Social Sciences and Politics, University of Indonesia. Her other publications include Golon an Etnis lion hoa di Indonesia; Suatu Masalah e Et n1c 1nese 1n Pempinaan Kesatuan angsa, e. Me y G. an Indonesia; A Problem of Nation-Building, 1981); "Women in Jakarta: Family Life and Family Planning" in Cultural Factors and Population in Developing Countries, co-author Hanna Papanek, et al. (1976). She 1s currently studying inter-generational change in value systems among ethnic Chinese in four Indonesian towns, and societal development on Batam Island. BUDI SOERADJI, Ph.D., is Chief, Family Planning and Population Bureau, National Development Planning Agency and Director, Academy of Statistics, Central Bureau of Statistics. His other publications include Fertility Differentials in Jawa (1982), Contrace~tive Use in Jawa-Ball (1982), Reg10nal D1fferentials in Age at 1rst Marr1a~e, co-author Sri Har1jat1 HatmadJl (1982); Soc1o-Econom1c D1f erentials in Fertility and Childhood Mortal1ty, co-author Bondan Supraptilah (1979); On the Apphcat1on of Stable Population Theory (1980), co-author Nathan Keyfitz. H1s current research 1nterests include family planning evaluation, fertility, mortality and population and development.