Children at the millennium : Where have we come from? Where are we going? 9780762307760, 0762307765

1,399 67 2MB

English Pages 323 [337] Year 2001

Report DMCA / Copyright


Polecaj historie

Children at the millennium : Where have we come from? Where are we going?
 9780762307760, 0762307765

Citation preview


ADVANCES IN LIFE COURSE RESEARCH Series Editor: Timothy J. Owens Recent volumes: Volume 1: Volume 2: Volume 3: Volume 4: Volume 5:

Work, Retirement and Social Policy, 1986 Family Relations in Life Course, 1986 Personal History Through Life Course, 1990 Delinquency and Disrepute Life Course, 1995 Self and Identity Through the Life Course in Cross-Cultural Perspective, 2000




SANDRA L. HOFFERTH University of Michigan, Michigan, USA

TIMOTHY J. OWENS Purdue University, West Lafayette, USA


JAI An Imprint of Elsevier Science Amsterdam – London – New York – Oxford – Paris – Shannon – Tokyo

ELSEVIER SCIENCE Ltd The Boulevard, Langford Lane Kidlington, Oxford OX5 1GB, UK © 2001 Elsevier Science Ltd. All rights reserved. This work is protected under copyright by Elsevier Science, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier Science Global Rights Department, PO Box 800, Oxford OX5 1DX, UK; phone: ( + 44) 1865 843830, fax: ( + 44) 1865 853333, e-mail: [email protected]. You may also contact Global Rights directly through Elsevier’s home page (, by selecting ‘Obtaining Permissions’. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: ( + 1) (978) 7508400, fax: ( + 1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: ( + 44) 207 631 5555; fax: ( + 44) 207 631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier Science is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier Science Global Rights Department, at the mail, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. First edition 2001 Library of Congress Cataloging in Publication Data A catalog record from the Library of Congress has been applied for. British Library Cataloguing in Publication Data A catalogue record from the British Library has been applied for. ISBN: 0-7623-0776-5 ∞ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of 䊊 Paper). Printed in The Netherlands.



















7. CHANGES IN AMERICAN CHILDREN’S TIME, 1981–1997 Sandra L. Hofferth and John F. Sandberg







Department of Sociology and Institute for Social Research, University of Michigan, USA

Jeanne Brooks-Gunn

Center for Children and Families, Columbia University, USA

Frances Goldscheider

Department of Sociology, Brown University, USA

Tamara G. Halle

Child Trends, Washington, D.C., USA

Sandra L. Hofferth

Institute for Social Research, University of Michigan, USA

Dennis P. Hogan

Department of Sociology, Brown University, USA

Heather Beth Johnson

Department of Sociology and Anthropology, Lehigh University, USA

Monica Kirkpatrick Johnson

Carolina Population Center, University of North Carolina at Chapel Hill, USA

Tama Leventhal

Center for Children and Families, Columbia University, USA

Penelope L. Maza

Administration for Children and Families, United States Department of Health and Human Services, USA

Kristin A. Moore

Child Trends, Washington, D.C., USA

Jeylan T. Mortimer

Department of Sociology, University of Minnesota, USA

Sabrina Oesterle

Department of Sociology, University of Minnesota, USA vii


Timothy J. Owens

Department of Sociology and Anthropology, Purdue University, USA

John F. Sandberg

Department of Sociology and Institute for Social Research, University of Michigan, USA

Juergen Zinnecker

Department of Education and Psychology, University of Siegen, Germany


SERIES PURPOSE AND ORIENTATION Advances in Life Course Research publishes original theoretical analyses, integrative reviews, policy analyses and position papers, and theory-based empirical papers on issues involving all aspects of the human life course. Adopting a broad conception of the life course, it invites and welcomes contributions from all disciplines and fields of study interested in understanding, describing, and predicting the antecedents of and consequences for the course that human lives take from birth to death, within and across time and cultures, regardless of methodology, theoretical orientation, or disciplinary affiliation. Each volume is organized around a unifying theme. Queries and suggestions for future volumes are most welcome. Please see ~ towens/ALCR/Pages/volumes_description.htm

PURPOSE OF THIS VOLUME The present volume (number 6) is subtitled Children at the Millennium: Where Have We Come From, Where Are We Going? It was produced under the auspices of the Section on the Sociology of Children and Youth of the American Sociological Association. A combination of invited and author initiated papers – all anonymously peer reviewed – this volume seeks to produce a cohesive though wide ranging source of information on the life courses of children and youths. We do this by digging deeply and critically into our social scientific understanding of childhood and adolescence with the aim of guiding the course and future development of research on children into a new century. Contributions reflect: • Demographic analyses and projections • Qualitative aspects of children’s lives ix


• • • •


Children and youth in historical and cross-cultural perspective Issues of development in social context Children and public policy Social and psychological dynamics of childhood and adolescence Timothy J. Owens Series Editor

ACKNOWLEDGEMENTS We wish to thank Diana Blackford for her diligent and cheerful copyediting assistance, often under tight deadlines. Our conscientious and gracious reviewers are also gratefully acknowledged: Peter and Patti Adler, Randal Day, David J. Eggebeen, David Kinney, Kristin Moore, Jeffrey Morenoff, Toby Parcel, Michael Shanahan, and Jane Waldfogel. The other reviewers, who requested anonymity, are also heartily thanked. Finally, we thank the contributors themselves for choosing the series as an outlet for their work. Sandra L. Hofferth Timothy J. Owens Volume Co-Editors


This Page Intentionally Left Blank



Timothy J. Owens and Sandra L. Hofferth The volume includes 10 chapters, including the introductory chapter, organized around four sectional themes: (1) (2) (3) (4)

Historical and Comparative Perspectives on Children and Childhood, From Values and Beliefs to Positive Behavior, Children’s Changing Experiences, and Children’s Living Arrangements.

SECTION I: HISTORICAL AND COMPARATIVE PERSPECTIVES ON CHILDREN AND CHILDHOOD The first section is composed of two chapters: “Children in Young and Aging Societies: The Order of Generations and Models of Childhood in Comparative Perspective” by Juergen Zinnecker and “From the Chicago School to the New Sociology of Children: The Sociology of Children and Childhood in the U.S., 1900–1999” by Heather Beth Johnson. The former provides an international perspective on children and childhood and the latter a historical overview. With respect to Zinnecker’s piece, since childhood can only be understood in relation to conceptions of adulthood, a society’s age-structure must be considered. Changes in these structures have made the contrast between childhood and adulthood more vivid than ever. As the proportion of a society’s under-18 population declines, and, consequently, fewer of its adults’ lives are devoted to Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 1–7. 2001 by Elsevier Science Ltd. ISBN: 0-7623-0776-5




raising children or even to living in households in which children are present (Hogan & Goldscheider, this volume), knowledge of and familiarity with children and childhood weakens. Additionally, childrearing tends to undergo increased professionalization, as entire occupational specializations develop to address the younger age-groups’ needs. According to Zinnecker, as societies age, children and their qualities become even more highly valued. This may lead to prolonging childhood, to attempting to remain young as long as possible, and to replacing children by technology (or immigration). Why, then, are children frequently less well-off than older members of society? Zinnecker argues that it is due to children’s lack of political power. Taking a sociology of knowledge perspective, Zinnecker goes on to outline four conceptual models of childhood: traditional, fundamentalist, modern, and postmodern. The traditional and fundamentalist models are conservative and centered on protecting children through the strong maintenance of cultural and ethical traditions. Religious and other beliefs-oriented communities exemplify these models, such as the home-schooling movement. The modern model is exemplified by contemporary socialization and educational careers paradigms, as well as timebudget studies. The postmodern model, in contrast, permits children to be more active agents in their own socialization. Children’s rights movements are very much a product of this model. In this view, professional agents should have little or no power to set society’s agendas and objectives concerning children. Zinnecker ties each of these perspectives to current research traditions in the social sciences. Finally, the chapter addresses an important topical concern: the role that productive, income-producing activities among children play in this life course era. Is childhood a moratorium from the concerns of adult life or preparation for it? Should children be permitted to be economically active or is “play” their work? The question of whether or not children should be allowed (or even encouraged) to fully participate in society depend upon one’s theoretical perspective and whether one envisions childhood as a time free from adult responsibilities or as a time to train for adult responsibilities? This intriguing and challenging paper is a fitting piece to kick-off our volume of research on children and childhood in the 21st century. Johnson’s paper shifts our primary focus from Europe to the U.S. as it draws an historical overview of American research on children over the past century. Starting with the Chicago School in the early decades of the 20th century and some of its path-breaking work on children (e.g. The Unadjusted Girl, Childhood in America, The Gang, and The Jack-Roller: A Delinquent Boy’s Own Story) and moving into the re-emergence of important sociological investigations of childhood in later decades (e.g. Childhood Socialization, Gender Play: Boys and Girls at School, and With the Boys: Little League

Children at the Millennium: A Report from the Social Sciences


Baseball and Preadolescent Culture), Heather Johnson takes a sweeping look at the origins, growth, decline and re-emergence of the sociological study of children and childhood. While tacitly agreeing with others that the sociology of children spent a substantial period on the margins of mainstream sociology, Johnson demonstrates how important prior research emanating from the first third of the 1900s paved the way for the emergence of a “new” sociology of children in the 1980s. The field is now maturing, and this chapter helps us take stock of where we have been and where we are going.

SECTION II: FROM VALUES AND BELIEFS TO POSITIVE BEHAVIOR This section consists of two chapters: “Parental Values, Beliefs, and Behavior: A Review and Promulga for Research into the New Century” by Duane F. Alwin and “Preventing Problems vs. Promoting the Positive: What Do We Want for Our Children?” by Kristin Moore and Tamara Halle. Alwin’s chapter examines how social structure and culture influence the way parents impact their children. He argues that despite some changes in the nature of the family and the increased role over the past century of other social institutions in the socialization of children, the family (despite some notable claims to the contrary) clearly remains the primary social institution responsible for the care and nurture of children in all societies. While parents’ effects on their children are becoming increasing complex and perhaps more subtle, though certainly not less consequential, the choices parents make with respect to their children are obviously constrained by culture and society and mediated by parental value orientations and beliefs. This chapter aims to clarify our understanding of two key aspects of parent-child interactions (activities and relationships), first, by examining the role that social structure and culture play in shaping parental behavior and orientations with respect to raising children, and, second, by developing a framework that will make it possible ultimately to assess their developmental impacts. To meet these ends, the chapter is organized around five key goals: (1) A clarification of the concept of values and the role values play in childrearing and well-being. (2) A literature review on how parental values and related behaviors are connected to the socialization of children. (3) An overview of research on historical changes in parental values and childrearing. (4) A consideration of the relationship between parental values and childrearing behaviors.



(5) A discussion of several possible research agendas on the role of parental beliefs and behaviors on child outcomes. Kristin Moore and Tamara Halle’s paper focuses on children’s behaviors, especially positive ones. Parents, teachers, and policymakers want children to grow up to be happy, to develop their own special talents, and to relate well to others. Most also want them to develop character, a spiritual side, and tolerance, while also participating in physical, social and cultural activities, being environmentally sensitive, and being helpful to others. However, the indicators gathered by researchers and the government on child well-being are typically limited to measures of problem behaviors and difficulties, such as behavioral troubles, school tribulations, and drug use. This piece provides a useful beginning for thinking about ways to measure and monitor the attainment of these societal goals through better measurement of positive behavior. Moore and Halle review the literature linking a variety of positive behaviors and characteristics to desirable long-term outcomes for children and society alike. Their discussion also raises issues about how such constructs might be measured and monitored. Few current data-collection efforts contain a broad set of positive indicators; it will be a challenge to convince new collections to do so, but this chapter provides justifications for just such activity.

SECTION III: CHILDREN’S CHANGING EXPERIENCES This section consists of two chapters: “Men’s Flight from Children in the U.S.? A Historical Perspective” by Dennis Hogan and Frances Goldscheider, and “Changes in American Children’s Time, 1981–1997” by Sandra Hofferth and John F. Sandberg. A key question addressed in this section is to what extent children’s experiences have been changing and what the implications of these changes are for the future? Hogan and Goldscheider assess whether men’s abandonment of families and, in particular, their children will continue or be reversed. In any given recent year, one-quarter of white and Hispanic children and one-half of black children do not live with their fathers. Surprisingly, however, there have been no systematic examinations of historical data on men living with children. And since men may live with children in cohabiting or in step relationships, and with grandchildren as well as children of other relatives, the end of a first marriage does not necessarily mean the end of their living with children. Hogan and Goldscheider use Census data from 1880 to 1990 to examine historical trends in nonimmigrant men and women living with children in the U.S. The results show that, rather than a new development, recent declines in male coresidence are simple extensions of a long-term trend

Children at the Millennium: A Report from the Social Sciences


and that such trends are similar for women as well. They reason that the trends are due to a decreased proportion of time spent married and to declines in farm residence. Fatherhood is still important, but for a smaller period of the life course. Yet even though the proportions have declined, two out of five adults still lived with children under age 15 in 1990. These proportions would likely increase if immigrant families were included in the analyses since marriage and childbearing often take a larger proportion of their life courses. Next, Hofferth and Sandberg use data from 24-hour diaries of young children’s time to examine how children’s lives have changed over the past decade and a half. Taking a shorter-term view of children’s experiences than Hogan and Goldscheider, they give a picture of how children’s time within and outside the home has changed over the last part of the 20th century. Thinking back to the beginning of the past century when a substantial proportion of families lived on farms, children presumably spent much more time working and a lot less time in school and play activities. Today, however, children’s free time (after removing school, eating, sleeping, and personal care) amounts to a substantial 30% of their day, a small decline from 34% in 1981. The authors show that increases have occurred in structured activities such as sports and art activities while reductions have occurred in play, television viewing, visiting, and passive leisure. Of this discretionary time, about half is nevertheless spent watching television and playing. While a few of these changes are related to increased maternal employment, most are related to such demographic characteristics as increased education and reduced family size. How children spend time reflects both parental values and current cultural norms.

SECTION IV: THE CONTEXTS OF CHILDREN’S LIVES The final section focuses on work, neighborhoods, and social and economic disadvantage. It includes three chapters: “Adolescents’ Anticipations of WorkFamily Conflict in a Changing Societal Context” by Monica Kirkpatrick Johnson, Sabrina Oesterle, and Jeylan T. Mortimer; “Changing Neighborhoods and Child Well-being: Understanding How Children May be Affected in the Coming Century” by Tama Leventhal and Jeanne Brooks-Gunn; and “Confronting the Dilemmas of Child Welfare: Past, Present and Future” by Penelope L. Maza. Johnson, Oesterle, and Mortimer focus on the role that adolescent work plays in the lives of children and their families. Their discussion is advanced by considering the connections between macro-level changes in work and family structures in the U.S. and the micro-level attitudes adolescents hold regarding their future work and family lives. Using longitudinal data spanning



more than a decade, the authors focus on two key though interrelated research questions: (1) Have adolescents’ orientations toward work and family changed in the last two decades? (2) Do adolescents anticipate conflict between work and family roles? If so, how do they respond to such concerns? What difficulties do they face when trying to integrate work and family during early adulthood? Johnson and associates find that while adolescents remain highly interested in work and family roles and have rising aspirations for educational attainment and extrinsic work rewards, as they enter young adulthood they anticipate more interference between their family roles and their career plans. For those who have actually made a family transition, the incidence of work-family conflict is high. This research should be of considerable interest to researchers, educators, and policymakers alike. Leventhal and Brooks-Gunn take up the topic of children’s neighborhoods and child development, particularly well-being. After exploring the current research literature they introduce models of neighborhood change that can be used to address the issue of how childhood contexts influence child development. In doing so they incorporate both static and dynamic influences. One noteworthy contribution is their focus on theoretical mechanisms whereby neighborhood contexts affect children – institutional resources, relationships, and norms/collective efficacy. In particular, they argue that the changes in neighborhoods that children experience are often as important if not more important than neighborhood characteristics at a given time. For example, although most families remain in similar neighborhoods over time, neighborhood changes may nevertheless occur. And while nearly half of all families have ever lived in a poor neighborhood and the same proportion has lived in an affluent one, few remain in a poor or affluent neighborhood all their lives. Leventhal and Brooks-Gunn report that conditions around the time of school entry (age 5) appear to be consistently and strongly associated with children’s IQ scores at age 8. The authors argue that incorporating neighborhood measures will require longitudinal data, while also arguing the need for measures that go beyond the traditional variables presented in decennial U.S. censuses. Maza’s paper completes the volume by discussing the history and current context of child welfare, focusing particularly on the foster care system in the U.S. One of the author’s major concerns is the institution’s fairness, particularly the overrepresentation of low-income families and AfricanAmerican children in the child welfare system. African-American children are

Children at the Millennium: A Report from the Social Sciences


more likely to come from low-income and disadvantaged families, which put them at greater risk of foster-care placement than other children. She offers two explanations for this poverty-child welfare system risk scenario. On one hand, poverty leads to greater household stress, and therefore to increased risk of child maltreatment. On the other hand, poor families are more likely to be scrutinized by representatives of government agencies, which in turn may lead to greater identification of risk and thus more encounters with the child welfare system. The literature seems to support the former more than the latter explanation. While noting the system’s more recent shift from long-term foster care to concern with reunification or adoption, the chapter discusses the general development of the child welfare system from almshouses and orphanages to the more contemporary foster care system and family reunification and adoption initiatives. The growth of family support services and family preservation services are also examined, along with such current trends and controversies as time-limited welfare legislation (TANF) passed in the mid1990s. Finally, a new national study of child welfare, currently in the data collection stage, is discussed. The National Survey of Child and Adolescent Well-Being is expected to have a major impact on child welfare policy and practice in the coming years. Taken as a whole, Children at the Millennium: Where Have We Come From, Where Are We Going? presents a diverse voice on our contemporary understanding of childhood and youth, and is aimed at provoking thought, orienting thinking, and guiding research at the dawn of our new millennium. In the spirit of maintaining a useful dialog, comments and criticisms directed to the editors’ attention is most welcome.

This Page Intentionally Left Blank


This Page Intentionally Left Blank



Juergen Zinnecker INTRODUCTION: CHILDHOOD AND THE SOCIAL ORDER OF GENERATIONS We have learned from recent childhood research in Europe and North America that we cannot speak, write, or research about children and childhood if we are not willing to discuss the status of adulthood in our societies. This means that we must consciously acknowledge when we are doing research on children that we carry with us our specific adult-centered perspectives, values, and ideologies. This is the case even if we are willing to give children their voice in our research, and even if we try faithfully to reconstruct their perspective on themselves and the world. In a more general sense, we have to consciously acknowledge that childhood is embedded in a systemic relationship with adulthood. Childhood and adulthood form opposing social codes that are not only mutually dependent, but which also define and imply each other. For example, if we believe in innocent children, we imply that adults are not so innocent. Children and Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 11–52. Copyright © 2001 by Elsevier Science Ltd. All rights of reproduction in any form reserved. ISBN: 0-7623-0776-5




Table 1. Examples of Binary Codes for Children and Adults in Western History. Children/young


Positive or Negative Value for Children

Primitive Innocent Learning Instinctive

Civilized Knowing Working Controlled

Negative pole Negative pole Negative pole Negative pole

Authentic Creative God-gifted

Socialized Standardized Secular

Positive pole Positive pole Positive pole

adults, young and old, exist semantically for us as binary codes. We are well aware that the history of Western society has produced a number of opposing semantic markers to characterize young and old, children and adults. In most cases, the young pole of the age continuum is negatively coded, though in many instances the young define the positively valued pole. This ambivalence reflects the asymmetrical status of the two groups in Western history as well as a continuing tendency to devalue and discriminate against the younger group. In the case of educational or social reform movements we often find an attempt to socially construct the youthful pole in a way that favors children, by casting them as representatives of a positive alternative to the values of mainstream society (Baader, 1996). In a more general theoretical sense, we suggest that there is a social order of generations that regulates the relationship between childhood and adulthood. Every discourse about childhood, and every research endeavor on childhood conducted by adults is part of the prevailing social order of generations. By their nature, such discussions contain an implicit bias, whether in favor of maintaining or diminishing the social hegemony of adults. Feminists among the new childhood researchers were among the first to propose that the existing social order of generations be conceptualized as analogous to the hierarchical social ordering of gender, race, or class (see Alanen, 1994; Bardy, 1994; Qvortrup, 1991, p. 17; Oakley, 1993). Hence childhood, too, is a relational concept: childhood exists only in relation to adulthood. Social consciousness depicts the child in relation to a conception of the adult just as a sense of adulthood is constructed on the positing of the child. This leads to the suggestions that parallel to a ‘gender agenda’ we can also imagine a ‘generational agenda’ being at work-a particular social order that organizes children’s relations to the world in a

Children in Young and Aging Societies


systematic way, allocates them positions from which to act and a view and knowledge about themselves and their social relations. A generational system, analogously to the gender system? (Alanen, 1994, p. 37).

This idea has much to recommend it. As we will see, it might be useful, at least in the case of childhood, not to speak of a single social order of generations, but to keep in mind that different generational orders coexist at the same time and are sometimes in conflict. Contemporary childhood research orients itself toward different patterns of generational order. In the following discussion we review the most common of these patterns and suggest a typology for mapping models of childhood in modern “aging” societies. In the following section, I discuss the different status of the generations in societies that are “young” and societies that are “aging.”

“MINORITY” AND “MAJORITY” GENERATIONS: THE DEMOGRAPHIC PERSPECTIVE ABOUT YOUNG AND AGING SOCIETIES AND GROUPS We start with a consideration of the quantitative distribution of young and old between and within societies. Since the beginning of the century we find a continuum of aging population, which has produced a shift in the proportion of children and older persons in society (Qvortrup, 1991, p. 20f.). More developed world regions like North America and Europe have been leading this process. According to the Population Division of the United Nations: In 1950 the proportion of children in these (i.e. developed) regions was 27% while that of older persons was 12%. In 1998, for the first time the proportion of older persons exceeded that of children, respectively at 19.1 and 18.8%. By 2050, the more developed regions will have a very old population, with the proportion of older persons projected to increase to 33% in 2050, while the proportion of children will decline to 15% (UN Population Division. Internet-April 22, 1999).

On a global level, we have to keep in mind two aspects of aging: • Aging of societies is an accelerating historical process in the twentieth and twenty-first centuries; • Aging of societies is connected with socioeconomic development. Aging societies are also more developed societies (or regions of the world). The empirical correlation between an aging population and socioeconomic development becomes evident when we compare the age pyramids of more and less developed regions of the world in 1998 (UN data, see above). Another useful statistical measure is the median age of the population – the age at which half the population is older and half is younger. In less developed



Fig. 1. Age pyramids for more and less developed regions, 1998 and 2050 (Medium variant projections). Source: United Nations Population Division, World Population Prospects: The 1998 Revision, forthcoming.

Children in Young and Aging Societies


Table 2. Median Age by Major Area, 1950, 2000, and 2050. (Source: United Nations Population Division). Median age (years)

World total Less developed regions More developed regions Least developed countries Africa Europe Northern America




23.6 21.4 28.6 19.5 19.0 29.2 29.8

26.5 24.3 37.4 18.2 18.4 37.7 35.6

36.2 35.0 46.4 26.5 27.4 49.5 41.0

regions, the median age is 24.3 and in more developed regions it is 37.4 in 2000 (according to United Nations Population Division). The youngest societies are currently the least developed countries, mostly in Africa, with a median age of 18.2 years. The oldest region is in Europe with a median age of 37.7 years. Living in an Aging Society The social demographic distinction between young and aging societies is relevant for articulating the kind of global or world-level theory of generational order that we currently lack. That topic, however, would require separate treatment to do it justice. Here I concentrate on what I believe to be three consequences of living in an aging society. These consequences are important for the way we view childhood and childhood research. First, those defined by their chronological age as “minors” become, for the first time in history, a numerical minority in comparison to the number of adults. This demographic tendency will strengthen further in the next decades. Therefore, it seems useful to take this figure into account as one possible starting point for theorizing about the nature of childhood in advanced modernized societies. It appears that until now this phenomenon is primarily discussed in a very restricted sense; for example, in public debate about participation of young people in voting and other political activity (Palentin & Hurrelmann, 1998; Qvortrup, 1991, p. 22). The problem of this changing demographic balance (Qvortrup, 1994) among age groups also plays a role in the debate about distributive justice between generations (Wintersberger, 1993, p. 8) or about the division of labor between young and old in Europe (Qvortrup,



1993, p. 14). But however important these discussions are, I believe the theoretical meaning of childhood in the social order has a much more substantial importance. Second, the general prolongation of life courses in developed societies has meant an increase in the numbers of generations living together at the same time within a society. In particular, the life phase of “growing up” and the phase of “growing old” are lasting longer. It is also more likely that the vast majority of the newborn population will experience both these life phases. The main stages of life are increasingly differentiated from each other and tend to develop recognizable cultures and lifestyles of their own. Therefore, it seems increasingly inappropriate to speak only of two generations (young and old, childhood and adulthood) and to adopt a more differentiated conception of life phases. In the present political debate over children’s rights, we find a tendency to lump together the status of preschoolers, schoolchildren, and adolescents. The justification for this odd categorization seems to be that all persons in these different stages of life are children up to the age of 18 years because they are dependent according to the UNESCO classification. By this misleading categorization it appears that political elites are trying to capitalize on the positive image of modern childhood and avoid the negative stigmatization of youth (adolescence). But even though this labeling strategy may have its political uses, we find that the reality is for these life phases to be increasingly differentiated in lifestyle and important aspects of attitude and feelings. A special case is the coexistence of different birth cohorts in modern “aged” societies. Within the broad generational distinction between “young” and “older,” we find a number of different historical birth cohorts that were formed by such major social events as war, economic crisis, technological invention, shifts in mass consumption, and rapid change in the traditional and class-based social milieus. It is possible to distinguish at least a dozen political or cultural generations, in the sense of Mannheim or Ryder (1965), that have been formed over the course of cultural or political history. Thus, in modern aging societies there is a greater presence of history among its generations than is true for younger societies. Third, living as a child in an aging society is not just a statistical fact on the macro-sociological level. It also has effects in the everyday lives of children. One significant difference is the increasing age of persons who care for and teach dependent children. For example, in schools we find a shift from younger to older teachers. In family life we find a diminishing influence of siblings and a growing influence of grandparents. The sociodemographic composition of families is also changing. The aging of society means that there

Children in Young and Aging Societies


is a parallel historic shift from young to aging families. Children now are increasingly likely to live in three or even four-generation families and to have the opportunity to interact with a larger number of grandparents and greatgrandparents. Due to the greater life expectancy of women, this is particularly the case for grandmothers and great-grandmothers (Lauterbach, 1995). This shift has long been invisible, because national statistics are based on households and do not count the availability of social networks that extend beyond the single household (Bien & Marbach, 1991; Cochran, Larner, Riley, Gunnursson & Henderson, 1990). Since grandparents, particularly in Europe, normally live in separate households (often only a short distance away) their availability to younger gene is typically undercounted by census data. Social Interaction With Other Generations What does it mean sociologically if children are minorities and adults majorities? According to Peter Blau (1977) and his theory about “Inequality and Heterogeneity,” the relationships between young and old follow elementary rules that regulate interactions between groups of diverse size. Smaller groups such as children have a greater chance to interact with adults who are more numerous. Conversely, the adult numerical majority has fewer opportunities to interact with children. In aging societies thus we find fewer adults or older persons who have an opportunity to share time and build up personal relationships with the young. To illustrate this relationship on a microsociological level, consider the contrasting situations in family groups in young and aging family groups: Case A: In this hypothetical case there are 12 children and two parents – a “young family” form. Parents in this constellation have a good chance to build up personal relationships with at least some of their children, while not all of the 12 children will have the same chance to be intimate with their parents. This is a common story in the social history of family relations in the past (as in historical biographies). Parents in this situation tend to handle the siblings as a group or collective, though they are likely to emotionally favor selected children (for example, the oldest son). In such a young family, or more broadly, in young societies, parents and adults generally are a scarce resource. Members of the younger generation must compete among themselves for attention from adults. Case B: In this hypothetical case, let us take a balanced configuration of two children and two parents. Obviously, both sides have an equal opportunity for contact with the members of the other generation; mathematically, the social resources are balanced.



Case C: In the case of “aging families,” the generational relationships are quite different. If, for example, there are two children and six adults (two parents and four grandparents), members of the younger generation benefit from the availability of a proportionally large network of adult contacts. In this case, the young are a scarce resource and the older generation competes for contact and affection of the younger generation. Even these adults are fortunate in comparison with a certain percentage of modern couples who have no children and may therefore lack social contact with children completely. Social Knowledge: What Each Generation Knows About the Other The social knowledge each generation has about the other varies greatly between hypothetical Cases A and C, if we consider the unequal chances of both sides to interact directly with persons of the other generation. In young families (and, by extension, in young societies), we would expect the older generation to be more informed about the younger generation than vice versa. In aging families (and aging societies) the younger generation should know more about the older generation, while the older generation will be less informed about the problems, lives, and personalities of youth. In aging families and societies the proportion of adults and older persons who have no firsthand experience with the minority of children is growing. The younger generation seems increasingly to represent an outgroup for them. In the worst case, youth and particularly most recent birth cohorts are simply strangers. The knowledge deficit of the older generation in aging societies seems even more dramatic if we take into account the restrictions built up in modernized societies against ”unauthorized” adult contacts with minors. The childhood moratorium (discussed in more detail later) in modernized societies is twofold: First, it comprises a private family moratorium and, second, a semipublic educational moratorium. This means in essence that only parents and professional educational staff are authorized to have meaningful contact with children. A variety of public spaces are taboo for both children and for unauthorized adults (Zinnecker, 1990). By contrast, public space in both urban and rural communities served as an important zone for contact between younger and older generations in the past (Behnken & Zinnecker, 1989, 1992; Lang, 1998). If there is the potential for the young to become an unknown and alien population from the perspective of the older generation, there is a need for developing compensating techniques that serve to redefine and maintain contact with youth. One such technique in aging societies is a public discourse on the younger generation carried out in the public media and in the social

Children in Young and Aging Societies


sciences. This discourse enables the adult generation to compensate their lack of firsthand knowledge about the latest developments and changes in the younger generation. Educational advice for parents in books, magazines, and television has been a booming media industry since the 1950s. We read and hear about troubling children and children in trouble (Best, 1990, 1994). The continuous public discourse on these topics helps create the political motivation both to defend against and to save the younger generation through social work. What role does contemporary childhood research play in this context of aging society? One central function of childhood research is to construct for the adult audience images of the young generation, another is to evaluate the lifestyles and social problems of youth. These adults may lack firsthand knowledge of the younger generation, but wish to learn how to deal with these youngsters. This means that the public interest for new (and old-fashioned) childhood research is not restricted to professional experts (e.g. those who manage children within institutional settings such as schools). The research reaches a greater public starved for information about the younger generation. Which Generation is a Scarce Resource? Which is Valued? In comparing the models of the young with aging society, we suggest: In young societies (both historically and in less developed regions today), members of older generation represent a demographic minority and are therefore a scarce social resource. Not only are the adult persons that the younger generation may interact with relatively scarce, but the specific values and competencies of this older generation are scarce, highly valued goods, whether it is historical knowledge, wisdom, or simply life experience. Conversely, in aging societies (the case of highly developed and modernized societies), the younger generation takes on the quality of a scarce resource. The young persons and the qualities one can attribute to the younger generation become highly valued, whether it be health, beauty, mobility, quickness to learn, optimism, initiative, or even rebelliousness. Three Social Strategies to Compensate for the Scarcity of Youth Aging societies have developed typical strategies to compensate for the scarcity of youth and the younger generation. The problem is that even when youth are relatively scarce, the qualities and values associated with the younger generation are still needed. There are three typical compensatory strategies: the first is the prolongation of being young; the second is to create a lifestyle of



“being young” that is independent of actual chronological age; and the third strategy is to create a “virtual” young generation by means of technology. “Being Young” Longer The first strategy is to enlarge the amount of time one is able to be young in the course of a single life. Instead of ending the social and economic dependencies of childhood and youth around the twentieth birthday, it is possible for an increasing proportion of young people to prolong this phase until the age of thirty or more. The times of learning and examination in schools, universities and other institutions of the educational system are prolonged. Where society once prescribed normative times for the occurrence of such status passages as getting married and beginning a family, these norms are less binding than in the past. There is more room for individualized choice, and these events take place at a later time in life for many young people. Even though the choices are individual ones, the social effect solution seems to compensate the declining proportion of the younger generation. Though there are proportionally fewer young people, the period of youth is extended. The result is an effective increase in the number of dependents, quasi-children and quasi-adolescents. The effect can be calculated simply: If 100 young people spend 10 years (say, between the ages of 10 and 20) as learning, creative, experimenting adolescents, society gets 1,000 life-years of youth. If that number is halved, the same effect can be achieved by having those 50 people stay in a dependent or quasi-adolescent for 20 years (say, between their tenth and thirtieth birthdays). Either strategy produces 1,000 life-years of youth for the society. This social and cultural technique changes the Lebenswelt (life space) of the younger generation and, because the life courses are increasingly individualized, this tactic results in a number of unanticipated consequences and costs. Creating the “Young” Regardless of Age A second solution for increasing the amount of youth in society is to transfer many of the qualities and values of being young from younger to older generations. In this strategy, the adult generation (or at least a noticeable proportion of them) takes on some of the qualities of youth and redefines the nature of adulthood and old age. As a result, more adults try to emulate an exploring, adventuring, learning lifestyle – a lifestyle that also usually emphasizes physical beauty and strength as well. A consequence of this development is that it leads to a weakening of the traditional demarcation between childhood/adolescence/dependence on the one hand and adulthood/ independence on the other. The critical generation discourse that arises from this situation is about a “crisis of adulthood” – led by a tendency for the older

Children in Young and Aging Societies


generation to become “childish” and immature (Lenzen, 1985). The obverse of the argument is a perceived tendency for children to hurry up childhood culturally and psychologically (Elkind, 1981) to become prematurely selfdirected and independent of the adult generation. Creating a “Virtual” Young Generation Through Technology The third solution of aging societies for the problem of declining proportions of youth is far more radical. Modernized societies reconstruct many of the virtues and skills of young people through technology. They create a virtual childhood and youth by inventing machines and robotics that possess the productive qualities and labor power of young people. This process began in early periods of industrialization when the physical strength of young male workers and the handicraft of young female workers were replaced by machines. The same historic development occurs in the case of family household work which substitutes for having young females in adult households. In this respect, aging societies demonstrate their economic independence from the younger generation. They have created technologically a very effective virtual manpower that is more productive than that of the youthful manpower in economically undeveloped societies. By this strategy, adults may benefit from the fact that the invented virtual youth is less deviant and less rebellious than real young people are. This substitution effect holds true not only for peacetime economic productivity, but also for waging war. Historically, young men have served as warriors, particularly in young societies. Since the nineteenth century, national societies in Europe have recruited young men for the army (Loriga, 1997). Their current value has diminished by the introduction of technological weapons, which only need the assistance of a relatively small professional army (which may now include young women), instead of whole generations of young men (Gillis, 1993). It is worth mentioning one significant failure in replacing the real younger generation by a virtual younger generation. Modern societies have been far more effective in replacing adolescents and youth than children. Children possess certain emotional qualities and social psychological power to be meaningful for adults (as shown in the “values of children” studies, (Zelitzer, 1985; Nauck, 1989, p. 53ff). The emotional rewards provided by children cannot as yet be replaced technologically. This constitutes a major difference in the social value and status of children – as opposed to youth – in modern “aged” societies. The qualities of children are highly valued, the qualities of youth are increasingly devalued. It is not by chance that children, not youth, are



most effective as models in marketing consumer goods. It is also no coincidence that the UN has declared the Rights of Children and not the Rights of Youth. (Although according to the UN, adolescents up to eighteen years of age are included in the Rights of Children). Children embody a positive image while youth embodies a negative one. In some respects we suspect and predict for that in the future – the youth as a stage of life will become superfluous and undesired. Gillis, the historian of youth, goes so far as to speak of “vanishing youth” and an “uncertain place of the young in a global age” (1993). One mitigating circumstance is that modern aging societies attract many young people from less developed (and younger) societies. These young immigrants replace the native born youth in positive and negative ways, and often become part of the public discourse on youth as a social problem. The combination of generation and ethnic differences produces among the older generation a doubly motivated fear of young people who are also aliens. Social Policies for the Younger Generation: Why is the Social Status of Children and Adolescents Deteriorating in Modern Societies? Most comments on childhood in aging societies underline the hostility of these societies towards children. At the same time, however, modern societies claim to be more child-centered than their predecessors in history. These contradictory perspectives are not easy to disentangle. The minority status of childhood in modernized societies may help us to understand this ambivalent process. A comparable age group in modern aging societies, which might help us better understand the mechanisms at work regarding children (and adolescents), is the older segment of the population, those who are no longer part of the work force. Both population groups depend on the active, working adult population to care for them in the form of money transfers and other material or immaterial services. The question arises whether both population groups, the older and the younger, profit from society’s services to the same extent, or whether one group enjoys advantages over the other. The answer to this question can help us define the relative status of the given age group in the sociopolitical system. Surprisingly, one has noticed a shift of transfer services in favor of the older population during the second half of the twentieth century. While their relative economic and social psychological status improved, that of younger age groups fell behind. Probably the first one to discover this historic shift in the recent past, and to document it extensively, was the demographer

Children in Young and Aging Societies


Preston (1984). He compared the development of the status of old and young age groups in the U.S. between the 1960s and 1980s and discovered that during these years, the long familiar risk of old-age poverty had become a new risk of childhood poverty. Apart from an improved economic situation, Preston also revealed a shift in the social psychological well-being of the traditionally more problem-fraught age group of the elderly. They were feeling better and better, while the well-being of the young deteriorated. By comparing both groups’ suicide rates, sociopsychological health, and subjective feelings of satisfaction with life, Preston was able to validate his observations. The support and quality of the two most significant institutions for each age group developed in opposing directions as well. While spending on schools and their efficiency (important for the young generation) stagnated, the efficiency and benefits of the public health system increased and improved dramatically, the main ones to benefit being the elderly (as their increased life-expectancy rate proves). Demographically, this contrasting development of the relative status of the young and the old may at first seem paradoxical. The longevity of the elderly is objectively increasing the burden on society, whereas the diminishing population of young people objectively lightens society’s load. Yet, old people are generally better off, while the situation of the young is becoming more and more problematic. How can this contradiction be resolved? The improvement or deterioration of a given age status can be traced back to changes in the private, familial sector as well as to changes in the public transfer system. During the second half of the twentieth century, the family system has changed in a way that has caused disadvantages for the young generation. An increased divorce rate, a growing number of single parent homes and a strongly developed individualization of the private value system and way of life of adults have become risk factors for the young. Their chance of becoming the victims of neglect within their family has increased. A nontrivial segment of the adult population decides not to have children at all, avoiding this burden altogether. In comparison, the situation of the old remains relatively stable in a developed social welfare state that offers public pension plans. Their welfare does not depend as much on the stability of family systems. In the public sector, amount and quality of transfer services depend largely on the climate and conditions of political processes and decisions (see Preston, 1984, p. 446). Young and old profit from these transfer services to the extent to which they can influence politics and gain the politically active public’s support and votes. In this respect, the large population group of old people clearly has an advantage. They not only have the right to vote in favor of the kinds of transfer services they desire, they can also rely on a strong lobby among the



adults still in the work force. These are motivated to support the older population group for two reasons: they know that they are shaping their own future as elderly people when they establish and maintain a system of financial support for the old, plus they are aware of the fact that transfer services for the older generation makes it easier for them to care for elderly family members. When their policies favor the old, adults can transfer resources back to themselves. In comparison, children find themselves in a much weaker position. They are not allowed to vote, and their parents only make up a small fraction of the adult population. Those adults who do not have any children are less interested in transfer services to children or families with children. These are “other people’s children” (Preston, 1984, p. 452) who additionally are often (immigrant) “minority children.” One also has to keep in mind that these childless singles and couples are often the better educated, wealthier and politically more active adults. The already-discussed social psychological alienation of the older generation regarding the young minority intensifies the problem. Many adults are not even familiar with problems the young generation has. What reaches them is only the often-distorted image of childhood and youth conveyed by the media. Here, children are often portrayed as affluent consumers who terrorize their parents with their demands and interests in “conspicuous consumption” and who don’t hesitate to use violence among each other. Therefore, we find aging societies structurally indifferent or even hostile towards specific interests and needs of the young minority. At the same time, there are more opportunities than ever to compensate for the consequences of this structural indifference and inconsiderate behavior of the adult majority. Because of the relatively small number of children and the affluence of the adult population, economically developed societies can invest in every single child – more than any social structure of the past or any young society of the present. To apply this principle, for example, to the chronic social problem of childhood poverty within developed aging societies, we have to analyze why this type of generational inequality is easily generated in modern affluent societies. At the same time we have to keep in mind that children’s problems and their low status are politically and structurally provoked in aging societies, but are by no means a necessary or unchangeable feature within modern aging societies, as is certainly the case in undeveloped young societies. The problem of low access rates of the young to the resources of modern society is caused by social policies and may become the focus of a social movement to fight the uneven material and immaterial distribution between the younger and older generations (distributional justice).

Children in Young and Aging Societies


MULTIPLE MODELS OF GENERATIONAL ORDER AND OF CHILDHOOD IN MODERN AGING SOCIETIES Modern aging societies do not follow one single model of a generation order. They have produced different social orders of generations, which often include different ideological representations. These different conceptions of generational order coexist in society and compete with each other. To distinguish among them, we suggest the following question as a criterion: How do the models of childhood models interact with the phenomenon of modernity? By doing so we get four fundamental (ideal) types of orientation towards the historic process of modernity and toward the acceleration of modernization in economically developed societies: The typology I suggest has four ideal types: postmodern, modern, traditional, and fundamentalist orientations and lifestyles. All of them are deeply rooted in the social history of modern industrial societies and can be understood as fundamentally different ways of dealing with the sociocultural transformation, paradoxes, and social problems that are integral parts of the process of modernization in society. The postmodern orientation is associated with the postindustrial sector of modern societies, with its service industries, advanced and hedonistic consumerism, electronic communications and virtual realities created by modern media. The modern orientation represents features of the classic industrial world; it is based on the ascetic Protestant work ethic as described by Max Weber, and it focuses on individual achievement, mobility and career development. The traditional orientation focuses on the assimilation of values and lifestyles of the preindustrial world and can be understood as a partial rejection of the culture of advanced modern society. The fundamentalist orientation is dogmatic, religious, and sectarian in its adherence to idealized premodern values and lifestyles; it actively opposes modernity by building islands or niches of the old world into modernized societies. The Four Models of Childhood and Their Images of Society If we apply this typological scheme to the topic of generational orders in modern society, it is possible to distinguish among four types of childhood (Zinnecker, 1996; see Table 3, column A). We refer to them as: I Postmodern Childhood II Advanced Modernized Childhood III Traditional Modernized Childhood IV Fundamentalist Childhood

B Description of Generational Model (Types of Childhood Moratoria) (Patterns of Adult-Childhood Relationships)

C Strategies for Handling Modernity (Politics/Pedagogy of Childhood) (Actor Models of Childhood)

D Social Construction of Childhood in Scientific Discourse

AI Postmodern Childhood

BI Children as a minority group have to fight for their right to participate, not to be discriminated against by age/dependency, and to build up identity, self-esteem. Adults have to be “politically correct” in not discriminating against children.

CI (Reintegration of children into society) Empowerment to survive as a minority group and to protect themselves. Children’s rights in society as a political movement.

DI (New childhood research) Self-constructive child (Autopoietic constructivism) Minority child (Social Policy) Social structural child (Demography) Tribal child (Ethnography)

A II Advanced Modernized Childhood

B II Children live in open institutions specialized for their needs and interests. Children are self-responsive for careers and well-being in moratoria – as adults are individually responsible for their own world. Children gain access to commercialized zones. Changing balance of power between adults-children

C II (Open and de-educated childhood moratorium) Individualizing children within institutions – Learn to handle ambivalent situations in imperfect protected moratorium and in future life course

D II (Socialization research) Parent-socialized child Peer-socialized child Media-Socialized child Self-socialized child

A III Traditional Modernized Childhood

B III Children live in closed and safe institutions of “pedagogical moratorium” (family, school, church, neighborhood), separated from adult work and life. Differences between children and adults in development, knowledge, and power are stressed. Adults have to protect children from adult modernity.

C III (Safe developmental niche) Protecting children by welfare and education. Adults as professional and voluntary child savers.

D III (Pedagogical-psychological research) School- and family-educated child Developing child (endogenic perspective)

A IV Fundamentalist Childhood in Modernity

B IV Children live together with adults in separated cultures and territories, opposing and retreating from modernity in a fundamentalist way. Children and adults as tribal minority or subcultures.

C IV (Tribal niche with adults) Retreat and opposition against modernity in an archaic tribal world and generation community

D IV (Research from the outside perspective on political and religious movements) (Little self-representation in childhood research) ?


A Childhood Label (modernity)


Table 3. Mapping of Four Childhood Models Within Different Social Orders of Generations, Coexisting in Modern Aging Societies.

Children in Young and Aging Societies


Each of these four childhood models (column A) implies a different ordering of the relationship between young and adult generations (column B), differing notions of the actions of children and of strategies for dealing politically and pedagogically with childhood problems (column C), and with differing constructions of childhood in social scientific discourse (column D). Postmodern Childhood Postmodern children are media kids and consumer kids (Del Veccio, 1997), and it is on this image that many of the hopes and fears concerning the future of modernity are focused. These children understand toll-free numbers, cyberspace dating on the Internet, video films of horror and violence, the latest fast-food news, and have an elaborate knowledge of brand names of clothing for their age set. They already live in a mass media hyperreality. They use the everyday world as a stage for subculture performances and the presentation of themselves – a privilege formerly connected only with older youth generations. This model casts children as postmodern urban characters: ironic, skeptic, morally relativistic, not at all innocent. For critics of modern culture, postmodern children are proof that childhood, as a historic project of European Enlightenment, has come to an end (Hengst, 1996; Cross, 1997). These critics denounce “the disappearance of childhood” (Postman, 1982), the emergence of “children without childhood” (Winn, 1981), “childhood as a fiction” (Hengst, Koehler, Riedmueller & Waumbach, 1981) or the modern “mythology of childhood” (Lenzen, 1985). It is argued that the real existence of postmodern kids contradicts the historic model of childhood as protected, innocent beings in a separate, peaceful garden of education (Higonnet, 1998). Children are no longer separated from the worlds of adult knowledge and adult recreation. Instead, they often serve as an avant-garde in inventing, detecting, and using newest consumer fashions. In this model, children are no longer served and controlled by the agents of the educational establishment. Childhood is constructed by media, toys, food, and other childrelated industries. Postmodern childhood is directed and encouraged by corporate commercial interests. Pedagogical sites are those places where power is organized and deployed, including libraries, TV, movies, newspapers, magazines, toys, advertisements, video games, books, sports, and so on. The organizations that create this cultural curriculum are not educational agencies but rather commercial concerns that operate not for the social good but for individual gain. Cultural pedagogy is structured by commercial dynamics, forces that impose themselves on all aspects of our own and our children’s private lives. Corporate cultural pedagogy has ‘done its homework’ – it has produced educational forms that are wildly successful when judged on the basis of their capitalist intent. Replacing traditional classroom lectures and seatwork with dolls with a history, magic kingdoms, animated


JUERGEN ZINNECKER fantasies, interactive videos, virtual realities, kick-boxing TV heroes, spine-tingling horror books, and an entire array of entertainment forms produced ostensibly for adults but eagerly consumed by children, corporate America has revolutionized childhood (Steinberg & Kincheloe, 1997, pp. 3–4).

Modern Childhood Modern and postmodern models of childhood are similar in that they both blur and transgress the traditional line separating the worlds of childhood and adulthood. In the case of modern childhood, children participate in the highly organized and stressful urban life of adult civilization. They are little managers in their own world of schooling, household, traffic, malls, culture and sport clubs. In searching for characteristic features of modern childhood, we hypothesize that we will find in the everyday world of children essentially the same dimensions of modernity which constitute the everyday life of adults in developed industrialized societies (Buechner & Fuhs, 1994, p. 66).

Modern children have a busy, fully booked schedule. They need watches and calendars for organizing their appointments at an early age – in the case of Germany, time consciousness and time management starts for most kids within the first two grades in elementary school (Schorch, 1982). Modern children no longer stroll around the neighborhood on foot as they did generations ago. Instead they commute by car (i.e. their mother’s car) or, if necessary, by bicycle. They settle between home and a diverse set of institutionalized “isles” or sites in the community that have been developed to serve the special needs of the children (Zeiher, Büchner & Zinnecker, 1996; Zeiher & Zeiher, 1994). A U.S. survey on children’s time schedules in 1997 reports a decline (since 1981) in unstructured play activities among 3- to 12-year-olds. It also finds that an increasing amount of time is spent in institutions with preplanned and supervised activities, particularly in school and in sports (Hofferth, 1999; Hofferth & Sandberg, 2001). The author explains the changing time obligations of children as a result of the growing time stress in households where both adults are employed. (Between the 1980s and the 1990s the amount of time both parents had to work to meet the family budget has grown substantially.) A second explanation for the changing timetables of the children may be the growing competitive pressure among modern children to acquire educational degrees and other elements of cultural and economic advancement (Zinnecker, 1995; Buechner, 1996). The acquisition of cultural capital (such as knowledge, “taste,” and educational degrees) creates youthful careers in pursuit of privileging credentials. Children and their parents alike set off on a chase for the most aristocratic credentials education has to offer, that is, advanced secondary

Children in Young and Aging Societies


education diplomas and university degrees. For example, a recent representative survey of German children aged 10 to 13 (Zinnecker & Silbereisen, 1996) found that 50 to 60% of all children hope to achieve the Abitur, the highest level of secondary school education in Germany, a necessary precondition for entering university studies. Modern children regard successful school careers as the most important pathway towards successful adulthood (not to mention an important marriage strategy). A majority of them report that school reports, homework, and their annual school performance are the subjects of many family discussions and are taken very seriously. In the case of the U.S., Schneider and Stevenson (1999) compared the educational ambitions of young adolescents in the 1950s and the 1990s and found convincing empirical evidence that the later cohort is much more ambitious concerning their educational careers than were the baby boomers of the 1950s. Apart from educational qualifications and credentials, childhood careers related to leisure and consumption activities are gaining in importance. These careers may be in sports, extracurricular activities involving creative and artistic skills, careers in the media system and roles as fashion leaders. These leisure careers are capable of producing additional leisure stress. Recent German data indicate (Zinnecker & Silbereisen, 1996) that about half the children say that they are active members of clubs and organizations. About 50% of the children practice particular skills, especially in sports and music (usually playing an instrument). One-fourth of them attend workshops or study groups at school in the afternoons; two-thirds have already acquired extracurricular diplomas and prizes. Leisure activities and hobbies pursued by parents and children together form part of the weekend schedule in a significant number of families. There is also a growing process of “scholarization,” which represents the centrality of the educational process in organizing the lives of children. Figure 2 shows this development in the schedules of modern children (Zinnecker, 1995, p. 89). It demonstrates that the historical tendency toward scholarization among young people of all ages not only takes place in schools, but also within families, extracurricular activities, and in clubs. The developmental psychologist David Elkind has criticized the modern type of childhood under the heading The Hurried Child (1981), which is “growing up too fast too soon.” Today’s pressures on middle-class children to grow up fast begin in early childhood. Chief among them is the pressure for early intellectual attainment, deriving from a changed perception of precocity. Several decades ago precocity was looked upon with great suspicion. The child prodigy, it was thought, turned out to be a neurotic adult; thus the phrase ‘early ripe, early rot!’ Trying to accelerate children’s acquisition of academic skills



Fig. 2. The process of schooling in childhood, adolescence and young adulthood. Source: Zinnecker, 1995, p. 68. was seen as evidence of bad parenting . . . . The change . . . reflects the new attitude that the years of childhood are not to be frittered away by engaging in activities merely for fun. Rather, the years are to be used to perfect skills and abilities that are the same as those of adults. Children are early initiated into the rigors of adult competition (Elkind, 1981, pp. 6, 9).

Traditional Childhood Postmodern and modern models of childhood both place positive emphasis on the transforming processes taking place in contemporary societies. They even see children as an integral part and even the avant-garde of these transformations. Traditional and fundamentalist patterns of childhood take an entirely

Children in Young and Aging Societies


different position on modernization, projecting a model of childhood that is in opposition to mainstream contemporary developments. Fundamentalist childhood is designed as anti-modernist life trajectory, while the model of traditional childhood follows a strategy that might be described as “defensive modernization.” The traditionalist model seeks to maintain reliable traditions by adapting them to the changing societal environment. As family groups and their networks attempt to transmit cultural traditions to the next generation, they have a natural interest in keeping the primary group and their specific culture alive. They are suspicious of changing processes in society and their possible negative effects on the continuity of their own microsocial environment. Therefore, family groups and networks try to control and preselect the impacts of new values, commodities, or electronic media before they are allowed to cross the threshold of the family home. Research on the effects of media, especially TV, on the life and personal development of children, underestimates the effect this family constellation has on guiding the media habits of children. This form of research often assumes that children have to deal alone with the risks and temptation of media consumption. A more realistic assumption would be that most children are guided, and to a certain extent protected, by specific family traditions in managing the impact of exogenous media stimuli. Family groups also develop informal as well as institutional networks as cultural reference groups to protect their own traditional milieu against the dominant urban-corporate culture. They are interested in the synergistic effects that neighborhoods, local parish, and private (religious) schooling may exert on the education of their children. They want to assure that those institutions are following the same values and aiming at the same educational goals. One window on this phenomenon comes from the strong empirical tradition of social capital research, which strongly favors the integration of family and educational institutions (Coleman, 1987). Surprisingly, in light of strong historical processes of secularization in Northern and Western European societies, we find an astonishing large number of children with religious and church affiliations. Half of the children in Germany, for example, define their family education as religious, believe in a personal Christian god, pray regularly, and participate in religious services on a regular basis (Zinnecker & Silbereisen, 1996). Our research with German youth indicates further that religious beliefs and practices are among the cultural traditions that parents most successfully transmit to the young generations. Families are much more effective in this area than, for example, in transmitting orientations in music, education or sports (Zinnecker & Hasenberg, 1999). Research suggests the



same strong transmission of religious culture to be present in the U.S. as well (Myers, 1996). Fundamentalist Childhood In the fundamentalist childhood model, the religious component plays a central role, although nonreligious movements may also propagate a fundamentally belief-oriented childhood. It is crucial for this model that the children should fully be educated in alternative communities, apart from state schools, apart from the mundane economic- and consumption-oriented world, and protected from other secularized institutions and influences. Fundamentalist childhood models played a role in the development of alternative lifestyles of the 1960s student movement. In the framework of the nonauthoritarian and alternative movements of the seventies, alternatives to standardized middle-class education and to public schools were developed (e.g. Free School Movement). A number of utopian educational models served as ideological models; for example, the collective education model of the kibbutz in Israel, anarchical school programs like that of Leo Tolstoy in Russia, or the Free School of Summerhill (A. S. Neill) in Great Britain (for more documentation on Europe as well for North America, see the evaluation of childhood commune education in Eiduson, Kornfein, Zimmerman & Weisner 1982; Cyprian & Wurzbacher 1979). As a result of the fundamental opposition against the modern way of life, new religious movements like Hare Krishna or Sannya grew sharply in the seventies. At least since the early 1980s, the children of these movements have lived a fundamentalist childhood in Europe and North America. (For recent ethnographic and theological research on these groups, see Palmer & Hardman, 1999.) In the long term, however, the conservative fundamentalist childhood models, which are aligned with conservative fundamentalist and Christian religious traditions, have been much more successful and influential. The wellknown Amish groups in North America (Spindler, Enders-Huntington & Hostetler, 1991) offer a living museum of such a childhood. In that case, the religiously grounded community life of what was originally a medieval European sect has been preserved. Also prominent today are educational programs that originated with conservative evangelical Protestants in North America (Coleman, 1999). Comparing these conservative religious groups with mainstream experts, Bartkowski and Ellison (1995) . . . argue that these Christians have consciously promoted ‘traditional’ parenting techniques in opposition to those of the mainstream. Literalist views of the Bible as providing fail-safe, sufficient truths to guide the conduct of all affairs, including

Children in Young and Aging Societies


childrearing, are generally maintained. Also evident are beliefs in the sinfulness of human nature, leading to concerns with punishment and salvation. Overall, an affinity seems to exist between literalism and authority-mindedness in the texts they examine, and this has an impact on childrearing. Writers imply that to succeed in adult roles, children must embrace principles of parental authority and family hierarchy. Virtually all of the Christian parenting specialists reviewed claim that children derive their initial image of God from the behavior of parents, so the cultivation of a correct image of parent as divinity must be maintained. Such views, to the extent that they stress the importance of authority, generally fit in with conventional depictions of conservative Protestant families as idealizing a family form that involves rigid role demarcation according to gender and age: the father acting as spiritual head of the family and chief provider, in an economic sense; the mother providing a stable home base for their offspring; and the children learning obediently to adopt roles as mirror images of their parents (Abstracted by Coleman, 1999, p. 77).

From the description it seems clear that Protestant fundamentalism directs itself against the widespread crisis of parental authority and the debate over the ability of the parents to educate their children. (For example, see the fundamental critique of popular beliefs among family socialization researchers by Harris (1999). The success and the credibility of this book may be seen as a scientific mirror image of fears about the weakness of everyday parenting today.) In the case of European societies, we find fundamentalist tendencies among some groups of immigrant parents and their offspring. This applies specifically to German-Turkish Muslim fundamentalism (Heitmeyer, Müller & Schröder, 1997). Muslim parents are immigrants from relatively traditional (and young) societies in Asia Minor or North Africa. They have strong reason to fear and to counteract the decay of their authority in postmodern European consumption societies. In 1983, a homeschooling movement began in North America under the influence of conservative Christians. A variety of different groups assembled in this movement, united by the idea of breaking the state monopoly on the schooling of their children and teenagers, sought to recapture the right to instruct children at home. This movement came to involve school opponents of the seventies, advocates of open classrooms, and many interested in community education. In the center of that movement, however, stands the idea of “Christian homeschooling.” The Bible-loyal Protestant families attempt to find a Biblical Foundation for Christian homeschooling. An Internet home page of a Christian family from August 16, 1999, sets aside the following statement of faith and purpose: Thank you for visiting our homeschooling center. We are the Moore family from Ponca City, Oklahoma. My name is Robert. My wife Christina, and I homeschool our 3 children – Susanna (12), Elizabeth (10), and David (8). We teach our children in response to God’s


JUERGEN ZINNECKER instruction in Deuteronomy 6:5–7. ‘And thou shalt love the LORD thy God with all thine heart, and with all thy soul, and with all thy might. And these words, which I command thee this day, shall be in thine heart: And thou shalt teach them diligently unto thy children, and shalt talk of them when thou sittest in thine house, and when thou walkest by the way, and when thou liest down, and when thou risest up.’ This homepage is dedicated to families who desire to pray together, worship together, minister together, and learn together. One of our primary goals is to be an encouragement to other Christian homeschooling families ( hmsc/home.htm, August 16, 1999).

The homeschooling movement has a very effective national organization, the Home-School Legal Defense Association (HSLDA). This organization enforced since the eighties some test cases in different federal states in order to change existing school laws in favor of homeschooling. HSLDA sees as its main opponent the educational establishment and particularly the National Education Association (NEA). The association emphasizes as its future tasks: “Protecting Parental Rights,” “Regaining and Restoring Religious Freedom,” and concretely “Fighting State regulations,” which are directed against the homeschooling of the families (On the Edge of the 21st Century (, August 16, 1999). According to the HSLDA, it presently represents over 55,000 families and has a staff of 58 persons. On the Internet, it is easy to find at least 100 book titles in which the programs for homeschooling are explained, didactic hints are given, and personal experience is reported. It is estimated that in the mid-1990s approximately 300,000 children were being taught at home. With homeschooling, parents have revived a model of childhood education that played an important role between the sixteenth and nineteenth centuries (prior to the beginning of what we now call modernized childhood). Throughout those centuries, homeschooling was particularly popular among the aristocracy and wealthy bourgeoisie. It was commonly known as the home teacher system, and was especially influential in the education of girls (keeping them within the private sphere and out of the public schools). Reinvented at the end of twentieth century as a fundamentalist model of modern childhood, homeschooling tries to reestablish the hegemony of the traditional family constellation. The objective is to protect children from the diversity of postmodern value systems and from modern bureaucratic school systems (as the embodiment of a dehumanized and secularized society). Childhood Models and Social Science How are the different models of childhood reflected in different scientific representations of childhood? (see Table 3, column D). From the perspective of the sociology of knowledge, it seems to be legitimate and useful to look for

Children in Young and Aging Societies


homologies between both dimensions. We assume that researchers who follow the different patterns or paradigms of childhood research identify more or less consciously with one of the four models. Of course, we do not mean that the community of childhood researchers would agree precisely on a typology like the one I propose in this paper. However, childhood research normally identifies implicitly or even explicitly with one or the other basic forms I have described. Typically the research takes one model of childhood for granted on an axiomatic level. At a more subtle level, every childhood model follows a specific “anthropology of childhood” that has a particular perspective on the structure and culture of childhood. New childhood researchers normally are in favor of some version of the postmodern childhood model, while pedagogicalpsychological research usually votes for a more traditional model of modernized childhood. This diversity of approaches means that the different realities of childhood are taken for granted both as objects of research identification and public portrayal. In this sense, different types of childhood research are co-constructors of quite different childhood worlds. There is an obvious correspondence between a postmodern childhood model (AI) and a new childhood research (DI). According to James, Jenks and Prout (1998), the new paradigm includes several types of research defined by the image of childhood these different researchers produce. The self-constructive child is a scientific image produced by autopoietic constructivism; the minority child is the center of social policy research; the social structural child is favored by sociodemographic research; and the tribal child figures in ethnographical childhood research. In each case, despite the methodological and political differences, we find a basic orientation to the model of children as modernized, active social actors. This conception has promoted the notion of a social policy that is developed both for and with children – a policy orientation in which specialists do not colonize children but enter an active researcher-child relationship with them. The paradigm of socialization research (DII) seems to be strongly connected with the model of advanced modernized childhood (AII) (du Bois-Reymond, Büchner, Krüger, Ecarius & Fuhs, 1994). The socialization paradigm began to have success in political, educational, and academic market during the post-war era. During the early period between the 1940s and 1960s, there was also a modernizing of the traditional model of childhood. The socialization paradigm acknowledges that the process of growing up takes place not only within traditional educational institutions like the family, school, or church. Children are influenced by a wider array of socialization agents, including the consumer industry, the mass media, and peer groups. Social learning takes place at



different institutions and occasions, and socialization research does not assume that the process is fully controlled by educational authorities. Instead the paradigm of socialization is especially sensitive to learning processes and learning arrangements that are hidden behind the stated goals of the adult authorities (the so-called “hidden curriculum”). The result of this perspective is to view children as having an active role in choosing the settings in which much significant learning takes place. Socialization research also takes into account individual differences and the individualizing processes involved in modernized childhood, one major effect of this perspective is to deemphasize the role of educational interventions within formal institutions. On the other hand, the paradigm of socialization research is traditional insofar as it emphasizes learning processes as being essentially one-sided – the model strongly implies that the flow of socialization is from adult role models to child. Interactive (reciprocal) and recursive (from child to adult) learning processes were only introduced to the paradigm around the end of the 1960s (Bell, 1968; Bell & Harper, 1977; Rheingold, 1969; Stafford & Bayer, 1993). No doubt this change in perspective was influenced by the challenges to adult dominance associated with the student and youth movements during that decade. This inclusion of reciprocal and recursive socialization models makes a difference between socialization versus new childhood research approaches, and between the advanced/modernized versus postmodern childhood models. In the latter case, children are believed to influence adults at least as much as adults impose their influence on children. Another demarcation line concerns the goals or results of growing up. Socialization research is quite sure that it is possible in principle to find and formulate definite goals for successful socialization of children by adults. Socialization research is therefore interested in the ordering of childhood careers and in the acquisition of skills and competencies by children. Socialization research has also been interested in social restrictions and constraints in the children’s lives that need to be overcome in order to optimize the socialization of children. New childhood research, as well as the model of postmodern childhood, both differ from the socialization approach in seeing the process as essentially open-ended. In this view, the coming generation and each individual person within it has the responsibility for finding and formulating their goals of growing up. This approach does not assume that adults are capable of finding these goals, or even that they should try to search for them without the active participation of children in the process. The model of the traditional modernized childhood (DI) is connected with the mainstream of pedagogical-psychological research on childhood development (DIII) that dominated the scientific advice for teachers, parents, and politicians for at least the first half of the twentieth century. The development

Children in Young and Aging Societies


of children is seen within the mainstream of this paradigm as a rather naturally determined endogenic process of the individual. It is a process that takes place in relatively predictable stages, and research can find the laws of their development and describe their outcome to professional and lay educators. It is the task of education to construct and preserve developmental niches for children according to the laws of optimal development. If processes of modernization intervene to disturb the process of development, one has to protect the children from their negative effects (whether the source is urban life, weakened family ties or consequences of poverty and social discrimination). Adults need this type of research, because they are physically, socially, and psychologically remote from the life world of children. Adult reference persons for children have the legal right and power to intervene with protective strategies in the children’s lives. Pedagogical-psychological research therefore sets itself the task of developing the necessary instruments of intervention through sound and empirically grounded scientific knowledge. For practical pedagogical interventionists, the conception of children as actors in their own right is not the main focus. Instead, the underlying assumption of traditional modernized childhood and the homologous research tradition is that children are tractable objects who are the beneficiaries of well-meaning interventions of powerful adults in a patriarchic and social welfare world. As far as I can see, there is no adequate, scientific self-representation of fundamentalist childhood (AIV). Research is done more or less from the perspective of the outsider who is interested in the social integration (as well as control) of sectarian political and social movements (Palmer & Hardmann, 1999). Some of the fundamentalist childhood movements are explicitly antiscientific in orientation so little research is to be expected from inside this tradition. From the standpoint of a sociology of knowledge, it is unwise to take sides for, say, advanced/modernized or postmodern childhood. I prefer to argue that all four models of childhood and generational order are involved in the historic struggle to define the meaning and course of modernity; all supply more or less convincing strategies for handling modernity within their perspectives (column C). Key elements are empowering children to survive as a minority group within society (postmodern), individualizing children so that they can act autonomously (advanced modern), protecting children (traditional modern) or withdrawing them from society (fundamentalist modern). There exist no convincing arguments that only one of these childhood models will survive within the next decades. There may be historic tendencies, which are in favor of postmodern or advanced modern childhood, but this means only a quantitative shift in the percentage of children (and families) living according



to models I, II, III or IV. It does not mean inevitably that one or the other has hegemony as the exclusive model for contemporary childhood. Because childhood models and scientific representations of childhood are strongly connected, we may risk a prediction for the future. As long as all four childhood models survive historically, we can predict that each of the research paradigms will survive as well. Therefore, it seems to me only a matter of selfpromotion when, for example, British representatives of the new social studies of childhood claim a historic progression from presociological via transitional to sociological approaches in childhood studies (James et al., 1998). By this hierarchical ordering we learn only that the representatives of the new social studies of childhood are strongly in favor of the postmodern model of childhood. From an analytical point of view, it may be of interest which paradigm of childhood research will be dominant in the scientific field and whether this possible dominance parallels the social and cultural history of childhood models. The pedagogical-psychological research (and the vision of a traditional modernized childhood as well as the developmental perspective) dominated the first half of the twentieth century. This situation changed beginning in the 1950s. The dominant pattern now seems to be the socialization research paradigm, at least up until the 1980s. This may indicate that the advanced/modernized childhood model became dominant in the second half of the century. Since the 1980s and 1990s, we find a scientific shift in the direction of new childhood research that may parallel the rise of postmodern childhood models in societies characterized by the growth service information sector employment and by growth in modern media. We may or may not like such speculative constructions of history. However, this perspective directs our attention to the necessity of analyzing the process of modernization that appears, in one form or another, in all four models of childhood and childhood research. Each has a somewhat different history. Developmental models differ considerably over time if we compare the models from the beginning of the twentieth century and those of the 1930s or the 1970s. The same holds true for the traditional modernized model of childhood. In the case of socialization research, we find a development of at least five decades, during which time there was considerable change in topics, methods, methodologies, and basic assumptions about children as actors (Hurrelmann & Ulich, 1991). It is of little scientific use to compare the paradigm of socialization research as it has developed historically during the fifties and sixties with the present day paradigm of new childhood research (Zinnecker, 1997b). A more useful connection is to compare the model of the individualized modern childhood with recent innovations in the field of socialization research. Here, an important concern is how far the theory of

Children in Young and Aging Societies


socialization has gone in developing new models of the child as actor. What has been achieved by shifting the focus from socialization in educational institutions to peer socialization processes or, a more recent trend, selfsocialization? Childhood Moratoria and Child-Adult Relationships Column B in Table 3 combines models of childhood and models of generational order. This approach relies on two central criteria – types of childhood moratoria on the one hand, and patterns of adult-children relationships on the other. These characteristics are useful in distinguishing among the four models. Childhood models II and III (advanced versus traditional modernized childhood) both rely on the principle of generational segregation as an ordering principle. In this sense they are consistent with the European history of childhood and civilization since the sixteenth century. These models are distinguished by the type of moratorium they constitute for childhood. Traditional modernized childhood implies a moratorium that provides safe developmental niche for the younger generation. This moratorium is constituted by changing pedagogical expertise that became increasingly aware of psychological principles (replacing a moral/theological framework). Advanced modernized childhood relies on a more open form of generational segregation in which the role of educational experts is partially modulated by recognition of the self-management competence of children and youth. The model of postmodern childhood strives to overcome this historic legacy of generational separation. It tends to view children as a minority group. Like other minority groups, they have the challenge of achieving social survival and recognition as a group with its own valid political, social, and psychological needs. Lastly, fundamentalist ideas center on archaic tribal niches in which children as well as adults are obliged to retreat from the hegemony of mainstream society by constituting separate cultures and territories.

THE CONCEPT OF A SOCIAL AND CULTURAL MORATORIUM IN THE EUROPEAN-WESTERN MODEL OF CHILDHOOD Thus far in the discussion I have used the concept of a childhood moratorium without describing in much detail what is implied in the concept. It might be



useful, therefore, to explain more about the meaning and status of the term. In youth research, as in life course research in general, the term is largely influenced by Erikson’s notion of the psychosocial moratorium (Erikson, 1959/1966). In this context, it means that there are phases during the life course (especially during adolescence), when society grants individuals a break from practical responsibilities in order to find themselves and to develop and test their personal identity. Particularly for adolescents, this societal offer (if they choose to accept it) opens the opportunity to live through a productive phase of identity crisis, after which one presumably emerges with a developed and elaborated personal and social identity. As used by Erikson and his followers, the concept was strongly connected to clinical psychological work. When used as a sociological concept, Erikson’s notion of a moratorium provides the beginnings of a social and historical theory of childhood and youth. To better apply to the problem of life course in society, Erikson’s concept needs to be expanded and generalized. We, therefore, propose another somewhat differently situated version of the moratorium concept. In doing so, we enter an under-explored terrain of social science. That is regrettable since the notion of moratorium may provide part of the key to the way society regulates participation and nonparticipation of individual society members. The theoretical neglect of this question may have to do with the fact that social analysis from the beginning has been preoccupied with the productive aspects of social life (not the nonproductive phases). Even more important may be that social analysis in the past tended to neglect social time, which is important in any social action and particularly in any regulation of social moratoria (Giddens, 1984/1998, Ch. 3). The moratorium – as retreat from self-determined action and retirement from social interaction – is a matter of timing. It is connected to the temporal dimension of social life and social order. Consider the analogy to daily cycles of exhaustion, relaxation, sleep, and other withdrawals from everyday life. On a longer time scale we have periods of vacation times, of illness, and phases of social exclusion at the end and at the beginning of life. We consider it meaningful to develop an integrated theoretical concept that includes these and other conditions of social denial, exemption, withdrawal, and exclusion from social life at various stages of the life course. One key concept for this purpose is an expanded version of the moratorium (or perhaps different subtypes of moratoria). Useful here is Parsons’ notion of the moratorium of illness, a type of socially legitimated break that allows the relaxation of the usual social norm obligations for a period of time until health, and social norms, are reestablished (Parsons, 1951, pp. 297–320).

Children in Young and Aging Societies


Social Moratoria for Children and Adolescents One type of moratorium is reserved by society for children and adolescents and is integrated into the social system of age groups. This system contains agesegmented regulations and prescriptive norms that generally regulate the times of legitimate retreat for the standard life course. In modern societies such moratoria are narrowly connected to the distinction between age classes that are economically productive and independent versus those that are economically unproductive and dependent. There seems to be a fundamental ambivalence about availability of these social exemptions to children, adolescents, and the elderly – an ambivalence that is reflected in the social and political value put on this phenomenon. On the one hand, we find apologists of the exemption who emphasize the personal opportunities and social privileges that are interconnected with age-graded moratoria. Critics of age-related social moratoria accentuate the compulsory character of this exemption and are often arguing that it reflects an ascribed age status and, therefore, psychological, social, and economic discriminations. In the controversy about modern childhood, we find defenders of the European-Western model of childhood (which includes the moratorium). Among them are the protagonists of traditional modernized and advanced modernized childhood models (see A II and A III). Both models emphasize the advantages of such a social exemption, which they see as a positive cultural achievement produced by modern developed societies. Even the culture-critical complaints over the end of pedagogic childhood moratoria share this general positive assumption about the European-Western model of childhood. On the other side, we find Marxist-inspired theoreticians of childhood side by side with representatives of the Children’s Rights Movement. They are in favor of a postmodern childhood model (A I) where (they argue) the strict connection between age grading and moratoria should be lifted. They typically argue that children should ultimately be reintegrated into the productive/ independent age classes. The fundamentalist childhood model represents a special case (A IV). Its supporters claim a kind of spiritual moratorium for both children and parents. As members of a Christian family or alternative community, both sides wish the right to withdraw from certain social roles and obligations – for example, from the consumption of secular and profane culture. In this example, we can speak of an additional spiritual moratorium that also applies to adults. From these examples, it is obvious that the way in which generations are ordered in the social hierarchy makes a crucial difference to whether children and grown-ups are conceptualized as opposite age classes, (i.e. unproductive



and dependent children on the one hand, and productive and independent adults on the other) or whether the age groups are seen as complementary. In either case, childhood and youth research since the 1960s has generally agreed that these moratoria are to be seen as historic inventions of European-Western societies. Generally speaking, what characteristics mark the childhood moratorium? We can give a negative and a positive definition. Children as an age group are excluded from two central areas of activity that contribute to the social reproduction of society over time and across generations. In the first area, the childhood moratorium is defined as the time of life during which people are freed from working activities. That primarily refers to the economic market of paid labor. In the other area, children should be excluded from any activity of biological reproduction. That primarily refers to the social institution of parenthood and reflects the general taboo against a sexualized childhood. Childhood can also be characterized positively according to the institutions created for children that serve the personality development and learning activities of the young. These two forms of activities give rise to childhood moratoria and are represented by the family and the school (including nursery schools). Childhood becomes domesticated by pedagogical institutions. The status of childhood removes children from all kinds of productive actions usually done in the world of grown-ups. Their mode of acting is called playing, which means free imagination and purity of action. This stands strictly in contrast to the purposive rationality that characterizes the economic action of adult society members. The interpretation of this life phase takes its meaning from the value placed on activities in the present, but also from the value of those activities for which the child is being prepared. The historical discourses on these attributions of value, however, have not produced a consensus. In one type of interpretation, the activities of the children are seen primarily as preparation for a future life. Personality formation, cognitive learning and play are meaningful because of what they prepare children to do as adults. They serve the acquisition of abilities and qualities that are useful for the productive stages of life. This lends a teleological character to children’s activities and appears most clearly in didactic models of teaching and learning in school. It is also present in advice to parents, especially if parental figures are motivated by the idea of upward mobility via education. The teleological model of children’s activities is also a central focus within mainstream socialization research. Within the traditional European-Western discourse on childhood, we find also an idea rather contrary to the teleological one. According to this

Children in Young and Aging Societies


perspective, the activities of children are meaningful and valuable by themselves; they do not need further justification by reference to the economic or social purposes of the adult world. The childhood moratorium functions as a type of counterworld to the adult world of the production. During the years of childhood, creative human potential can be tapped that is typically neglected later when the person is dedicated to the more serious duties of the vocational and familial life. Children have the leisure to ask philosophical questions of the world. Children can afford playful contact with the world. Children can devote themselves to the fine arts, music, theater, and literature. Among adults, all these activities are typically of the lifestyle of a cultural leisure class, though in this case they are promoted in the first years of the life course. Adult society customarily is used to delegate creative activities to these early years and to project feelings of happiness onto that golden age of life. In its most elaborated form, this constitutes the romantic idea of childhood that developed around the eighteenth century. Both ideas about childhood have in common that they understand childhood as a lifetime devoted to cultural activities. It is a cultural moratorium that may have intrinsic value in the present, or may be justified through its instrumental/ teleological value in preparing the child for the future. In either case, this historic model of childhood is the time in life where learning of culture and self-directed creativity take place. With regard to this interpretation of childhood, a general consensus about its major epochs has emerged among historians. Three phases are conventionally distinguished by them: The first is the epoch of the invention of the EuropeanWestern childhood model (situated roughly between the sixteenth and eighteenth centuries, prior to the industrial phase of the nineteenth century). The century between 1850 and 1950 is seen as high point in the spread and domination of this model of modern childhood. It is the historic phase that coincides with the high points both of classic industrial society and the contemporary urbanization. The crisis of modern childhood, which some of the authors proclaim to be the end of this model, is located within the second half of the twentieth century and hence parallels the emergence of the postindustrial service society (Honig, 1999). As I suggested earlier in the discussion, I propose that one of four models of childhood dominates in each of these historical phases. The model of the traditional childhood moratorium dominates in the founding phase of childhood. During the high phase, the advanced model of modernized childhood takes center stage in the discourse about childhood. In the years of crisis, the debate about postmodern childhood stands in the foreground of the media discourse.



The End of the Adult Monopoly on Educating Youth A change in the nature of the childhood moratorium brings with it a change in the educational relationship between younger and older generations. With the allocation of the younger to an unproductive moratorium, the older persons take the position of teachers and educators of the children. They attain a monopoly in principle, but one that is not fully employed by all adults – it is delegated in large part to those who are authorized and professionally trained as parents, teachers, or social workers. For pedagogy as practice and science, it becomes constitutive of their roles that they have children as the main “medium” of their professional action (to borrow a term from the social system theory (Luhmann, 1991)). This monopoly of adults over educational activities declines with the crisis of the classic childhood moratorium. Education increasingly is directed to all age classes and children are increasing seen as simply one group of learners among others. Children in turn are increasingly conceded the ability to influence the older age classes (by day-to-day interaction), and even to take on the role as teachers from whom adults can learn (Stafford & Bayer, 1993). The Childhood Moratorium and the Changing Nature of Labor It is part of the European-Western model of childhood to idealize it as a life phase in which one doesn’t work. Obviously, even in the high phase of this childhood model, between 1850 and 1950, society did not succeed in keeping all groups of children away from work. This is well documented in the social history of childhood (de Coninck-Smith, Sandin & Schrumpf, 1997; Nasaw, 1985). Rural childhood, especially in the farming sector, has been connected more frequently and more intensively with forms of paid labor and unpaid family work than urban childhood. Children of working-class families contributed in these years to the support of their families, while girls traditionally did more housework. However, in spite of these well-documented differences by social class and by region, the dominant idea of a work-free childhood moratorium remained valid during this epoch. Social reformers and professional educators tried to adapt the lives of underprivileged children to that norm. Their historical success coincided with the processes of urbanization and suburbanization as the proportion of farming and working-class families declined throughout the twentieth century. After 1950, the release of children from paid employment and from family work reached a peak. The then popular model of nonworking housewives and mothers contributed to that development, as did the mechanization of households by means of labor-saving

Children in Young and Aging Societies


machines. With the process of domestic mechanization, the help of children in housekeeping activities became unnecessary. The working activities during childhood moratorium are disguised as pedagogical ideologies. In the case of children’s work at home, this means that the work of children is redefined as part of a useful educational custom, by which children may learn to cooperate, to feel responsible, or to show good manners. Another type of pedagogic masking involves school performance. Learning is not understood as a type of work, whereby children contribute productively to the future social and economic development of the society. Only the adult work of teachers is emphasized as a productive contribution to the development of human capital. The corresponding learning activities of pupils are thus defined, not as work but as a form of intellectual consumption. Even the so-called working-education movement, which aimed at freeing children from the role of passive consumers of education, conceptualized the working activities of pupils as part of educational training. Instead of being paid for learning as an intellectual type of working activity, parents and children have to pay for the privilege of consuming the educational “product” (for a recent critique see Oldman, 1994a, b). With the crisis of the childhood model, the work-free character of the classic European-American childhood moratorium is also put into question. That crisis generates a variety of different forms of discourse: (1) One discourse accentuates the active, productive side of learning during childhood. That parallels the economic debate about the character of consumption. Theories of the service society tend to emphasize the agency, or active participation of the consumers, in the act of consumption. Analogously, learning activities are seen as two-sided – children are consumers in their role as learners, but at the same time they rearrange the subject matter actively and produce something new. Because of these subjective activities of the learners, some constructivist theories deny the assumption underlying professional teacher training that teaching could be planned and controlled systematically. According to them, every socialization is, in the end, self-socialization. Children invent – or at least co-construct – their own learning process and events. By these redefinitions the social image of modern childhood changes. It takes its inspiration from the image of an unproductive leisure class, one that is related to age. Some followers of the new sociology of childhood are in favor of paying children for learning in schools. This debate follows a path similar to that laid out in the debate about paying housewives for their private work at home.


JUERGEN ZINNECKER The fact that children’s work in schools obtains exchange value only when they come of age is no argument against the instrumentality of their actual use of time and energy whilst being in school. It is only to say that their relationship to the market has changed from being a synchronous one to being a diachronous one: only when they become adult, they are paid back for the efforts done in school. The close relationship between children’s work in school and the functioning of modern economy cannot be denied. What remains, therefore, is to show that children’s school activities are not acknowledged by adult society while they are children, and by the same token the value of these activities is blurred for the children themselves (Qvortrup, 1991, p. 27).

(2) The historic shift from societies based on industrial labor to ones based on occupation in the service sector changes the social and economic functions of age-related social moratoria. Private consumer activities play an important role in economies that rely on high levels of mass consumption by all age groups. Consuming activities of children and adolescents play an increasing important role in economy. As they become more and more recognized as independent participants in buying and consuming commodities, they become an important target population for marketing campaigns. This implies a reintegration of the childhood moratorium in economic processes and a further weakening of the semiautonomous management of childhood by educational institutions and authorities. (3) Contemporary childhood researchers have come to focus on a double burden that children and adolescents carry in modern society. They spend far more time than previously in institutions of secondary and tertiary education (sometimes into their thirties). Simultaneously, however, they also pursue various types of paid labor such as part-time and holiday jobs. Comparative research demonstrates that these working burdens increased over the past decades (for the U.S., see Modell, 1989; Schneider & Stevenson, 1999, or Hofferth, 2001; for Germany, Zinnecker & Silbereisen, 1996). Modell (1989, p. 38), for example, shows that the percentage of 16- to 18-years olds who participated in schooling as well in paid labor, increased substantially in the U.S. between 1930 and 1970. While in 1930 only 5–10% participated in both activities, the percentage had increased by 1950 to 10–19% and reached 20–29% by 1970. The percentage of young men combining school and work has also been increasing. This development continued until the end of the twentieth century as we can see from the data presented by Schneider and Stevenson (1999). Therefore, it makes sense to put this dual burden of learning and working activities of young people on the contemporary agenda, just like the women’s movement focused attention on the double burden of adult women in the sixties and seventies.

Children in Young and Aging Societies


To discuss this issue effectively will require that we overcome traditional notions about the image of social moratoria for children and adolescents. (4) The European-Western model of childhood has been challenged by the worldwide debate over child labor in poor and less developed societies. In the past, international institutions (like the International Labor Organization in Geneva) attempted to forbid child labor in poor countries by national and international law. This policy has been criticized as Eurocentric, because it follows the historic line of the Western childhood model as a labor-free moratorium. Children’s organizations in poor countries are in favor of a more flexible policy (Liebel, 1994), arguing that children should have a right to work and to earn money for their families and for themselves. In this view, only oppressing and harmful types of labor like child prostitution or slave labor should be forbidden by law. The emerging shift in global policing of child labor obviously has reciprocal effects on the idea of Western childhood moratorium, which is nowadays seen as more culture-bound than in the past. Taken together, these developments undermine the traditional binary code of childhood versus adulthood. It no longer makes sense to confront childhood as a learning age with adulthood as a teaching age, as children’s play versus adults’ work. This means that the applicability of European-Western model of childhood must be reevaluated, although this does not necessarily mean that this type of moratorium will come to an end. The effect of these changes differs for the four types of childhood that have developed historically within this general childhood model. I predict that the percentages of children living according to these four models will change, as will the ideological popularity of each model. Postmodern childhood (moratorium) will become more important and the traditional-modern childhood (educational moratorium) will become less important. On the other hand, fundamentalist and various forms of open-ended moratoria may well become more important in the future. Therefore, it makes little sense to speak generally of a historical end of childhood or adolescence moratoria.

REFERENCES Alanen, L. (1994). Gender and Generation: Feminism and the ‘Child Question’. In: J. Qvortrup, M. Bardy, G. Sgritta & H. Wintersberger (Eds), Childhood Matters (pp. 27–42). Aldershot, England: Avebury. Baader, M. S. (1996). Unterlegene Erwachsene, überlegene Kinder: Der romantische Blick auf das Kind und die Kindheit. In: E. Liebau & C. Wulf (Eds), Generation: Versuche über eine pädagogisch-anthropologische Grundbedingung (pp. 190–200). Weinheim, Germany: Deutscher Studienverlag.



Bardy, M. (1994). The Manuscript of the 100-Years Project –Towards a Revision. In: J. Qvortrup, M. Bardy, G. Sgritta & H. Wintersberger (Eds), Childhood Matters (pp. 299–317). Aldershot, Germany: Avebury. Bartkowski, J. P., & Ellison, C. G. (1995). Divergent Models of Childrearing in Popular Manuals: Conservative Protestants vs. the Mainstream Experts. Sociology of Religion, 56, 21–34. Behnken, I., & Zinnecker, J. (1989). Soziale Entwöhnung der Straßenkinder oder Härtetest für junge Stadtbewohner [Social Weaning of Street Children or Testing the Toughness of Young Urbanites]. In: C. Büttner & A. Ende (Eds), Lebensräume für Kinder [Lifespace for Children], Bd. 6, Jahrbuch der Kindheit (pp. 37–66). Weinheim and Basel, Germany: Beltz. Behnken, I., & Zinnecker, J. (1992). Straßenkinder und ihre Wächter: Eine Fallstudie zur städtischen Kindheit um 1900 [Street Children and Their Guards: A Case Study on Urban Childhood in 1900]. Die alte Stadt [The Old City], 19, 117–136. Bell, R. Q. (1968). A reinterpretation of the direction of effects in studies of socialization. Psychological Review, 75, 81–95. Bell, R. Q., & Harper, L. V. (1977). Child Effects on Adults. Hillsdale, NJ: Lawrence Erlbaum Associates. Best, J. (1990). Threatened children. Rhetoric and concern about child-victims. Chicago, IL: University of Chicago Press. Best, J. (Ed.) (1994). Troubling Children: Studies of Children and Social Problems. New York: Aldine de Gruyter. Bien, W., & Marbach, J. (1991). Haushalt – Verwandtschaft – Beziehungen. Familienleben als Netzwerk [Households – Relatives – Relationships: Family Life as Social Network]. In: H. Betram (Ed.), Die Familie in Westdeutschland: Stabilität und Wandel familialer Lebensformen [The Family in West Germany. Stability and Change in Family Lifestyles]. DJI-Familien-Survey (Vol. 1, pp. 3–44). Opladen, Germany: Leske + Budrich. Blau, P. M. (1977). Inequality and Heterogeneity. A Primitive Theory of Social Structure. New York: The Free Press. Buechner, P. (1996). Das Kind als Schülerin oder Schüler: Über die gesellschaftliche Wahrnehmung der Kindheit als Schulkindheit und damit verbundene Forschungsprobleme. In: H. Zeiher, P. Buechner & J. Zinnecker (Eds), Kinder als Außenseiter? Umbrüche in der gesellschaftlichen Wahrnehmung von Kindern und Kindheit [Children as Outsiders? Social Perception of Childhood in a Time of Upheaval] (pp. 157–187). Weinheim, Germany: Juventa. Buechner, P., & Fuhs, B. (1993). Kindersport [Sport for Children]. In: M. Markefka & B. Nauck (Eds), Handbuch der Kindheitsforschung (pp. 491–500). Neuwied, Germany: Luchterhand. Buechner, P., & Fuhs, B. (1994). Kinderkulturelle Praxis: Kindliche Handlungskontexte und Aktivitätsprofile im außerschulischen Lebensalltag. In: M. du Bois-Reymond, P. Buechner, H.-H. Krueger, J. Ecarius & B. Fuhs (Eds), Kinderleben: Modernisierung von Kindheit im interkulturellen Vergleich (pp. 63–136). Opladen, Germany: Leske + Budrich. Cochran, M., Larner, M., Riley, D., Gunnarsson, L., & Henderson, Jr., C. R. (1990). Extending Families: The Social Networks of Parents and Their Children. Cambridge, England: Cambridge University Press. Coleman, J. S. (1987). Families and Schools. Educational Researcher, 16, 32–38. Coleman, S. (1999). God’s Children: Physical and Spiritual Growth Among Evangelical Christians. In: S. Palmer & C. Hardman (Eds), Children in New Religions (pp. 71–87). New Brunswick, NJ, and London: Rutgers University Press.

Children in Young and Aging Societies


Cross, G. (1997). Kids’ Stuff: Toys and the Changing World of American Childhood. Cambridge, MA: Harvard University Press. Cyprian, G., & Wurzbacher, G. (1979). Strukturbedingungen frühkindlicher Sozialisation in Wohnkollektiven. In: F. Neidhardt (Ed.), Frühkindliche Sozialisation (2nd ed., pp. 321– 373). Stuttgart, Germany: Enke. de Coninck-Smith, N., Sandin, B., & Schrumpf, E. (Eds) (1997). Industrious Children: Work and Childhood in the Nordic Countries 1850–1990. Odense, Denmark: Odense University Press. Del Veccio, G. (1997). Creating Ever-Cool: A Marketer’s Guide to a Kid’s Heart. Gretna, Louisiana: Pelican Publishing Company. du Bois-Reymond, M., Büchner, P., Krüger, H.-H., Ecarius, J., & Fuhs, B. (1994). Kinderleben. Modernisierung von Kindheit im interkulturellen Vergleich [Children’s Lives: Modernizing Childhood in Crosscultural Perspective]. Opladen, Germany [country?]: Leske + Budrich. Eiduson, B. T., Kornfein, M., Zimmerman, I. L., & Weisner, T. S. (1988). Comparative Socialization Practices in Traditional and Alternative Families. In: G. Handel (Ed.), Childhood Socialization (pp. 73–101). New York: Aldine de Gruyter. Elkind, D. (1981). The Hurried Child: Growing Up Too Fast Too Soon. Reading, MA: AddisonWesley. Erikson, E. H. (1959/1966). Identität und Lebenszyklus [Identity and the Life Cycle]. Frankfurt, Germany: Suhrkamp. [Originally published in 1959.] Gillis, J. R. (1993). Vanishing Youth: The Uncertain Place of the Young in a Global Age. Young, 1(1), 1–15. Giddens, A. (1984/1988). Die Konstitution der Gesellschaft. Grundzüge einer Theorie der Strukturierung [The Constitution of Society: Outline of the Theory of Structuration]. Frankfurt, Germany, and New York: Campus. [Originally published in 1984.] Harris, J. R. (1999). The Nurture Assumption: Why Children Turn Out the Way They Do. New York: The Free Press. Heitmeyer, W., Müller, J., & Schröder, H. (1997). Verlockender Fundamentalismus: Türkische Jugendliche in Deutschland [Tempting Fundamentalism: Turkish Youth in Germany]. Frankfurt, Germany: Suhrkamp. Hengst, H. (1996). Kinder an die Macht! Der Rückzug des Marktes aus dem Kindheitsprojekt der Moderne [All the Power to the Kids! The Retreatment of the market from the Pedagogical Project of Enlightment]. In: H. Zeiher, P. Büchner & J. Zinnecker (Eds), Kinder als Außenseiter? Umbrüche in der gesellschaftlichen Wahrnehmung von Kindern und Kindheit [Children as Outsider?] (pp. 117–133). Weinheim and München, Germany: Juventa. Hengst, H., Koehler, M., Riedmueller, B., & Wambach, M. M. (1981). Kindheit als Fiktion [Childhood as Fiction]. Frankfurt, Germany: Suhrkamp. Hernandez, D. J. (1993). America’s Children: Resources from Family, Government, and the Economy. New York: Russell Sage Foundation. Higonnet, A. (1998). Pictures of Innocence: The History and Crisis of Ideal Childhood. New York: Thames and Hudson. Hofferth, S. L., & Sandberg, J. F. (2001). How American Children Spend Their Time. Journal of Marriage and the Family, 63(2), May. Hofferth, S. L. (1999). Changes in American Children’s Time, 1981–1997. Childnews (Newsletter of the Sociology of Children Section of ASA), (Spring), 1–4. Honig, M. S. (1999). Entwurf einer Theorie der Kindheit [Draft for a theory of childhood]. Frankfurt, Germany: Suhrkamp.



Hurrelmann, K., & Ulich, D. (Eds) (1991). Neues Handbuch der Sozialisationsforschung (4th ed.). Weinheim and Basel: Beltz. James, A., Jenks, C., & Prout, A. (1998). Theorizing Childhood. Cambridge, England: Polity Press. Lang, F. R. (1998). The young and the old in the city: Developing intergenerational relationships in urban environments. In: D. Görlitz, H. J. Harloff, G. Mey & J. Valsiner (Eds), Children, Cities, and Psychological Theories (pp. 598–628). Berlin and New York: Walter de Gruyter. Lauterbach, W. (1995). Die gemeinsame Lebenszeit von Familiengeneratione [The Common Lifetime of Generations in Families]. Zeitschrift für Soziologie, 24, 20–39. Lenzen, D. (1985). Mythologie der Kindheit: Die Verewigung des Kindlichen in der Erwachsenenkultur [Mythology of Childhood: Perpetual Childishness in Adult Culture]. Reinbek, Germany: Rowohlt. Liebel, M. (1994). Wir sind die Gegenwart: Kinderarbeit und Kinderbewegungen in Lateinamerika [We are the Present: Child Labour and Children’s Social Movements in Latin America]. Frankfurt, Germany: Verlag für Interkulturelle Kommunikation. Loriga, S. (1997). Die Militärerfahrung [The Experience of Military Service]. In: G. Levi & J. C. Schmitt (Eds), Geschichte der Jugend [History of Youth from Enlightenment to the Present], Von der Aufklärung bis zur Gegenwart (Vol. 2, pp. 56–96). Frankfurt: S. Fischer. Luhmann, N. (1991). Das Kind als Medium der Erziehung [The Child as Medium in Education]. Zeitschrift für Pädagogik, 37, 19–40. Modell, J. (1989). Into One’s Own: From Youth to Adulthood in the U.S. 1920–1975. Berkeley: University of California Press. Myers, S. M. (1996). An Interactive Model of Religiosity Inheritance: The Importance of Family Context. American Sociological Review, 61, 858–866. Nasaw, D. (1985). Children of the City at Work and at Play. Garden City, NY: Anchor Press. Nauck, B. (1989). Individualistische Erklärungsansätze in der Familienforschung: die rationalchoice-Basis von Familienökonomie, Ressourcen- und Austauschtheorien [Individualistic Explanations in Family Research: Rational-Choice-Approaches in Family Economy, Theories of Resourciveness and Exchange]. In: R. Nave-Herz & M. Markefka (Eds), Handbuch der Familien- und Jugendforschung (Vol. 1, pp. 45–62). Neuwied, Germany: Luchterhand. Oakley, A. (1993). Women and Children First and Last: Parallels and Differences between Children’s and Women’s Studies. In: J. Qvortrup (Ed.), Childhood as a Social Phenomenon: Lessons from an International Project (Eurosocial Report 47, pp 51–69). Vienna: European Centre. Oldman, D. (1994a). Childhood as a Mode of Production. In: B. Mayall (Ed.), Children’s Childhoods: Observed and Experienced. London: Falmer. Oldman, D. (1994b). Adult-Child Relations as Class Relations. In: J. Qvortrup, M. Bardy, G. Sgritta & H. Wintersberger (Eds), Childhood Matters: Social Theory, Practice and Politics (pp. 43–58). Aldershot, England: Avebury. Palentin, C., & Hurrelmann, K. (1998). Jugend und Politik [Youth and Politics]. Neuwied, Germany: Luchterhand. Palmer, S. J., & Hardman, C. E. (Eds) (1999). Children in New Religions. New Brunswick, NJ and London: Rutgers University Press. Parsons, T. (1951). The Social System. New York: Macmillan Company. Postman, N. (1982). The Disappearance of Childhood. New York: Delacorte Press.

Children in Young and Aging Societies


Preston, S. H. (1984). Children and the Elderly: Divergent Paths for America’s Dependents. Demography, 21(4), 435–457. Qvortrup, J. (1991). Childhood as a Social Phenomenon – An Introduction to a Series of National Reports (Eurosocial Report 36). Vienna: European Centre. Qvortrup, J. (Ed.) (1993). Childhood as a Social Phenomenon: Lessons from an International Project (Eurosocial Report 47). Vienna: European Centre. Qvortrup, J., Bardy, M., Sgritta, G, & Wintersberger, H. (Eds.) (1994). Childhood Matters. Social Theory, Practice and Politics. Aldershot, England: Avebury. Rheingold, H. L. (1969). The Social and Socializing Infant. In: D. A. Goslin (Ed.), Handbook of Socialization Theory and Research (pp. 779–790). Chicago, IL: Rand McNally. Ryder, N. B. (1965). The Cohort as a Concept in the Study of Social Change. American Sociological Review, 30, 843–861. Schneider, B. and Stevenson, D. (1999). The Ambitious Generation. America’s Teenagers: Motivated but Directionless. New Haven and London: Yale University Press. Schorch, G. (1982). Kind und Zeit: Entwicklung und schulische Förderung des Zeitbewußtseins [Children and Time: Development of Time Consciousness and its Support by Schools]. Bad Heilbrunn/Obb, Germany: Klinkhardt. Spindler, L. S., Enders-Huntington, G., & Hostetler, J. (1991). Amish Children. Education in the Family, School, and Community (Case Studies in Cultural Anthropology). New York: Harcourt Trade Publishers. Stafford, L., & Bayer, C. L. (1993). Interaction between parents and children. Newbury Park, CA: Sage Publications. Steinberg, S. R., & Kincheloe, J. L. (Eds) (1997). Kinderculture: The Corporate Construction of Childhood. Boulder, CO: Westview Press. United Nations, Population Division (1999). Aging. [] – (May 22, 1999). Winn, M. (1981). Children Without Childhood. New York: Pantheon Books. Wintersberger, H. (1993). Preface. In: J. Qvortrup (Ed.), Childhood as a Social Phenomenon: Lessons from an International Project (Eurosocial Report 47, pp. 7–10). Vienna: European Centre. Zeiher, H., Büchner, P., & Zinnecker, J. (Eds) (1996). Kinder als Außenseiter? Umbrüche in der gesellschaftlichen Wahrnehmung von Kindern und Kindheit [Children as Outsiders? Social Perception of Childhood in a Time of Upheaval]. Weinheim, Germany: Juventa. Zeiher, H., & Zeiher, H. (1994). Orte und Zeiten der Kinder: Soziales Leben im Alltag von Großstadtkindern [Places and Times for Children: Everyday Social Life of City Kids]. Weinheim and München, Germany: Juventa. Zelitzer, V. A. (1985). Pricing the Priceless Child: The Changing Social Value of Children. New York: Basic Books. Zinnecker, J. (1990). Vom Straßenkind zum verhäuslichten Kind: Kindheitsgeschichte im Prozeß der Zivilisation [From Streetchild to the Domesticated Child: History of Childhood in the Process of Civilization]. In: I. Behnken (Ed.), Stadtgesellschaft im Prozeß der Zivilisation [Urban Society in the Process of Civilization] (pp. 142–162). Opladen, Germany: Leske + Budrich. Zinnecker, J. (1995). The Cultural Modernization of Childhood. In: L. Chisholm, P. Büchner, H.H. Krüger & M. du Bois-Reymond (Eds), Growing Up in Europe: Contemporary Horizons in Childhood and Youth Studies (pp. 85–94). Berlin and New York: Walter de Gruyter. Zinnecker, J. (1996). Kindheit in der Postmoderne: Fragen, Modelle, Lösungen [Childhood in Postmodernity: Questions, Models, Solutions]. In: P. Dillig & H. Schilling (Eds),



Erziehungsberatung in der Postmoderne [Educational Advice in Postmodernity] (pp. 35– 58). Mainz, Germany: Matthias-Grünewald-Verlag. Zinnecker, J. (1997a). Metamorphosen im Zeitraffer: Jungsein in der zweiten Hälfte des 20. Jahrhunderts [Metamorphosis in Quick Motion: Being Young in the Second Half of the 20th Century]. In: G. Levi & J. C. Schmitt (Eds), Geschichte der Jugend (Vol. 2). Von der Aufklärung bis zur Gegenwart (pp. 460–505). Frankfurt, Germany: S. Fischer. Zinnecker, J. (1997b). Children as Agents of Change: How to Conceptualize the Process of (Re)producing Culture and Society Between Generations. Paper presented at the International Conference on Childhood and Children’s Culture at Esbjerg, Denmark (May 30–June 2). Zinnecker, J., & Silbereisen, R. (1996). Kindheit in Deutschland: Aktueller Survey über Kinder und ihre Eltern [Childhood in Germany: A National Survey About Children and Their Parents]. Weinheim, Germany: Juventa. Zinnecker, J., & Hasenberg, R. (1999). Religiöse Eltern und religiöse Kinder: Die Übertragung von Religion auf die nachfolgende Generation in der Familie [Religious Parents and Religious Children: The transmission of religious culture to the next generation within families]. In: R. Silbereisen & J. Zinnecker (Eds), Entwicklung im sozialen Wandel [Development and Social Change] (pp. 445–458). Weinheim, Germany: Psychologie Verlags Union.



Heather Beth Johnson The past 100 years have seen the development, establishment, and, in the past decade, the rapid expansion of the Sociology of Children and Childhood in U.S. sociology. As we enter a new millennium we wonder with excitement where this field is headed. However, before we can begin to anticipate where we are going, we must understand where we have come from. The aim of this paper is to trace the development of the Sociology of Children, as a field of study, by looking back over the past 100 years of research and literature on the subject and asking, “Where have we come from?”

STRONG BEGINNINGS FOR THE CHICAGO SCHOOL AND THE SOCIOLOGICAL STUDY OF CHILDREN: 1900–1930 In the early 1900s the study of children in sociology was virtually nonexistent. However, beginning in the early 1920s, children begin to clearly emerge in the sociology literature in the U.S. The 1920s were a time of rapid expansion for

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 53–93. 2001 by Elsevier Science Ltd. ISBN: 0-7623-0776-5




American sociology (Coser, 1978). The influence of the University of Chicago’s Department of Sociology, or “The Chicago School” as it is commonly referred to, dominated the sociological scene of the time and was the “center of the discipline” during those years (Ritzer, 1992: 45). The role that the Chicago School played in the development of the field of Sociology of Children is a large one. Founded in 1892 by Albion Small, the department of sociology at the University of Chicago was the first sociology department to be established in an American university. Focusing on empirical social research in and around the city of Chicago, the Chicago School became the pioneering approach to urban sociological research and theory. Flourishing during the years between World War I and World War II, and peaking in the 1920s, the Chicago school had a major impact on U.S. sociology, which can still be seen today. Several prominent figures led the Chicago School during its most significant years and wrote major works that came to exemplify the classic qualitative urban sociological work for which Chicago has become so well known. Major players included Charles Horton Cooley, Edward Franklin Frazier, George Herbert Mead, Robert E. Park, and W. I. Thomas, among many others. Examples of some of their most influential works include W. I. Thomas and Florian Znaniecki’s The Polish Peasant in Europe and America (1918), and Robert E. Park and Ernest W. Burgess’s The City (1925). These sociologists and their work had a distinctively American flavor; that is, they had a symbolic interactionism/social-psychological orientation. Much of the work looked at macrosociological phenomena and how it impacted social life on the microsociological level, and vice versa. Generally, these sociologists were interested in not just the positive influence (as many of the European functionalist theorists before them), but also the negative influence of culture and social structure on individual and collective behavior. One of the legacies of the Chicago School was its commitment to the understanding of how historical, social, and cultural factors impact individual and group behavior. Another significant aspect of the work of Chicago School sociologists was their explicit interest in pragmatism, social change, and social reform. Many of the school’s members were ministers, or sons of ministers, and were dedicated to scientific study for the purpose of “social amelioration” (Ritzer, 1992: 45). They believed that studying the poor or otherwise disadvantaged would reveal social facts that would then lead to an understanding of how to reform conditions that create poverty and other urban problems. This dedication to social justice and social policy was an integral part of the work of the Chicago School.

The Sociology of Children and Childhood in the United States, 1900–1999


Also significant to the Chicago School was its emphasis on distinctively qualitative methodologies. Methods were geared toward the collection of subjective, ethnographic, descriptive data. Many of the studies utilized a variety of methods including participant observation, interviewing, content analysis, and case studies. The Chicago School’s approach was to “get their hands dirty,” to study real people in their natural habitats, and to view social life in action – literally, not abstractly. During the 1920s, several subdisciplines arose which had not previously been established in the field of sociology. These subdisciplines were firmly rooted in the Chicago School in their theoretical perspective, pragmatic orientation, and methodological approach, one of which was the sociological study of children, childhood, and child development. James W. Trent outlines the growth and decline of early studies of children and child development in his article entitled: “A Decade of Declining Involvement: American Sociology in the Field of Child Development, The 1920s” (1987). In it he states that, “after World War I, several sociologists were writing and lecturing on issues pertinent to child development, and a few were involved in funded research efforts and multidisciplinary associations” (Trent, 1987: 12). Many of these sociologists were major players in the Chicago School, and significant thinkers in American Sociology in general at the time. Trent documents that as the U.S. emerged as a major industrial nation at the turn of the century, problems associated with children came to the forefront. Child labor, mandatory education (and mental deficiency), and juvenile delinquency all became significant issues in the public policy arena. Given the Chicago School’s interest in social reform, these policy issues became central to their work. There were several key players influential in the emergence of child development as a field of study within sociology. We will discuss three of the most noteworthy: William I. Thomas, Frederic Thrasher, and Clifford R. Shaw. W. I. Thomas and his wife, Dorothy S. Thomas, contributed to the direction of the field in the 1920s. W. I. Thomas’ classic book, The Unadjusted Girl, was first published in 1923. Using the case study method, Thomas put forth a descriptive, illustrative analysis of the “problemed” adolescent girl in the context of the conflicts of a changing society (an “unadjusted world”). The Unadjusted Girl is a landmark book because it was the first widely read sociological approach to juvenile delinquency with an explicit focus on girls. The basic argument Thomas makes though is more general: He asserts that the complexity of human lives and character make it impossible to explain social behavior clearly and absolutely, but general trends can be seen in that behavior



is an attempt to adjust to one’s environment based on one’s own “definition of the situation.” The book emphasizes not the “sick individual,” but rather the “sick society,” which is the context in which social behavior occurs. At the time, it was a provocative interpretation of human behavior because Thomas’ argument was an attack on the popular social-Darwinist, functionalist and biological-determinist beliefs of the time. Thomas’ claim that “natural” and cultural factors combine to influence behavior, and that the two are intrinsically inseparable, was quite radical. In The Unadjusted Girl, Thomas asserts that humans are social (and thus always influenced, and to some extent controlled by, social forces) and they have the ability to make decisions based on their own definition of the situation (thus, there is always room for human agency to some extent). Norms and models of behavior are always changing; thus, the girl is always confronted with some operational meanings different from those her grandmother was confronted with. It is this constant flux and decision making that explains social change, while social forces, cultural norms and others, structure society in relatively inflexible ways: “There is therefore always a rivalry between the spontaneous definitions of the situation made by the member of an organized society and the definitions which the society has provided for him” (Thomas, 1923: 42). The individual is a product of culture but is always making adjustive choices that may run counter to dominant norms. The book documents the process through which individuals who have “failed” to adapt to socially dominant standards of behavior turn to delinquency, crime, prostitution, or other forms of “social maladjustment.” Five years later, the Thomas’ amplified their views in The Child in America: Behavior Problems and Programs (1928). The authors stated in clear detail their now-famous claim that, “If men define situations as real, they are real in their consequences” (1928: 572). In this book emphasis is placed on what people think and how this affects what they do. Again, using the case study method, one can see the complexity of individual-social interaction in the various ways people adapt, or fail to adapt, to the requirements of the social order. One’s definition of the situation impacts one’s behavior in that situation, and in return, that behavior impacts the actual situation. This behavioral approach to analyzing child development and child behavior made the Thomas’s approach to studying children distinctively sociological and has had a lasting impact on the field. Another important participant in the Chicago School was Frederic M. Thrasher. His 1927 study, The Gang, is still considered the most comprehensive study of adolescent gangs ever undertaken, having taken over seven years to

The Sociology of Children and Childhood in the United States, 1900–1999


research, and including over 25,000 gang members in Chicago. In The Gang, Thrasher puts forth a sociology of the adolescent male gang as a type of human social group, a specific form of social organization. The study explores the relation between the gang and problems of juvenile delinquency, demoralization, crime, and politics. Thrasher reveals that “gangland” is “interstitial,” that it is not totally isolated to one demographic area, but in fact covers many areas of the city and involves boys from various types of lower-class neighborhoods. Thrasher asserts that boys join gangs (and form gangs) in their “quest for new experience,” excitement, adventure, and thrill-seeking: The gangs dwell among the shadows of the slum. Yet, dreary and repellent as their external environment must seem to the casual observer, their life is to the initiated at once vivid and fascinating. They live in a world distinctly their own – far removed from the humdrum existence of the average citizen (1927: 3).

The gang offers a substitute for what society fails to give to disadvantaged boys and provides relief for them. Thrasher did not believe that gangs caused delinquency and crime, but he did view them as a major influence on their members. His focus on the development of the gang, the social organization of it, and the implications of gang life for boys’ behavior acknowledged that children and adolescents have unique pressures and desires of their own. In this way, The Gang, paved the way for later sociological studies of youth culture and childhood peer culture. Another Chicago School sociologist, Clifford R. Shaw, wrote The JackRoller: A Delinquent Boy’s Own Story, a 1930s classic case study of “Stanley” based on his “own story.” Compared with The Gang, a massive, comprehensive study of thousands of boys in Chicago, The Jack-Roller is the opposite: an indepth look at one boy’s life. Shaw uses one intensive case study as illustrative of a broader group (in this case, delinquent children). Shaw’s focus in The Jack-Roller is on how the situation (the conditions a child finds himself or herself in) impacts the child’s “definition of the situation” (how the child perceives and conceptualizes himself or herself). Through Stanley’s story we see how a change in one’s situation can impact selfperception and thus can bring about a change in the situation. Since the child acts “as if” his or her interpretations and definitions are true, we must take his or her definitions of the situation seriously. For example, Shaw argues that “Even if it were true that Stanley’s interpretation of the family situation were somewhat exaggerated, it cannot be doubted that he acted ‘as if’ these



interpretations were true” (1930: 55). Shaw stresses that the sincerity of a person’s “own story” should not be questioned: The story should be read with a view to getting insight into the boy’s attitudes, typical reactions, and the social and moral world in which he lived. From this standpoint, as previously indicated, rationalizations, prejudices, exaggerations are quite as valuable as objective description (1930: 47).

The implications of Shaw’s work are far-reaching. The Jack-Roller reveals the importance of giving voice to a child’s “own story,” and proves that useful sociological information can be revealed by looking at the child’s point of view, the child’s social world and the history of one’s unique set of life experiences. Furthermore, through Shaw’s work we see the importance of studying children in relation to the social and cultural context they are in. Thomas, Thrasher, Shaw and others of the Chicago School played large roles in the emergence of a subdiscipline in sociology dedicated to studying children. During the 1920s this new field of study was active and influential in mainstream U.S. Sociology. But by the mid 1930s, the Chicago School had begun its demise and there were significantly fewer sociological studies of children being published. An increased focus on quantitative and “hard, scientific methods” of research began to dominate American Sociology. The Ivy League’s (most notably, Harvard’s) emphasis on structural functionalism and its preoccupation with statistics propelled the fall of the Chicago School. Soon the Chicago School’s qualitative methods and dedication to social justice came to be thought of as “soft,” at least in the eyes of positivists who began to dominate mainstream sociology (Ritzer, 1992: 45). The Chicago School quickly lost its position of central importance in American sociology. In addition, the 1930s decline in sociological interest in children was a combination of several factors, including: the failure of sociology to develop a clinical component to the discipline (as did the fields of psychology and social work), and a lack of funding for sociological studies due to the absorption of funds by quantitative studies and clinical areas (Trent, 1987). Talcott Parsons’ publishing of The Structure of Social Action in 1937 signified the culminating point of the loss of the Chicago School’s dominance of American sociology (Trent, 1987). Given the fact that the Chicago School had cultivated a sociological interest in children, the Sociology of Children as a subdiscipline also waned. Although it never completely vanished, in the decades following the 1920s and 30s, work specifically focused on children as the subject of study was seriously neglected. The Chicago School’s emphasis on descriptive, ethnographic studies, often focusing on the child’s perspectives (their “definitions of the situation”) was an

The Sociology of Children and Childhood in the United States, 1900–1999


important contribution from the Chicago School sociologists to the Sociology of Children. But, like all schools of thought or theoretical models, the work that came out of the Chicago School was problematic in many ways. Here we will discuss just a few. One of the flaws of the Chicago School’s work was its general reliance on typologies. Their work was often ideal-typical, seeking to categorize whole groups of people in uniform ways. In doing this, they did not acknowledge the varieties of people within each group. For example, in Thrasher’s The Gang, we see only lower-class boys’ gangs and do not have a sense of the various sorts of other gangs in the city. Another problem was the complex relationship between research and action so often at the heart of so much of the Chicago School sociologists’ work. Their dedication to social justice focused most of their work on how to solve social problems and redirect behavior. Often the work from this period is heavily weighted with suggestions for public policy to address the problems revealed in the research. While this is not necessarily negative itself, it often meant that the researcher was overwhelmingly focused on how to get people to act in ways that they deemed appropriate. The perspective was generally that of a middleclass man trying to figure out how to get the “other” to assimilate more successfully to their way of life. The assumption was usually that the researcher knew best. This problem was especially noticeable in studies of children, because while they saw children as creating problems for society, they also saw them as being society’s greatest hope for a better future. And lastly, the Chicago School’s work with children can be criticized for its often overstated claim that it was based on the authentic child’s voice. Rather, the work was based on court records, interviews with social workers and others working directly with children, analyses of newspaper reports, etc. Even in The Jack-Roller, which is supposedly based on a “boy’s own story,” the data is primarily made up of an adult’s remembrance of his childhood. While this is important data, it is clearly different from data revealed in direct interviews with children or observations of them. Although opponents and critics have launched serious attacks on the Chicago School’s approach and some of the theoretical perspectives that were developed, the lasting impact the Chicago School had on American sociology should not be underestimated. The Chicago School left a legacy in that they did “. . . succeed in elaborating a flexible, theoretical and macrosociologicallyoriented frame of reference for the many empirical studies of phenomena of everyday life in the modern (American) big city” (Joas, 1987: 102) in which children’s lives were frequently considered and explicated.



FOCAL CONCERNS AND JUVENILE DELINQUENCY: THE 1940s AND 1950s The 1940s and 1950s were characterized by a strong emphasis on functionalism in mainstream U.S. sociology. Both methodologically and theoretically, sociology during this time was located on the opposite end of the spectrum from the Chicago School approaches of the 1920s and 1930s. A strong tradition of survey research took hold and pushed qualitative work to the margins. Parsonian theoretical models, rooted in structural functionalism, dominated the sociological scene. At the same time, though, it was during the 1940s and 1950s that qualitative sociology became widely accepted as a legitimate and valid source of information. The 1954 Brown v. Board of Education ruling, for example, was “surrounded with top social scientist participation” (Rapoport, 1985: 17), including sociologists and anthropologists whose work focused on children and child development. A case in point is Mary Ellen Goodman’s seminal work on race awareness in young children which was highly influential in the Brown decision. Goodman’s work revealed that children are exposed to “race-related cultural patterns” which reinforce the “white over brown formula” in U.S. society (Goodman, 1952). Her data, along with others’ (e.g. Clark & Clark, 1939, 1947) was accepted by the court as evidence for desegregating America’s public schools. In her 1952 book, Race Awareness in Young Children, Goodman “introduced to the scientific community ways to explore social concepts in young children and inspired other researchers to investigate the topics of race and ethnicity in young subjects” (Holmes, 1995: 1). She showed that children as young as four years old are doing the following kinds of things: (1) They are perceiving (registering) the objective features of people, things, and behavior, and making classifications on the basis of these perceptions; (2) They are becoming used to a great number of doings and ways of doing, and are increasingly practicing these ways themselves; (3) They are learning to like the things that other people like, and to dislike the things other people dislike (Goodman, 1952: 41). Goodman’s book was the foundation for future research on how children see and feel race differences, the nature of early race awareness and attitudes, and the personal and social contexts in which children’s racial beliefs develop. But, as discussed above, during the 1930s, sociological research on children had waned. By the 1940s and 1950s, very little sociological work was being published about children and childhood. Work such as Goodman’s was not

The Sociology of Children and Childhood in the United States, 1900–1999


widespread. Much of the work that did appear tended to be centered on juvenile delinquency. An important exception is Erik Erikson’s 1950 publication of his first book, Childhood and Society. This book has proved to be a classic in the study of the social significance of childhood. In it, Erikson argued that inseparable links exist between the physical (the body as a biological organism), the mental (the ego, the psychological), and the social (human interactions and relations). The somatic processes, the ego processes, and the societal processes make up the whole of the process of human life (Erikson, 1950). Childhood and Society was groundbreaking in that it laid out in clear terms a theory of socialization which took into account the multiple factors of human social life. Erikson’s perspective was clearly rooted in socialization (not biological determinism or social Darwinism). He saw a clear relationship between childhood “training” and “cultural accomplishment.” From his perspective, children play a central role in society, for, “every adult, whether he is a follower or a leader, a member of a mass or of an elite, was once a child” (Erikson, 1950: 404). Every society consists of men in the process of developing from children into parents. To assure continuity of tradition, society must early prepare for parenthood in its children; and it must take care of the unavoidable remnants of infantility in its adults . . . . This is a large order, especially since a society needs many beings who can follow, a few who can lead, and some who can do both, alternately or in different areas of life (Erikson, 1950: 405).

As we will see later, this sort of socialization perspective has become the heart of a major theoretical debate in the Sociology of Children. But regardless of its flaws, Erikson’s perspective was important in the development of the Sociology of Children in that his theory was an authentically sociological one – one that attacked the psychological and biological theories of its time and argued for a contextualized understanding of social processes. Erikson’s ideas have become vastly influential in our understanding of human development and socialization. Childhood and Society changed the ways society was perceived by giving insight into such important concepts as identity, the life cycle and societal change. Erikson’s insight into the relationship between childhood and society was groundbreaking for sociology. But, as mentioned earlier, in the 1940s and 1950s, what little work on children that existed in sociology tended to be focused on juvenile delinquency (mainly delinquent boys). Sociologists came to conceptualize children as ’problems’ more and more. Specifically, middle-class ‘experts’ began to problematize ‘lower class culture’, basically with the assumption that the lower



class is factually different from the rest of society. Here we begin to see a clear class differentiation in the literature begin to emerge. During the 1940s, Clifford Shaw and Henry McKay began to build on Ernest Burgess’s “Concentric Zone Theory” by charting the spatial distribution of crime and delinquency on maps of Chicago. This work revealed patterns of concentration in juvenile delinquency in certain urban areas such as those adjacent to central business districts and heavy industrial areas (Shaw & McKay, 1942). They argued that rates of delinquency vary inversely with distance from the center of the city. Thus, the inner city was not only characterized by slum housing, poor immigrants and minorities, but by “high delinquency areas” (Shaw & McKay, 1942). They began to theorize about the development of local subcultures and their relationship to delinquency rates. They proposed that delinquency rates reflect the kinds of neighborhoods that kids grow up in, claiming that deteriorated, poverty-stricken areas lead to social disorganization which leads to delinquency. They argued that the local subculture transmits localized values and norms to children that are sometimes contrary to the best interests of the wider society. Walter B. Miller’s 1958 piece, “Lower Class Culture as a Generating Milieu of Gang Delinquency,” built on Shaw and McKay’s theory. Miller argued that lower class culture contains unique values, norms and expectations which lower class kids internalize, just as middle class kids internalize their own specific class values. Miller proposed that lower class kids’ delinquency is a reflection of their own class cultural values based on six focal concerns. These class-specific focal concerns develop due to segregated social, economic, and spatial conditions. Miller’s main propositions were that: (1) the lower class is characterized by distinctive values; (2) these values vary distinctively from middle-class values which undergird the legal code; and (3) the result is that conforming with certain lower-class values may automatically result in the violation of legal codes (1958). He summarizes his argument as follows: The primary thesis of this paper is that the dominant component of the motivation of ‘delinquent’ behavior engaged in by members of lower class corner groups involves a positive effort to achieve states, conditions, or qualities valued within the actor’s most significant cultural milieu (Miller, 1958: 18).

Miller asserted that lower class culture is not just middle class culture “turned upside down,” but is “a distinctive tradition many centuries old with an integrity of its own” (1958: 19).

The Sociology of Children and Childhood in the United States, 1900–1999


In their 1959 edited volume, Delinquent Behavior: Culture and the Individual, Kvaraceus and Miller attempt to offer an “integrated theory of delinquency” based on interdisciplinary studies. Here they clarify the lowerclass culture theory even further by arguing that lower class ways of life run counter to the definitions of conformity that prevail in middle or upper class communities. Their perspective was: . . . the conclusion is set forth that delinquency as norm-violating behavior usually represents adaptive behavior for the individual delinquent; indeed, the delinquent act may often represent the only resolution the youngster can find to his personal-social problem (Kvaraceus & Miller, 1959: 15).

Kvaraceus and Miller’s “cultural transmission theory” (Bynum & Thompson, 1996) was an explanation for delinquent behavior based on participant observation, tape recordings of group activities and discussions, and content analysis of social workers’ contact reports. During a time of increasing focus on quantitative methods, these sociologists held fast to their beliefs in studying children in their own environments and attempting to understand how they perceive their social worlds. As opposed to sociologists such as Whyte (1955), Miller and his colleagues saw slum neighborhoods as being highly socially disorganized. This assumption of social disorganization led them to imply that lower class culture is not just different from middle- or upper-class culture, but that it is actually worse. Furthermore, they conceptualized the lower class was one social group and did not account for multiple race, ethnic and gender variations within classes. For example, the work of the 1940s and 1950s was almost exclusively centered on boys – we see very little acknowledgment of the heterogeneity of class groups in regards to gender. But most critically, cultural transmission theories are flawed because they do not account for those individuals who do actually move upward in status. Their cultural determinism leaves little room for human agency or social mobility. And their focus on culture often obscures some of the systemic structural problems that contribute to such phenomena as juvenile delinquency.

INTERVENTION AND REHABILITATION: THE 1960s AND 1970s As with the 1940s and 1950s, sociological studies of children in the 1960s and 70s continued to be focused on delinquency. Richard Cloward and Lloyd Ohlin published Delinquency and Opportunity: A Theory of Delinquent Gangs in 1960. This work built on the theories of Durkheim (1897), Robert K. Merton (1938), and Albert Cohen (1955) before them. In their work, Cloward and



Ohlin argued against the theories Miller and others presented in the 1950s. Whereas Miller had seen lower class delinquency as a result of boys’ internalization of lower class cultural values, Cloward and Ohlin saw delinquency as a result of the status frustration that develops from lower class boys’ internalization of middle class values. They argued that lower class boys internalize middle- and upper-class success goals but have limited opportunity for attaining them. This differential opportunity creates social strain and anomie, which leads to the development of reactive subcultures: Our hypothesis can be summarized as follows: The disparity between what lower-class youth are led to want and what is actually available to them is the source of a major problem of adjustment. Adolescents who form delinquent subcultures, we suggest, have internalized an emphasis upon conventional goals. Faced with limitations on legitimate avenues of access to these goals, and unable to revise their aspirations downward, they experience intense frustrations; the exploration of nonconformist alternatives may be the result (Cloward & Ohlin, 1960: 86).

Put simply, Cloward and Ohlin’s theory suggests that delinquent subcultures develop as a response to the contradiction between internalized goals and socially structured means of attaining them. Social strain theories (Bynum & Thompson, 1996) such as that of Cloward and Ohlin, were popular in the 1960s and 1970s. These theories saw delinquency as mainly a result of the social structure in which it occurs. They assumed that lower class kids conformed to dominant middle- and upper-class ideas related to aspirations. This assumption has been severely criticized and shown to be empirically false in more recent years (see, for example, MacLeod, 1987). Perhaps more problematic though is the insistence, on the part of such theorists as Cloward and Ohlin, on focusing on only lower-class delinquency. These theories do not explain the delinquency of middle- or upper-class children. Furthermore, based on social strain theories, we have no explanation of why not all boys and girls growing up in lower-class neighborhoods become delinquents. As stated above, mainstream sociological studies of children in the 1960s and 1970s continued to be focused on delinquency. In fact, during these decades, studies of deviant, alternative, countercultural and illegal groups flourished more than ever. But, unlike earlier work, the 1960s and 1970s work became increasingly focused on intervention and rehabilitation. The decades can be characterized by a preoccupation with problems (such as youth drug use, delinquency and child mental health), and intervention. An awareness of child abuse and neglect began to emerge in the literature, along with the sociological characterization of children as victims in need of help. Money and public policy began to be funneled into programs to rehabilitate the individual.

The Sociology of Children and Childhood in the United States, 1900–1999


This focus on the individual implied the assumption that the fault was in the individual, not the social structural forces acting on the individual. Officials began looking for easy policy solutions that would protect and change the kid. By the late 1970s, this therapeutic milieu was increasingly focused on providing custom services for children. At the same time, this preoccupation with social problems and intervention prompted much political action on the macro-level. The War on Poverty was proclaimed, The Great Society Programs were established, and publicly funded programs such as the Peace Corps were developed (Rapoport, 1985). Similarly, several sociologists began looking at the big picture, and taking their sociological analysis of children and childhood to a more macroscopic level. Erik Erikson’s 1961 book, The Challenge of Youth (originally published as Youth: Change and Challenge), Philippe Aries’s Centuries of Childhood (1962), and Edgar Friedenberg’s Coming of Age in America (1963) are examples of such work. All of these books were widely read in the field of sociology and all of them helped foster increasing anxiety about the “defiant” young people of the time. Also coming from a more macro-perspective was pioneering work, such as that of Glen Elder, Jr. (1974), which began to explore the relationships between social change, society, families, and children’s lives in historical context.

SOCIOLOGY OF CHILDREN AND CHILDHOOD: THE 1980s AND 1990s In the aftermath of the 1960s and 1970s “rehab revolution,” the 1980s and 1990s were characterized by an expansion of child abuse literature in sociology. As Spencer Cahill has stated, during the 80s “fear for children eclipsed fear of them” (1992: 662). The early 1980s in the U.S. witnessed a shift in the entire culture toward abused, missing, and otherwise “threatened” children. As Joel Best has documented, “child-saver” rhetoric flourished during this time (1990). Throughout the 1980s and 1990s sociologists provided statistics and extreme examples describing the blameless young victims of the moral corruption of the times. Books such as Neil Postman’s The Disappearance of Childhood (1982), Sylvia Ann Hewlett’s When the Bough Breaks: The Costs of Neglecting Our Children (1991), and Marie Winn’s Children Without Childhood (1983), asserted that permissive parents, an irresponsible media, and a corrupt society were stealing away childhood innocence. Their plea was that we are all guilty of neglecting children’s needs and that we must save the children of our country from the horrors of the society before it is too late.



The 1980s and 1990s were also characterized by a return to more theoretical work on children and childhood, with a renewed emphasis on qualitative research and focus on social-structural problems (reminiscent of the earlier Chicago School work from the 1920s and 1930s). And in the late 1980s we began to see an emergence of a “new” Sociology of Children (discussed later). In this decade, sociological studies of children and childhood began to move from the margins and more toward the mainstream than they had been in years. The “Big Picture”: Analyzing Children and Childhood as a Construct Building on the tradition of Philippe Aries’s classic work, Centuries of Childhood (1962), sociologists in the 1980s and 1990s began seriously analyzing children and childhood as a social construct. Aries had made a solid argument that over time children have been conceptualized as qualitatively different from adults. His book tells the history of how the idea of childhood has changed throughout the centuries. His argument – that broad social processes influence the social construction of the concept “childhood,” and that childhood is the result of particular historical conditions, decisions and actions, and economic, political and cultural struggles – strongly influenced social constructionists analyzing children and childhood. Building on the social constructionist approach, John Allan Lee (1982) argues that there are three theoretical paradigms of the social construction of childhood: (1) children as property (a social definition originating in pre-industrial society); (2) protection of children (a concept legitimated by industrialism); and (3) the child as a person (advocated by recent liberation movements). These three models, or paradigms, are the ways in which society has organized its perceptions of childhood, but in reality, they are purely social constructs. This perspective, like many others of this time, is grounded in Peter Berger and Thomas Luckmann’s famous assertion that “reality is socially constructed” (1966). Similarly, Anthony Synnott explores some of the themes in the development and construction of thinking about children and childhood from the early modern age to the present in his piece, “Little Angels, Little Devils: A Sociology of Children” (1983). He describes several variations of the concept, “childhood,” over time and attempts to explain why they changed based on social, cultural and historical factors. He argues that,

The Sociology of Children and Childhood in the United States, 1900–1999


Surely these constructions of childhood tell us more about the constructors than they do about childhood. Childhood is not a given; it is not a ‘natural’ category. There are no natural categories, only social categories with different meanings imposed and developed by every age, and by different populations within every age (Synnott, 1983: 92).

Synnott documents the ways that children have been regarded: as “sponges, clean slates, little angels, little devils, little monkeys,” etc., to prove that childhood is a socially constructed category (1983: 92). In a similar vein, Joel Best’s Threatened Children: Rhetoric and Concern about Child-Victims, documents the cultural construction of the “social problem” of the “child-victim” (1990). Best looks beyond the 1980s’ “children at risk rhetoric” and problematizes that claim by revealing the socially constructed nature of the notion. He asks why the social construction of “threatened children” took hold in the 1970s and 1980s and answers that it did so because “the campaigns against threats to children benefited from an ongoing wave of social reform, institutional developments conducive to claimsmaking, and the meanings ascribed to childhood in the late twentieth century” (Best, 1990: 20). Work such as these, as well as other historical approaches (Elder et al., 1993) take macro perspectives and look at childhood as a social construct. They show not only how the definition of childhood changes over time, but how each generation of children grows up in a different world of realities. By looking at the big picture, we see how children’s lives and the ever-changing social world are intricately linked. Studies of Structural Disadvantage and Institutional Settings During the late 1980s and the 1990s, we begin to see a noticeable increase in studies of structural disadvantage and institutions related to children’s lives. Much of this work is explicitly focused on the interplay between research and action in programs and institutional settings aimed at benefiting children, youth and families (see, for example, Rapoport, 1985). Many of these books are a clear plea for social change on the part of the researcher. Often they are written based on ethnographies or quasi-ethnographies of children in settings such as group homes (Marek, 1987), schools or training centers (Kozol, 1991; Ryan, 1995), impoverished neighborhoods (Kotlowitz, 1991; Kozol, 1995), or juvenile detention centers (Ayers, 1997). Other times, they are explicitly focused on poor, neglected, or abused children with an emphasis on their structurally disadvantaged conditions (Garbarino, Kostelney & Dubrow, 1991; Garbarino, Dubrow, Kostelney & Pardo, 1992; Gelles, 1996).



These books aim to reveal the hidden truths of the harsh and predetermined conditions many children find themselves in. Many draw connections between biography (life experience), social and cultural conditions, and history. Books such as these attempt to “give voice” to children’s concerns, and in many ways they do (they often include portions of interviews with children, or the authors’ recollection of what children said or did in specific situations). But, the voices heard are mostly those of the author, and often we hear more about what the authors think than about what the children themselves think. Most of the work is focused on the researchers’ interpretations of the situation. How children make sense of their own situations in these circumstances is not clear. Furthermore, most of these books are extremely heavy on the structural aspects of the analysis. They are often descriptive, in-depth analyses of how desperate and horrifying the conditions that many children are living in are. Such books by Alex Kotlowitz (1991) and Jonathan Kozol (1991, 1995), while informative and compelling, are extremely hopeless. As Cahill has commented, work such as that of Kotlowitz, for example, tends to portray kids as “prisoners of a history that had destroyed a community and the spirit of those adults with whose lives their own were hopelessly entangled” (1992: 671). These are dark, pessimistic, and depressing books; in them we see children being structurally imprisoned, boxed into hopelessly predetermined constraints. We see little of children’s creativity, imagination, and hopefulness. While these books often portray children as resilient and surprisingly strong, they give no clear explanation of how this resilience can transform children’s lives. Theoretically this type of work is problematic because it leaves little room for understanding human agency; however, these books do reach a broad audience and provide non-sociologists with a sense of the powerful social forces which negatively impact on children growing up in economically and socially distressed situations in America. A Return to Theory: Socialization Expanded and Agency Uncovered During the 1980s and 1990s we see a return to sociological theory explicitly focused on children and childhood (as opposed to being focused on children indirectly through deviance/delinquency, “the family,” etc.). Much of this work is geared toward expanding the socialization theories first presented by some of the classical sociological theorists. An example of a major work is Gerald Handel’s edited volume, Childhood Socialization (1988). In it, Handel clearly defines socialization as “. . . the process by which the newborn human organism is transformed into a social person, a person capable of interacting with others” (1988: xi). He continues with the following:

The Sociology of Children and Childhood in the United States, 1900–1999


Socialization is a process that is carried out by persons and organizations that may have official designation as agents of socialization or that may be allowed to function as such an agent. They are agents in a double sense of that term: (1) They act upon the child; and (2) they act on behalf of the larger society (or some particular segment of it that may be acceptable or unacceptable to the larger society). The principal agents in our society, and in many others, are family, school, and peer group (1988: xi).

Handel’s work shows that socializing agents have particular positions in society that affect their socializing activity. Thus, he makes room for various socializations, acknowledging gender socialization, class socialization, race socialization, etc., all occurring within various social groups at various times. While Childhood Socialization is a basic socialization text, it is well regarded in the field for its close and thoughtful explanation of socialization as a process that continues throughout life and not just during childhood. While this may seem obvious, Handel’s assertion is important because it acknowledges that socialization is not a process uniquely situated in childhood. Furthermore, socialization theory is expanded by Handel when he states clearly that socialization, “. . . cannot be understood as simple molding . . . the child is an active agent in its socialization and does not turn out exactly as socializing adults wish” (1988: 1). For socialization theory (a tradition firmly rooted in functional models and highly criticized for its social-determinism orientation), this acknowledgment of children as active agents is a major theoretical expansion. Likewise, in the 1990s we begin to see, for the first time, serious scholarly attention being paid to children’s agency in empirical work. Books such as Deborah Meier’s The Power of Their Ideas (1995), Lillian Rubin’s The Transcendent Child (1996) and Valerie Hey’s The Company She Keeps (1997) give deep insight into the resilience and agency of children as human actors. We see researchers pay attention to, and showcase, the autonomy of children and their roles as social agents and resisters. In the late 1990s we begin to see children emerge in the literature, not just as victims or failures but as potentially powerful “transcenders” (as in Rubin, 1996). Children are seen as quite socially sophisticated (Hey, 1997), and infinitely capable of learning when given the chance (Meier, 1995). Unlike the work mentioned earlier, these books, while studying children of disadvantage, pay special attention to agency and acknowledge the power and resilience of children. Such books are theoretically important because they “push the envelope” of socialization theory by refusing to see children purely as socialized beings, but also to show that they are acting in their childhoods as well as being acted on.



While these books are based on empirical research they are theoretically focused as well. They are using empirical data to inform their theoretical perspective to extend theories of children and childhood more generally. But perhaps the most influential in reconceptualizing socialization theory to include children’s agency has been William Corsaro. Beginning in the late 1970s, extending through the 1980s, and especially in the 1990s, the work of Corsaro was making important contributions to the development of our understanding of children as active agents in their own socialization (Corsaro, 1977, 1979a, b; 1985, 1986, 1992, 1994, 1997). As Gil Musolf has rightly noted, Corsaro, along with Spencer Cahill (1986) and Norman Denzin (1977), not only pushed childhood theory forward, but made leaps for symbolic interaction theory by arguing that children contribute to the social construction of their own worlds and by viewing socialization as a collective process (1996). By accentuating children’s agency and focusing on children’s social interactions (especially with their own peer groups), Corsaro in particular made major theoretical contributions to the sociology of children and childhood. Contemporaries, including Eder, Fine, Goodwin, and Thorne, among others, have shared Corsaro’s structure-agency perspective and have contributed significantly to the expansion and reconceptualization of socialization theory. As we will see in the next sections, Corsaro and his colleagues have had a major impact on the field. Children’s “Voices” and Interdisciplinary Work While the work of the Chicago School in the 1920s and 1930s made a serious effort to bring children’s voices to the forefront, since then their authentic voices have rarely been heard. But in the 1990s, children’s voices begin to reemerge in the literature. Much of this is from the reemergence of qualitative methodologies aimed at capturing the perspective of the child. In the 1990s, qualitative research with children increased. Not only does this work come to be seen more in the sociological literature (see the section “New Sociology of Children” below), but it enters on the interdisciplinary scene as well. Perhaps one of the most well-known examples is from the field of psychology. Carol Gilligan’s Making Connections (1990) involved work with girls at the Emma Willard School. Gilligan and her colleagues use an ethnographic approach, utilizing multiple interviews, questionnaires, focus groups and observations to collect data regarding adolescent girls’ meanings of self, relationship, and morality. Gilligan’s work has been influential in reconceptualizing our understanding of the relationships between girls and their peers, their schooling, their social-psychological development, and their

The Sociology of Children and Childhood in the United States, 1900–1999


culture. In Making Connections, we see girls acting and interpreting their social worlds, speaking and articulating their views. This book, as well as Gilligan’s other books, are clearly psychological in their approach: they tend to emphasize the individual, internal psychological processes and obscure socialstructural ones. Nonetheless, Gilligan’s work has been a major contribution to many fields, including the Sociology of Children. Others in the field of psychology have contributed as well and their in-depth research with children and adolescents has furthered our understanding of kids’ lives in significant ways (Coles, 1977, 1986a, b, 1992; Way, 1998). Other fields have also contributed to our appreciation of children’s voices. By providing a forum for kids to be heard, work from fields such as literature and social work (Krementz, 1996; Desetta, 1996) reveal the broad range of feelings and attitudes among children from specific social groups. And work from a social-historical perspective brings a valuable perspective as well (Nightingale, 1993). Works such as these inform our understanding of children as a heterogeneous, richly diverse group, and falsify the notion of “the child” so dominant in much of the earlier work. Lastly, research from the field of education has explored in creative, qualitative ways the inner lives of children. Often teacher-researchers, these scholars are able to give a unique perspective to studies of children. Books such as Sally Middlebrooks’ Getting to Know City Kids: Understanding Their Thinking, Imagining, and Socializing (1998), and Karen Gallas’s Sometimes I Can Be Anything: Power, Gender and Identity in a Primary Classroom (1998), have truly contributed to a sociological understanding of children and childhood by offering in-depth analyses of children’s everyday lives and social interactions, and by including rich and lengthy segments of interviews with children and recordings of children’s talk in their books. They reveal children as active, innovative social actors with quite sophisticated understandings of their worlds.

THE STATE OF THE ART: THE “NEW SOCIOLOGY OF CHILDREN” The goal thus far has been to establish the major themes and directions in work on children in the field of sociology. The history of the development of the field as described touches on many important scholars and their work related to children. But in documenting the historical development of the Sociology of Children, it is easy to lose sight of just how few studies related to the field there really have been. While it is important to acknowledge the work that has been produced, it is just as important to acknowledge that relative to other



sub-disciplines and more popular fields of study, literature on children has been slim. As explained, after the 1920s and 1930s, very few sociologists considered themselves to be specifically focused on children and childhood. Studies related to children and childhood (such as those touched on in the preceding pages) were small in number compared to other areas of study. And usually the literature that was produced was not explicitly dedicated to looking at children themselves as the subject of study (as we discussed previously, most literature has been on areas related to children: delinquency and gangs, institutional and organizational settings, or the concept of childhood as a social construct). Furthermore, before the 1980s, much of the work on children was focused solely on adolescents. It is important to keep this in mind as we look at sociology’s rediscovery of childhood in the late 1980s (Corsaro, 1997). The Emergence of a “New” Field of Study in the Late 1980s While it may seem from the preceding pages that much work on children existed all along, we must understand the context in which a claim of “rediscovery” was made in the late 1980s: studies explicitly focused on children had been very rare since the 1920s, and when they did exist they usually were relegated to the margins of the field of sociology and most definitely did not enjoy mainstream popularity. Much of the work that claimed to be about children really did not take seriously the child’s perspective, or place children as central (Cahill, 1992). When a researcher such as Anne-Marie Ambert claims that as recently as the early 1980s there was a “near absence of studies on children in mainstream sociology” (1986), we must understand that she is referring to widely read and respected sociological studies aimed specifically at taking children as the subject of research. It is within this context that we can appreciate the emergence of a “new” Sociology of Children. In the late 1970s there were very few sociologists studying children as the primary subject of research. But the few who were working in the field, although definitely marginalized in the larger discipline, began publishing work in the 1980s that became very influential. Patricia and Peter Adler, Spencer Cahill, William Corsaro, Norman Denzin, Donna Eder, Gary Fine, Nancy Mandell, and Barrie Thorne, for example, were some of the pioneers who struggled to have children seen as a serious and important social group deserving of legitimate sociological attention. These researchers believed that children, in and of themselves, deserved first-rate scholarly attention from sociologists. It was during the late 1980s that the effects of the work of American sociologists began to really be felt in the discipline.

The Sociology of Children and Childhood in the United States, 1900–1999


In 1986, Patricia and Peter Adler started an annual review of sociological work on children entitled, Sociological Studies of Child Development. This edited volume became the first real attempt at creating a sense of camaraderie among sociologists interested in work with kids. In their introduction to the first volume, the Adlers argued that “the study of children and child development is empirically and theoretically central to the discipline of sociology” (1986: 3). Empirically, they said, all of us are children in our early years in the world and we all accumulate roles and experiences throughout our childhood which we continue to draw on throughout our life. Theoretically, the Adlers asserted, understanding how humans develop socially and come to behave and interact as they do in group life and social organizations is a critical goal of sociology. Whether from a social interactionism, structural functionalist, or conflict perspective, children are theoretically central to understanding how social life is reproduced and how it changes over time. Yet, they said, there has been a “paucity” of sociological literature in this area. They argued that the neglect of sociological studies focused specifically on children is due to the fact that children are generally conceptualized as being different and inferior to adults and thus unimportant to the study of “adult society.” Furthermore, they asserted that there is no established mainstream tradition of studying children in sociology, thus young sociologists interested in the subject get little support for pursuing their interests (and as a result, often people interested in children turn to other disciplines such as clinical psychology or social work). The Adlers’ volume established a forum for the dissemination of research and information related to the Sociology of Children and Childhood. Contributors were not only adamant about the importance of the re-emergence of the field, but were committed to documenting and explaining sociology’s neglect of studies of children. A 1986 piece by Anne-Marie Ambert examines the historical and sociodemographic context of this neglect and gives several explanations for it. These explanations include the following: the dominance of psychology in the field of research with children and child development focusing on clinical research and “treating” children; sociology as being an historically male-dominated discipline wherein children have traditionally been seen as unimportant; the methodological “road block” encountered in research with children due to mainstream sociology’s heavy leaning toward quantitative work; and the deeply embedded theoretical assumption of children as passive recipients of socialization, and thus unimportant to research in their own right (Ambert, 1986). Ambert also argued that, “researchers tend to study topics which are encouraged in their discipline or which contribute to advance their careers, especially in terms of publication opportunities,” thus, studying children may have been self-defeating, even for young, upwardly mobile



feminist scholars (1986: 12). Like others during the late 1980s, Ambert argues that the establishment of a Sociology of Children, as a field of study, is longoverdue and important to the advancement of social theory: “. . . this very complex world is lived in by children, perceived and negotiated by them, and we do not have nearly enough empirical research or adequate theories covering these important phenomena” (1986: 23). In the late 1980s a string of work such as that of Ambert’s emerged that clearly documented the absence of children in sociology and argued for a Sociology of Children and Childhood dedicated to making children the central subject of study (these include: Corsaro, 1985; Adler & Adler, 1986; Ambert, 1986; Adler & Adler, 1987; Martinson, 1987; Trent, 1987; Thorne, 1987; Alanen, 1990). This work was instrumental in the establishment of a real field of study. Another body of work instrumental in the development and advancement of a Sociology of Children in the U.S. was that of the European Sociology of Childhood. In contrast to the U.S., in many European countries (especially pioneered in the Scandinavian countries), the sociology of childhood has been well established in the social science agenda. Throughout the 1980s European sociologists were contributing in significant ways to the American conversation about children and childhood. At the forefront of this has been the European Centre for Social Welfare Policy and Research’s “Childhood Project” which has been active since 1986 and is based in Vienna. The work coming out of this centre is primarily “aimed to build up a picture of childhood in western society by giving voice to children and promoting their sociological visibility” (Brannen & O’Brien, 1995: 730). Perhaps most notable is the work of Jens Qvortrup, whose sociohistorical analyses and work on childhood as a social form has greatly influenced the American sociological study of children (Qvortrup, 1995, 1996; Qvortrup, Bardy, Sgritta & Wintersberger, 1994). Corsaro and others have drawn heavily on Qvortrup’s work and continue to do so. European sociologists have played a major role in pushing theory forward to conceptualize children as autonomous and theoretically important in their own right. The European influence has been especially great in terms of reconceptualizing childhood itself as a distinct structural form in the life course. A major contribution of the European work has been to highlight the political processes and relationships between children, childhood, and society and to locate these processes in historical, international, comparative, and intergenerational perspective. Furthermore, the European Sociology of Childhood has made major advances in putting forth a political and social policy

The Sociology of Children and Childhood in the United States, 1900–1999


perspective, a more macro-level perspective, and an interdisciplinary orientation. The European Sociology of Children research and literature has been well established by the work of Berentzen (1984), Church and Summerfield (1994), Frønes (1995), James and Prout (1990), Jenks (1982, 1996), Oswald (1992), Oswald, Krappman, Chowdahuri and von Salisch (1987) and others. The 1994 volume, Childhood Matters: Social Theory, Practice, and Politics, is a good example of the work representing this school of thought. Recently, the journal Childhood, affiliated with the Norwegian Centre for Child Research, has begun to play a major role in the U.S. by disseminating research which represents much of the work coming out of the European Sociology of Childhood genre. As a result of the push of some American sociologists to move children to the forefront of the research agenda and the influence of the European work, throughout the 1980s in the U.S. a new Sociology of Children began to emerge. Sociologists involved in the field were contemplating the direction their work should take: Both feminist and traditional knowledge remain deeply and unreflectively centered around the experiences of adults. Our understanding of children tends to be filtered through adult perspectives and interest. How can we bring children more fully into our understanding of social life, including processes of social change (Thorne, 1987: 86)?

By the early 1990s, these researchers were answering their own questions. They became unrelentlessly focused on extending traditional qualitative methodologies for use in research with children. Several sociologists went into the “field” doing ethnographic work with kids and began producing solid, groundbreaking, empirical research on the cutting-edge of the discipline. By 1992, the Sociology of Children had been officially recognized as a “legitimate” subdiscipline; in February of that year the American Sociological Association formally recognized Sociology of Children as an official section in the association. The field became a small but exciting hotbed of scholarly activity focusing on fresh, new perspectives on children and childhood and dedicated to expanding social theory through qualitative research. Those working in this area were focused on moving the study of children from the margins to the center of U.S. Sociology. Major Empirical Work in the 1990s In the 1990s, major empirical works began to appear which solidified the establishment of Sociology of Children as a serious subdiscipline. Some of this literature was widely read throughout the larger discipline and thus gave increasing recognition to the study of children in the field. Although the Sociology of Children continues to be made up of a relatively small group of



sociologists, the work that has begun to be published has been solid and has contributed greatly to sociology and other fields. The new literature both extends methods (by creating innovative qualitative research methodologies for use with children) and social theory (by critiquing and extending traditional theories of child development and socialization based on empirical work with children). Here we turn to take a closer look at a few examples of this work. In 1991, Frances Chaput Waksler’s edited volume, Studying the Social Worlds of Children: Sociological Readings was published as a collection of papers based on empirical studies of children. The book outlines the concept of socialization and provides an in-depth critique of it. Waksler argues that the concept of socialization, if taken as the assumed and only perspective, obscures the empirical reality of children and childhood. Through qualitative research, we see that children are capable and competent social agents and that they are acting in and on the social world. They are not simply passive recipients of socialization processes. Waksler’s main argument is that children must not be seen as “incomplete adults,” but as human beings and social members of society. She asserts that researchers must attempt to not study children as children, but to look at them simply as people (in other words, to “suspend the notion of children” so that the researcher is a sociologist studying people, not an adult studying kids). Waksler claims that “by suspending ‘adult’ assumptions about children and looking at them as we look at other social actors, we can find far more that children are doing than our everyday life attitudes as adults could ever have led us to expect” (1991: 157). Barrie Thorne’s, Gender Play: Girls and Boys in School (1993), was a landmark book which has come to be regarded as a contemporary Sociology of Children classic. It was the first major ethnography produced in the “new” Sociology of Children genre to reach a wide academic audience. Thorne’s feminist methodology involved daily observations and participation in the everyday social worlds of fourth and fifth graders in school. Thorne is “. . . primarily interested in the ways kids construct their own worlds, with and apart from adults” (1993: 20). Her focus is on children and gender and the unique gendered dimensions of children’s social worlds. She asks how children help create and challenge gender structures and meanings: How do children actively come together to help create, and sometimes challenge, gender structures and meanings? Of course, children are strongly influenced by cultural beliefs and by parents, teachers, and other adults. But children’s collective activities should weigh more fully in our overall understanding of gender and social life. One of my goals is to help bring children from the margins and into the center of sociological and feminist thought (Thorne, 1993: 4).

The Sociology of Children and Childhood in the United States, 1900–1999


Thorne’s data reveals girls and boys as active creators and resisters of gender, not simply as passive recipients of adult socialization as previous theories had suggested. Her extensive look at the “play of gender” as “power play” in the social organization of groups of kids and their activities reveals the fluidity of gender and the role of children as agents in its creation. We see how children use gender to propel a cycle of hierarchical but shifting domination and subservience. Furthermore, Thorne’s work shows how the organization and meaning of gender are influenced by age, ethnicity, race, sexuality and social class. We see how the meaning and structure of gender shift within various social contexts. In Troubling Children: Studies of Children and Social Problems (1994), Joel Best presents an edited volume of case studies exploring the relationship between preadolescent children and social problems. The book offers a range of current research in the field to compensate for the fact that, by largely ignoring younger children, sociologists have virtually surrendered the study of those children to child psychologists, who concentrate on the individual, psychological processes that characterize childhood development, rather than focusing on children as social beings (Best, 1994: 4).

Best takes an explicitly social constructionist approach in that he clarifies the ways in which the concepts of “children,” “childhood,” and “social problem” are subjectively defined and located in ways which vary over time and across societies in historically, culturally, and social structurally specific ways. The chapters of the book focus on various aspects of the relationship between children and social problems: the impact on children of societal crisis, pregnancy and infancy, and children’s relationships within institutions such as families and schools. The last chapter of the book, a piece by Donna Lee King, analyzes children’s perspectives on social problems (King, 1994), and paves the way for the publication of her own book the next year. Donna Lee King’s book, Doing Their Share to Save the Planet: Children and Environmental Crisis (1995), like those above, takes children as the sociological subject of analysis. She explores the relationship between children and environmentalism by looking at the roles that children play in the “environmental crisis” and “environmental crisis management.” Her qualitative methods draw on interviews, participant observation, and content analysis of children’s drawings and essays. King reveals a “liberal environmental paradox” which: calls for kids to conserve and consume, diffuses responsibility to support the notion that the crisis is everyone’s fault, and offers clear, simple, individualistic “lifestyle” solutions to what are in reality complex, socialstructural problems of global proportions. She argues that the environmental crisis’ mass-mediated rhetoric distracts from the deeper causes of the problems



and what it would take to truly solve them. Furthermore, she argues that this rhetoric unreflectively reproduces ideologies and social relations of patriarchy, racism and capitalism that are at the root of the problems the rhetoric claims to seek to change. King’s empirical research with children reveals that most kids, – girls and boys, black and white, gifted and ‘special education’ – have developed a clear sense of empowerment, a buoyant belief that they can do something about the problem. However, how empowering is it to teach children about radioactive waste or massive destruction of the rain forests and then tell them the answer is to plant a tree in the backyard or to pick up litter (1995: 5)?

King reveals children’s complex and problematic perspective on the environmental crisis and argues that it requires serious scrutiny and critical attention. She shows children to be actively thinking and theorizing about the problem of environmental crisis and both internalizing and resisting dominant ideological rhetoric. Anne-Marie Ambert’s, Parents, Children, and Adolescents: Interactive Relationships and Development in Context (1997), a family sociology text, is groundbreaking because it is the first to pay close attention to the interactive relationships between adults and children in families. Ambert takes an interactive perspective on the subject of socialization by incorporating the ways in which children affect parents into the sociological analysis of the family (see as well Ambert, 1992). Previous to Ambert’s work, most literature on the family was linear and uni-dimensional; it saw parents as socializing children, and did not acknowledge the various ways in which children, as social actors, also socialize adults. Ambert is interested in children’s resilience and social strength and sees children “. . . not as passive recipients of parental effect, but as active participants in a relationship, as co-producers of their own development, and as co-agents in the production of their family’s quality of life” (1997: 2). Ambert also pays close attention to the context of relational development. She is interested in the ways in which family relations are embedded in social contexts that impact the individuals involved and viceversa. Her work counters “the assumption of parental determinism” by arguing that “we give far too much credit to parents when their children do well and far too much blame when their children have problems” (1997: 248). Ambert’s book is an important contribution to the Sociology of Children because it takes the sociological analysis of the family to a new level, one that recognizes children’s active involvement. Patricia A. Adler and Peter Adler’s long awaited book, Peer Power: Preadolescent Culture and Identity (1998), will no doubt be considered a classic ethnography of children’s peer cultures. The Adlers explicitly locate

The Sociology of Children and Childhood in the United States, 1900–1999


their work in the context of a “New Sociology of Children” which takes an interpretive and social constructivist approach to analyzing children’s lives in the context of their social worlds. Peer Power focuses on children’s friendships and preadolescent peer culture as a subculture. The study followed 8–12 year old children in one community over a period of eight years: “It is about the way these children live their lives when they are unencumbered by the guidance of adults” (Adler & Adler, 1998: 4). The Adlers argue that preadolescent peer culture is a distinctive sub-unit of the broader American culture. They attempt to gain an understanding of the world from the preadolescent’s point of view. They analyze in depth the dimensions of social life that the children in their study see as important: friends and friendship, popularity and social status, free time and activities, and relationships between girls and boys. Children’s peer cultures are given central importance based on empirical evidence which shows that: Preadolescents do not perceive, interpret, form opinions about, or act on the world as unconnected individuals. Rather, they do all these things in concert with their peers, as they collectively experience the world, encounter problems, share their perceptions, and forge joint solutions to those problems (Adler & Adler, 1998: 206).

Their work reveals “. . . the vital and often undervalued role of preadolescent peer culture in mediating between the individual and society, and in powerfully shaping preadolescents’ socialization and identity” (1998: 18). As these examples demonstrate, during the 1990s major empirical works on the sociological study of children emerged. This new body of literature is distinctively well grounded both theoretically and methodologically. As major works began to appear on the scene, they influenced other researchers in the field. By the late 1990s, it was clear in the literature that the “New Sociology of Children” was having an important impact. Smaller pieces, journal articles, and interdisciplinary work has begun to take on the perspective advanced by those working in the field of Sociology of Children, and sociologists have begun to take children seriously (Risman & Myers, 1997). Interpretive Reproduction Theory In 1997, William Corsaro published The Sociology of Childhood. In it, Corsaro presents a broad, yet in depth, synthesis of contemporary theoretical ideas and empirical research on children. This is the first comprehensive Sociology of Children text to provide such a complete overview of the field. This book articulates the theoretical perspective of what Corsaro and others have come to label the “New Sociology of Children.” Corsaro stresses the importance of “interpretive ethnographic methods” to uncover meanings and understandings



of social processes from children’s perspectives. And his approach is unabashedly child-centered; he argues that “children are the best sources for understanding childhood” (Corsaro, 1997: 103). Like the authors of the major empirical works described above, Corsaro asserts that children are worthy of study and documentation in their own right. But the most important contribution that Corsaro’s book makes to the field is its clear and in-depth explanation of the empirically grounded, contemporary social theory evolving out of work in the “New Sociology of Children.” As a representative of the field, he articulates a newly developing theoretical perspective that has arisen from his own and others’ empirical research on children and childhood. Corsaro labels this new theoretical perspective “Interpretive Reproduction” and breaks new ground by offering a sophisticated conceptualization of the theory as it has been developed so far. Although Corsaro himself had been developing this theoretical paradigm for over two decades (see especially Corsaro, 1992; Corsaro & Rizzo, 1988), it is in his 1997 book that the theory is so fully articulated. The basic premise of the theory is that children contribute to both social stability and social change through a process of “interpretive reproduction.” Interpretive Reproduction can be summarized as follows: Children are conceptualized as active social agents who collectively participate in the production of culture and society and who are not just passive, adults-in-thewaiting. Corsaro gives three types of children’s collective action which “occur both in the moment and over time” (1997: 41): (1) Children’s creative appropriation of information and knowledge from the adult world; (2) Children’s production and participation in a series of peer cultures; and (3) Children’s contribution to the reproduction and extension of the adult culture (1997: 41). Children think about, interpret and creatively act on and in the social world. They resist, challenge, negotiate, and create in social life; they are innovative in their participation in cultural production and social change (thus, the “interpretive” in “interpretive reproduction”). At the same time, children are understood to be members of a social group that is very much subordinated in the social structure. Childhood is considered a structural form (a part of society), interrelated with other forms (such as class, race, gender, etc.) which changes over time and varies among cultures and social structures. As members of a subordinate social group, children’s agency is severely limited by the existing social structure. Their social, cultural, physical and economic power is seriously constrained (more so than most other social groups). They are

The Sociology of Children and Childhood in the United States, 1900–1999


impacted by dominating social forces, cultural practices, and individuals who hold more power than them to act in ways which contribute to the reproduction of the social structure over time (thus, the “reproduction” in “interpretive reproduction”). Corsaro defines children’s peer culture as “a stable set of activities or routines, artifacts, values, and concerns that children produce and share in interaction with peers” (1997: 114). Peer culture is an important component of the sociology of childhood because “. . . children create and participate in their own unique peer cultures by creatively taking or appropriating information from the adult world to address their own peer concerns” and “individual development is embedded in the collective production of a series of peer cultures which in turn contribute to reproduction and change in the wider adult society or culture” (1997: 18, 26). Generally speaking, Interpretive Reproduction is the theoretical basis underlying the perspective of the New Sociology of Children. Although Corsaro is the first to so clearly articulate the theory it has emerged and evolved in part from the work of several sociologists who have worked in the field of Sociology of Children (including Adler & Adler, 1998; Ambert, 1997; Eder, 1995; Fine, 1987; Goodwin, 1990; King, 1995; Mandell, 1988; Qvortrup, 1993; Thorne, 1993; Waksler, 1991). It is a new theoretical model which stands in opposition to traditional theories of child development and socialization which basically viewed children as going through a process of slowly internalizing society over time. Interpretive Reproduction criticizes traditional socialization theories as: being overly deterministic and not leaving enough room for human agency and the active capacity of children; being too focused on outcome (adulthood), often neglecting to understand children as being in a state-of-being; being too abstractly theoretical and overlooking the empirical reality of children’s lives; being overly individualistic, neglecting an historical perspective; and simplifying complex processes to a linear, one-dimensional model. In this way, the sociologists working to advance Interpretive Reproduction theory are reconceptualizing childhood socialization by emphasizing its creative and collective aspects and seeing children as active in their own development. The implications of this theoretical advancement are vast and far-reaching both within the field and beyond it. Corsaro’s work throughout his career, especially his work on Interpretive Reproduction, has made major contributions to the Sociology of Children and has influenced the work of other major players in the field and vice versa. His 1997 articulation of this cutting edge paradigm sets the stage for major theoretical advances in the field as researchers enter a new decade.



Major Tenets and Basic Research Agendas The New Sociology of Children seeks to understand children from a uniquely sociological perspective. Its major goal is to advance our empirical and theoretical understanding of children and childhood. While the work emerging from this newly developing field is diverse in its subject matter and content, it does have several things in common which characterize its major tenets and goals. In many ways it is a developing school of thought much as the work on children coming out of the Chicago School during the 1920s and 1930s was a certain school of thought. First, the New Sociology of Children takes children as the basic unit of study. Children are the explicit subject of analysis. Unlike much of the work before it, the New Sociology of Children focuses on kids – just kids. Previously much of the literature tended to focus on topics associated with children such as the family, education, or adolescent crime and delinquency and children had “rarely been studied as a topic in their own right” (Brannen & O’Brien, 1995: 729). But for the New Sociology of Children, children, often preadolescent or very young children, are the sociological topic of research not the object of it. Even more specifically, the New Sociology of Children has a certain conception of children as being active. More than ever before, children are being viewed as active members of society, possessing roles and agency and making an impact on society. The focus has turned to, . . . the place of children in society, the roles they occupy, and their agency. Children are seen as active members of society who act upon it and react to it. Children are seen as individuals who contribute to society, and childhood is an integral structural element of any society” (Ambert, 1995: 249).

This conceptualization is more than just a new way of looking at kids, it is a critique of traditional models of development and socialization which have “rendered children dependent, relegating them to ‘waiting rooms’ and to being adults in the making rather than children in the state of being” (Brannen & O’Brien, 1995: 730). The New Sociology of Children attacks traditional models for their “assumptions of determinism, linearity, functionality and harmony in the process of socialization” (Alanen, 1990: 19). Children are seen as being quite “accomplished” in negotiating their worlds and are being presented as more sophisticated than ever before in the literature (Adler & Adler, 1987). Another major characteristic of the New Sociology of Children is its use of qualitative methodologies aimed at revealing data on the child’s perspective. The work seeks to understand how children themselves perceive their world, how they construct meaning, and make sense of social life. Methods are

The Sociology of Children and Childhood in the United States, 1900–1999


constructed in order to elucidate children’s own experiences and their understandings of their daily lives. Much of the work on children outside of New Sociology of Children often relies on adults’ interpretations of children, or on adults’ recollections of their own childhood. But researchers working in the New Sociology of Children work directly with children, often utilizing methods such as participant observation, interviewing, analysis of children’s artwork or play, or conversation analysis. These researchers tend to support an ethnographic approach to studying children because it makes more real-life sense to the kids (as opposed to laboratory experiments or statistical surveys, etc.). While research methodology is still a major hurdle in sociological work with children (sociologists are still trying to construct creative, new methodologies that are child-centered), the qualitative methodologies being advanced by New Sociology of Children are paving new ground. Work focusing explicitly on methodology and discussing the challenges and controversies surrounding sociological research on children has been especially influential in helping researchers create productive research designs. Methods of studying children have become a significant topic in their own right. Examples of this important dimension of the literature can be found in the work of Adler and Adler (1996), Fine and Sandstrom (1988), Graue and Walsh (1998), Holmes (1998), and others. But while the New Sociology of Children often takes an ethnographic, micro-interactional approach to empirical research, this does not stand in the way of acknowledging macro-level phenomena at play. A characteristic of the New Sociology of Children is its attention to historical, social, and cultural patterns and forces in the lives of individuals. Childhood is understood to be a socially constructed structural form interrelated with other structural forms. An emphasis is placed on the fact that children (nor any other member of society for that matter) never operate within a vacuum. It is understood that micro- and macro-level phenomena are interrelated in significant and meaningful ways, and one cannot be understood without the context of the other. Here is where we’ve seen major contributions by quantitative researchers studying children. Social demographers have paved new ground by making children the central unit of analysis, putting forth statistical information, showing historical demographic trends and portraying childhood in sociodemographic perspective. This work, including that of Entwisle and Astone (1994), Hauser (1994), Hernandez (1995; 1997), and Lichter (1997) has played a significant role in developing some of the macro-sociological dimensions of the field. Furthermore, empirical research that relies on quantitative methods is a piece of the New Sociology of Children literature that should not be overlooked. Maureen Hallinan’s work on children’s friendships (see her work with Eder



1978; with Smith, 1989; and with Williams, 1990, 1989, 1987), Jeylan Mortimer’s work on adolescent work experience (see her work with Shanahan, 1994; and with Finch et al., 1990), and Owens’ work on youth self-esteem (see 1994; and with Mortimer & Finch, 1996) are good examples of the strengths of quantitative methodologies. The New Sociology of Children emphasizes the interaction between structure and agency and the dynamic tension that exists between them. This insistence upon focusing on the micro-macro links and the complexity of human behavior in the context of social life makes it difficult to make simple generalizations but makes for a more realistic, grounded sociology of children. Researchers working from both qualitative and quantitative perspectives have helped to advance the field and shape the research agenda. The New Sociology of Children can also be characterized by a focus on how power plays a part in children’s lives. One of the themes that runs throughout the work is an emphasis on the variousness and diversity of children along race, class, and gender lines (see, for example, Rosier & Corsaro, 1993). Much of the work is specifically oriented toward furthering understanding of how complex power dynamics play a part in children’s lives and are used and understood by them. Lastly, the New Sociology of Children, like the Chicago School, makes a clear association between research and action (Trent, 1987). The goal is not just to understand children, but to understand how to use sociological knowledge for the advancement of social justice: “Much of the driving force behind the new sociology of childhood is political and its underpinnings are a concern with children’s rights” (Brannen & O’Brien, 1995: 730). Thus, we see some connections between the contemporary work on children coming out of the New Sociology of Children and the early work on children that came out of the Chicago School. Sociological studies of children and childhood have, in a sense, come full circle, but with a new expanded perspective.

CONTEMPORARY DEBATES Since the New Sociology of Children is still in its infancy, theoretical perspectives have just begun to be articulated, and research methods are still being developed, it is no surprise that although contemporary debates exist in the literature they are not clearly defined. Nevertheless, several contemporary issues face those interested in working in this area of study, some of them quite controversial in the field. Here we will look at a few of these contemporary debates.

The Sociology of Children and Childhood in the United States, 1900–1999


One challenge facing the field is the conceptual problem of what exactly is the appropriate subject of study. While sociologists working both in and outside of the “New Sociology of Children” claim to be focused directly on the child, often their work has a tendency to conflate the concepts of child, childhood, children’s lives, etc. without making distinctions. For example, Corsaro (1997) clearly identifies his work as being within the New Sociology of Children, and he explicitly states his commitment to research that puts children as central. But this is all articulated in a book entitled, The Sociology of Childhood (note, he chooses “Childhood”, not “Children” for the title). It is not always clear whether his focus is on kids or on the stage of social life called “childhood.” While it may seem irrelevant, it really is not: The boundaries between the concepts are not clear, and while many researchers’ work does emphasis children themselves as the subject, many others’ work is devoted to the history of childhood as a social construction, or the social arrangements of children’s lives. The questions of what the actual conceptual subject of research should be, and how important distinctions between the terms are, are ones that remain contested in the field. Should the focus be the study of children (such as in Adler & Adler, 1998; and in King, 1995), or the study of childhood (such as in Lee, 1982; Synnott, 1983; Qvortrup et al., 1994)? Or should the research agenda of the field include both? If we are to include both is it necessary to differentiate the terms and delineate the attendant studies of both? Or are there ways to combine the two and weave the concepts in our research and writing? Another issue in the literature is the debate on the relationship between structure and agency – specifically, the question of whether to view children as social agents or objects of socialization. While much of the work within the New Sociology of Children attempts to link structure and agency and understand how the two interact together, the conflation and blurring of the two concepts is commonplace. In the literature on children it is often difficult to discern to what extent children have the capacity to act in meaningful ways, and to what extent they are constrained by pre-determined conditions. While many strive to see children as a social group which is subject to hierarchies and social control and to recognize the autonomy and agency of children, usually there is a leaning toward one side or the other in the work. What are the limits to children’s power? To what extent should society (and adults) be held responsible for children and to what extent should children be made responsible for themselves? What exactly defines agency when one is considering a severely constrained social group, such as children? Does deviance translate into resistance on the part of a social agent, or does it translate into re-inscription? To what extent do children really play a role in social change? It is in this area that researchers have and will continue to draw



on Corsaro’s work which makes genuine and successful attempts at understanding the structure-agency complexities in the lived experience of children. Here again Corsaro’s theoretical contributions have been and will continue to be influential in the field (Corsaro & Rizzo, 1988; Corsaro & Eder, 1990; Corsaro, 1992; Corsaro, 1997). Yet, we are still faced with the theoretical challenge of advancing our understanding of the structure-agency relationship. This leads to another, somewhat related, debate in the contemporary literature. That is the question of whether children should be viewed as agents of their own desires or victims of others’ desires. To what extent are children products of the society or free agents? Sociologists such as Joel Best have argued that the concept of child as “victim” is a socially constructed form of rhetoric that has been much exaggerated (1990). At the same time, we know that children often are in relatively powerless positions, unable to protect or defend themselves, let alone acquire their own desires. We don’t want to underestimate their vulnerability and inability to act due to constraints external to them. On the other hand, seeing children as a completely disenfranchised, powerless group blinds us to seeing their ability to act in the world. Should sociological research on children be focused on children’s ability to further their own cause, or should the goal be to further children’s causes on behalf of them given their “victim” status? Or, should the research agenda find successful ways to portray children as both? Is childhood a time of agency, exploitation, or a complex combination of both? If it is both, how can we focus the research agenda in ways that will help us better understand how these concepts interact and play themselves out in real life? Lastly is the issue of the nature of childhood: To what extent are children different from, or the same as, adults? Is childhood purely a social construct or are children somehow qualitatively different in kind from other human beings? On one side of the debate are sociologists such as Waksler, who argues that we should view children as actors in the social world like any other person (1991; see also 1987 and 1996). She argues that by viewing children as conceptually different from adults we obscure what they can do and are doing. We must, she argues, at least for the sake of our research methodology, view children as being just “little people.” On the other side of the debate are sociologists such as Corsaro (1997), Thorne (1993), and Fine (1987), who argue that by seeing kids simply as actors (as Waksler asserts), we obscure what is being done to them and the larger, structural consequences of what they do. They argue for a perspective that sees children as a “special case,” as different from adults in many ways.

The Sociology of Children and Childhood in the United States, 1900–1999


THE SOCIOLOGY OF CHILDREN AT THE MILLENNIUM In the opening to his book The Sociology of Childhood (1997), William Corsaro states the following: “Sociology has no tradition for studying children and childhood; until recently those areas have been relegated to the margins of the field” (p. xiii). As we have seen, starting in the mid-1980s sociology did indeed witness the emergence of a “New” Sociology of Children which has begun to successfully push children from the margins to the center of the sociological lens. For the first time in the history of American Sociology sociologists are studying children from an interpretive perspective and are making kids the central focus of analysis. However, as we have also seen, throughout the past one hundred years there have been sociologists attempting to establish research and literature on children in the U.S. In many cases they were pioneers: their work paved new ground in empirical research and expanded the theoretical literature for those who followed them to build upon. As we move into a new millennium, we who study children are witnessing and taking part in the development of a new theoretical and substantive area in the Sociology of Children. As has been shown here, the New Sociology of Children is innovative and progressive in fundamentally important ways and the advances in the sociological study of children are truly impressive. At the same time, many of the questions that earlier work on children addressed are quite similar to contemporary research agendas: How do children perceive the social world (what are children’s definitions of the situation)? What role do children play in social reproduction and social change? What are the relationships between children and society? And many of the earlier researchers’ methodological challenges were similar to contemporary researchers’ as well: How can an adult sociologist best approach research with child “subjects”? How can we create methodologies to allow children to express themselves in ways most natural and comfortable for them? What are the unique ethical dilemmas involved in research with children (for example, can children give informed consent on their own behalf?). Understanding the questions that have been asked allows us to better refine our own questions and formulate new ones. By carefully reviewing the earlier work, we see that we actually share a lot in common with sociologists who studied kids before us. We can see how our field has evolved over time. While it is exciting to know that we are on the cutting edge and pioneering in new territory it is also comforting to know that others have walked some of the road before us. It centers us working in the field



today to know that we are building on a foundation, not creating something from nothing. Now, as we look back on where we have come from, we can’t help but be enthusiastically excited about where we are going.

ACKNOWLEDGMENTS Thanks to Tom Shapiro, Gordana Rabrenovic, Peter and Patricia Adler, Fran Waksler, and especially Maureen Kelleher for advice and support throughout the development of this paper. Thanks also to Timothy Owens, Sandra Hofferth, and two anonymous reviewers for criticisms and comments that helped to strengthen this paper immensely.

REFERENCES Adler, P. A., & Adler, P. (1986). Introduction. Sociological Studies of Child Development, 1, 3–9. Adler, P. A., & Adler, P. (1987). Introduction. Sociological Studies of Child Development, 2, 3–9. Adler, P. A., & Adler, P. (1996). Parent-as-Researcher: The Politics of Researching in the Personal Life. Qualitative Sociology, 19, 35–58. Adler, P. A., & Adler, P. (1998). Peer Power: Preadolescent Culture and Identity. New Brunswick, NJ: Rutgers University Press. Alanen, L. (1990). Rethinking Socialization, The Family, and Childhood. Sociological Studies of Child Development, 3, 13–28. Ambert, A.-M. (1986). Sociology of Sociology: The Place of Children in North American Sociology. Sociological Studies of Child Development, 1, 11–31. Ambert, A.-M. (1995). Sociological Theorizing on Children: Concluding Thoughts. Sociological Studies of Children, 7, 247–255. Ambert, A.-M. (1997). Parents, Children, and Adolescents: Interactive Relationships and Development in Context. New York: Haworth Press. Ambert, A. M. (1992). The Effects of Children on Parents. New York: Haworth Press. Aries, P. (1962). Centuries of Childhood: A Social History of Family Life. Translated by Robert Baldick. New York: Vintage Books. Ayers, W. (1997). A Kind and Just Parent: The Children of Juvenile Court. Boston: Beacon Press. Berentzen, S. (1984). Children Constructing their Social World. Bergen Studies in Social Anthropology, 36. Berger, P. L., & Luckmann, T. (1966). The Social Construction of Reality. New York: Doubleday. Best, J. (1990). Threatened Children: Rhetoric and Concern about Child-Victims. Chicago: University of Chicago Press. Best, J. (1994). Troubling Children: Studies of Children and Social Problems. New York: Aldine de Gruyter. Brannen, J., & O’Brien, M. (1995). Childhood and the Sociological Gaze: Paradigms and Paradoxes. Sociology, 29, 729–737. Burgess, E. W. (1925). The Growth of the City. In: R. E. Park, E. W. Burgess & R. D. McKenzie (Eds), The City (pp. 47–62). Chicago: University of Chicago Press. Bynum, J. E., & Thompson, W. E. (1996). Juvenile Delinquency: A Sociological Approach. Boston: Allyn and Bacon.

The Sociology of Children and Childhood in the United States, 1900–1999


Cahill, S. (1986). Childhood Socialization as a Recruitment Process: Some Lessons from the Study of Gender Development. Sociological Studies of Child Development, 1, 163–186. Cahill, S. (1992). The Sociology of Childhood at and in an Uncertain Age. Contemporary Sociology, 21, 669–672. Church, J., & Summerfield, C. (1994). Social Focus on Children. London: HMSO. Clark, K. B., & Clark, M. K. (1939). The Development of Consciousness of Self and the Emergence of Racial Identification in Negro Preschool Children. The Journal of Social Psychology, 10, 591–599. Clark, K. B., & Clark, M. P. (1947). Racial Identification and Preference in Negro Children. In: T. M. N. Comb & E. L. Hartley (Eds), Readings in Social Psychology. New York: Holt. Cloward, R., & Ohlin, L. (1960). Delinquency and Opportunity: A Theory of Delinquent Gangs. New York: The Free Press. Cohen, A. K. (1955). Delinquent Boys: The Culture of the Gang. New York: The Free Press. Coles, R. (1977). Privileged Ones: The Well-Off and the Rich in America (Vol. V of Children of Crisis). Boston: Little, Brown & Company. Coles, R. (1986). The Moral Life of Children. Boston: Houghton Mifflin. Coles, R. (1986). The Political Life of Children. Boston: Atlantic Monthly Press. Coles, R. (1992). Their Eyes Meeting the World: The Drawings and Paintings of Children. Boston: Houghton Mifflin. Corsaro, W. A. (1977). The Clarification Request as a Feature of Adult Interactive Styles with Young Children. Language in Society, 6, 183–207. Corsaro, W. A. (1979). We’re Friends, Right?: Children’s Use of Access Rituals in a Nursery School. Language in Society, 8, 315–336. Corsaro, W. A. (1979). Young Children’s Conception of Status and Role. Sociology of Education, 52, 46–59. Corsaro, W. A. (1985). Friendship and Peer Culture in the Early Years. Norwood, NJ: Ablex. Corsaro, W. A. (1986). Discourse Processes within Peer Culture: From a Constructivist to an Interpretive Approach to Childhood Socialization. Sociological Studies of Child Development, 1, 81–101. Corsaro, W. A. (1992). Interpretive Reproduction in Children’s Peer Cultures. Social Psychology Quarterly, 55, 160–177. Corsaro, W. A. (1994). Discussion, Debate, and Friendship Processes: Peer Discourse in U.S. and Italian Nursery Schools. Sociology of Education, 67, 1–26. Corsaro, W. A. (1997). The Sociology of Childhood. Thousand Oaks, CA: Pine Forge Press. Corsaro, W. A., & Eder, D. (1990). Children’s Peer Culture. Annual Review of Sociology, 16, 197–220. Corsaro, W. A., & Rizzo, T. A. (1988). Discussion and Friendship: Socialization Processes in the Peer Culture of Italian Nursery School Children. American Sociological Review, 53, 879–894. Denzin, N. K. (1977). Childhood Socialization. San Francisco: Jossey-Bass. Desetta, A. (1996). The Heart Knows Something Different: Teenage Voices from the Foster Care System. New York: Persea Books. Durkheim, E. (1897/1951). Suicide: A Study in Sociology (Translated by J. A. Spaulding & G. Simpson). New York: The Free Press. Eder, D., Evans, C. C., & Parker, S. (1995). School Talk: Gender and Adolescent School Culture. New Brunswick, NJ: Rutgers University Press. Eder, D., & Hallinan, M. T. (1978). Sex Differences in Children’s Friendships. American Sociological Review, 43, 237–250.



Elder, G. H., Jr. (1974). Children of the Great Depression: Social Changes in Life Experience. Chicago: University of Chicago Press. Elder, G. H., Jr., Modell, J., & Parke, R. D. (1993). Children in Time and Place: Development and Historical Insights. New York: Cambridge University Press. Entwisle, D. R., & Astone, N. M. (1994). Some Practical Guidelines for Measuring Youth’s Race/ Ethnicity and Socioeconomic Status. Child Development, 65, 1521–1541. Erikson, E. H. (1950). Childhood and Society. New York: W. W. Norton and Company. Erikson, E. H. (1961). The Challenge of Youth. New York: Anchor Books. Fine, G. A. (1987). With the Boys. Chicago: University of Chicago Press. Fine, G. A., & Sandstrom, K. L. (1988). Knowing Children: Participant Observation with Minors. Beverly Hills, CA: Sage Publications. Friedenberg, E. Z. (1963). Coming of Age in America: Growth and Acquiescence. New York: Vintage Books. Frønes, I. (1995). Among Peers: On the Meaning of Peers in the Process of Socialization. Oslo, Norway: Scandinavian University Press. Gallas, K. (1998). Sometimes I Can Be Anything: Power, Gender and Identity in a Primary Classroom. New York: Teachers College Press. Garbarino, J., Dubrow, N., Kostelny, K., & Pardo, C. (1992). Children in Danger: Coping with the Consequences of Community Violence. San Francisco: Jossey-Bass. Garbarino, J., Kostelny, K., & Dubrow, N. (1991). No Place to be a Child: Growing up in a War Zone. Lexington: Lexington Books. Gelles, R. J. (1996). The Book of David: How Preserving Families can Cost Children’s Lives. New York: Basic Books. Gilligan, C., Lyons, N., & Hanmer, T. (1990). Making Connections: The Relational Worlds of Adolescent Girls at Emma Willard School. Cambridge, MA: Harvard University Press. Goodman, M. E. (1952). Race Awareness in Young Children. New York: Collier Books. Goodwin, M. H. (1990). He-Said-She-Said: Talk as Social Organization Among Black Children. Bloomington, IN: Indiana University Press. Graue, M. E., & Walsh, D. J. (1998). Studying Children in Context: Theories, Methods, and Ethics. Thousand Oaks, CA: Sage Publications. Hallinan, M. T., & Smith, S. S. (1989). Classroom Characteristics and Student Friendship Cliques. Social Forces, 67, 898–919. Hallinan, M. T., & Williams R. A. (1987). The Stability of Students’ Interracial Friendships. American Sociologial Review, 52, 653–664. Hallinan, M. T., & Williams R. A. (1989). Interracial Friendship Choices in Secondary Schools. American Sociological Review, 54, 67–78. Hallinan, M. T., & Williams R. A. (1990). Students’ Characteristics and the Peer Influence Process. Sociology of Education, 63, 122–132. Handel, G. (1988). Childhood Socialization. New York: Aldine de Gruyter. Hauser, R. M. (1994). Measuring Socioeconomic Status in Studies of Child Development. Child Development, 65, 1541–1560. Hernandez, D. J. (1995). America’s Children: Resources from Family, Government and the Economy. New York: Russell Sage Foundation. Hernandez, D. J. (1997). Child Development and the Social Demography of Childhood. Child Development, 68, 149–153. Hewlett, S. A. (1991). When the Bough Breaks: the Costs of Neglecting Our Children. New York: Basic Books.

The Sociology of Children and Childhood in the United States, 1900–1999


Hey, V. (1997). The Company She Keeps: An Ethnography of Girls’ Friendship. Buckingham: Open University Press. Holmes, R. M. (1995). How Young Children Perceive Race. Thousand Oaks: Sage. Holmes, R. M. (1998). Fieldwork with Children. Thousand Oaks, CA: Sage. James, A., & Prout, A. (1990). Constructing and Reconstructing Childhood: Contemporary Issues in the Sociological Study of Childhood. London: Falmer Press. Jenks, C. (1982). The Sociology of Childhood. London: Batsford Academic. Jenks, C. (1996). Childhood. London: Routledge. Joas, H. (1987). Symbolic Interactionism. In: A. Giddens & J. Turner (Eds), Social Theory Today (pp. 82–115). Stanford, CA: Stanford University Press. King, D. L. (1994). If We Don’t Do Anything Now, There Won’t Be Anything Left: Categories of Concern in Children’s Drawings of Environmental Crisis. In: J. Best (Ed.), Troubling Children: Studies of Children and Social Problems (pp. 221–246). New York: Aldine de Gruyter. King, D. L. (1995). Doing Their Share to Save the Planet: Children and Environmental Crisis. New Brunswick, NJ: Rutgers University Press. Kotlowitz, A. (1991). There Are No Children Here: The Story of Two Boys Growing Up in the Other America. New York: Anchor Books, Doubleday. Kozol, J. (1991). Savage Inequalities: Children in America’s Schools. New York: Crown. Kozol, J. (1995). Amazing Grace: The Lives of Children and the Conscience of a Nation. New York: Crown Publishers. Krementz, J. (1996). How it Feels to be Adopted. New York: Alfred A. Knopf. Kvaraceus, W. C., & Miller, W. B. (1959). Delinquent Behavior: Culture and the Individual. Westport: Greenwood Press. Lee, J. A. (1982). Three Paradigms of Childhood. Canadian Review of Sociology and Anthropology, 19, 591–608. Lichter, D. T. (1997). Poverty and Inequality Among Children. Annual Review of Sociology, 23, 121–145. MacLeod, J. (1987). Ain’t No Makin’ It: Aspirations and Attainments in a Low-Income Neighborhood. Boulder, CO: Westview Press. Mandell, N. (1984). Children’s Negotiation of Meaning. Symbolic Interaction, 7, 191–211. Mandell, N. (1988). The Least Adult Role in Studying Children. Journal of Contemporary Ethnography, 16, 433–467. Marek, E. (1987). The Children at Santa Clara. New York: Elisabeth Sifton Books: Viking. Martinson, F. M. (1987). Children and Family Sociology Texts: United States and Sweden. Wisconsin Sociologist, 24, 49–56. Meier, D. (1995). The Power of Their Ideas: Lessons for America from a Small School in Harlem. Boston: Beacon Press. Merton, R. K. (1938). Social Structure and Anomie. American Sociological Review, 3 (October), 672–682. Middlebrooks, S. (1998). Getting to Know City Kids: Undertanding Their Thinking, Imagining, and Socializing. New York: Teachers College Press. Miller, W. B. (1958). Lower Class Culture as a Generating Milieu of Gang Delinquency. Journal of Social Issues, 14, 5–19. Mortimer, J. T., Finch, M. D., Owens, T. J., & Shanahan, M. (1990). Gender and Work in Adolescence. Youth and Society, 22, 201–224. Mortimer, J. T., & Shanahan, M. J. (1994). Adolescent Work Experience and Family Relationships. Work and Occupations, 21, 369–384.



Musolf, G. R. (1996). Interactionism and the Child: Cahill, Corsaro, and Denzin on Childhood Socialization. Symbolic Interaction, 19, 303–321. Nightingale, C. H. (1993). On the Edge: A History of Poor Black Children and Their American Dreams. New York: Basic Books. Oswald, H. (1992). Negotiations of Norms and Sanctions among Children. Sociological Studies of Child Development, 5, 99–114. Oswald, H., Krappman, L., Chowdahuri, I., & von Salisch, M. (1987). Gaps and Bridges: Interactions Between Girls and Boys in Elementary School. In: P. Adler & P. A. Adler (Eds), Sociological Studies of Child Development (Vol. 2, pp. 205–223). Greenwich, CT: JAI Press. Owens, T. J. (1994). Two Dimensions of Self-Esteem: Reciprocal Effects of Positive Self-Worth and Self-Deprecation on Adolescent Problems. American Sociological Review, 59, 391–407. Owens, T. J., Mortimer, J. T., & Finch, M. D. (1996). Self-Determination as a Source of SelfEsteem in Adolescence. Social Forces, 74, 1377–1404. Park, R. E., & Burgess E. W. (1925). The City. Chicago, IL: University of Chicago Press. Parsons, T. (1937). The Structure of Social Action. New York: McGraw-Hill. Postman, N. (1982). The Disappearance of Childhood. New York: Dela Cote Press. Qvortrup, J. (1993). Societal Position of Childhood: The International Project Childhood as a Social Phenomenon. Childhood, 1, 119–124. Qvortrup, J. (1995). From Useful to Useful: The Historical Continuity of Children’s Constructive Participation. Sociological Studies of Children, 7, 49–76. Qvortrup, J. (1996). Childhood in a Post-Industrial World. Development, 1, 64–67. Qvortrup, J., Bardy, M., Sgritta, G., & Wintersberger, H. (1994). Childhood Matters: Social Theory, Practice, and Politics. Brookfield, VT: Avebury. Rapoport, R. N. (1985). Research and Action. In: R. N. Rapoport (Ed.), Children, Youth, and Families: The Action-Research Relationship (pp. 1–25). New York: Cambridge University Press. Risman, B. J., & Myers, K. (1997). As the Twig is Bent: Children Reared in Feminist Households. Qualitative Sociology, 20, 229–252. Ritzer, G. (1992). Classical Sociological Theory. New York: McGraw-Hill. Rosier, K. B., & Corsaro W. A. (1993). Competent Parents, Complex Lives: Managing Parenthood in Poverty. Journal of Contemporary Ethnography, 22, 171–204. Rubin, L. (1996). The Transcendent Child: Tales of Triumph Over the Past. New York: Basic Books. Ryan, J. (1995). Little Girls in Pretty Boxes: The Making and Breaking of Elite Gymnasts and Figure Skaters. New York: Warner Books. Shaw, C. R. (1930). The Jack-Roller: A Delinquent Boy’s Own Story. Chicago: University of Chicago Press. Shaw, C. R., & McKay, H. D. (1942). Juvenile Delinquency and Urban Areas. Chicago: University of Chicago Press. Synnott, A. (1983). Little Angels, Little Devils: A Sociology of Children. The Canadian Review of Sociology and Anthropology, 20, 79–95. Thomas, W. I. (1923). The Unadjusted Girl: With Cases and Standpoints for Behavior Analysis. New York: Harper TorchBooks Harper and Row. Thomas, W. I., & Thomas, D. S. (1928). The Child in America: Behavior Problems and Programs. New York: Knopf.

The Sociology of Children and Childhood in the United States, 1900–1999


Thomas, W. I., & Znaniecki, F. (1918). The Polish Peasant in Europe and America. New York: University of Illinois Press. Thorne, B. (1987). Re-visioning Women and Social Change: Where are the Children? Gender and Society, I, 85–109. Thorne, B. (1993). Gender Play: Girls and Boys in School. New Brunswick, NJ: Rutgers University Press. Thrasher, F. M. (1927). The Gang: A Study of 1,313 Gangs in Chicago. Chicago: University of Chicago Press. Trent, J. W. (1987). A Decade of Declining Involvement: American Sociology in the Field of Child Development, The 1920s. Sociological Studies of Child Development, 2, 11–37. Waksler, F. C. (1987). Dancing When the Music is Over: A Study of Deviance in a Kindergarten Classroom. Sociological Studies of Child Development, 2, 139–158. New York: The Falmer Press. Waksler, F. C. (1996). The Little Trials of Childhood. Bristol: Falmer Press. Way, N. (1998). Everyday Courage: The Lives and Stories of Urban Teenagers. New York: New York University Press. Whyte, W. F. (1955). Street Corner Society. Chicago: University of Chicago Press. Winn, M. (1983). Children Without Childhood. New York: Random House.

This Page Intentionally Left Blank


This Page Intentionally Left Blank



Duane F. Alwin ABSTRACT In this chapter the author reviews the research literature on parental values, beliefs and behavior over the past several decades and turns his attention to issues that can be profitably addressed in future research on this topic. This review is framed against the backdrop of the massive changes experienced by the family in its makeup and functioning over that time in Europe and North America, and the implications of these changes for understanding the lives of children and their parents. The chapter emphasizes the importance of conceptualizing values within a social psychological framework of cultural and cognitive organization, which views parental behavior as activity organized to satisfy basic material, psychological and social needs, and parental values as the standards of desirability that govern the choices that parents make in choosing approaches to child-rearing. This framework is coupled with Bronfenbrenner’s conceptualization of the social environment as a dynamic set of interconnections of social settings, embedded in a multi-layered social and cultural context, that have major implications for child-rearing behavior. A review of the literature on parental values identifies several Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 97–139. Copyright © 2001 by Elsevier Science Ltd. All rights of reproduction in any form reserved. ISBN: 0-7623-0776-5




themes, particularly those having to do with socioeconomic inequalities in parental values, religio-ethnic differences in historical changes in values, and the linkages between parental values and child-rearing behavior. In particular, research has shown that there have been major changes in parental values over the past century in the direction of greater preference for autonomy versus obedience in children and that these value orientations are connected to aspects of parental styles of behavior which may also be changing in the contemporary neo-traditional family. The chapter concludes with some thoughts about how research on parenting in future years can not only benefit from what is already known, but can also be invigorated by new theoretical and methodological approaches.

1. BACKGROUND There is widespread agreement among scholars, laypersons, and policy makers that the “traditional” family in Europe and North America has experienced fundamental changes in its makeup and functioning over the past two centuries. Of particular interest is the fact that families have changed in a number of critical ways that significantly affect the lives of children and their parents, particularly in their typical size, their composition and the activities of parents. One need only consider a few bedrock social and demographic changes that have been occurring systematically during the past century to understand how profound are the changes affecting children and the family environments that nurture them, e.g. technological and industrial change, fertility decline, and changing norms involving marriage and the sexual division of labor, among other things (Westoff, 1986). Parenting among humans occurs in a cultural and historical context. For example, the historical tradeoffs between child-bearing and child-rearing underscores the importance of looking at the link between reproduction and socialization – not only at the numbers, timing and spacing of children, but also the kinds of children families want to produce. It may be increasingly the case that people inhabit what Furstenberg (1992) called the neo-traditional family – “a renovated model of the gender-based division of labor when women share a greater measure of economic responsibility and men may assume a greater share of domestic chores than was deemed appropriate in previous times”. This is, of course, a classic example of the potential disparity between attitudes or beliefs and actual behavior, for while many would assert a preference for a more egalitarian family life, it may be that few people can actually bring it into

Parental Values, Beliefs, and Behavior


existence. But, whatever is the case, this raises the question of the most desirable child qualities that fit best in this smaller, more egalitarian family? There are well-known linkages between socioeconomic circumstances and reproductive strategies which have implications for child-rearing (Caldwell, 1976, 1982). Some have argued that with fewer children to care for, for example, parental investment in children is not so “diluted” as it once was (Blake, 1989). Theoretically, at least, it is now possible for children to receive greater parental investments of time, resources and attention to their development as individuals. Recent statistics on the well-being of children may suggest the opposite, with scores on standardized tests at an all-time low, with suicide rates and various forms of deviant behavior on the rise. Some have interpreted these trends in adolescent life as directly due to a reduced commitment to the parental role (Uhlenberg & Eggebeen, 1986), but it is far from clear what is going on. Hofferth and Sandberg (2001a, b), by contrast, report the opposite, an increase in time spent in structured activities, such as school, day care, sports, and art activities, along with a decrease in children’s time spent in unstructured activities, such as watching television, visiting or passive leisure. In harmony with the suggestions of family decline have come the voices of sociologists decrying the state of modern life as one devoid of communal ties, resulting from too great an emphasis on individualistic values (Bellah et al., 1985). There is little question that fundamental shifts in cultural values have undermined the authority of traditional institutions, like the church. Indeed, the Hofferth and Sandberg (2001b) time-use study indicates that while TV viewing is declining, so are such things as time spent in youth groups, household work, and church activities. This reflects I believe what Lesthaeghe and Surkyn (1988) argue were the two most salient features of Western ideational change over the past two centuries, which have been linked to processes of secularization and individualism, referring specifically to the growth in the importance of personal freedom and the rights of the individual. What is not clear is whether individualistic values have increased at the expense of family values, or whether these changes result from the emergence of new norms that are completely compatible with a newly shaped set of family values. An alternative point of view is that the family has not changed in its fundamental institutional functions, but what has changed is the sociodemographic context in which families are formed, maintained and changed, and this has brought with it an increase in the value of individualism and autonomy of family members. Thus, the changes in levels of family-linked demographic and social indicators, e.g. fewer children in families, more



opportunities for women to work outside the home, or the changing number of single parents, may affect the possibilities of what family life may be like, and under a new regime of shifting preferences, the vitality of the modern family remains. This view holds that many of the shifts in family makeup and functioning now being witnessed need not be characterized in negative terms – it would eschew such terms as “family disintegration”, “family breakdown”, and the like – but there is an issue of the “strength” of the traditional family as an institution. And in this sense the term “family decline” (Popenoe’s term) is not necessarily evaluative. Indeed, the weakening of the traditional authority of the family, itself yielding to the demands for individual autonomy and equality, may be what is occurring. Indeed, what I believe we are witnessing in the changing family and the emergence of new forms of organizing family life are some fundamental changes in our culture, in the expression of our basic values, and in some areas a clear shift in values, which are reflected in the changing features of childrearing. It appears that parents may in fact be investing greater amounts of time in children’s self development, and if there are declines in their level of investment in time spent with their children it may be more than compensated for by increases in structured activities with other adults and/or institutions responsible for their socialization. There seems to be an increased emphasis on the development of self-discipline in children through activities that stress the importance of autonomy and self-reliance. It is essential to realize that in addition to temporal changes choices about child-rearing are constrained by culture, social structure and ecological factors, whose effects are mediated through parental value orientations and beliefs about what qualities are desired in children and how to go about producing them. I argue that what is changing is linked to demographic change, but fertility shifts, rather than being an underlying cause of family change are, if anything, just one reflection of deeper changes in culture. The parental generation is motivated by the joint desires of preserving elements of the social environment and the need to adapt to prepare children for the future. I argue that parents are motivated to prepare their children for a future life and they make their child-rearing choices within the framework of the constraints and opportunities posed by history, culture and social structure. Given this motivational assumption, there is, thus, a strong reason to expect parents to adapt their values to their beliefs about the kind of qualities that will be required of their children in a future world. And thus, there is strong justification for looking at adult orientations toward children as one important indicator of social change (see Inkeles, 1955/1983).

Parental Values, Beliefs, and Behavior


2. OBJECTIVES OF THE CHAPTER Despite changes in the nature of the family and the increased role over the past century of other social institutions in the socialization of children, clearly the family is still the primary social institution responsible for the care and nurture of children in all societies. Recent claims to the contrary (e.g. Harris, 1998), I believe there is plenty of evidence that parents have considerable influence on their children, although these influences are extremely complex, and may have grown more subtle in recent years. Adults make literally thousands of choices that affect the development of their children and their well-being. Those choices are obviously constrained by culture and society and mediated by parental value orientations and beliefs. The goals of this chapter are five: (1) to clarify the concept of values and distinguish it from other concepts, such as attitudes and beliefs, and their role in child-rearing and child wellbeing; (2) to review research on parental values and related behaviors connected to the socialization of children; (3) to review research on historical changes in parental values and childrearing; (4) to consider the relationship between parental values and child-rearing behavior; and (5) to discuss several possible desiderata for future research on parental beliefs and behaviors and their relation to child outcomes. In working toward these aims the chapter hopefully offers a clarification of the central factors needed to understand the ways in which social structure and culture influence the behavior and orientations of parents with respect to raising their children and to develop a framework that will make it possible ultimately to assess their developmental impact. In the text below I argue that cultural factors help govern the choices parents make in carrying out the socialization of children. I focus primarily on the concepts of “values” and “beliefs” and their role in understanding parental behaviors, especially in making choices that affect their children’s lives, e.g. choice of activities and the use of time. Ultimately, we are concerned with what roles parental values and beliefs play in shaping their children’s environments, how they are expressed in activities parents choose for their children, how they are related to what parents teach their children, and to their approaches to childrearing. Such concepts are often referred to as “subjective” or “ideational” because they are presumed to reflect content that differs from behavior. The



present discussion takes a much more behavioral approach, defining such concepts as “cognitive representations” of culture, which are very much linked to behavior and the influence of the social environment. The emphasis here is on developing a framework for the applicability of these and related concepts to understanding parental factors that affect the well-being of children and their families.

3. CULTURAL VALUES AND PREFERENCES Behavior entails making choices between alternative courses of action. Here I am concerned specifically with parental behavior and orientations to children, but the topic can benefit from a more general discussion of behavioral processes. Naturally, behavior often follows from (or is determined by) certain stable preferences about what ends, including behavioral outcomes, are desirable, as well as about the desirable means of achieving them. Although they are often used to refer to the same things, a distinction should be drawn between “preferences” on the one hand and “values” on the other. Preferences are observed regularities in behavioral choices – sometimes these are inferred from behavior, and sometimes they are inferred from questions posed by survey researchers. Very often, preferences are expressed in terms of a relative ranking of behavioral choices or end-states of existence. As such, they reflect underlying latent dimensions, which are less likely to be observed directly, but which are so basic to human life that they often escape our attention. We can define values as cognitive representations, or beliefs, about what is desired, as well as the standards of desirability that govern choices aimed at satisfying basic physical and social needs. Consistent with this definition, Rokeach (1973) made a distinction between “terminal” and “instrumental” values. Terminal values are desired end-states of existence. Instrumental values are desirable means of achieving them. These may not necessarily correlate in any preconceived way. Parents, may value “autonomy” in their children, for example, but may differ dramatically in terms of the ways in which they believe they should go about achieving that quality and related ones. One parent may choose a permissive approach to child-rearing in order to teach her child to think for themselves, for example, while another may choose an “authoritative” approach to child-rearing to instill the same value of autonomy, and yet another might choose an “authoritarian” approach. For example, Diana Baumrind (1989, 1991a, b) contrasts parental behavior that is “authoritarian” (behavior that is demanding and directive, but not responsive, stressing obedience and respect for authority) with that which is “authoritative” (behavior that is both demanding and responsive, assertive, but not intrusive or restrictive).2

Parental Values, Beliefs, and Behavior


It is useful in this context to distinguish among several concepts, specifically between “evaluation” as a psychological mechanism, “values” as standards of desirability used in the evaluation process, and the “valuables” which have inherent “value” and which are the desired ends of behavior.3 Evaluation is the process involved in choosing among behavioral alternatives and of assessing them in terms consequences for maximizing desired outcomes. An evaluation of preferred outcomes rests on standards that help define the relation among various end-states, or a structure of preferences. For example, child autonomy is a valued end, as is obedience in children, and one could make a very long list of desired outcomes, but they may not in all likelihood be valued to the same degree and in the same way. Indeed, the unlikelihood that all valued ends can be achieved quickly and simultaneously requires some ordering of preferences and delay of gratification. As standards, value orientations help define which desired ends are most desirable, and which are least desired, and are assumed therefore to have a powerful role in governing choices among valued alternatives. Thus, as above, values are defined as the standards that underlie behavioral preferences. There is widespread use of the concepts of value and preferences by rational choice theory, which is at the basis of modern economics. This theory assumes that humans primarily act to maximize utility, or maximize benefits relative to costs, and that choices among alternatives are made on the basis of a rational calculation of costs and benefits with respect to maximizing certain desired ends. According to Becker (1996), though there may be some cognitive limitations on rationality, and other cultural and situational factors that shape behavior, in the main “individuals are assumed to make forward-looking, maximizing, and consistent choices.” Values are, thus, stable expressions of individual and collective beliefs about what ends to seek, the standards used to choose among desired end-states and means to achieve them. Even if behavior has a “rational” component, this does not mean that people are not affected by other social actors, and that norms and pressures from other people do not influence behavior. Indeed, social psychologists have known for a long time that the social situation, as well as culture and traditions, have an important role in shaping behavior (see Seeman, 1997). Regularities in social behavior reflect shared values and preferences, and embody “social norms” created as solutions to problems inherited from previous generations. Taking children to church, for example, may in some cases reflect a type of rational choice for maximizing the likelihood of one’s (or one’s child’s) salvation, but it is also likely to represent habitual behavior in response to the expectations of others. We return to a discussion of social norms and their linkage to the value concept in a subsequent discussion. First, however, we need to understand the



ways in which values are related to other concepts, as part of the cognitive organization of the individual. Values, Beliefs and Attitudes Social psychologists define values as central to other cognitive phenomena, and it is important for present purposes to clarify the relationship of values to other “ideational” or “subjective” phenomena. Beliefs, for example, are cognitive representations of “what is” – basic information that produce states of “expectancy” about the physical and social environment. The “subjects” of beliefs – what beliefs are about – include a wide variety of phenomena, including the physical world, the self, other persons, and society. Examples of belief statements are: “A child should work hard and learn as much as they can.” “A working mother can establish just as warm and secure a relationship with her child as a woman who does not work.” “Children should be seen and not heard.” All of these statements embody a common characteristic – all express information about the subject, which is taken by the actor – the “believer” – to be true. Of course, they may or may not be true, or for that matter, there may be no way of verifying the truth-value of all that is believed. The point is, that whatever is real to individuals is real in its consequences (Thomas & Znaniecki, 1927). If parents believe children are capable of learning and developing a wide variety of skills, they will likely behave in ways compatible with that belief, providing as many learning opportunities as possible. The acquisition of information about the characteristics of objects in the physical and social environment or the relation among such characteristics is thought to be formed and developed relatively early in life. As the child develops, she “learns that there are certain (things) that virtually all others believe, other (things) that are true for her even though no one else believes them, other important beliefs about which (people) differ, and other beliefs that are arbitrary matters of taste” (Rokeach, 1970, p. 11). Some beliefs are acquired through direct experience, while others are derived from authority, e.g. parents, teachers, priests and ministers, or encyclopedias. Beliefs about the desired qualities of children are undoubtedly dependent upon all or most of these factors. Whatever their source, beliefs become a relatively stable part of the individual’s cognitive organization. Of course, beliefs also change, and given their dependence on need and experience, some aspects of the belief system do not develop until adulthood, e.g. political identities and views on social issues, or beliefs about the best approaches for child-rearing.

Parental Values, Beliefs, and Behavior


Individuals hold hundreds of beliefs, which vary along a number of dimensions, e.g. the extent to which they have important consequences, the extent to which they are verifiable, the extent to which they are rooted in existential reality, or the extent to which they may be abandoned in favor of other beliefs, to name just a few. Two important aspects of beliefs are their centrality and their stability. The centrality of beliefs refers to the extent to which beliefs are linked to other beliefs. A belief whose change results in a consequent change in other beliefs is more central to the structure of beliefs than one whose change has no implications for changes in other beliefs (Rokeach, 1970; Converse, 1964). The stability of beliefs is often thought to be linked to their centrality to the overall cognitive organization of the individual, however, we are now learning that the stability of beliefs arise as much from the stability of the environment as they do the stability of cognitive dispositions themselves (Alwin, Cohen & Newcomb, 1991). The stability of beliefs, thus, becomes an important focus for research, as it can tell us a considerable amount about the dynamic stability of the inter-relation of the individual and her environment (Alwin, 1994a, 1995a). Beliefs also differ in the extent to which they are shared with others (Rokeach, 1970). Typically we use the concept of culture to refer to learned and shared beliefs and behaviors (to which we will return below). Another phenomenon that is often confused with beliefs and values, which is not central to the present development, but which deserves some discussion, is the concept of attitude. The attitude concept is often described as “the most distinctive and indispensable concept in contemporary social psychology” (Allport, 1968). Attitudes are predispositions to respond or behave in particular ways toward social objects, along a positive or negative dimension (e.g. approval vs. disapproval, approach vs. avoidance, satisfaction vs. dissatisfaction). Attitudes are often thought to have emotional, cognitive, and behavioral dimensions, all of which are “evaluative” in nature. Such evaluations are often easily manipulated and are subject to situational factors. Some researchers have concluded that there is little evidence that stable, underlying attitudes can be said to exist (e.g. Abelson, 1972; Wicker, 1969), and even those who accept the theoretical legitimacy of the attitude concept, there is considerable skepticism that the concept applies to all members of the population (Converse, 1964, 1970, 1980). While the attitude concept is an important one in its own right, in part because it is often easier to measure than other aspects of cognitive functioning, it is often viewed as epiphenomenal because it is derived from things more basic, namely values and beliefs. There are some clear differences between “attitude” on the one hand and “value” on the other. First, virtually all attitude



theorists admit that the concept of attitude is derived from other things, and therefore is “not a basic, irreducible element of personalty” (Rokeach, 1968, p. 112). Attitudes are dispositional in the same sense as we have defined values, but values are more basic than an attitude, often thought to underlie it. Rokeach (1970, p. 112) defines attitudes as “a relatively enduring organization of beliefs around an object or situation predisposing one to respond in some preferential manner.” Also, attitudes are conceived of with respect to specific objects and situations, values can be thought of as “abstract ideals, positive or negative, not tied to any specific attitude object or situation, representing a person’s beliefs about ideal modes of conduct and ideal terminal goals” (Rokeach, 1970, p. 124). Although the distinctions can often be fuzzy, there is kind of syllogistic logic that is often used to conceptualize the relation between attitudes, beliefs and values (Glenn, 1980, pp. 597–598). For example, the statement “human life should be preserved” expresses a value affirming the sanctity of human life. Another statement, “personal freedom is good” also expresses a value, asserting the importance of the abstract notion that liberty and choice for individuals is a desired end-state. These two values are often in conflict, say in the cases of abortion and assisted suicide. If one places the value of life above the right of a woman to decide whether to have an abortion, one will be unlikely to favor policies that permit legal abortions. On the other hand, if one places personal freedom above the right to life, at least in some cases, a positive attitude toward a policy of allowing legal abortions is likely. The movement from values to attitudes as a sort of deductive process is, however, problematic; mainly because it ignores the role of other beliefs. Consider the statements: “I believe abortion involves the taking of human life, but is justified in the case of rape, incest, or if the pregnancy is a danger to the mother’s health.” This is an example of a complex expression of a set of beliefs about what abortion is and when the taking of human life is justified. Values are, thus, defined as essential “standards” that govern behavioral choices. They may be viewed as a subclass of beliefs, in that values are assertions about what is “good” or “desirable”. Though they are difficult to measure, in part due to their level complexity, and due to their abstract nature, they are extremely important in shaping behavior choices and human action. Clearly, values combine with other beliefs to do this. Attitudes are somewhat less consequential, although they share some of the same cognitive elements as values. Attitudes are clearly an important facet of social life and certainly worth studying, however, we suggest that it is the underlying components of attitude – namely values and beliefs – that should also be of particular importance in the study of family and child well-being.

Parental Values, Beliefs, and Behavior


Values and other beliefs are linked to behavior and intentions. For example, to the extent that women’s labor-force participation is a matter of choice, it is reasonable to assume that values about the desirability of women’s employment is a factor influencing the labor-force behavior of women. At the same time, regardless of values, employment experiences under a regime that defines women’s labor-force participation as necessary, may well change people’s beliefs and attitudes about the desirability of women’s employment and the compatibility of work and family life (see Braun et al., 1994). More importantly, people’s own beliefs and those of their immediate family members, help determine whether work and family goals are viewed as being compatible. If people believe that employment is incompatible with “being a good mother,” then mothers who work are likely to feel considerable roleconflict and strain, promoting a negative view of the desirability of juggling family and work (Scott & Duncombe, 1992). Needs, Activities and Choices Economists typically formulate the problem of value(s) in terms of their linkage to satisfying basic material needs, but they also recognize that human behavior is motivated by the satisfaction of other needs as well (e.g. Coleman, 1990; Becker, 1996). Although much human activity is organized around the satisfaction of basic biological needs, people are also highly motivated to obtain social goods that are symbolic of value, such as social acceptance, selffulfillment, or power. Clearly, some needs, e.g. for food, sleep, sex, clothing and shelter, are more “basic” and must be satisfied before others can be pursued. Needs for acceptance by the social group, obtaining respect from others, and self-actualization, are also important basic needs, which in some instances compete for satisfaction. Maslow (1954) argued that needs exist on a dimension of importance and that their satisfaction was governed by a basic “hierarchy of needs.” This is now the accepted framework for understanding the primacy of some needs over others. According to the Maslowian framework, “physical needs” override all other basic needs when individuals experience physical deprivation. The “social-affectional” needs, or the needs for love and affection, are second, but only to physical ones. Third is “selfesteem,” or the need for dignity, which have considerable significance to the well-being of children and their families, and fourth are “self-actualization” needs, which are met only after satisfaction of those above in the hierarchy. Whatever one takes to be the relationship among the types of basic needs, most agree that needs help define the end-states that motivate human behavior, and therefore help govern behavioral choices. They play a strong role in shaping



behavior; however, it would be tautological to argue that all behavior results from an effort to satisfy needs. Clearly people “want” things that may not be in service of basic needs, or they may want more than they need. Inglehart (1977) adopted the Maslowian framework in his conceptualization of preferences for materialist vs. post-materialist political goals, and found that birth cohorts varied systematically in their relative preferences for things like security and economic stability, versus freedom and democracy (see also Inglehart (1991, 1997). Understanding human needs helps understand the nature of many human activities. Indeed, activities, or actions, are the behavioral linkage between needs and values. Put simply, activities are engaged in to meet needs. In their general theory of action, Talcott Parsons and Edward Shils (1951) along with other social theorists have long-recognized that activities represent the linkage between needs and their satisfaction. Activities, chosen on the basis of fundamental social and personal values, are organized in most cases by social institutions in a way that certain basic needs are satisfied, e.g. the organization of a religious service, a day-care setting, or a family meal. Some activities are spontaneous and unstructured, but what is of most interest here are the activities societies define as valued and valuable. We are especially interested in those activities to which social institutions devote substantial amounts of time. Bronfenbrenner (1979, p. 45) defines a molar activity as “an ongoing behavior possessing a momentum of its own and perceived as having meaning or intent by the participants in the setting.” Engagement in such “molar” activities reflects a continuing process of behavior that has a momentum of its own. As something that is engaged in to satisfy basic human needs, an activity helps create motivation and what Bronfenbrenner calls a “tension system” that “makes for persistence through time and resistance to interruption until the activity is completed . . . produced by intent (and) the desire to continue what one is doing either for its own sake or as a means to and end.” Intent is a very important aspect of this process because it “creates a motive for closure, which in turn leads to perseverance and resistance to interruption” (1979, p. 46). The stability of such behavior over specified units of time is referred to as molar stability (see Alwin, 1994a, 1995a). Activities are, thus, a crucial aspect of the social environment because they help motivate and maintain behavior, behavior aimed at meeting basic needs and desires. Activities also constitute the major ways in which institutions, such as the family, regulate the consumption of time. Although a consideration of the nature of the activities which engage human action is crucial for understanding the basis for behavioral choices, they are, of

Parental Values, Beliefs, and Behavior


course, not the only aspects of the environment. Again, turning to Bronfenbrenner’s (1979) conceptualization of the social environment, we can see it as a set of interconnected social settings, embedded in a multi-layered social and cultural context. The immediate settings in which actors are embedded – what Bronfenbrenner calls a micro-system – contains four main elements: the physical attributes of the setting, molar activities, roles and relationships. We return to a consideration of roles and relationships when we discuss social norms below, but suffice it to say that these are extremely important aspects of the environment, because roles constrain behavior and carry with them prescriptions that set expectations for behavior. Similarly, we cannot assign enough importance to the concept of “relationships” in shaping the well-being of children. In addition to the micro-system, the environment is composed of several outer layers – meso-system, exo-system, and macro-system. The mesosystem contains all micro-systems in which ego is an actor, and their interconnections. The exo-system contains all other micro-systems that affect the individual, but in which ego is not an actor. And the macro-system contains broad cultural values and norms which exist at the macro-level. One approach to studying child and family well-being would focus on molar activities, as critical elements of the micro-environment, and the time spent on them. This approach would raise questions about the effects of attributes of activities on child development. For example, Bronfenbrenner (1979, p. 46) suggests that activities vary in terms of their complexity and goal structure, among other things. A second aspect of activities important for child development is the amount of time spent on them. Choices are made on the basis of a variety and combinations of factors, but the success of environments for learning are clearly affected by the time devoted to the activities chosen to promote learning and development (Draper & Harpending, 1987). Cultural Norms Stable patterns of activities, stability of environments, and stable sets of needs and goals for the social group all lead to the stability of individuals. While psychologists have tended to use the concept of personality to understand the stability of individual differences, sociologists with an interest in culture often rely on somewhat less ‘trait-like’ concepts like values. The concept of value(s) as it is applied to the understanding of human behavior is useful in considering both the individual and cultural levels of analysis. We typically think of the cultural counterpart to individual-level values as “shared values”. We refer to such shared values as “social norms.” Norms are distinctively “social” in nature and as such are reflected in cultural or societal solutions to problems, rather



than something that exists at the level of individuals, though individuals are influenced by norms and their behavior frequently embodies those norms.4 This is not to say that individuals’ values are not acquired from the society or culture, via the influence of social norms, or that individuals do not have some role in promoting social change through the creation and development of new normative frameworks. As a concept, “social norm” has two distinct features, which may at times seem incompatible with one another, depending upon one’s reference point. One primary feature of norms is their “behavioral” component. This involves the regularity or patterning of behavior. It involves the “typical” or “modal” behavior, although there may be many competing norms in the sense conveyed by the concept of cultural pluralism. There are, for example, religious norms for Catholics, Moslems, Protestants, Jews etc. which are distinctly different, e.g. with respect to the daily practice of prayer or the blessing of food. So, what is typical of one group or subgroup in no way implies what is normative for others. The other primary feature of norms involves their “moral” elements, that is, the component of norms that carry with them an “ought” character to behavior, which makes the actor believe what s/he is doing is “right” or “correct”. Thus, social norms play a major role in the development of individual-level values. These two essential features of norms are often associated, that is, values tend to be influenced by behavioral norms. As Homans (1974, p. 250) observed, “what is has a way of fast becoming what ought to be” (italics added) (see also Heider, 1958, p. 235). The relationship between values and behavior is in flux in part because of changes going on at the macro-social level (see below). However, clearly values and norms may be out of sync. Under a regime of social change, in which social norms are changing, individuals confront new situations which require solutions that may conflict with traditional value orientations (Kohn et al., 2000). In other words, behavioral demands emerging from changes in social arrangements may tax the relevance of certain of the individual’s values and beliefs. For example, an individual brought up under a set of norms where women with young children did not work may find it difficult to accept massive levels of extra-familial child care in contemporary society. Also, it should be noted that apparently conflicting values are often held in tandem – the notion that obedience and independence are opposites at some abstract level may not square with the fact that most parents want their children to be both obedient and to show autonomy. There is a broad base of support for the view that beliefs and belief systems function as prescriptions for behavior. The empirical evidence for this view is a bit murky because of the nature of the lag between various aspects of social change, that is, changes in norms tend to anticipate changes in individual-level

Parental Values, Beliefs, and Behavior


values. Still, the most prevalent view of the link between values and behavior is that value orientations at the individual level commit the person to observe certain norms and standards in situations which allow or require her to make a choice among behavioral outcomes. One of the ways that social norms operate on the individual is through a system of roles. Roles are made up of sets of expectations for behavior associated with a position within society. Of most relevance here are those roles associated with kinship positions, e.g. father, mother, grandparent. Although in any society at a given time, there may be quite different standards for parental conduct, we would argue that at some level there are dominant values or “ideology” of child-rearing, which specify role definitions. Some of these are codified in law – e.g. the principle of child support – but the extent of conformance with legal obligations is actually quite variable. Some aspects of parental roles, then, tap into broad-based cultural values, shared by most persons living within a particular tradition. Other aspects of parental role behavior can be quite idiosyncratic and responsive, not to tradition and the expectations of others, but to situational contingencies. This broader view fits well with Alex Inkeles’s (1955) observation that the family is an important mechanism for the mediation of social change, and that parental approaches to child-rearing are as much an effort to prepare children for a life in society’s future as they are reflections of current life circumstances. One can only speculate about the motivational basis for parental behavior, as it is no doubt strongly affected by “current” conditions of life (Elder, 1974), but it is also clearly the case that in theory parents would prefer their children to survive in the future society, as well as the present one. Thus, what parents perceive the future to be potentially has a direct bearing on their approaches to the socialization of children and their perceptions of desirable qualities of children. Elder (1974, p. 13) suggests such a perspective assumes too great a degree of “future awareness, rationality, and choice in parental behavior”, and may not be relevant in all situations. He argues specifically that during times when family survival is at stake, the parental generation focuses upon the immediate needs of children, not on “future adult roles”. His research into the impact of the economic deprivation experiences connected to the Depression suggest that “the socialization environment and the response of parents to children in deprived situations during the 1930s [in America] had much less to do with their anticipation of life in the future than with the immediacy of survival requirements” (Elder, 1974: 13). However, if only from the point of view of self-preservation of the family grouping, one can assume that parents have some considerable motivation to prepare their children for the future, as well as



the present, so in a very real sense parents contribute to social change through the socialization of their children. Presumably, in addition to the conditions of life they face during their parental years, parents are also affected by their own experiences as children, as adolescents, or as young adults. Thus, their past experiences contribute greatly to their present orientations to children. Parental experiences, thus, reflect changes carried forward by previous generations, and given that with “biographic” time, the most basic of human orientations (e.g. beliefs and values) tend to stabilize over individual lifecourse trajectories (see Alwin, Cohen & Newcomb, 1991; Alwin, 1994a), we would expect that parental orientations to children will reflect something “stable” about the individual. That is to say, while we normally assume that parents are relatively adaptive and orient themselves toward preparing their children for a future life in the society, it may also to some extent be true that there may be a point at which parents are unable to adapt to change, given the crystallization of beliefs and values in their own cognitive organization and world view. Therefore in order to comprehend the nature of social change, it is also important to conceptualize the influence of time along another dimension – the biographic dimension of the parental life-cycle. It is often assumed that parental orientations are relatively persistent over the biographies of individuals, but this is a relatively unexplored dimension (Alwin, 1990b, 1996a). Do parental orientations to children remain stable, even though children are changing through time, due to developmental and historical factors? This raises a complex set of questions regarding the sources of parental child-rearing orientations. Do they originate exclusively in the parents’ own socialization experiences, and current social class position, or are they responsive to the nature and direction of their childrens’ development? To what extent is there flexibility in the preferences of parents for what they want to see embodied in their children? This raises the possibility that, although parents may maintain considerable continuity in their underlying values, the external manifestation of those values may differ across the child’s development. For example, how one socializes a four-year old for autonomy is presumably different from how one socializes an adolescent toward the embodiment of the same value. There has been extensive work on the link between life-span development and parentchild relationships, but these issues have largely been ignored (for an exception see chapters by Altmann and Rossi in Lancaster et al., 1987). There are several serious methodological difficulties that arise in verifying the truth of the above assertions – that values and beliefs play a role in transmitting the influences of social structure and culture on parental behavior. Designing research in a way that demonstrates the importance of this linkage is crucial to understanding these linkages. For example, it is difficult to make

Parental Values, Beliefs, and Behavior


the argument that values cause behavior, if one does not have independent assessments of the two within the framework of a design that permits some assessment of cause-effect relations. There are difficult measurement problems as well. We know, for example, that to the extent we rely on self-reports of beliefs, we can go wrong, in that what people say is not necessarily related to what they do. We have a basic problem – words and deeds are not always consonant (Wicker, 1969). This is, in part, because of the “culture lag”problem mentioned above, which suggests that society changes more rapidly than individuals, and that under a regime of change, parents find that their value orientations do not always fit in with new social norms. A complex research design is needed to capture patterns of stability and change at both the individual level as well as the macro-social level.

4. RESEARCH ON PARENTAL VALUES AND BELIEFS The foregoing discussion is important inasmuch as it helps to develop the background for the second major objective of this chapter, namely a clarification of the role of parental values and beliefs in child-rearing and the well-being of children. Parental values are assumed to underlie social preferences that help shape behavior with respect to children. Conceived of as standards, as we defined values above, parental values are assumed to be the criteria used as the basis for asserting what are the desirable qualities of children (Kohn, 1969), what are desirable approaches to child-rearing, and what are the best environments in which to raise children. Of course, there is a wide range of desirable child qualities, and tradeoffs or choices need to be made. There is also a wide range of desirable environments. Operationally, when adults express beliefs about desirable qualities of children, or attitudes about preferred ways of behaving to acquire or achieve these qualities, they express something about their values with respect to children. Parental values may include such broadly defined cultural goals as “education” or “knowledge,” as well as the criteria used to choose among educational options (e.g. attending a four-year college vs. a 2-year community college), and they may also include specific qualities of children (e.g. being trustworthy, loyal, helpful, friendly, courteous, kind, obedient, cheerful, brave, clean and reverent). Choosing among such educational goals, or such child qualities, entail the use of “standards of desirability” which parents use to evaluate or prefer in the behavior and attitudes of their children. Following from our earlier definitions, parental values are the standards that serve as a basis for making choices about desirable qualities of children.



Values for Obedience vs. Autonomy in Children A principle contrast of interest here, which is in line with several scholarly treatments of parental socialization values, is the contrast between autonomy vs. obedience, or put simply, the contrast between “thinking for oneself” versus “obeying” adult authority. This contrast has been written about widely. Regardless of the particular language used, it is remarkable that a number of scholars have all pinpointed the same dimensions of relevance in describing the nature of values underlying parent-child relations. Lynd and Lynd (1929: 131–152) contrasted parental values for “independence” and “obedience”. Duvall (1946) distinguished “traditional” from “nontraditional” values, aligned along this dimension. Miller and Swanson (1958: 55–58) came even closer to a dimension of sociological relevance in contrasting parental orientations that were “entrepreneurial”, which emphasize child self-control and a manipulative stance toward the environment, with those that were “bureaucratic”, which stress reliance on external forms of behavioral control and an accommodation to the environment. Lenski (1961) used the terms “intellectual autonomy” and “intellectual heteronomy” to contrast parental preferences that stressed “thinking for oneself” vs. “obedience”. Similarly, Kohn and his colleagues (Kohn, 1959, 1963, 1969, 1977, 1981; Pearlin & Kohn, 1966; Kohn, Schooler et al., 1983; Kohn & Slomczynski, 1990) popularized the contrast between “self-direction” and “conformity to external authority” to refer to this same dimension. More recently, Schaffer and his colleagues (e.g. Schaefer, 1987; Schaefer & Edgerton, 1985) have used the term “parental modernity” to refer to an orientation among parents, which stresses self-reliance, self-control and independence. Such contrasts in values are central to the dimensions along which substantial social change has occurred in American society, although the two concepts of “autonomy” and “obedience” should probably not just be viewed in terms of their contrast, but separately, especially to the extent they are changing independently. In any event, this contrast is definitely linked to the central dimension of historical social development in Western industrialized societies, which have increasingly required the exercise of independent thought and action (Stone, 1977; LeVine & White, 1986, 1987). In addition, this set of distinctions is also developmentally relevant, since it is normally believed that children progress from a relatively “obedient” to a relatively “autonomous” state as they mature cognitively, emotionally, and socially (see Piaget, 1932). This is particularly interesting, as mentioned earlier, to the extent that “developmental” trajectories and “historical” trajectories interact to produce different requirements of children at different historical

Parental Values, Beliefs, and Behavior


periods. The study of parental values over historical time is especially important for what it can reveal about the extent that “developmental” trajectories and “historical” trajectories interact to produce different requirements of children at different historical periods. Indeed, the very conception of childhood, as an extended period of dependence on adults, is a relatively modern invention, tied to particular historical circumstances. Aries’ (1962) provocative essays on the history of childhood should remind the contemporary observer of the necessity of viewing childhood, and all stages of human development, in historical perspective. Social historians have pointed to differences over time in conceptions and treatment of children in Europe and America over the past few centuries, resulting from cultural changes, as well as the evolution of technology, demography, and social organization (Elias, 1978; Greven, 1970; Schlumbohm, 1980; Stone, 1977; Vinovskis, 1987; Zelizer, 1985). A third reason for the importance of studying this dimension should be noted. The contrast between the emphasis on “autonomy” vs. “obedience” is also relevant to social stratification. Research on class differences in childrearing orientations in U.S. society, in part due to the early work of the Lynds (1929), appropriately has focused on modes of promoting child development. In the 1940s and 1950s, for example, researchers debated whether the working class was more permissive in child-rearing than the middle class (see Davis & Havighurst, 1948; Havighurst & Davis, 1955), or vice-versa (Sears, Maccoby & Levin, 1957). Later interpretations (e.g. Bronfenbrenner, 1958; Kohn, 1959a) focused explicitly on the goals of child-rearing rather than the means. Kohn (1963, 1969, 1981), for example, emphasized the concept of parental values as embodying the standards of desirability parents use in evaluating the behavior of their children, regardless of the particular child-rearing practices, that is, regardless of the means they use to achieve their child-rearing goals. In this tradition, the principle contrast in parental values used almost universally is again the contrast between “autonomy” and “obedience”. The results of a vast amount of research on class differences in child-rearing orientations have done much to confirm the observations of the Lynds (1929), namely that the middle and upper classes tend to express a greater preference for autonomy or self-direction in children, while the working classes give a relatively greater emphasis to obedience and conformity to authority and tradition (see reviews by Alwin, 1989a; Gecas, 1979; Kerckhoff, 1972; Kohn, 1977). Parental values have also historically been linked to gender, and socialization processes were believed to mold the child differently depending on gender. During the post-World War II period, with the changes in maternal labor force participation, adult sex roles have in theory been converging (Giele, 1988).



With these changes it is often hypothesized that the socialization of children has produced fewer sex differences among children and that in future generations gender differences among adults can be expected to diminish (Hoffman, 1988). If the adult roles of men and women converge, in theory at least, sex-based differences in child-rearing patterns will diminish. There have been some changes in the sexual division of labor (Gershuny & Robinson, 1988), but the change is minuscule compared to the overall differences in domestic time expenditures. Moreover, to date there have been very few differences in parental values or child-rearing behaviors linked to gender of either parents of children (Alwin 1991, 1995b), despite strong theoretical reasons to expect parental socialization to depend on the gender of the child (Block, 1978, 1983). The Measurement of Values We commonly infer the values of persons from a variety of different things – we infer values from behavior, from expressed attitudes, and from the choices of desired end-states and means of achieving them. Considerable social psychological theory has been concerned with the processes by which people infer the attributes and intentions of others, as well as themselves, by posing questions or behavioral situations and watching how people respond. In surveys research there are several possibilities, but it is difficult to pose questions about the behavior or intended behavior of parents that might give clues regarding their underlying values, and survey researchers have tended not to rely on behavioral self-reports in the measurement of values. In the case of parental values, measurement typically takes place by asking people how important it is to instill certain qualities in children, or alternatively, what qualities of children are deemed to be most desirable to prepare children for life. Because many child qualities are highly desirable, most value researchers insist on using rankings of importance rather than ratings (see review by Alwin & Krosnick, 1985). Such global measures of values, such as those developed by Lenski (1961), Kohn (1969), and others (e.g. Duvall, 1946), are all useful for charting trends in parental values. They are, however, not without limitations and potential lack of clarity, and it is important to understand the variety of possible interpretations to such survey questions. First, valued qualities of children may be seen as absolutes, in the sense that certain qualities may reflect “markers” along a dimension of social and/or moral development. Some developmental outcomes, as exemplified by certain child qualities, may be seen as important as objectives or goals of child-rearing. There may, obviously, be no consensus

Parental Values, Beliefs, and Behavior


on what is developmentally desirable, and there may be criteria other than developmental ones that govern parental choices of desirable child qualities. For example, a parent may desire obedience to family customs or to religious teachings as being most important in an absolute sense, but this may bear no explicit relationship to any objectively-defined level of moral or social development. A second way one might interpret parental valuation of child qualities is in their reflection of “unmet” desires or objectives, in contrast to values as absolutes. Thus, as statements of preferences, values may reflect the disparity between what one has in reality achieved and that which one aspires to or hopes for. For example, a parent whose child is well-behaved, orderly and generally obedient may not choose such a response in a survey, not because they do not enjoy such a childhood state, but since the child’s obedience is plentiful, the parent may take it for granted and desire other qualities, e.g. independence, not necessarily because they are more desirable in an absolute sense, but because they represent desired outcomes yet to be attained. On the other hand, a parent who may desire independence equally in an absolute sense, but whose child is not very obedient, may rank obedience higher than independence on the scale of importance, mainly because it is relatively harder to come by. Finally, parental responses to questions concerning desirable child qualities may in some instances reflect a fait accompli. Children may acquire certain qualities through exposure to the larger culture, and parents may simply accept these as realities they must accept. Thus, a parent may desire independence in children, not just because it is desirable in some absolute or relative sense, but because she has observed it in her child’s behavior and has learned to accept it. It becomes a goal of child-rearing through a process of absorption of values already embodied in children’s behavior. There is yet another problem with the interpretation of social changes in rankings of parental preferences for child qualities. Because of the fact that most research on parental values uses rankings of the importance of child qualities, it is possible only to know the relative preferences and not absolute ones (see Alwin & Krosnick, 1985). Rankings are ipsative, and thus, the sum of the scores for an individual (or the sample as a whole) is always the same. Aside from the more technical measurement problems (e.g. ratings vs. rankings for measuring values; see Alwin & Krosnick, 1985; Krosnick & Alwin, 1987, 1988), because of these several conflicting interpretations of responses reflecting parental preferences, the results of research on parental values are often ambiguous. In short, if we admit that responses to survey questions have more than a single interpretation, and we do not know in what proportion particular interpretations exist at any one time or at different times, then there may be little basis for arriving at intertemporal assessments of changes in



parental values. Or, if different cohorts respond to such survey questions in different ways, then the situation for assessing social change via the replication of surveys may be even more hopeless. If we cannot unambiguously assess change over time in measures of parental values, due to ambiguities in the meaning of responses to such questions, then conclusions may be risky. At the same time, however, despite these limitations, there is still some basis for drawing inferences about social change, even if our current level of understanding is limited due to lack of complete precision. Researchers employing such measures should proceed therefore both with anticipation regarding the revelations they may find in patterns of variation in these measures over time, as well as with warranted caution in what meanings they should place on the discovered patterns.

5. HISTORICAL CHANGES IN PARENTAL VALUES The third major objective of this chapter is to review research on the historical changes in parental values. There is considerable empirical evidence that relationships between parents and their children have changed in important ways in industrialized countries over major periods of this century. One of the best early quantitative descriptions of the family and parent-child relations in the early 20th century is Robert and Helen Lynds’ ethnographic account, embodied in Middletown (1929). Whereas the Middletown of 19th-century America had given emphasis to the importance of child-bearing, because of its link to the agrarian economy of middle America, the more urban Middletown of the 1920s, having shifted its technological base, showed evidence of a greater emphasis on child-rearing. Parental orientations to children had changed from a concentration on fitting children into society to one of providing for children in a way that would enhance their development (1929, p. 131). They observed that the traditional conception of child-rearing in the late 19th century had consisted primarily of “making children conform to the approved ways of the group”, securing the maximum of obedience from them. The Middletown of the 1920s, by contrast, was seen as more differentiated in responsibilities toward children and the family as a less potent force in securing adherence to established group sanctions. The decline of the influence of the traditional institutions of church and family and the increasing role during this period of the school in the socialization of young people in American society is confirmed by the observations of Ogburn (1922), Sorokin (1927), and Thomas and Thomas (1929). The concerns of parents were often with the fact that their children had too much independence and that their socialization through the formalized

Parental Values, Beliefs, and Behavior


social system of the school made many demands upon children for independence of action for which parents did not believe they were prepared (Lynd & Lynd, 1929, pp. 131–152). Thus, adults often wanted young people to pay greater attention to their parents and to behave in ways consistent with institutional definitions for appropriate behavior. From the perspective of social scientists, young persons were often viewed as having to adapt to rapid social change, and were particularly vulnerable given that the new social forms of regulation and preparation for life had not yet stabilized. This situation led W. I. and D. S. Thomas (1929) to comment that it was “widely felt that the demoralization of young persons, the prevalence of delinquency and crime, and profound mental disturbances are very serious problems, and that the situation is growing worse instead of better” (1929, p. xiii). Parental emphases in childrearing, thus, often stressed obedience to parents and loyalty to institutional authority, although perhaps somewhat less so than was true in the earlier era. However, according to the Lynds (1929, p. 149), this type of concern was motivated by a developmental orientation toward children. They remarked, for example, that “one cannot talk with Middletown mothers without being continually impressed by the eagerness of many to lay hold of every available resource for help in training their children” (1929, p. 149). Such an emphasis on child development did not necessarily translate into self-actualization with respect to the values expressed by the same mothers. Despite rather dramatic changes in the nature of the modern family (see e.g. Bumpass, 1990; Giddens, 1992; Goldscheider & Waite, 1991; Kobrin, 1976; Lesthaeghe, 1995; Popenoe, 1988; Thornton, 1990; Thornton & Freedman, 1983), many of the basic functions regarding socialization of children have remained intact. The family recognizes the child as an object of affection and attention from the time of its birth, and, in one way or another, the child is given a place in the family grouping and this helps shape her/his behavior. Of course, what constitutes the family grouping and the nature of the family structure in any given culture at any particular time is variable, and the existence of the parent-child relationship, lodged in the nature of the domestic environment of the family, is subject to change historically. Generally speaking, what is done with and to the child depends upon a number of factors. It depends in part on the level of physical, psychological and emotional maturity of the child, but as we suggested earlier, it also depends upon the social, cultural and historical factors. What the family defines as the “ideal” child, the associated behavioral qualities or traits expected of the child, and modes of child-rearing aimed at attaining those ideals are given meaning through the ways in which children are treated (see Ariés, 1962).



Within the past few decades research has been conducted on historical changes in parental values, and from several countries, over a variety of different time periods and using a variety of different measures, all show that greater value has been placed on independence or autonomy in children and declining importance has been attached to the obedience of children to traditional institutional authority. Alwin (1988a, b), for example, used data from the replication of the famous Middletown Studies described earlier (Lynd & Lynd, 1929; Caplow et al., 1982; Caplow & Chadwick, 1979) to show there were dramatic increases in parental valuation of independence in children and a decline in an emphasis on obedience between 1924 and 1978. These results indicate a fairly clear pattern of increasing preferences of autonomy in children and a decline in the valuation of obedience – increases in preferences for “independence” and “tolerance” in children are accompanied by contrasting levels of decline in “strict obedience” and “loyalty to church”, suggesting a substantial shift in the nature of parental socialization values over this more than 50-year period. The same conclusion was reached using data from the Detroit metropolitan area, although gathered over a much shorter period of time. Several studies of the Detroit area have revealed similar patterns of an increase in the valuation of autonomy, specifically the value of children “thinking for themselves”, and a decrease in preferences for “to obey” as a factor parents would chose as the most important quality to emphasize to prepare children for life. (Alwin, 1984, 1986; Duncan, Schuman & Duncan, 1973; Duncan, 1985). For explainable socio-historical reasons, the pattern is most evident among Catholics, and religio-ethnic category seems to interact in an interesting way with social change (see Alba, 1981; Alwin, 1984, 1986, 1988c; Lenski, 1971). In the late 1950s respondents identifying themselves as white Protestants exceeded Catholics in the extent to which they preferred autonomy over obedience, but by the 1970s these earlier differences had vanished, and by the early 1980s they had been reversed (see the results presented in Fig. 1). Recent research suggests that these changes have continued into the late 1980s (Alwin, 1996a, b, 2001). These patterns are also evident in national-level data for the U.S. Wright and Wright (1976) used two NORC-based studies – Kohn’s 1964 survey and the 1973 General Social Survey (GSS) – to try to document the nature of change in parental values. They found, over the decade of the 1960s, the percentage of fathers valuing the quality “obeys parents well” as the most important quality declined from 24% to 13% in the short space of a single decade. And the quality “good sense and sound judgement” increased in importance from 10 to 15%.” Other indicators showed the same pattern, and Wright and Wright (1976: 531) concluded that these data were indicative of a trend toward less

Parental Values, Beliefs, and Behavior

Fig. 1. Value Preferences of Religio-Ethnic Groups by Year – Detroit Metropolitan Area, 1958–1983. 121

Source: D. F. Alwin (1986)



“authoritarian or conformist values, or alternatively, an overall population increase in the value of self-direction”. Alwin (1989b, 1990a) re-examined the data used for comparison and tabulated data from later GSS surveys as well, confirming the results showed strong support for the change interpretation of the data examined by the Wrights, and that the change lasted into the 1970s. These data show a systematic historical shift in the expected direction from the mid-1960s through the late 1970s, with valuation of qualities reflecting obedience declining and those reflecting autonomy increasing. The trend may have abated into the early 1980s, as the GSS data reflect little apparent further change into the 1980s (Alwin, 1994b). The source of such a reversal may be located in differences in the experiences of the youngest cohorts, as there was some evidence that the very youngest cohorts in the GSS data showed more valuation of obedience and less emphasis on independence (see Alwin, 1989b). These patterns have also been demonstrated outside the United States, particularly in Germany (Mohler, 1989; Reuband, 1988), and Japan (Tromsdorff, 1983), testifying to the pervasiveness and far-reaching nature of these trends. Data from Germany are available since 1951 and show that, while independence and obedience were similarly endorsed in the early 1950s, there has been a significant and systematic trend in the direction of favoring independence in children, and an accompanying decline in preferences for obedience (see EMNID, 1983). It seems therefore, on the basis of the vast majority of research into this question, there is considerable evidence that a major historical trend has been occurring in the child-rearing values of parents – from values stressing obedience to those giving more emphasis to autonomy. The evidence spans many continents, many different time periods, and many different types of measures. And while one may question the methodological adequacy of any given study, a reasonable reading of the available research would suggest that important changes have occurred in parental values for autonomy and obedience, among other things.

6. PARENTAL VALUES AND CHILD-REARING BEHAVIOR The purpose of this section of the chapter is to examine the relationship between parental values and child-rearing behavior. The above-documented trends raise a number of questions about child-rearing behavior, and in particular, whether child-rearing behavior is also changing. By contrast to the above well-documented trends, we know very little about trends in actual child-

Parental Values, Beliefs, and Behavior


rearing behavior. It is very difficult to find extant survey data that provide comparable assessments of behaviors across time. There are some possible exceptions to this conclusion. There has been a dramatic increase, for example, in attention to issues of child abuse and violence directed toward children, and there is some evidence that the most extreme forms of maladaptive parental behavior may have declined (see review by Straus, 1991). National data on beliefs about the importance of a “good hard spanking” from the General Social Surveys (see Alwin, 2001) also show a significant decline between 1986 and 1998 in the prevalence of this view. Recently, surveys conducted by the National Committee for Prevention of Child Abuse between 1988 and 1992 suggest that some forms of severe punishment, namely spanking/hitting and insulting/swearing, have declined somewhat (see Doro & Gelles, 1992). However, when these studies are compared to results from the earliest National Family Violence surveys, conducted in 1975 and 1985, the results are mixed. It is clearly one thing to conceptualize dimensions of what parents desire in their children. It is quite another to connect such values to actual parenting and parental approaches to raising their children. How do parents achieve the endstates manifest in their preferences for particular child qualities? Parents clearly differ in the manner by which they go about achieving what they value, and it is difficult to sort out the nature of the causal dynamics linking values and behavior. The classic statement of the inherent implausibility of any causal interpretation was made by Kohn (1969: 146). He noted that the punishment of children by spanking illustrates how difficult it is to link a particular mode of child-rearing to any particular parental value orientation. Parents may spank children regardless of whether they value autonomy or obedience – it is the reason behind the punishment that gives clues as to the nature of the parental value, not the behavior itself. The causal arrows in this case may actually best be drawn the other way, that is, we might better think of values as being expressed by or reflected in behavior. On the other hand, some would argue that punitiveness as a child-rearing approach, e.g. spanking children, is an embodiment of certain values for obedience. How is behavior linked to values, if it is not a reflection of underlying values, or if they are not reflections of the same underlying phenomenon? It is useful in this regard to consider the approach taken by Diana Baumrind (1989, 1991a, b), who defines certain pure types of parental behavior in a four-fold classification based on the dimensions of parental demandingness and parental responsiveness. Punitiveness is one of the characteristics of authoritarian parenting styles. According to Baumrind, this classification describes how parents reconcile their approach to the joint needs of children for



nurturance and limit-setting. The definitions of the four types, taken from Baumrind’s text (1991b), are as follows: Authoritative – Parents who are both demanding and responsive. They impart clear standards for their children’s conduct. They are assertive, but not intrusive or restrictive. Their disciplinary methods are supportive rather than punitive. They want their children to be assertive as well as socially responsible, and self-regulated as well as cooperative. Authoritarian – Parents who are demanding and directive, but not responsive. They are obedience and status-oriented, expecting their orders to be obeyed without explanation. They provide an orderly environment and a clear set of regulations, monitoring their children’s activities carefully. Permissive – Nondirective parents, who are more responsive than they are demanding. They are lenient, do not require mature behavior, allow considerable self-regulation, and avoid confrontation. Rejecting-neglecting – Disengaged parents who are neither demanding, nor responsive. They do not structure and monitor their children’s behavior, and are not supportive. They may be actively rejecting or neglect their child-rearing responsibilities altogether.

Although this scheme is intended to have broad applicability to the description of parents’ behavior, Baumrind (1991b, p. 62) indicates that the “operational definitions of these four prototypes – authoritative, authoritarian, permissive, and rejecting-neglecting – differ somewhat depending on social context, developmental period, and method of assessment, but share certain essential features”. One of the main contrasts in Baumrind’s work in describing parenting behavior, which is of particular interest here is the contrast in the behavior of the authoritative vs. authoritarian parenting style. The former are more likely to instill autonomy, as an aspect of competence, in children. Her study of adolescent outcomes (Baumrind, 1991b) indicates that more than any other type, “the success of authoritative parents in protecting their adolescents from problem drug use and in generating competence should be emphasized” (p. 91). She suggests that “authoritative upbringing” consistently generates adolescent competence and deters problem behavior in both boys and girls at all developmental stages. Other researchers have reinforced this conclusion (see e.g. Steinberg, Darling, Fletcher et al., 1995). It becomes of considerable interest in this context to see whether and how these various dimensions are in fact related to parental values. We can look briefly at some relevant data on these matters by exploring data collected in the 1982–1983 Child-Rearing in Detroit Project (Alwin, 1994b). This survey included several measures of parental values – measures of values for “obedience” and “autonomy.” The data were gathered from two successive sample surveys of parents (who were raising children between the ages of 2 and 18) in the Detroit metropolitan area during 1982 and 1983. The data were

Parental Values, Beliefs, and Behavior


gathered face-to-face in an interview of approximately one hour in duration, focusing specifically on respondents’ experiences as a parent and their ideas and plans regarding their childrens’ lives. The interviews in the 1983 study were conducted by professional interviewers, employed by the Survey Research Center of the University of Michigan, and trained specifically for this study. Approximately one-half of the interviews in the 1982 study were obtained by trained graduate student interviewers, and the balance were conducted by the SRC interviewers.5 We here focus on a measure of parental values developed by Lenski (1961) in which respondents were asked to consider what things were important to emphasize in preparing children for life and rank-order the following qualities of children, as follows: “If you had to choose, which thing on this list would you pick as the most important for a child to learn to prepare him (or her) for life? (a) To Obey, (b) To be well-liked or popular, (c) To think for himself or herself, (d) To work hard, (e) To help others when they need help. Which comes next in importance? Which comes third? Which comes fourth?” With regard to operationalizing Baumrind’s classification scheme, the dimensions of parental behavior of theoretical interest are as follows: (1) the extent of restrictions and/or constraints – the degree of parental restrictiveness as shown by the setting of limits on children’s activities and (2) parental warmth – the extent to which the parent expressed support in dealing with child’s activities and problems.6 For purposes of the present analysis we divided the distribution of the measures of each of these variables at their respective medians and used their cross-classification as an approximation of the four types of parenting styles identified by Baumrind. Chart 1 presents levels of parental preferences for autonomy versus obedience by categories of Baumrind’s parenting style classification scheme. Higher numbers reflect greater valuation of “thinks for self” over “obedience.” The results in Fig. 2 show that reports of parental behavior clearly predict parental values. These results show that high levels of autonomy and low levels of obedience are associated with the combination of low demandingness and low nurturance.7 Generally speaking, and not surprisingly, parents who score high on the measure of demandingness tend to prefer autonomy over obedience, but there appears to be a small interaction with the parental warmth dimension. This is particularly true for the trait “think for self ” where there is a clear ordering of parenting styles. Those parents who are neither demanding nor supportive are the ones who emphasize “think for self ” the most, followed by those who are high on nurturance and low on demandingness. The reverse pattern is true for our measure of parental valuation of obedience. Those


Fig. 2. Proportion of Parents Favoring Autonomy over Obedience by Parenting Style – Child-Rearing in Detroit Study, 1983.

Parental Values, Beliefs, and Behavior


parents who are high on demandingness clearly prefer obedience in their children.8 These results lend themselves to more than one interpretation. Are parental child-rearing behaviors embodiments of parental values? Or are parental values merely a reflection of behavioral norms? There is certainly support for both such interpretations at a theoretical level, but there is no real empirical basis for ruling one or the other out. Researchers should adapt to interpretations of such relationships between values and behavior are inherently open with respect to causation. The important thing to take from this research is that there are predictable relationships among measures of parental behavior, as indicated by our operationalization of Baumrind’s parenting styles, and parental values, as measured by Lenski’s measure of parental values for autonomy versus obedience.

7. A PROMULGA FOR FURTHER RESEARCH Having taken stock of what we know about changes in the family and parental values, let me turn briefly to several possible desiderata for future research. This discussion focuses on what I see to be the most important challenges to future research on parental child-rearing behavior and parental influences on child development. As I suggested above, any human science needs to understand the nature of human needs, and should be able to understand social activities, or actions, as the behavioral linkage between needs and values. Humans engage in social activities to meet needs and we can safely assume that most social behavior is need-driven in some fundamental way. Sociologists have long-recognized that activities represent the linkage between needs and their satisfaction. Activities, chosen on the basis of fundamental social and personal values, are organized in most cases by social institutions in a way that certain basic needs are satisfied, e.g. the organization of a religious service, a day-care setting, or a family meal. Activities, as well as actions, embody values. We are especially interested in those activities to which social institutions devote substantial amounts of time. Some activities are spontaneous and unstructured, but what is of most interest here are the activities societies define as valued and valuable. This emphasis in future research on the activities of child-rearing is stimulated by the work of Bronfenbrenner (1979) whose ecological approach to human development offers a great deal of potential for prediction and further understanding of parental child-rearing preferences and behaviors. Bronfenbrenner (1979, p. 45) gives a central status to the concept of molar activity reflecting a “tension system” that tends to persist through time and resist



interruption. Such activities constitute the major ways in which institutions, such as the family, regulate the consumption of time and influence the social content to which children are exposed. Thus, it becomes extremely important for research into parental behavior and child-rearing to focus on the amount of time devoted to the various types of activities parents and children engage in. We need to better understand the nature of the culture(s) of child-rearing in present, future and past societies. The social worlds of children are subject to vast changes in meaning and content – even the slightest pondering of the “virtual” worlds to which children are exposed in our contemporary “internet society,” leads one to quickly conclude that previous generations of children were subject to different influences. One approach to studying child and family well-being would focus on molar activities, as critical elements of the micro-environment, and the time spent on them. This approach would raise questions about the effects of attributes of these settings and their activities on child development. For example, Bronfenbrenner (1979, p. 46) suggests that activities vary in terms of their complexity and goal structure, among other things, and as already mentioned, another aspect of activities important for child development is the amount of time spent on them. Choices are made on the basis of a variety and combinations of factors, but the success of environments for learning are clearly affected by the time devoted to the activities chosen to promote learning and development. We should make better use of the “time use” studies that exist in the archival record as a way of understanding the past (e.g. Gershuny & Robinson, 1988; Hofferth & Sandberg, 2001a), but we should also continue to collect data on how children and their parents spend their time, together and apart (see Yeung, Sandberg, Davis-Kean & Hofferth, 1999). Not only do we need more quantitative research documenting frequencies and patterns of time spent with children, we need more research of an ethnographic nature into the major portions of time children and parents are spending on various sorts of social and personal activities (see Weisner, 1996). In order to evaluate the role parents play in the socialization of their children we need to develop research designs that permit a broad assessment of parental influences. Improved research in this area can better respond to the kind of argument made by Harris (1998) that parents do not play a very important role in their children’s lives, that children are socialized primarily by other people. This may be true to some extent, but by paying better attention to the full array of environmental factors, we can better grasp the ways in which parents structure the behavior and activities of their children and ways that they do not, and as a result we can position ourselves to better understand the subtle ways in which parents exert their influences on children. Parental influences may, for

Parental Values, Beliefs, and Behavior


example, may not be as direct as it is indirect, and it is important to assess the exogenous influences as well as those operating directly. Activities are just one important piece of Bronfenbrenner’s concept of microsystem. Roles, relationships and the social norms that define and constrain them, are important aspects of the environment, and they carry with them prescriptions that set expectations for behavior. We cannot assign enough importance to the concept of “relationships” in shaping the well-being of children. This is particularly important when it is considered within the context of modern family forms in which children are often straddled between different homes and subject simultaneously to separated or divorced parents, and step parents who have different standards and values. We have to be careful to avoid the tendency to treat the concept of “parental units” as if they are embedded solely within the traditional nuclear family. Bronfenbrenner’s “ecology” of human development also takes into account the interconnections among behavior settings and the accommodation to social situations of which the actor is not a part. Recall that in addition to the microsystem, the environment is composed of several outer layers, organized like a set of Russian matrioska dolls. The meso-system contains all micro-systems in which ego is an actor, and their interconnections. The concept of the mesosystem, as the collection of all micro-systems or behavior settings that a given actor (a parent or child) is a part, provides fruitful territory for exploring the linkages among social situations and the potential conflicts or contradictions that arise in social life. Values compete with one another, and there are many instances where the value conflicts experienced by families reside at the intersection of micro-systems with one another. Bronfenbrenner (1990) has theorized that the effective functioning of child-rearing processes “requires establishing ongoing patterns of exchange of information, two-way communication, mutual accommodation, and mutual trust between the principal settings in which children and their parents live their lives” (p. 36). In other words there is a developmental payoff to the accommodation and coordination of behavior settings. This logic also extends to the exo-system, that part of the environment which contains all those behavior settings which affect a persons life but of which s/he is not a part. Again, Bronfenbrenner (1990) has theorized that the effective functioning of child-rearing processes in the family “requires public policies and practices that provide place, time, stability, status, recognition, belief systems, customs, and actions in support of child-rearing activities” on the part of the entire society, including economic, social and political institutions. Finally, we should pay better attention to the heterogeneity of values that exist at the macro-system level. This is the level at which most of the literature



on parental child-rearing has focused its attention. It is this level – the outer layer in Bronfenbrenner’s system that contains broad cultural values and norms which impinge on all environmental systems within. We should recognize that these cultural factors are in no sense homogeneous, but represent a great deal of variegation, given the pluralistic nature of our culture and the great variety there is in values for behavior. Also, we think of the influences of these factors as originating outside of behavioral settings, but in many ways the emanate from within, as reflections of behavior and choices with respect to the consumption of time and the direction of activities. Past research in this area has focused considerable attention to social stratification and inequality – which is a subject quite worthy of our attention – but the principle dimension located in parental values that is relevant to social class and its effects may not be generally applicable to all domains of parental child-rearing behavior. The literature on this subject has been dominated by the contrast between values for autonomy versus values for obedience, and while this focus has yielded vast amounts of information on the relation between this dimension and and several dimensions of social stratification, it has yielded little else. Such an emphasis may over-simplify the complexity of the social environmental factors operating at the level of the macro-system. There needs to be a renewed focus on other dimensions of parental values. For example, the results or our recent research reveal that an emphais of working hard is gaining ground in our historical analyses of parental value change in Detroit. Recent writings have also suggested that American society is increasingly characterized as alienated and uninvolved in “civic affairs”. Voluntarism and civic engagement are at an alltime low. Robert D. Putnam, a Harvard political scientist, whose work over the past two decades has focused on the scientific study of civic engagement and community life, has argued that there has been a decline in social capital – showing that Americans have become increasingly dissociated from family, friends, neighbors and voluntary associations – which he captures with the metaphor that Americans nowadays are more likely to be “bowling alone.” This kind of critique of modern American society and its tendencies toward the autonomy of the individual and the weakness of communal ties harkens back to the claims of sociologist Robert Bellah and his colleagues’ Habits of the Heart (1985), but Putnam takes a different tack. His analysis is based on considerable ethnographic materials, but the focal point of his treatment is the quantitative analysis of massive amounts of overtime survey data, which permit him to more objectively analyze social change. Part of Putnam’s (2000) explanation for this decline is that there has been a succession of generations, and that more recent cohorts are much less civically engaged than those born earlier, so in addition to stimulating the widespread discussion of issues

Parental Values, Beliefs, and Behavior


connected to civic engagement, Putnam’s work is also raising the question of the viability of generational or cohort interpretations of social change. Putnam’s work suggests that parents may decreasingly emphasize the importance of social relationships and the importance of community life, in part because their own experiences have led them to depend much more on themselves for the satisfaction of basic needs and much less on the community. A renewed focus on these additional dimensions of valuation may promote a more complete understanding of the relationship between parental values on the one hand and parental behavior on the other.

NOTES 1. Preparation of portions of this work was supported by grant funds from the National Institute of Mental Health (MH37289 and MH39761), the National Science Foundation (SES-8712119), and the National Institute on Aging (AG04743). The author acknowledges the research assistance of Merilynn Dielman, Tom Carson, Xiaohe Xu, David Klingel, Ryan McCammon, Tim Manning, and Pauline Mitchell. 2. Baumrind’s classification describes how parents reconcile the joint needs of children for nurturance and limit-setting, and is based on the fourfold classification of the dimensions of parental demandingness and parental responsiveness, dimensions she claims result from factor analyses of parents’ behavior, citing Maccoby and Martin (1983). The resulting four types are authoritarian, authoritative, permissive, and rejecting-neglecting. I return to this conceptual scheme later in the chapter (see section 6). 3. A problem arises when the term “value” is used to refer to all three of these concepts. The main confusion is between values as standards and values as end-states. The previous discussion of “preferences” and “values” is not central here, as that distinction is primarily one between observed ranked preferences and the underlying values or standards by which preferences are organized. 4. Sometimes the concept of “personal norm” is used to refer to regularities in the behavior of an individual. We prefer to use the term “value” when considering the individual, and the term “social norm” to refer to the group-level. Clearly, values and norms are linked, but one is conceptualized at the individual level and the other is conceptualized as existing at the “social” level. 5. The 1982 sample was a standard cross-sectional area probability sample, stratified on race and geographical location, obtained from a sampling frame SRC developed representing the three-county area of southeastern Michigan (Wayne, Oakland and Macomb counties) making up the Detroit metropolitan area. The 1983 sample was somewhat different, involving a disproportionate over-sampling of households in highdensity black areas. The response rates for the two surveys were approximately 80%, with resulting sample sizes of 540 and 520 in 1982 and 1983 respectively. In a few instances measures were obtained in only one of the two studies, but for most of the analyses data exist in both portions of the study. In the latter case, the data are combined and are weighted to take into account the disproportionate between-race sampling fractions.



6. Restrictions and/or Constraints were measured using parental responses to questions involving the extent to which children were restricted in the following: (a) restrictions on the amount of time the child watches television, (b) limits on the type of television programs the child can watch, (c) whether television viewing is permitted during evening meals, (d) limits on how late the child can stay up at night, (e) limits on the amount of candy, sweets, and snacks the child has, (f) the extent of control the parent tries to exercise in selecting playmates, and (g) for school aged children, the extent to which the parent controls free time after school, and (h) whether the parent sets a time for homework. Parental warmth was assessed by four questions asking the parent (a) how well they get along with the child, (b) how often they compliment the child, (c) how often the parent believes they can help the child when s/he is unhappy, and (d) how often they tell the child they love them. 7. Baumrind’s “rejecting-neglecting” parenting style is relabelled as the “disengaged” approach to child-rearing in Chart 1. This is not only more accurate, it also involves a less-evaluative descriptor. 8. Parental demandingness is the more important of the two variables in Baumrind’s scheme as measured here, as measured by the statistical significance of the differences. Differences in parental values across levels of demandingness are highly significant. Also, the interaction between demandingness and nurturance in the prediction of “think minus obey” is marginally significant.

REFERENCES Abelson, R. P. (1972). Are Attitudes Necessary? In: B. T. King & E. McGinnies (Eds), Attitudes, Conflict and Social Change (pp. 19–32). New York: Academic Press. Alba, R. (1981). The Twilight of Ethnicity among American Catholics of European Ancestry. Annals of the American Academy of Political and Social Science, 454, 86–97. Allport, G. (1968). The Historical Background of Modern Social Psychology. In: G. Lindzey & E. Aronson (Eds), The Handbook of Social Psychology (pp. 1–80, 2nd ed.). Reading MA: Addison-Wesley. Altmann, J. (1987). Life Span Aspects of Reproduction and Parental Care in Anthropoid Primates. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-span: Biosocial Dimensions (pp. 15–29). New York: Aldine de Gruyter. Alwin, D. F. (1984). Trends in Parental Socialization Values: Detroit, 1958 to 1983. American Journal of Sociology, 90, 359–382. Alwin, D. F. (1986). Religion and Parental Child-Rearing Orientations: Evidence of a CatholicProtestant Convergence. American Journal of Sociology, 92, 412–440. Alwin, D. F. (1988a). From Obedience to Autonomy: Changes in Traits Desired in Children. Public Opinion Quarterly, 52, 33–52. Alwin, D. F. (1988b). “The Times They Are A-Changin’ ”: Reflections on the Changing Culture of Child-Rearing in American Society”. Paper presented at the annual meetings of the American Sociological Association. Atlanta. Alwin, D. F. (1988c). Religion in Detroit. Unpublished paper. Institute for Social Research. University of Michigan. Ann Arbor. Alwin, D. F. (1989a). Social Stratification, Conditions of Work and Parental Socialization Values. In: N. Eisenberg, J. Reykowski & E. Staub (Eds), Social and Moral Values: Individual and Societal Perspectives (pp. 327–346). New York: Erlbaum.

Parental Values, Beliefs, and Behavior


Alwin, D. F. (1989b). Changes in Qualities Valued in Children in the United States, 1964 to 1984. Social Science Research, 18, 195–236. Alwin, D. F. (1990a). Historical Changes in Parental Orientations to Children. In: N. Mandell & S. Cahill (Eds), Sociological Studies of Child Development (pp. 65–86, Vol. 3). Greenwich, CT: JAI Press. Alwin, D. F. (1990b). Cohort Replacement and Changes in Parental Socialization Values. Journal of Marriage and the Family, 52, 347–360. Alwin, D. F. (1991). Changes in Family Roles and Gender Differences in Parental Socialization Values. In: S. Cahill (Ed.), Sociological Studies of Childhood (pp. 201–224, Vol. 4). Greenwich, CT: JAI Press. Alwin, D. F. (1994a). Aging, Personality and Social Change: The Stability of Individual Differences Over the Life-Span. In: D. L. Featherman, R. M. Lerner & M. Perlmutter (Eds), Life-Span Development and Behavior (pp. 135–185). Hillsdale, N.J.: Lawrence Erlbaum. Alwin, D. F. (1994b). Child-Rearing in Detroit Project. 1982–83 Detroit Survey Data Sets. Ann Arbor, MI: Inter-University Consortium for Political and Social Research. Alwin, D. F. (1995a). Taking Time Seriously: Studying Social Change, Social Structure, and Human Lives. In: P. Moen, G. H. Elder, Jr. & K. Lüscher (Eds), Examining Lives in Context: Perspectives on the Ecology of Human Development (pp. 211–262). Washington, D.C.: American Psychological Association. Alwin, D. F. (1995b). Parental Shaping of Gender in Child-Rearing. In: J. J. Hox, B. F. van der Meulen, J. M. A. M. Janssens, J. J. F. ter Laak & L. W. C. Tavecchio (Eds), Advances in Family Research (pp. 17–32). Amsterdam: Thesis Publishers. Alwin, D. F. (1996a). Parental Socialization in Historical Perspective. In: C. Ryff & M. M. Seltzer (Eds), The Parental Experience at Midlife (pp. 105–167). Chicago: University of Chicago Press. Alwin, D. F. (1996b). From Childbearing to Childrearing: The Link Between Declines in Fertility and Changes in the Socialization of Children. In: J. B. Casterline, R. D.Lee & K. A. Foote (Eds), Fertility in the United States: New Patterns, New Theories (pp. 176–196). New York: The Population Council. Alwin, D. F. (2001). The Disciplined Self: Transformations of Child-Rearing Over the 20th Century. Unpublished manuscript. Institute for Social Research. Ann Arbor MI. Alwin, D. F., & Krosnick, J. A. (1985). The Measurement of Values: A Comparison of Ratings and Rankings. Public Opinion Quarterly, 49, 535–552. Alwin, D. F., & Krosnick, J. A. (1991). Aging, Cohorts, and the Stability of Socio-Political Orientations Over the Life-Span. American Journal of Sociology, 97, 169–195. Alwin, D. F., Cohen, R. L., & Newcomb, T. M. (1991). Political Attitudes Over the Life-Span: The Bennington Women After Fifty Years. Madison, WI: University of Wisconsin Press. Ariés, P. (1962). Centuries of Childhood: A Social History of Family Life. New York: Knopf. Baumrind, D. (1989). Rearing Competent Children. In: W. Damon (Ed.), Child Development Today and Tomorrow (pp. 349–78). San Francisco: Jossey-Bass. Baumrind, D. (1991a). Parenting Styles and Adolescent Development. In: R. M. Lerner, A. C. Peterson & J. Brooks-Gunn (Eds), Encyclopedia of Adolescence (pp. 758–72). New York: Garland. Baumrind, D. (1991b). The Influence of Parenting Style on Adolescent Competence and Substance Use. Journal of Early Adolescence, 11, 56–95. Becker, G. S. (1981). A Treatise on the Family. Cambridge, MA: Harvard University Press. Becker, G. S. (1996). Accounting for Tastes. Cambridge, MA: Harvard University Press.



Bellah, R. et al. (1985). Habits of the Heart: Individualism and Commitment in American Life. Berkeley: University of California Press. Blake, J. (1989). Family Size and Achievement. Berkeley/London: University of California Press. Block, J. H. (1978). Another Look at Sex Differentiation in the Socialization Behaviors of Mothers and Fathers”. In: J. A. Sherman & F. L. Denmark (Eds), The Psychology of Women: Future Directions in Research (pp. 29–87). New York: Psychological Dimensions, Inc. Block, J. H. (1983). Differential Premises Arising From Differential Socialization of the Sexes: Some Conjectures. Child Development, 54, 1335–1354. Braun, M., Scott, J. L., & Alwin, D. F. (1994). Economic Necessity or Self-Actualization? Attitudes Toward Women’s Labor-Force Participation in Germany, Great Britain and the United States. European Sociological Review, 10, 29–47. Bronfenbrenner, U. (1958). Socialization and Social Class Through Time and Space. In: E. E. Maccoby, T. M. Newcomb & E. L. Hartley (Eds), Readings in Social Psychology (pp. 400– 425). New York: Holt, Rinehart and Winston. Bronfenbrenner, U. (1979). The Ecology of Human Development. Cambridge, MA: Harvard University Press. Bronfenbrenner, U. (1990). Discovering What Families Do. In: D. Blankenhorn, S. Bayme & J. B. Elshtain (Eds), Rebuilding the Nest: A New Commitment to the American Family (pp. 27– 51). Milwaukee, WI: Family Service America. Bumpass, L. L. (1990). What’s Happening to the Family? Interactions Between Demographic and Institutional Change. Demography, 27, 483–498. Caldwell, J. C. (1976). Toward a Restatement of Demographic Transition Theory. Population and Development Review, 2, 321–366. Caldwell, J. C. (1982). Theory of Fertility Decline. New York: Academic Press. Caplow, T. et al. (1982). Middletown Families: Fifty Years of Change and Continuity. Minneapolis, MN: University of Minnesota Press. Caplow, T., & Chadwick, B. A. (1979). Inequality and Life Style in Middletown, 1920–1978. Social Science Quarterly, 60, 367–386. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge, MA and London, U.K.: Belknap Press of the Harvard University Press. Converse, P. E. (1964). The nature of belief systems in the mass public. In: D. E. Apter (Ed.), Ideology and Discontent (pp. 206–61). New York: Free Press. Converse, P. E. (1970). Attitudes and Non-Attitudes: Continuation of a Dialogue. In: E. R. Tufte (Ed.), The Quantitative Analysis of Social Problems (pp. 168–189). Reading, MA: Addison-Wesley. Converse, P. E. (1980). Rejoinder to Judd and Milburn. American Sociological Review, 73, 32–49. Daro, D., & Gelles, R. J. (1992). Public Attitudes and Behaviors With Respect to Child Abuse Prevention. Journal of Interpersonal Violence, 7, 517–531. Davis, A., & Havighurst, R. J.. (1948). Social Class and Color Differences in Child-Rearing. American Sociological Review, 11, 698–710. Draper, P., & Harpending, H. (1987). Parent Investment and the Child’s Environment. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-Span: Biosocial Dimensions (pp. 207–235). New York: Aldine de Gruyter. Duncan, O. D. (1985). Generations, cohorts and conformity. In: W. M. Mason & S. E. Fienberg (Eds), Cohort Analysis in Social Research (pp. 289–321). New York: Springer. Duncan, O. D., Schuman, H., & Duncan, B. (1973). Social Change in a Metropolitan Community. New York: Russell Sage Foundation.

Parental Values, Beliefs, and Behavior


Duvall, E. M. (1946). Conceptions of Parenthood. American Journal of Sociology, 52, 193–203. Elder, G. H., Jr. (1974). Children of the Great Depression. Chicago: University of Chicago Press. Elder, G. H., Jr., Modell, J., & Parke, R. D. (1993). Studying Children in a Changing World. In: G. H. Elder, Jr., J. Model & R. D.Parke (Eds), Children in Time and Place (pp. 3–21). Cambridge: Cambridge University Press. Elias, N. (1978). The History of Manners: The Civilizing Process (Vol. 1). New York: Random House. First German Edition (1936). EMNID (1983/1986). Goals of Child-Rearing: Stagnation in the Trend Toward Non-Authoritarian Child-Rearing? EMNID Informationen, 8–9/1983; 3–4/1986. Furstenberg, F. F., Jr. (1992). Family Change and the Welfare of Children: What Do We Know and What Can We Do About It? Paper presented at the Seminar on Gender and Family Change in Industrialized Countries. Rome, Italy. January. Gecas, V. (1979). The Influence of Social Class on Socialization. In: W. R. Burr, R. Hill, F. I. Nye & I. L. Reiss (Eds), Contemporary Theories About the Family (pp. 365–404). New York: Free Press. Gershuny, J., & Robinson, J. P. (1988). Historical Change in the Household Division of Labor. Demography, 25, 537–52. Giddens, A. (1992). The Transformation of Intimacy: Sexuality, Love and Eroticism in Modern Societies. Stanford, CA: Stanford University Press. Giele, J. Z. (1988). Gender and Sex-Roles. In: N. J. Smelser (Ed.), Handbook of Sociology (pp. 291–323). Sage Publications: Beverly Hills, CA. Glenn, N. D. (1980). Values, Attitudes and Beliefs. In: O. G. Brim, Jr. & J. Kagan (Eds), Constancy and Change in Human Development (pp. 596–640). Cambridge, MA: Harvard University Press. Greeley, A. M. (1977). The American Catholic. New York: Basic Books. Goldscheider, F. K., & Waite, L. J. (1991). New Families, No Families? The Transformation of the American Home. Berkely: University of California Press. Greven, P. J., Jr. (1970). Four generations: Population, land,and family in colonial Andover, Massachusetts. Ithaca, NY: Cornell University Press. Harris, J. R. (1998). The nurture assumption: Why children turn out the way they do. New York, NY: Touchstone Books. Havighurst, R. J., & Davis, A. (1955). A Comparison of the Chicago and Harvard Studies of Social Class Differences in Child-Rearing. American Sociological Review, 20, 438–442. Heider, F. 1958. The Psychology of Interpersonal Relations. New York: John Wiley & Sons. Hernandez, D. J. (1993). America’s Children: Resources from Family, Government and the Economy. New York: Russell Sate Foundation. Hofferth, S. L., & Sandberg, J. F. (2001a). How American Children Spend their Time? Journal of Marriage and the Family, 63, xx-xx. Hofferth, S. L., & Sandberg, J. F. (2001b). Children at the Millenium: Where Have We Come From, Where are We Going? Advances in Life Course Research (pp. xx-xx). New York: Elsevier Science. Hoffman, L. W. (1977). Changes in Family Roles, Socialization, and Sex Differences. American Psychologist, 32, 644–657. Hogan, D. P. (1987). Demographic Trends in Human Fertility and Parenting Across the Life-Span. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-Span: Biosocial Dimensions (pp. 315–349). New York: Aldine de Gruyter. Homans, G. C. (1974). Social Behavior: Its Elementary Forms. New York: Harcourt Brace Jovanovich.



Inglehart, R. (1977). The Silent Revolution: Changing Values and Political Styles. Princeton, NJ: Princeton University Press. Inglehart, R. (1990). Culture Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press. Inglehart, R. (1997). Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies. Princeton, NJ: Princeton University Press. Inkeles, A.. (1955/1983). Social Change and Social Character: The Role of Parental Mediation. Journal of Social Issues, 11, 12–23; Reprinted, Journal of Social Issues, 39, 179–191. Kerckhoff, A. C. (1972). Socialization and Social Class. Englewood Cliffs, NJ: Prentice-Hall. Kobrin, F. (1976). The Primary Individual and the Family: Changes in Living Arrangements Since 1940. Journal of Marriage and the Family, 38, 233–239. Kohn, M. L. (1959). Social Class and Parental Values. American Journal of Sociology, 64, 337–366. Kohn, M. L. (1963). Social Class and Parent-Child Relationships: An Interpretation. American Journal of Sociology, 68, 471–480. Kohn, M. L. (1969). Class and Conformity: A Study in Values. Homewood, IL: Dorsey Press (2nd ed.) (1977). Chicago: University of Chicago Press. Kohn, M. L. (1976). Social Class and Parental Values: Another Confirmation of the Relationship. American Sociological Review, 41, 538–545. Kohn, M. L. (1977). Reassessment 1977. In: Class and Conformity: A Study of Values (pp. xxv-lx, 2nd ed.). Chicago: University of Chicago Press. Kohn, M. L. (1981). Personality, Occupation, and Social Stratification: A Frame of Reference. In: D. J. Treiman & R. V. Robinson (Eds), Research in Social Stratification and Mobility (pp. 267–297). Greenwich, CN: JAI Press. Kohn, M. L., & K. M. Slomczynski. (1990). Social Structure and Self-Direction: A Comparative Analysis of the United States and Poland. Cambridge, MA & Oxford, U.K.: Basil Blackwell. Kohn, M. L., & Schooler, C. et al. (1983). Work and Personality: An Inquiry into the Impact of Social Stratification. Norwood, NJ: Ablex. Kohn, M. L., Zaborowski, W., Janicka, K., Mach, B., Khmelko, V. Slomczynski, K. M., Heyman, C., & Podobnik, B. (2000). Complexity of Activities and Personality Under Conditions of Radical Social Change: A Comparative Analysis of Poland and Ukraine. Social Psychology Quarterly, 63, 187–207. Krosnick, J. A., & Alwin, D. F. (1987). An Evaluation of a Cognitive Theory of Response-Order Effects in Survey Measurement. Public Opinion Quarterly, 51, 201–219. Krosnick, J. A., & Alwin, D. F. (1988). A Test of the Form-Resistant Correlation Hypothesis: Ratings, Rankings and the Measurement of Values. Public Opinion Quarterly, 52, 526–538. Lancaster, J. B. Altmann,, J., Rossi, A. S., & Sherrod, L. R. (Eds) (1987). Parenting Across the Life-Span: Biosocial Dimensions. New York: Aldine de Gruyter. Lenski, G. (1961). The Religious Factor: A Sociological Study of Religion’s Impact on Politics, Economics and Family Life. Garden City NY: Doubleday. Lenski, G. (1971). The Religious Factor in Detroit: Revisited. American Sociological Review, 36, 48–50. Lesthaeghe, R. (1995). The Second Demographic Transition in Western Countries: An Interpretation. In: K. O. Mason & A.-M. Jensen (Eds), Gender and Family Change in Industrialized Countries (pp. 17–62). Oxford: Clarendon Press.

Parental Values, Beliefs, and Behavior


Lesthaeghe, R., & Surkyn, J. (1988). Cultural Dynamics and Economic Theories of Fertility Change. Population and Development Review, 11, 1–45. LeVine, R. A., & White, M. (1986). Human Conditions: The Cultural Basis for Educational Development. London: Routledge and Kegan Paul. LeVine, R. A., & White, M. (1987). Parenthood in Social Transformation. In J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-Span: Biosocial Dimensions (pp. 271–293). New York: Aldine de Gruyter. Lynd, R. S., & Lynd, H. M. (1929). Middletown: A Study in Contemporary American Culture. New York: Harcourt, Brace & Co. Lynd, R. S., & Lynd, H. M. (1937). Middletown in Transition: A Study of Cultural Conflicts. New York: Harcourt, Brace & Co. Maccoby, E. E., & Martin, J. A. (1983). Socialization and the Context of the Family: Parent-Child Interaction. In: E. M. Hetherington (Ed.), Socialization, Personality, and Social Development (pp. 1–101). New York: John Wiley. Maslow, A. H. (1954). Motivation and Personality (2nd ed.). New York: Harper & Row. McKenna, J. J. (1987). Parental Supplements and Surrogates Among Primates: Cross-Species and Cross-Cultural Comparisions. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-span: Biosocial Dimensions (pp. 143–184). New York: Aldine de Gruyter. Miller, D. R., & Swanson, G. E. (1958). The Changing American Parent. New York: Wiley. Mohler, P. Ph. (1989). Wertkonflict oder Wertdiffusion? Ein Vergleich von Ergebnissen aus Bevölkerungsumfragen und einer Inhaltsanalyse von Leitartikeln der FAZ [Value conflict or diffusion of elite values? A comparison of results from the polls and a content analysis of editorials from the Frankfurter Algemeine Zeitung]. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 1, 95–122. Ogburn, W. F. (1922). Social Change. New York: Viking Press. Parsons, T., & Shils, E. A. (Eds). (1951). Toward a General Theory of Action. Cambridge, MA: Harvard University Press. Pearlin, L. I., & Kohn, M. L. (1966). Social Class, Occupation, and Parental Values: A CrossNational Study. American Sociological Review, 31:466–479. Piaget, J. (1932). The Moral Judgement of the Child. New York: The Free Press. Popenoe, D. (1988). Disturbing the Nest: Family Change and Decline in Modern Societies. New York: Aldine de Gruyter. Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster. Reuband, K.-H. (1988). Von äu erer Verhaltenskonformität zu selbständig Handeln. In: H. O. Luthe & H. Meulemann (Eds), Wertwandel – Facktum oder Fiktion? Bestandsaufnahmen und Diagnosen aus kultursoziologischer Sicht (pp. 73–97). Frankfurt/New York: Campus Verlag. Rokeach, M. (1970). Beliefs, Attitudes and Values. San Francisco: Jossey-Bass. Rokeach, M. (1973). The Nature of Human Values. New York: The Free Press. Rossi, A. S. (1987). Parenthood in Transition: From Lineage to Child to Self-Orientation. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-Span: Biosocial Dimensions (pp. 31–81). New York: Aldine de Gruyter. Schaefer, E. (1987). Parental Modernity and Child Academic Competence: Toward a Theory of Individual and Societal Development. Early Child Development and Care, 27, 373–389.



Schaefer, E., & Edgerton, M. (1985). Parent and Child Correlates of Parental Modernity. In: I. E. Sigel (Ed.), Parental Belief Systems: The Psychological Consequences for Children (pp. 287–318). Hillsdale, NJ and London, U.K.: Lawrence Erlbaum Associates. Schlumbohm, J. (1980). “Traditional” Collectivity and “Modern” Individuality: Some Questions and Suggestions for the Historical Study of Socialization: The Examples of the German Lower and Upper Bourgeoisie Around 1800. Social History, 5, 71–103. Schumpter, J. (1988). Decomposition. Population and Development Review, 14, 499–506. Original published in Capitalism, Socialism and Democracy. New York: Harper & Row. Scott, J., & Duncombe, J. (1992). Gender-role attitudes in Britain and the USA. In: S. Arber & N. Gilbert (Eds), Women and Working Life: Division and Change. London: Macmillan. Scott, J., Braun, M., & Alwin, D. F. (1993). The Family Way. In: R. Jowell, L. Brook & L. Dowds (Eds), British Social Attitudes: Special International Report. Aldershot, England: Gower. Sears, R. R., Maccoby, E. E., & Levin, H. (1957). Patterns of Child-Rearing. Evanston IL: Row, Peterson & Co. Seeman, M. (1997). The Elusive Situation in Social Psychology. Social Psychology Quarterly, 60, 4–13. Sorokin, P. A. (1927). Social Mobility. New York: Harper & Row. Stone, L. (1977). The Family, Sex and Marriage in England 1500–1800. New York: Harper and Row. Straus, M. A. (1991). Discipline and Deviance: Physical Punishment of Children and Violence and Other Crime in Adulthood. Social Problems, 38, 133–152. Thornton, A. (1990). Changing Attitudes Toward Family Issues in the United States. Journal of Marriage and the Family, 51, 873–893. Thornton, A., & Freedman, D. (1983). The Changing American Family. Population Bulletin, 38(4). Washington, D.C.: Population Reference Bureau. Trommsdorff, G. (1983). Value Change in Japan. International Journal of Intercultural Relations, 7, 337–360. Thomas, W. I., & Thomas, D. S. (1929). The Child in America. New York: Alfred Knopf. Thomas, W. I., & Znaniecki, F. (1927). The Polish Peasant in Europe and America. New York: Alfred A. Knopf. Uhlenberg, P., & Eggebeen, D. (1986). The Declining Well-Being of American Adolescents. The Public Interest, 82, 25–38. Vinovskis, M. A. (1987). Historical Perspectives on the Development of the Family and ParentChild Interactions. In: J. B. Lancaster, J. Altmann, A. S. Rossi & L. R. Sherrod (Eds), Parenting Across the Life-Span: Biosocial Dimensions (pp. 295–312). New York: Aldine de Gruyter. Weisner, T. S. (1996). Why Ethnography Should Be the Most Important Method in the Study of Human Development. In: R. Jessor, A. Colby & R. A. Shweder (Eds), Ethnography and Human Development – Context and Meaning in Social Inquiry (pp. 306–324). Chicago: University of Chicago Press. Westoff, C. (1986). Perspective on Nuptiality and Fertility. In: K. Davis, M. Bernstam & R. Ricardo-Campbell (Eds), Below-Replacement Fertility in Industrialized Societies (pp. 155– 70). Cambridge: Cambridge University Press. Wicker, A. W. (1969). Attitudes versus Actions: The Relationship of Verbal and Overt Behavioral Responses to Attitude Objects. Journal of Social Issues, 25, 41–78. Wright, J. D., & Wright, S. R. (1976). Social Class and Parental Values for Children: A Partial Replication and Extension of the Kohn Thesis. American Sociological Review, 41, 527–537.

Parental Values, Beliefs, and Behavior


Yeung, W. J., Sandberg, J. F., Kavis-Kean, P. E., & Hofferth, S. L. (2001). Children’s Time with Fathers in Intact Families. Journal of Marriage and the Family, 63. Zelizer, V. A. (1985). Pricing the Priceless Child: The Changing Social Value of Children. New York: Basic Books.

This Page Intentionally Left Blank



Kristin A. Moore and Tamara G. Halle ABSTRACT Despite the desires of parents and policymakers for America’s children to achieve positive life outcomes, there is surprisingly little national data that tracks explicitly positive child behaviors and outcomes. Rather, the indicators of child well-being that researchers and the government gather about children and youth are typically limited to measures of problem behaviors and difficulties. The reduction of these negative indicators is interpreted as developing positively. In this paper, we posit that positive development is not just the absence or reduction of negative behaviors but also the presence of real and desirable characteristics and behaviors across multiple child outcome domains. To have a balanced perspective on children’s outcomes, we believe it is essential to develop valid and reliable social indicators that explicitly assess positive development. Furthermore, we offer some suggestions on measures of positive development that could be monitored over time, including: character, civility, parent-child relationships and activities, peer relationships, social capacity, religiosity/ spirituality, tolerance, extra curricular activities, sports and exercise, participation in cultural and literary activities, environmental lifestyle,

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 141–170. 2001 by Elsevier Science Ltd. ISBN: 0-7623-0776-5




volunteering and collective efficacy. Data on these outcomes can help inform public opinion and public policy, help set national and program goals regarding positive youth development, and expand the range of multivariate and longitudinal studies conducted on child well-being. Parents generally want their children to avoid drugs, violence, and jail, and they don’t want them to drop out of high school or become a teen parent. However, they also want children to be happy and emotionally healthy, to have positive relationships with other people, and to contribute to the community. Despite the desires of parents for positive outcomes for their children, there is surprisingly little focus on positive youth behaviors in the public data systems used by politicians and the media to inform public opinion and public policy. The indicators of child well-being that researchers and the government typically track about children in general and about youth in particular tend to be limited to measures of problem behaviors and difficulties. In this paper, we posit that positive development goes beyond avoiding negative behaviors, and we discuss the need to define and measure indicators of positive child well-being. We argue that it is essential to develop valid and reliable indicators of positive attitudes, beliefs and outcomes so that positive development does not continue to be construed as merely the absence of negative behaviors and outcomes. We acknowledge the dearth of theory and conceptualization in this area, and draw upon the insights of practitioners working on youth development programs to identify positive constructs. We summarize the available research on this topic, and suggest a number of constructs that could be measured and tracked as indicators of positive development. Such a system of measurement is needed if we want to monitor children’s adherence to parents’ (and society’s) positive expectations of them.

CHILD WELL-BEING: POSITIVE AND NEGATIVE ASPECTS Many government agencies were instituted in order to address specific problems. Accordingly, they study and monitor trends in negative behaviors, such as substance abuse, crime, delinquency, teen childbearing, teen unemployment, and high school dropout. Moreover, reflecting their mandate and funding, they often take a rather narrow, categorical approach to child and adolescent development. Over the years, scattered articles have appeared documenting the tendency for risk-taking behaviors to cluster. For example, youth who use marijuana are also found to be more likely to smoke, drink, and have sex than

Preventing Problems vs. Promoting the Positive


youth who do not use marijuana (Mott & Haurin, 1988). In addition, literature reviews that examine the antecedents of risk-taking behaviors have found that various types of risk-taking share common antecedents, such as poverty, early school failure, and dysfunctional families (Coley & Chase-Lansdale, 1998; Hetherington, Bridges & Insabella, 1998; McLoyd, 1998; Moore, Miller, Morrison & Glei, 1995). Gradually, researchers have widened their lens to include multiple forms of risk-taking, and government-sponsored surveys such as the Youth Risk Behavior Surveillance Survey have come to include multiple problem behaviors within a single study. However, relatively little government support exists for studies of positive development, particularly positive development that crosscuts topical domains. Although research has been done on several isolated topics (such as moral development and health-promoting behaviors), many aspects of positive development (such as positive social and emotional development) have received much less attention by researchers (Larson, 2000). Furthermore, we have been able to find only one article (and it is cowritten by the first author: Moore & Glei, 1995) that simultaneously examines multiple indicators of positive development. Even though very little attention has been given to positive development by the research community, the practice community has been moving forward on identifying positive child behaviors to measure and promote as part of their program goals (Chalk & Phillips, 1996). One reason for the interest of the practice community in positive outcomes is the discovery that, in order to engage youth, positive goals are needed. Youth tend to be uninterested in programs designed by adults to squelch their negative behaviors. Rather, youth want programs that manifest an interest in the child or youth and his or her goals and interests. Programs that build competencies, relationships, and new experiences are more successful in attracting and involving children and youth. Practitioners’ views of positive development tend to stress positive behaviors and achieving a level of competence in one or more life skills. At this moment in time, the practice field appears to be leading the way in expanding the definition of positive development (United Way of America, 1998). A number of examples of positive outcomes that have been identified in the practice community suggest directions for discussion and research. For example, the International Youth Foundation (1998) developed a definition of youth development that incorporates several desirable youth outcomes: a sense of self-worth and confidence; a sense of accountability, responsibility, and control; and competence in the areas of physical and emotional health,



intellectual development, civic action, and employment. Also, the Council on Civil Society (1998) and Public Agenda (Frakas & Johnson, 1997) both identify civility and good citizenship as important, positive attributes for all youth. Several focus groups conducted for the Task Force for Child Survival and Development have yielded additional candidates for measures of positive development. Adults and children were asked to describe what a “successful 25-year-old” would be like. Their answers included such characteristics as: having self-confidence and self-esteem; learning from their own and other’s mistakes; and possessing faith, spirituality, or maintaining some form of religious practice. In fact, religiosity was one of the most important features of a successful 25-year-old; it was overwhelmingly endorsed by all age groups (Chervin, Reed & Dawkins, 1998). Another organization, the Carnegie Council on Adolescent Development, established a life-skills training working group in order to assess the effectiveness of programs designed to train youth in life skills. Life skills training was defined as “the formal teaching of requisite skills for surviving, living with others, and succeeding in a complex society” (Hamburg, 1990, p. 3). Skills that adolescents need in order to function in modern society included problem-solving, planning and decision-making skills; strategies to help them resist peer and media pressure; coping skills to deal with daily stressors; and skills to increase self-monitoring and self-regulation (Hamburg, 1990). Despite these insights from the practice community and the children that they serve, it is not straightforward to define and track positive development over time and across groups. This is true for several reasons. First, there is little agreement among policymakers, researchers, and the general public on the definition of positive development. In particular, groups that differ in religion, race or ethnicity, education, region, gender, and age may differ in the characteristics that they value. Second, there is relatively little theory or empirical research that focuses on the presence of positive behaviors, rather than the absence of negative behaviors (Larson, 2000; Ryff & Keyes, 1995). It would be helpful to know which positive behaviors during childhood are predictive of adjustment and productivity in adulthood, but few studies have addressed this important question. In addition, there is scant communication between the practice community – which has experience conceptualizing and measuring distinct positive behaviors – and the research community, so that there is currently only a limited

Preventing Problems vs. Promoting the Positive


common knowledge base. Moreover, parents and children themselves have only rarely been consulted for their perspective on positive development. Finally, the methodological work that researchers have done on the study of negative behaviors does not necessarily inform the methodological work that needs to be done on positive development. For example, researchers who design studies that examine negative behaviors must be concerned with underreporting and denials by respondents, whereas researchers who study positive behaviors need to be concerned about over-reporting. In addition, very little methodological work has been done to clarify several somewhat nebulous constructs associated with positive development, such as character and civility. While the challenges to defining and tracking positive development are substantial, we would suggest that it is past time for this important work to begin.

BASIC RESEARCH ON POSITIVE DEVELOPMENT A recent poll of American adults conducted by Public Agenda (Frakas & Johnson, 1997) indicated that adults have stridently negative views of contemporary teens, seeing them as “rude,” “irresponsible,” and “wild.” Just over one in ten adults described teens in positive terms, such as “smart” or “helpful.” Thus, there appears to be widespread attention to and agreement about the risky, unwanted, and annoying behavior in which children and adolescents engage. However, the frequency of these negative behaviors may be overestimated. There is a need to examine the popular beliefs about youth behavior with actual data – data that include measures of both positive and negative behaviors so that a comprehensive picture of child development and well-being can take shape. Currently, there is no collective agreement in the research community on what represents the set of measures of positive development. Indeed, there is generally more agreement regarding what constitutes negative development than what constitutes positive development in children and youth (Maynard, 1997; Yamaguchi & Kandel, 1987; Moore, Morrison & Glei, 1995). However, during the past decade, work has begun within the research community to remedy this situation. Researchers who participated in a 1994 conference convened to strengthen the field of child well-being indicators (Hauser, Brown & Prosser, 1992) suggested a number of measures of positive development. For example,



Takanishi, Mortimer, and McGourthy (1997) suggested educational attainment, including post-secondary education; perception of opportunity for future adult social and economic status; health-enhancing attitudes and behaviors, including exercise and fitness; use of seat belts, dental hygiene and a healthy diet; positive mental health; responsible parenthood; and responsible citizenship, including civics knowledge, voter registration, community service, and volunteer work. For younger children, Aber and Jones (1997) developed a set of positive indicators based on mastery of stage-salient tasks faced by young children (Sroufe, 1983). For example, children aged 0 to 3 need to develop “trusting relationships with primary caregivers and a sense of basic security” while children 4 to 7 need to develop “the ability to self-regulate thoughts, behavior and emotions.” At ages 6 to 9, children need to develop “skills to negotiate conflicts and solve interpersonal problems in nonaggressive ways” (Aber & Jones, 1997, p. 399). The stage-salient approach highlights an important concern: the need for positive measures across the stages of childhood. What outcomes should be sought during early and middle childhood? Are these the same general outcomes to be sought during adolescence? If so, can comparable measures be developed for middle childhood and adolescence? In addition to these socially oriented tasks, educators, researchers, and policy makers emphasize children’s “readiness for school” as an important component of positive development. However, there is little agreement in the research community on the set of criteria that constitutes school readiness (Kagan, 1995). Most researchers, however, acknowledge that school readiness encompasses skills and abilities in multiple domains, including the domains of physical well-being and motor development, emotional development, cognitive development, language and communication development, and approaches to learning (Forgione, 1998; Kagan, 1995; National Educational Goals Panel, 1997; Phillips & Love, 1997; West, Hausken & Collins, 1993). For example, Phillips and Love (1997) propose that children’s learning style – the degree to which they are curious, attentive, and persistent when faced with new tasks or challenges – are important markers of school readiness. These skills and positive approaches to learning are fostered through children’s interactions with supportive caregivers and environments and provide a base from which the challenges of school can be met. Indeed, successfully negotiating the transition to formal schooling is a critical task for children; adapting to the social and academic demands of school in the first few grades can set children on a positive educational trajectory (Entwistle, 1995; Entwistle & Alexander, 1993). For older children,

Preventing Problems vs. Promoting the Positive


school engagement has been preferred as a key indicator of scholastic achievement, school completion, and later life successes (Connell, Spencer & Aber, 1994; Hawkins, Catalano, Kosterman, Abbott & Hill, 1999; Wellborn & Connell, 1987). In addition to the 1994 conference on indicators of child well-being (Hauser, Brown & Prosser, 1992), several reports prepared by the research community have sought to define positive development. For example, Turning Points, prepared by the Carnegie Council on Adolescent Development (1989), defines desirable outcomes for youth as being intellectually reflective; being en route to a lifetime of meaningful work; being a good citizen; being a caring and ethical individual; and being healthy. Another report of the Carnegie Council on Adolescent Develop-ment A Matter of Time (Carnegie Corporation of New York, 1992), emphasizes the importance of constructive time use for young adolescents, particularly during the nonschool hours. Other research-based work has identified the components of youth programs associated with more positive development. For example, Roth, Brooks-Gunn and Galen (1997) have identified positive youth development programs as those which counter risk factors and enhance protective factors, which have an assetbased approach rather than a remedial approach, and which stress skill and competency development rather than the prevention of specific problem behaviors. Similarly, some years ago, Gisela Konopka (1973) identified opportunities needed by youth, including experience in decision-making; interacting with peers and acquiring a sense of belonging; experiencing success, especially in school; developing new skills and competencies; providing assistance to others; and enjoying sustained relationships with caring competent adults who recognize and reward pro-social behavior. In addition, the Search Institute (Benson, 1993; Leffert, Benson, Scales, Sharma, Drake & Blythe, 1998) has developed a widely used measure of assets, including the assets of commitment to learning, positive values, and positive identity. Researchers from the Search Institute also report that self-esteem has been one of the primary outcomes assessed in youth development intervention evaluations (Scales & Leffert, 1999). Most recently, an entire issue of the American Psychologist was devoted to the growing subfield of positive psychology. Although most of the focus in this issue is on adults, some attention is given to children (see Larson, 2000). Thus, researchers have given some thought to the topic of positive development over the years. However, compared with other topics, the volume of work concerning positive development has been slight and little rigorous empirical work has accumulated.



POSITIVE DEVELOPMENT ACROSS MULTIPLE DOMAINS At the 1994 indicators conference and in subsequent papers, Moore has called for a greater emphasis on research and data collection on positive outcomes (Moore, 1997; Moore & Glei, 1995; Moore, Evans, Brooks-Gunn & Roth, forthcoming). Noting that parents do not just want children who avoid mistakes and missteps, Moore and Glei (1995) used data from the National Survey of Children to examine positive development empirically. They developed a Positive Well-Being Index which includes six component measures from different domains of life experience: positive parent-child relationships, life satisfaction, the absence of depression, religiosity, placing importance on correcting social and economic inequalities, and community involvement. This paper was titled “Taking the Plunge” acknowledging the authors’ awareness that positing certain behaviors or characteristics to be desirable is a value-laden and risky, albeit essential, endeavor. Although selection of the variables for the Positive Well-Being Index was constrained by the realities of data availability as well as theoretical considerations, one criterion was that “the capacity of a youth to achieve success, as measured by this index, is not precluded by coming from a family with limited financial resources” (Moore & Glei, 1995, p. 20). This decision does not contradict the widespread agreement in America regarding the importance of socioeconomic attainment in the definition of success. Rather, it reflects an attempt to broaden the definition of success to include new constructs. Analyses revealed that youth who had close parent-child relationships as young children later had higher scores on the Positive Well-Being Index, as did youth with better-educated parents and youth who had experienced fewer marital transitions while growing up. In addition, the researchers found that youth who might otherwise be thought of as “at-risk” still had high scores on components of the positive well-being scale. Specifically, youth from lower-income families and black youth scored higher on the Index because of their greater concern with correcting social inequalities and their greater religiosity, while children with lower scores on a standardized test of verbal expression in grade school scored higher on the religiosity component of the scale. These analyses illustrate that some groups not generally seen as successful may in fact be successful on some developmental dimensions. The study also demonstrates that it is possible to create a composite measure of positive development that incorporates multiple aspects of positive behavior.

Preventing Problems vs. Promoting the Positive


We have not as yet been able to locate other empirical work that examines multiple aspects of positive development, such as that using the Positive WellBeing Index (Moore & Glei, 1995). Apart from the January 2000 issue of American Psychologist, there is a dearth of empirical research on positive development, broadly defined, particularly empirical work based on nationally representative samples. There does not appear to be a broad empirical social science knowledge base on positive outcomes among children or youth. The studies that exist appear to reflect narrow (albeit important) studies within disciplines, or within sub-disciplines, on one particular aspect of positive development. For example, educational attainment has received substantial research attention, as has health promotion (Millstein, Petersen & Nightingale, 1993). With the support of the Carnegie Foundation, an entire book was devoted to Promoting the Health of Adolescents, including socioemotional as well as physical health. Although, where possible, chapter authors focused on healthpromoting lifestyles (e.g. Elliott, 1993), much of the research on which the authors drew derives from “silo” research traditions, in which a deep but narrow knowledge accumulates about a particular topic, such as diet or exercise (Sallis, 1993). Moral development is another very positive and rich, yet rather narrow, research area in which considerable excellent work has been done (Piaget, 1932; Kohlberg, 1975, 1981; Gilligan, 1977, 1982; Damon & Gregory, 1997). However, many of these studies use local, non-representative samples, and data generally are not collected over time. Another source that can inform the discussion of positive development across domains is the fascinating and important work accumulating on positive outcomes among at-risk youth who are often referred to as “resilient” (Garmezy, 1985; Rutter, 1987; Werner, 1989; Winfield, 1991). One notable feature of the research on resiliency is its identification of the existence of protective factors across multiple domains. Another contribution of this research is the acknowledgment that individual risk factors do not necessarily lead to poor outcomes for children. Resilience researchers find that multiple factors distinguish resilient children from those who are not able to succeed in the face of adversity. These protective factors include individual attributes identifiable even among infants (e.g. being more active and affectionate), family characteristics (e.g. from smaller families), family relationships (e.g. having a close bond with at least one caregiver), and non-family relationships that provide emotional support (Zimmerman & Arunkumar, 1994). Family processes such as parent-child relationships and parenting behaviors such as monitoring, communication,



cognitive stimulation, and discipline are of particular interest since they are not fixed but can be affected by education or participation in a program. Moreover, resiliency research also provides some important caveats. One is that resilient children are not usually “invulnerable,” that is, somehow immune to all problems (Zimmerman & Arunkumar, 1994). Indeed, Luthar and Zigler (1991) studied a sample of urban ninth graders and found that those seen as academically resilient often showed evidence of emotional maladjustment. This may reflect adjustment in one sphere at the expense of another sphere, they suggest, or it could even be a reaction to peer rejection because of academic success. For our purposes, it signals that positive development in one domain cannot be presumed to imply positive development in all domains. Indeed, high achievement across all domains seems unrealistic even for advantaged children much less disadvantaged children. However, if success in multiple and non-traditional domains is allowed to “count” as success, the successes of children growing up under adverse circumstances may be more apparent. The resilience literature also provides experience in the definition and use of protective factors in analyses. Protective factors are often seen as the opposite end of a single continuum with risk factors, yet they tend to be treated as distinct constructs that add to or interact with risk factors to affect children’s outcomes (Zimmerman & Arunkumar, 1994; Moore et al., 1995). Further work is needed to explore critically the notion that risk and protective factors fall on a continuum. In addition, the research community needs to further articulate the factors that foster positive development and that contribute to resilience among at-risk populations. It is our belief that positive development is more than just the opposite of negative development, but this assumption needs to be tested.

POSITIVE INDICATORS OF CHILD WELL-BEING We have noted the lack of theory and data concerning positive development, broadly defined, in the fields of psychology, sociology, and the general social sciences. The social indicators field is another research arena in which positive development has been relatively underdeveloped. Social indicators are measures of well-being that are collected on a regular basis from a representative sample of the population, so that trends can be tracked over time and within subgroups (Moore, 1997). Social indicators have been used for many years to describe and monitor the state of our society. For example, social indicators are used to monitor fluctuations in population growth and infant mortality rates. They are also used to set standards and

Preventing Problems vs. Promoting the Positive


benchmarks, and to hold people or agencies accountable for improving the social well-being of individuals and communities (Brown & Corbett, 1997). Two recent publications, Trends in the Well-Being of America’s Children and Youth (U.S. Department of Health and Human Services, 1999) and America’s Children: Key National Indicators of Well-Being (Federal Interagency Forum on Child and Family Statistics, 2000), have become especially important in documenting children’s positive development. These annual publications compile information from multiple government statistical agencies on the condition of our nation’s children. They include information on children’s economic security, health and safety, educational achievement and cognitive attainment, and their social and emotional development. These categories represent the broad domains within which child well-being indicators are generally tracked (Child Trends, 1999). But, a careful examination of these publications reveals that many indicators of child well-being are actually monitoring negative rather than positive aspects of children’s development. The number of negative indicators (e.g. the percentage of students who have used illicit drugs in the last 30 days) far outweighs the number of positive indicators (e.g. the percentage of high school seniors in the U.S. who rate making a contribution to society as “extremely important”). However, if we only monitor trends in negative behaviors, discussions about children will continue to have a negative tone. The only good news will be a decline in bad outcomes. In fact, the need for more positive measures is noted in indicators reports themselves (Federal Interagency Forum on Child and Family Statistics, 1998, p. 40). However, those who produce these reports have encountered numerous obstacles in providing measures of positive development. Reflecting the lack of consensus regarding what constitutes positive development, available data are quite thin. This gap is particularly apparent in the socioemotional domain, where few positive indicators can be found. Few national databases currently contain a broad set of measures of positive development. Measures are, for the most part, scattered throughout several of the newer data sets (e.g. The National Longitudinal Study of Adolescent Health, referred to as Add Health; the Survey of Program Dynamics; the Early Childhood Longitudinal Study-Kindergarten Cohort; the Child Development Supplement to the Panel Study of Income Dynamics; and the National Longitudinal Survey of Youth, 1997 Cohort). Because many of the data sources that include measures of positive development are relatively new, the long-term tracking of positive development is just beginning. We would argue that it is essential to develop valid and reliable social indicators that assess positive development. If valid and reliable measures



cannot be produced, then conversations about children and families will continue to have an unbalanced focus on negative behavior, and positive development will continue to be construed as the mere absence of negative behaviors and outcomes.

AN EXPANDED APPROACH Indeed, it will be hard to move beyond goals comprised of the absence of problem behaviors to goals that also reflect the presence of desired behaviors unless new measures of positive development can be identified. The first task is to identify constructs of positive development. Below, we make a number of very concrete suggestions regarding the identification and measurement of positive development. One important caveat is that the set of positive constructs we highlight here is by no means an exhaustive list. While the constructs do represent a wide array of individual characteristics, positive behaviors and attitudes, they primarily capture elements of social development and do not address as directly the physical or cognitive domains of development as fully. This was by design. Given the importance of socioemotional development to child outcomes (National Research Council and Institute of Medicine, 2000), and the current dearth of socioemotional indicators in our national data sets, many of our suggested constructs fall into the domain of social and emotional outcomes. Obviously, positive development across all domains is important.

CHARACTER While an intuitive sense of what constitutes “character” is widespread and many people feel they can recognize a person of character, it is difficult to measure this construct in a concrete way. It encompasses the notions of responsibility, truthfulness, good values, and steadfast adherence to one’s principles. Character is not anticipated among young children; but during the preadolescent and adolescent years, enduring values develop and begin to be manifested. Public concern is high over a perceived loss of important values in the process of child rearing and over inadequate values among children. For example, a survey sponsored by the National Commission on Children (1991) found that adults, both parents and non-parents, expressed strong concern that important values were receiving too little emphasis among contemporary families. Thus, it would appear that the constellation of characteristics

Preventing Problems vs. Promoting the Positive


incorporated into the notion of character, and the absence of character, are very salient to the public.

CIVILITY An aspect of public and even private life that has become a very salient issue in recent years might be described as civility (Frakas & Johnson, 1997; Eberly, 1998). Obviously, having “good manners” is a reflection of civility, but the construct as intended here goes well beyond saying “please” and “thank you.” and has little to do with knowing which fork to use at a formal dinner party. Civility is manifested in the treatment of others with respect. Respect can be shown in a variety of ways. For example, respect can be shown by taking turns, by sharing, by not speaking harshly or rudely to other people, by listening to others who are speaking, and by not pushing or shoving or hurting others. Unfortunately, a recent survey conducted by Public Agenda (Frakas & Johnson, 1997) found only about 12% of adults think it is very common for teens to treat other people with respect. Although standards for civility become somewhat more demanding as children become older, the elements of civility seem more independent of age than is true for some other positive behaviors.

PARENT-CHILD RELATIONSHIPS AND ACTIVITIES Socializing children to be productive and compassionate citizens is not an easy endeavor. However, this task is one that parents and other socializing agents face every day. The social dynamics within a family are the first that children encounter, and are therefore an important model for how to behave in the larger society (Halle & Shatz, 1994). The messages that parents convey to their children through their actions and words are the basis for children’s understanding of appropriate social rights and responsibilities in the greater society. Indeed, patterns of parent-child interaction and activity influence the cognitive and social development of children from the first days of life onwards. For example, in infancy and early childhood, the amount of verbal interaction between parent and child spurs language development and reading ability, and can have long-ranging influence on academic skills such as mathematics in the school years (e.g. Bradley & Caldwell, 1980, 1984; Bradley et al., 1989). The emotional quality of the relationship between parent and child in the first years of life, often referred to as the quality of the “attachment” between parent and child (see Ainsworth, Blehar, Waters & Wall, 1978; Bretherton, 1985), also has immediate as well as long-term effects on a child’s



cognitive, social, and emotional functioning. When there is a “secure” attachment between infant and parent, the child is provided the necessary emotional security to explore his/her environment freely – both in the present moment and in the future (Cassidy, 1986). Secure attachments between mother and child in the first few years of life have been found to be associated with persistence at challenging activities (Frankel & Bates, 1990; Matas, Arend & Sroufe, 1978), social competence with peers in the preschool years (Suess, Grossman & Sroufe, 1992), and self-reliance and greater problem-solving ability in kindergarten (Arend, Gove & Srouf, 1979). Much has been written about the increasing influence of peers during the adolescent years (Brown, 1990). However, most adolescents retain strong bonds with their parents (National Commission on Children, 1991), and many continue to spend time with their parents. In general, having a positive relationship with parents constitutes an important indicator of positive youth development. Youth who disconnect from parental influence are at particular risk for delinquent activities and psychological problems (Resnick et al., 1997). Less clear are the implications of being close to one parent but not the other; but studies that have included assessments of both infant-mother and infantfather attachments have found that children who were securely attached to both parents in infancy are more sociable, and more socially competent than children who were securely attached to only one parent (Parke, 1996). Hence, as a measure of positive development, it is proposed that children be close to both of their biological parents. Parental report would not be adequate to assess this construct; it is essential that the child directly report on his or her feelings. While positive affect is clearly an important element of positive development, an additional and more behavioral indicator also seems appropriate. Specifically, do the parents and children spend time together and do things together? It may not matter what the activities are. Some families are oriented toward music, while others prefer sports, and others like to shop. The important aspect is that the child is positively engaged in the activity with a parent. Just sitting and talking ought to count. In fact, talking with a nonresident parent on the telephone or communicating regularly via e-mail might be included in a listing of parent-child activities, to equalize the opportunity for interaction across family types. In addition, research indicates that the quality of parent-child interactions may be just as important to children’s optimal development as the frequency of interactions (Pleck, 1997). That is, the positive, emotional content of shared activities is an important component of children’s positive development.

Preventing Problems vs. Promoting the Positive


A more difficult question is whether parallel but unengaged activities should be counted as a “parent-child activity.” Should time spent watching television together be considered as a parent-child activity? Should parental attendance at a child’s school performance or soccer game count as doing something together? In the absence of evidence that sitting together at a symphony is more beneficial than watching the child perform at a school concert, it is suggested that all types of activities be counted. However, it is suggested that researchers investigate whether watching television together with a child has positive developmental implications for that child, comparable to reading together, playing games together, or going for a hike together.

SIBLING RELATIONSHIPS Interactions with siblings are another important component of child development (Dunn, 1988; Dunn & Kendrick, 1982; Volling, forthcoming). Interactions with siblings influence the course of a child’s social and moral development, including the development of good citizenship and good character. For example, Dunn and Munn (1986) found that friendly behavior directed toward a younger sibling by an older sibling was associated with the younger sibling’s development of relatively mature behavior in both conflictual and cooperative situations. Also, the young sibling had a similar positive effect on the older sibling’s social behavior. Indeed, positive sibling relationships early in life are associated with higher quality social skills with peers (Mendelson, Aboud & Lanthier, 1994). Strong sibling relationships are also a source of fun and satisfaction, and represent a source of social support over the life course. As with parent-child relationships, both the amount of interaction and also the quality of the interpersonal relationship are important.

PEER RELATIONSHIPS The rules and norms of social interaction first encountered in the home environment through interactions with parents and siblings are further refined through social interactions with peers. Although many adults have a negative impression of the influence of peers on adolescents, the implications of peer relationships are often more positive than negative (Bearman & Bruckner, 1999). At any age, relationships between peers provide opportunities to hone social skills and prosocial behavior. In addition, peer relationships can provide cognitive, social, and physical stimulation through joint activities and



conversations. Indeed, developing cooperative and mutually beneficial relationships with peers is considered essential for successful negotiation of life (Borenstein, 1996). Children who have poor peer relationships are at risk for later life difficulties, especially school dropout and criminal behavior (Parker & Asher, 1987). Friendships in particular can provide emotional security and intimacy, and often serve as an additional source of support outside of the family, especially in times of crisis. Good friendships are one of life’s pleasures at any age, and they can provide a protective buffer against mental health problems and destructive behaviors.

SOCIAL CAPACITY Throughout history and across the world, people have lived in social groups such as families and communities. One characteristic of a successful human being is his or her ability to live and work peacefully and productively with others in these groups. Social capacity is the ability to interact positively within intimate and family relationships and also “the ability to demonstrate positive concern and caring in a larger social arena” (Moore, Evans, Brooks-Gunn & Roth, forthcoming). Social relationships form first with family members. Accordingly, attachment to parents, siblings, and other caregivers is a marker of a developing social capacity in very young children. However, even preschool children typically interact with non-family members. In all of their interactions, children can potentially gain experience in cooperation, sharing and empathy, and they can come to experience pleasure from the happiness of others. Social capacity should not be confused with popularity. A person high in social capacity may have many friends or only a few. Social capacity is, instead, the ability to feel concern and caring for a range of people, including but not limited to the immediate family. However, it includes an active element as well as empathy and caring. In terms of measurement, it seems that other constructs, such as volunteering and having positive relationships with family members and peers, overlap with the broad, underlying notion of social capacity. Where social capacity differs is in combining the cognitive and behavioral components. Thus, a person may have empathy and character, but may or may not actively live out these virtues. Alternatively, a person may be a social activist for reasons that have nothing to do with values or caring. Social capacity takes account of both action and the motivation for action.

Preventing Problems vs. Promoting the Positive


RELIGIOSITY/SPIRITUALITY Despite the legal separation of church and state, it might nevertheless be argued that involvement in some type of religious or spiritual activity would index positive youth development. This is indeed what many youth development workers have found in their work with adolescents (Chervin, Reed & Dawkins, 1998; Scales & Leffert, 1998). Attendance at services and other activities sponsored by religious organizations has been found to be correlated with better health and less risk-taking (Bearman & Bruckner, 1999; Moore, Manlove, Glei & Morrison, 1998). We assume that, beyond attendance, measures of belief and commitment also reflect positive development (e.g. frequency of prayer, reading, and meditation). As discussed below, an important caveat in a secular society might be an accompanying acceptance and tolerance of others who follow a different faith or tradition.

TOLERANCE Freedom of religion is guaranteed by the Constitution, and discrimination is proscribed by numerous laws. Yet, the successful integration of diverse groups into national life requires that the public absorb and live out these laws in their daily lives. Despite conspicuous exceptions, it would appear that tolerance for religious, racial, ethnic and other differences has increased in recent decades. For example, responses to a 1994 survey of American high school seniors indicate increased acceptance of racial integration (Bachman, Johnston & O’Malley, 1997). Tolerance is an attribute that encompasses respectful attitudes and nonviolent, nondiscriminatory behaviors toward others who differ from oneself. Tolerance does not imply, of course, giving up personal values or goals. Neither does it imply endorsing other people’s values and goals if they conflict with one’s own. Rather, what we mean by tolerance is consciously refraining from interfering with the rights of others who differ from oneself in terms of religious beliefs, demographic background, sexual orientation, values, and goals. Tolerance may also lead to respectful dialogue, or to participation in activities that celebrate differences. These latter behaviors, we suggest, reflect not only tolerance but also a certain level of true respect, inclusion and acceptance. An important reason for including tolerance in a listing of desirable characteristics is that tolerance complements and enhances other positive characteristics that are important for children and youth. For example, having



a strong sense of one’s own religion while also respecting other people’s contrasting religious beliefs and practices strengthens the value and meaning of religiosity and spirituality.

EXTRACURRICULAR ACTIVITIES In general, researchers have found that involvement in extracurricular activities is associated with positive development. For example, recent research has shown that involvement in at least one school club decreases the chances of a youth’s engagement in risk-taking behaviors (Entwisle, 1990), reduces the probability of a non-marital teen births (Moore, Manlove, Glei & Morrison, 1998) and increases the chances of high school completion for teens who are at risk for school dropout (Mahoney & Cairns, 1997). Noting these kinds of associations, efforts to provide youth development opportunities for disadvantaged and at-risk youth have been widely mounted in recent years. However, such approaches have a very long history, with organizations like Scouts, YMCA and YWCA having been active for many decades. Because there is no reason to assume that some types of activities are inherently superior to others, we posit that participation in any array of clubs, teams, and organizations is positive. However, marginal membership and inconsistent attendance are not expected to be meaningful. Thus, the level of participation needs to be considered, as does whether or not a child holds a leadership position. On the other hand, it is not assumed that more involvement is uniformly and linearly a good thing. Zill, Nord, and Loomis (1995) found that involvement in extracurricular activities of up to 5–19 hours per week was associated with less risky behavior, but that involvement of 20 hours or more a week was not as great a deterrent. Thus, regular, moderate involvement in an activity, perhaps in a leadership role or with real responsibilities such as Scouts have, at any level, is suggested as an indicator of positive development.

SPORTS AND EXERCISE The health advantages of exercise are widely known, yet many children and youth do not exercise frequently (National Center for Health Statistics, May 1996). Although participation in varsity athletics has been found to be associated with a higher incidence of binge drinking and, for males only, with a higher probability of teen parenthood, in general athletic participation is related to more positive outcomes (Zill et al., 1995). While many if not most children and adolescents will not participate in formal or varsity athletics, all kinds of regular athletic participation can

Preventing Problems vs. Promoting the Positive


contribute to positive development. At present, it is not clear what aspects of exercise and athletic participation are positive for social (as distinct from physical) development. Researchers need to examine whether and how often children engage in a variety of physical activities and whether activities are solitary or done with others, whether they are informal or done as part of an organized team, and whether adults are involved as coaches or supervisors. The amount of time expended in these activities over the course of a year and whether there are times of complete inactivity might also be explored.

PARTICIPATION IN CULTURAL AND LITERARY ACTIVITIES Information is obtained fairly regularly in our national data systems about the frequency with which children and adolescents read (other than for school), but relatively little information is available about other kinds of cultural participation in which children and adolescents might be engaged (see Zill & Winglee, 1989, for a review). How many children have attended a symphony concert, a dance performance, a play, or an opera during the past year? If children or adolescents have seen such a performance in the past year, how many times have they gone? While rock concerts and musicals may need to be considered separately, attendance at any kind of live performance may represent a significant cultural experience for children and adolescents whose primary exposure is television and movies. Even performances by other children and youth may be significant cultural events, and such experiences may be the only opportunities to see a live performance for children in rural areas and small towns. In addition, while it is useful to know that children and adolescents are reading anything at all, it would be even more useful to know what kinds of literature they are reading. Are they reading comic books, or books written for younger children? Or are they reading nonfiction and biographies as well as fiction, or books written for adults with demanding vocabularies and complex plots? In addition to being cultural consumers, children may be involved in performing. They may sing in a choir, act in a play, write poetry, or play a musical instrument. Whether they take lessons or are guided by an adult will vary across activities; accordingly, it may be most appropriate to identify whether a child engages in any cultural pursuits at all, and if so, how many hours a child invests in such activities over the course of a year.



Because access to opportunities varies with economic status and urban-rural residence, it would be appropriate to control for such differences in assessing how often children and adolescents participate in cultural activities.

ENVIRONMENTAL LIFESTYLE While agreement about what constitutes an environmental life style would be difficult to attain, certain basic behaviors that reflect a concern for the environment can undoubtedly be identified and accepted as indicators of positive development. Environmentally conscious behaviors relevant to a broad spectrum of the population would include recycling, cleaning up litter, and energy conservation. This construct might be assessed by measuring any number of concrete behaviors. Specifically, children might be asked whether and how often they personally recycle cans, bottles, and paper. They might also be asked whether and how often they litter and how often they clean up litter. Although these questions have a social desirability element, items might be developed that overcome this bias as much as possible. Youth might also be asked whether they ever engage in a series of activities, such as car pooling, taking public transportation, walking, biking, or reducing the heat or air conditioning, explicitly in order to save energy. Currently, such information is not available through national-level surveys of children.

VOLUNTEER COMMUNITY INVOLVEMENT As important as it is that children and youth be well adjusted and happy, it is also important that children ultimately become contributors to the communities in which they live. Serving one’s community – be it your neighborhood or your country – helps to foster a sense of compassion and responsibility for others. Children and youth learn the principles of tolerance, cooperation, and interdependence through participation in community organizations (Flanagan et al., in press). Perhaps most importantly, the exposure to community service engenders a sense of belonging and “connectedness” to the larger community (Flanagan & Gills, 1999). This is particularly important during adolescence, because it is a critical time for identity development (Youniss, McLellan, Su & Yates, 1999; Youniss & Yates, 1997). Nevertheless, even elementary age children can participate in varied kinds of community activities, such as food drives, visiting people in nursing homes, or cleaning up trash from a playground. The benefits of community service are far-reaching, not only for the child him/herself, but also for the community. In particular, young people

Preventing Problems vs. Promoting the Positive


bring energy and creativity to citizen service activities; this energy can be inspiring for all involved. Research shows that participation in civic activities such as student government or civil rights movements during one’s youth predicts to a social orientation in both thought and deed 10 to 25 years later (DeMartini, 1983; Fendrich, 1993; Hanks & Eckland, 1978). For example, membership in 4-H and other organizations during adolescence has been found to predict membership and leadership in community organizations in adulthood (Ladewig & Thomas, 1987). Recent research has found that participation in school- or adult-based activities predicts positively to adolescents’ political and religious involvement, and predicts negatively to adolescents’ substance use (Eccles & Barber, 1999; Youniss, Yates & Su, 1997; Youniss et al., 1999). In sum, civic activity during youth is positively related to concurrent positive behaviors, as well as predictive of long-term civic involvement. Children and youth may become involved in community activities through school, religious organizations, groups such as Scouts, or with their own families. Community involvement can occur regularly throughout the year; it can be an intense but relatively brief day or week of volunteering, or it can be varied and occasional. Moreover, community involvement may be engaged in alone or in groups. Community involvement can take many forms. It may be service to persons, to families, to organizations, to animals, or to the environment (such as parks). It may involve helping directly (e.g. tutoring) or indirectly (e.g. donating food). Recent research comparing voluntary and mandated service found that adolescents who volunteered their services were more likely than those who did mandatory service to be involved in more direct forms of service, such as tutoring, and less likely to be involved in functionary forms of service, such as reshelving books (Youniss, 1999). It may therefore be important to distinguish between service done as a requirement (e.g. for high school graduation) and volunteer work that is completely uncompensated. Because of the very great variety of activities that might be undertaken for the benefit of the community, and the varied ways in which service can be provided, the construct should be defined and measured broadly. The research literature may be able to guide the grouping together of different types of involvement that are more predictive of community involvement among adults or more strongly associated with positive youth outcomes than are other types of involvement. However, the amount of involvement rather than the type may prove to be especially important. While a distinction should be made between children and youth who have not engaged in volunteer community involvement and youth who have, the



amount of involvement represents an additional significant distinction. Ideally, it would be useful to obtain an estimate of the number of hours devoted to volunteer activities annually. Thus, regular commitments of several hours a week could add up to be equivalent to the hours invested by youth who give a week or more to a particular project. Also, youth whose activities occur in a particular season would not be overlooked, while youth currently engaged in volunteer activities receive full credit, if an annual accounting of hours can be obtained.

COLLECTIVE EFFICACY Some very important conceptual and empirical work done by Robert Sampson, Stephen Raudenbush, and Felton Earls (1997) explores the meaning and implications of neighborhood relationships. They define “collective efficacy” as “social cohesion among neighbors combined with their willingness to intervene on behalf of the common good” (Sampson et al., 1997, p. 918). To access this construct, survey questions were designed to ask adults how likely it would be for someone to intervene if a child skipped school, showed disrespect to an adult, or spray painted a building, and whether they felt people were willing to help their neighbors in their community. To explore the implications of collective efficacy, the research team assessed attitudes about social cohesion and informal social control in 343 neighborhood clusters in Chicago. They then compared these attitudes with measures of violence and crime in the neighborhoods. They found a strong link between residents’ feelings of collective efficacy with perceived neighborhood violence, with reported violent victimization, and with neighborhood homicide rates. This correlation held over and above the effects of other neighborhood characteristics, such as socioeconomic disadvantage and levels of residential mobility. While cross-sectional and limited to a single city, this study provides compelling evidence for a positive community-level construct, specifically collective efficacy.

MEASURING POSITIVE DEVELOPMENT Having proffered a set of positive characteristics, behaviors and outcomes for children, youth, and their families, a next task is to find appropriate measures to make these constructs concrete. As we have hinted throughout the last section, measurement is hindered by problems of social desirability. For example, it is difficult for researchers to devise survey questions that measure the possession of the attributes that constitute character. In particular, it is

Preventing Problems vs. Promoting the Positive


difficult to ask respondents directly if they are persons of character. However, more subtle or unobtrusive measures directed at children and adolescents might be developed for large-scale representative surveys. In addition, questions can be developed for other persons to answer about the child or adolescent. For instance, in the National Educational Longitudinal Study (NELS), both parents and teachers provide data about the student. In: the same way that parents and teachers rate negative behaviors, they can be asked to rate a child or adolescent with regard to his or her varied positive characteristics. It is desirable to obtain the perspectives of multiple reporters and to use multiple methods. For example, in addition to parent and teacher reports, children may be asked directly how they spend their time (Hofferth & Sandberg, 2000). Time use data might reveal time spent exercising, volunteering, and attending religious services. It is also important to assess positive constructs with measures that are age-appropriate. Researchers need to consider carefully when a measure should begin to be assessed and how to assess it at varied ages.

SUMMARY: THINKING POSITIVELY ABOUT THE FUTURE Currently, there is extensive coverage, both in the media and in our national survey data, of negative adolescent behaviors and poor child outcomes. The dearth of information on positive outcomes desired for and achieved by children contributes to the negative attitudes that the public seems to hold about children and youth. The scarcity of information on positive development is due, in part, to a lack of consensus among experts in the field regarding positive outcomes desired for children. Lack of agreement regarding what are positive outcomes undermines our collective capacity to raise healthy, high-achieving children, because parents, communities, youth leaders, and teachers lack a sense of what goals should be sought. We hope that the issues we have raised in this paper will spark productive conversations that will lead to a better understanding of the full range of development and behavior capable among children. We suggest that our ultimate goal, as a community, should be more balanced conversations, with an eye toward establishing positive goals and creating assessments of progress toward positive outcomes for children and youth, as well as recognizing and addressing negative behaviors. If our horizons extend only to preventing negative outcomes for individual children, our programs and policies and public discussions will be quite different than if our goals include positive



development for children as individuals, children within families, and children as members of and contributors to the community.

ACKNOWLEDGMENTS This paper was written with support from the Office of the Assistant Secretary for Planning and Evaluation, DHHS and the NICHD Family and Child WellBeing Research Network.

REFERENCES Aber, J. L., & Jones, S. M. (1994). Indicators of positive development in early childhood: Improving concepts and measures. Prepared originally for the Conference on Indicators of Children’s Well-Being, Bethesda, MD. Aber, J. L., & Jones, S. M. (1997). Indicators of positive development in early childhood: Improving concepts and measures. In: R. M. Hauser, B. V. Brown & W. R. Prosser (Eds), Indicators of Children’s Well-Being (pp. 395–408). New York: Russell Sage Foundation. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum. Arend, R., Gove, F. L., & Sroufe, L. A. (1979). Continuity of individual adaptation from infancy to kindergarten: A predictive study of ego-resiliency and curiosity in preschoolers. Child Development, 50, 950–959. Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1997). Monitoring the Future: Questionnaire responses from the nation’s high school seniors. Ann Arbor, MI: University of Michigan, Institute for Social Research. Bearman, P., & Bruckner, H. (1999). Peer effects on adolescent sexual debut and pregnancy. Paper for the National Campaign to Prevent Teen Pregnancy, Washington, D.C. Benson, P. L. (1993). The troubled journey: A portrait of 6th–12th grade youth. Minneapolis, MN: Search Institute. Borenstein, L. (1996). The development of friendship in childhood: A clinical conversation. Child and Adolescent Social Work Journal, 13, 213–224. Bradley, R. H., & Caldwell, B. M. (1980). The relation of home environment, cognitive competence, and IQ among males and females. Child Development, 51, 1140–1148. Bradley, R. H., & Caldwell, B. M. (1984). The relation of infants’ home environments to achievement test performance in first grade: A follow-up study. Child Development, 55, 803–809. Bradley, R. H., Caldwell, B. M., Rock, S. M., Barnard, K. E., Gray, C., Hammond, M. A., Mitchell, S., Siegel, L., Ramey, C. T., Gottfried, A. W., & Johnson, D. L. (1989). Home environment and cognitive development in the first 3 years of life: A collaborative study involving six sites and three ethnic groups in North America. Developmental Psychology, 25, 217–235. Bretherton, I. (1985). Attachment theory: Retrospect and prospect. In: I. Bretherton & E. Waters (Eds), Growing Points of Attachment Theory and Research. Monographs of the Society for Research in Child Development 50 (1–2, Serial No. 209), 167–193.

Preventing Problems vs. Promoting the Positive


Brooks-Gunn, J., Duncan, G., & Aber, J. L. (Eds) (1997). Neighborhood poverty: Context and consequences for children. New York, NY: Russell Sage Foundation. Brown, B. B. (1990). Peer groups and peer cultures. In: S. Feldman & G. Elliott (Eds), At the Threshold: The Developing Adolescent (pp. 171–196). Cambridge, MA: Harvard University Press. Brown, B. V., & Corbett, T. (1997). Social indicators and public policy in the age of devolution. Institute for Research on Poverty (Special Report Series SR #71). University of WisconsinMadison. Carnegie Corporation of New York (1992). A matter of time: risk and opportunity in the nonschool hours. New York: Author. Carnegie Council on Adolescent Development (1989). Turning points: Preparing American youth for the 21st century. New York: Author. Cassidy, J. (1986). The ability to negotiate the environment: An aspect of infant competence as related to quality of attachment. Child Development, 57(2), 103–106. Chalk, R., & Phillips, D. A. (Eds) (1996). Youth development and neighborhood influences: Challenges and opportunities. Washington, D.C.: National Academy Press. Chervin, D. D., Reed, E., & Dawkins, N. (1998). Identifying optimal outcomes for children and youth: A focus group study with parents and youth. Final Report to The Task Force for Child Survival and Development. Atlanta, GA: Macro International, Inc. Child Trends (1999). Children and welfare reform: A guide to evaluating the effects of state welfare policies on children. Washington, D.C.: Child Trends. Coley, R. L., & Chase-Lansdale, P. L. (1998). Adolescent pregnancy and parenthood: Recent evidence and future directions. American Psychologist, 53(2), 152–166. Compas, B. E. (1993). Promoting positive mental health during adolescents. In: S. G. Millstein, A. C. Petersen & E. O. Nightingale (Eds), Promoting the Health of Adolescents: New Directions for the Twenty-first Century (pp. 159–179). New York: Oxford University Press. Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience in AfricanAmerican youth: Context, self, action, and outcomes in school. Child Development, 65, 493–506. Council on Civil Society (1998). A call to Civil Society: Why democracy needs moral truths. New York: Institute for American Values. Damon, W., & Gregory, A. (1997). The youth charter: Towards the formation of adolescent moral identity. Journal of Moral Education, 26(2), 117–129. DeMartini, J. (1983). Social movements participation. Youth and Society, 15, 195–223. Dunn, J. (1988). The beginnings of social understanding. Cambridge, MA: Harvard University Press. Dunn, J., & Kendrick, C. (1982). Siblings: Love, envy, and understanding. Cambridge, MA: Harvard University Press. Dunn, J., & Munn, P. (1986). Siblings and the development of prosocial behavior. International Journal of Behavioral Development, 23, 791–798. Eberly, D. (1998). America’s Promise: Civic Society and The Renewal of American Culture. Lanham, MA.: Rowman & Littlefield Publisher’s, Inc. Eccles, J. S., & Barber, B. L. (1999). Student council, volunteering, basketball, or marching band: What kind of extracurricular involvement matters? Journal of Adolescent Research, 14, 10–43. Elliott, D. S. (1993). Health-enhancing and health-compromising lifestyles. In: S. G. Millstein, A. C. Petersen & E. O. Nightingale (Eds), Promoting the Health of Adolescents: New



Directions for the Twenty-first Century (pp. 119–145). New York: Oxford University Press. Entwistle, D. R. (1995). The role of schools in sustaining early childhood program benefits. Future of Children, 5(3), 133–144. Entwisle, D. R., & Alexander, K. L. (1993). Entry into school: The beginning school transition and educational stratification in the United States. Annual Review of Sociology, 19, 401–423. Federal Interagency Forum on Child and Family Statistics (1998). America’s children: Key national indicators of well-being, 1998. Federal Interagency Forum on Child and Family Statistics, Washington, D.C.: U.S. Government Printing Office. Federal Interagency Forum on Child and Family Statistics (1999). America’s children: Key national indicators of well-being, 1999. Washington, D.C.: U.S. Government Printing Office. Fendrich, J. (1993). Ideal citizens. Albany, NY: State University of New York. Flanagan, C. A., & Gills, S. (1999). Social integration, identity, and youth civic commitments. Paper presented at the Biennial Meeting of the Society for Research in Child Development, Albuquerque, NM. Flanagan, C., Jonsson, B., Botcheva, L., Csapo, B., Bowes, J., Macek, P., Averina, I., Sheblanova, E. (in press). Adolescents and the ‘Social Contract’: Developmental roots of citizenship in seven countries. In: M. Yates and J. Youniss (Eds), Community Service and Civic Engagement in Youth: International Perspectives (pp. 135–155). Cambridge, England: Cambridge University Press. Forgione, P. (December 1998). Early childhood education: Critical data needs for a critical period of child development. Paper presented at the U.S. Senate Committee on Labor and Human Resources hearing on “Are Our Children Ready to Learn?” Washington, D.C.: National Center for Education Statistics. Frakas, S., & Johnson, J. (1997). Kids these days: What Americans really think about the next generation. New York: Public Agenda. Frankel, K. A., & Bates, J. E. (1990). Mother-toddler problem solving: Antecedents of attachment, home behavior, and temperament. Child Development, 61, 810–819. Garmezy, N. (1985). Stress-resistant children: The search for protective factors. In: J. E. Stevenson (Ed.), Recent Research in Developmental Psychopathology (pp. 43–84). Oxford: Pergamon Press. Gilligan, C. (1977). In: a different voice: Women’s conception of self and morality. Harvard Educational Review, 47, 481–517. Gilligan, C. (1982). In: a different voice: Psychological theory and women’s development. Cambridge, MA: Harvard University Press. Halle, T., & Shatz, M. (1994). Mothers’ social regulatory language to young children in family settings. First Language, 14, 83–104. Hamburg, B. A. (1990). Life skills training: Preventive interventions for young adolescents (p. 3). New York: Carnegie Corporation of New York. Hanks, M., & Eckland, B. K. (1978). Adult voluntary associations and adolescent socialization. Sociological Quarterly, 19, 481–490. Hauser, R., Brown, B., & Prosser, W. (1992). Indicators of children’s well-being. New York: Russell Sage Foundation. Hawkins, J. D., Catalano, R. F., Kosterman, R., Abbott, R., & Hill, K. G. (1999). Preventing adolescent health-risk behaviors by strengthening protection during childhood. Archives of Pediatric Adolescent Medicine, 153, 226–234.

Preventing Problems vs. Promoting the Positive


Hetherington, E. M., Bridges, M., & Insabella, G. M. (1998). What matters? What does not? Five perspectives on the associations between marital transitions and children’s adjustment. American Psychologist, 53(2), 167–184. Hofferth, S. L., & Sandberg, J. F. (2000). Changes in American Children’s Time, 1981–1997. In: T. Owens & S. Hofferth (Eds), Children at the Millennium: Where Have We Come From, Where are We Going? Advance in Life Course Research. New York: Elsevier Science. International Youth Foundation (June 1998). What is youth development? In: Programs That Work [online]. Available: Kagan, S. L., Moore, E., & Bradekamp, S. (1995). Reconsidering children’s early development and learning: Toward common views and vocabulary. Washington, D.C.; National Education Goals Panel, Goal 1 Technical Planning Group. Kohlberg, L. (1975). The cognitive developmental approach to moral education. Phi Delta Kappa, 56, 670–77. Kohlberg, L. (1981). The philosophy of moral development. New York: Harper & Row. Konopka, G. (1973). Requirements for development of adolescent youth. Adolescence, 8(31), 1–26. Ladewig, H., & Thomas, J. K. (1987). Assessing the impact on former 4-H members. College Station, TX: Texas Tech University. Larson, R. W. (2000). Toward a psychology of positive youth development. American Psychologist, 55, 170–183. Leffert, N., Benson, P. L., Scales, P. C., Sharma, A. R., Drale, D. R., & Blyth, D. A. (1998). Developmental assets: Measurement and prediction of risk behaviors among adolescents. Applied Developmental Science, 2(4), 209–230. Luthar, S. S. & Zigler, E. (1991). Vulnerability and Competence: A Review of Research on Resilience in Childhood. American Journal of Orthopsychiatry, 61(1), 6–22. McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press. McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53(2), 185–204. Mahoney, J. L., & Cairns, R. B. (1997). Do extracurricular activities protect against early schoool dropout? Developmental Psychology, 33, 241–253. Matas, L., Arend, R. A., & Sroufe, L. A. (1978). Continuity of adaptation in the second year: The relationship between quality of attachment and later competence. Child Development, 49, 547–556. Maynard, R. A. (Ed.) (1997). Kids having kids: Economic costs and social consequences of teen pregnancy. Washington, D.C.: The Urban Institute Press. Mendelson, J., Aboud, F. E., & Lanthier, R. P. (1994). Kindergartners’ relationships with siblings, peers, and friends. Merrill-Palmer Quarterly, 40, 416–435. Millstein, S. G., Petersen, A. C., & Nightingale, E. O. (Eds) (1993). Promoting the health of adolescents: New directions for the twenty-first century. New York: Oxford University Press. Moore, K. A. (1997). Criteria for indicators of child well-being. In: R. M. Hauser, B. V. Brown & W. R. Prosser (Eds), Indicators of Children’s Well-Being (pp. 36–44). New York: Russell Sage Foundation. Moore, K. A., Evans, V. J., Brooks-Gunn, J., & Roth, J. (forthcoming). What Are Good Child Outcomes? In: A. Thornton (Ed.), The Well-being of Children and Families: Research and Data Needs. Ann Arbor, MI: University of Michigan Press.



Moore, K. A., & Glei, D. (1995). Taking the plunge: An examination of positive youth development. Journal of Adolescent Research, 10(1), 15–40. Moore, K., Manlove, J., Glei, D., & Morrison, D. R. (1998). Non-marital school-age motherhood: Family, individual, and school characteristics. Journal of Adolescent Research, 13(4), 433–457. Moore, K. A., Miller, B. C., Morrison, D. R., & Glei, D. A. (1995). Adolescent sex, contraception, and childbearing: A review of recent research. Washington, D.C.: Child Trends. Moore, K. A., Morrison, D. R., & Glei, D. A. (1995). Welfare and adolescent sex:The effects of family history, benefit levels, and community context. Journal of Family and Economic Issues, 16(2–3), 207–238. Mott, F. L., & Haurin, R. J. (1992). Linkages between sexual activity and drug use among American adolescents. Family Planning Perspectives, 20, 129–136. National Center for Health Statistics (1996). Health, United States: 1995. Hyattsville, MD: Public Health Service. National Commission on Children (1991). Speaking of kids: A national survey of children and parents. Washington, D.C.: U.S. Government Printing Office. National Education Goals Panel (1997). The National Education Goals Report: Building a nation of learners, 1997. Washington, D.C.: U.S. Government Printing Office. National Research Council and Institute of Medicine (2000). From neurons to neighborhoods: The science of early child development. Committee on Integrating the Science of Early Childhood Development, Jack P. Shonkoff and Deborah A. Phillips (Eds). Board of Children, Youth, and Families, Commission on Behavioral and Social Sciences and Education. Washington, DC: National Academy Press. Parke, R. D. (1996). Fatherhood. Cambridge, MA: Harvard University Press. Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357–389. Phillips, D. A., & Love, J. M. (1997). Indicators for school readiness, schooling, and child care in early to middle childhood. In: R. M. Hauser, B. V. Brown & W. R. Prosser (Eds), Indicators of Children’s Well-Being (pp. 125–151). New York: Russell Sage. Piaget, J. (1932). The moral judgment of the child. New York: Macmillan. Pleck, J. H. (1997). Paternal involvement: Levels, sources, and consequences. In: M. E. Lamb (Ed.), The Role of the Father in Child Development (pp. 66–103). New York: John Wiley and Sons, Inc. Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris, K. M., Jones, J., Tabor, J., Beuhring, T., Sieving, R., Shew, M., Ireland, M., Bearinger, L. H., & Udry, J. R. (1997). Protecting adolescents from harm. Journal of the American Medical Association, 278(10), 823–832. Roth, J., Brooks-Gunn, J., & Galen, B. (1997). Promoting healthy adolescence: Youth development frameworks and programs. Center for Children and Families, Teachers College, Columbia University, New York. Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American Journal of Orthopsychiatry, 57(3), 316–331. Ryff, C. D., & Keyes, C. L. (1995). The structure of psychological well-being revisited. Journal of Personality and Social Psychology, 69, 719–727. Sallis, J. F. (1993) Promoting healthful diet and physical activity. In: S. G. Millstein, A. C. Petersen & E. O. Nightingale (Eds), Promoting the Health of Adolescents: New Directions for the Twenty-first Century (pp. 209–241). New York: Oxford University Press.

Preventing Problems vs. Promoting the Positive


Sampson, R., Raudenbush, S., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Washington, D.C.: American Association for the Advancement of Science. Scales, P. C., & Leffert, N. (1998). Developing assets: A synthesis of the scientific research on adolescent development. Minneapolis, MN: Search Institute. Sroufe, L. A. (1983). Infant-caregiver attachment and patterns of adaptation in the preschool: The roots of maladaptation and competence. In: M. Perlmutter (Ed.), Minnesota Symposium in Child Psychology 16 (pp. 41–83). Hillsdale, NJ: Erlbaum. Suess, G., Grossman, K., & Sroufe, L. A. (1992). Effects of infant attachment to mother and father on quality of adaptation in preschool. International Journal of Behavioral Development, 15, 43–65. Sugland, B. W., Brown, B. V., & Moore, K. A. (1993). Pathways to achievement among at-risk youth. Washington, D.C.: Child Trends Takanishi, R., Mortimer, A. M., & McGourthy, T. J. (1997). Positive indicators of adolescent development: Redressing the negative image of American adolescents. In: R. M. Hauser, B. V. Brown & W. R. Prosser (Eds), Indicators of Children’s Well-Being. New York: Russell Sage Foundation. U.S. Department of Health and Human Services, Assistant Secretary for Planning and Evaluation (1998). Trends in the well-being of America’s children and youth, 1998. Washington, D.C.: U. S. Government Printing Office. United Way of America (1998). A preliminary analysis of outcome measurement advice in seventeen leading documents. A draft report prepared by James Bell Associates. Volliing, B. L. (forthcoming). Sibling Relationships and Children’s Well-Being. In: M. Bornstein, L. Davidson, C. Keyes & K. Moore (Eds), Well-Being: Positive Development across the lifespan. Hillsdale, NJ: Erlbaum. Wellborn, J., & Connell, J. P. (1987). Rochester assessment package for schools student report. Unpublished manuscript, University of Rochester, Rochester, NY. Werner, E. E. (1989). High-risk children in young adulthood: A longitudinal study from birth to 32 years. American Journal of Orthopsychiatry, 59, 72–81. West, J., Hausken, G. E., & Collins, M. (1993). Readiness for kindergarten: Parent and teacher beliefs (NCES 93–257). Washington, D.C.: U.S. Department of Education, National Center for Education Statistics. Winfield, L. F. (1991). Resilience, schooling, and development in African American youth: A conceptual framework. Education and Urban Society, 24, 5–14. Yamaguchi, D., & Kandel, D. B. (1987). Drug use and other determinants of premarital pregnancy and its outcome: A dynamic analysis of competing life events. Journal of Marriage and the Family, 49, 257–270. Youniss, J. (1999, April 16). Community service as a means to civic identity. Paper presentation at the Biennial Meeting of the Society for Research in Child Development, Albuquerque, NM. Youniss, J., & Yates, M. (1997). Community service and social responsibility in youth. Chicago, IL: University of Chicago Press. Youniss, J., Yates, M., & Su, Y. (1997). Social integration: Community service and marijuana use in high school seniors. Journal of Adolescent Research, 12, 245–262. Youniss, J., McLellan, J. A., Su, Y., & Yates, M. (1999). The role of community service in identity development: Normative, unconventional, and deviant orientations. Journal of Adolescent Research, 14, 248–261.



Zill, N., & Winglee, M. (1989). Literature reading in the United States: Data from nation surveys and their policy implications. Book Research Quarterly, 5(1), 24–58. Zill, N., Nord, C. W., & Loomis, L. S. (1995). Adolescent time use, risky behavior, and outcomes: An analysis of national data. Rockville, MD: Westat, Inc. Zimmerman, M. A., & Arunkumar, R. (1994). Resiliency research: Implications for schools and policy. Society for Research in Child Development Social Policy Report, 8(4), 1–19.


This Page Intentionally Left Blank



Dennis P. Hogan and Frances Goldscheider ABSTRACT The increase in the proportion of children living in female-headed families implies that men’s likelihood of living with their children has declined. However, this may understate men’s coresidence with children, as many female family heads live with other men, either with their fathers or in cohabiting relationships. Many of the absent fathers of these children live with children, their younger siblings or with stepchildren. Sex differences in living with children may not have increased as much as have femaleheaded families. In this paper, we examine patterns of coresidence with children under age 15 over the period 1880 to 1990 in the U.S., using the Integrated Public Use Samples (IPUMS) of the U.S. Census. We distinguish between “own children” and other children, and compare men and women. We examine the extent to which men and women’s living with children is a function of age, marital status, education, and farm residence. Our results are as follows: (1) Recent declines in male coresidential parenthood are simply extensions of the trend from 1880 onwards, rather than any new crisis in family life;

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 173–191. 2001 by Elsevier Science Ltd. ISBN: 0-7623-0776-5




(2) The long-term historical decline in parenting is quite similar for both men and women, with only a slight divergence towards fewer men engaged in coresidential parenting of an own child; (3) Coresidential parenting of children other than own children is somewhat higher among males, although differences were larger prior to 1950; and (4) The historical declines in fatherhood are associated with increasing non-farm residence and proportions unmarried. Fatherhood nevertheless continues to occupy a central, but more compact, role in the lives of American men.

INTRODUCTION Historical and contemporary research on the family has been nearly totally concerned with the lives of women (ignoring men) and of children (ignoring parenthood as a central adult role). These omissions have become particularly problematic at the end of the twentieth century. There have been dramatic changes in family patterns in most industrialized countries in the past three decades. The birth of children outside of marriage and increases in women’s labor force participation have weakened the father’s role in the family as provider, which is the only clearly defined element of the father role. Social scientists have largely embraced this view of fatherhood, culminating in research, conferences, and popular articles that ask whether (and how) fathers matter. Indeed, even as the father’s role as principal breadwinner in his family has been de-emphasized, the pursuit of child support payments from formerly married and unmarried “deadbeat” dads has reified this role of breadwinner as the only legally enforceable role of fathers. In fact, despite this seeming flood of research on families and children, little research has examined the parenthood experiences of men, how these have changed over time, and what implications these changes have for the future. This paper is intended to begin to close this research gap. In this paper, we identify changes in parenthood for native-born men (with comparisons to women), defined in terms of coresidence with persons under the age of 15 (what we call “coresidential parenthood”), for the period 1880 to 1990. To characterize the nature of the coresidential parenting experience we distinguish between coresidence with children who are “own children” (including stepchildren1) and “other (not-own) children,” who are primarily siblings and grandchildren and observe how the proportions of men and women experiencing parenthood have changed over time. We also trace real birth cohorts of men across the years to describe intercohort changes in the rough

Men’s Flight From Children in the U.S.: A Historical Perspective


shape of the coresidential parenting experience across the life course. The observed differences in the coresidential parenthood of men and women are placed in their social historical context, with attention to the effects of demographic composition, farm residence, education, and marital status.

APPROACH TO PARENTHOOD The major focus of research on family change has been on male-female relationships, which have become much less stable, both because of the rapid rise in divorce and because of the growth in cohabiting unions (which are even less stable than marriages). Given predominant mother custody, one littlenoticed result is that parenthood has become a less central and stable element in men’s lives, not only compared with the past, but perhaps also as compared with its role in the lives of women. The connections between men and children have become residentially complicated. With the rise in divorce and out-of-union childbearing, men are increasingly not living with the children they have fathered, but more may have the opportunity to be a father to other children (for example, their partner’s children) (Bumpass & Raley, 1995; Bumpass, Raley & Sweet, 1995; Juby & Lebourdais, 1998). Women, too, have been experiencing changes in the centrality of parenthood in their lives. The decline in mortality has extended men’s and women’s planning horizon for both their work and family lives. Coupled with the decline in fertility and changes in its timing, these factors have reshaped the life course so that coresidential parenthood is less likely to be a lifetime experience for men or women. Further, it is important to place this recent change in broader historical perspective. Much of our knowledge about families, parent-child linkages, and the experience of parenting is for the post-World War II period. The dramatic changes of the past three decades have dominated family scholarship, at least in part because there has been little systematic information available for earlier decades. Yet in studies of marriage and the timing and level of fertility, it is increasingly recognized that the 1950s and 1960s were an historical anomaly that provide a poor baseline for understanding more recent family change (Cherlin, 1992). Our approach to parenthood is via coresidence with children. We see parenthood as not simply a state but a role, an activity that has impact on both parents and children. As such, it is within the home that most effective parenting gets done. We argue that the social role of parenting considered in terms of economic support to the household, interaction with the child, and family life affected by the child, should be considered to include any adult



interaction with a child living in the same household. Given the recent changes in family life described above, men’s experiences are best considered from a broad “social parenting” perspective. Because our major focus is on longterm historical trends, we can capture the important component of social fatherhood men experience only through coresidence. Other studies that focus on noncoresidential families in recent periods cannot provide the longterm historical perspective that we believe is essential to understanding the current situation and prospects for men, children and families. Most extant demographic research on parenthood has focused on marriage and fertility. Although most analyses of changing marriage and fertility patterns have focused on women, it is clear that, overall, men and women have experienced fairly similar trajectories in the 1880–1990 period (Cherlin, 1992). Men’s ages at marriage may have fluctuated slightly more than women’s have. The decline in age at marriage during the 1940s and 1950s occurred faster for men than for women, reducing the average gap between spouses from more than four years in 1940 to about two years in 1960. Clearly, biological differences between men and women are not irrelevant to their parenthood experience. Men do have longer reproductive lives, and in the early part of the study period, when young adult mortality was still relatively high, men were more likely to have continued to have children at later ages than women due to early spouse death and remarriage. Changes in desired family size and contraceptive choice since then may have concentrated biological parenthood more narrowly in men’s lives, but this has not been a focus of demographic research. We take the position that the social parameters of parenthood for men should be studied even in the absence of evidence on biology, and that these same dimensions should also apply to women, permitting the study of male-female differences. Despite the large literature on changes in living arrangements, we know relatively little about changes in parental living arrangements. Most studies of living arrangements have focused on older adults and on unmarried persons. Such persons are the least likely to be living with children (Pampel, 1983; Kobrin, 1976; Michael et al., 1980; Sandefur & Tuma, 1987; Santi, 1990; Kramerow, 1995; Ruggles, 1996). An important exception to this neglect is the work of David Eggebeen and his coauthors (Eggebeen, 1999; Eggebeen & Uhlenberg, 1986), which has documented both the decline in coresidential fatherhood in the U.S. since 1960 and the racial differences therein. In this paper we build on this work. We extend the horizon back to 1880, focus on men in comparison with women, and distinguish between living with own children aged 14 or younger and living other children.

Men’s Flight From Children in the U.S.: A Historical Perspective


DATA We analyze data that cover most of the period from 1880–1990 drawn from IPUMS-98 – Integrated Public Use Microdata Series of the national U.S. decennial censuses created by Steven Ruggles and his colleagues at the University of Minnesota Department of History (Ruggles & Sobek, 1997). IPUMS-98 provides a series of compatible format individual-level samples of the census populations of 1880, 1900, 1920, 1940, 1950, 1960, 1970, 1980, and 1990. We did not use data from the 1850 census as the analysis includes both black and white men, and only free blacks were enumerated in 1850. After considerable investigation, we decided not to include the 1910 census in this analysis. This census sample treated boarders, lodgers, and employed individuals in households differently than was done for other census years, resulting in anomalous findings for 1910 relative to 1900 or 1920. The IPUMS group is now preparing a newly sampled census file for that year that will correct this problem. Each of these files is a cross-sectional sample for the census year, and provides exact data representative of men and women for that point in time. These census data are subject to undercount. In no year in the series is undercount greater than 6.7%, however, which is far better than generally available in recent population sample surveys that constitute the basis for much of the current analysis of parenting. Further, the level of undercount improved substantially over the period (Robinson, 1988). The population coverage for each census prior to 1960 is of all persons in the U.S., District of Colombia, and continental Territories that eventually became states. Beginning in 1960, Alaska and Hawaii are included. Given the small size of the populations in these states, overall statistics for the nation were largely unaffected. The data files are based on samples from the manuscript census archives, which vary in overall sampling fraction and in construction method, depending on the information available. Sampling fractions varied from 1 in 100 to 1 in 760. In all census years except 1900 there are more than 200,000 sample adults for analysis; the number of sample adults available for this analysis is nearly one million for each census from 1940 onward. Despite the very low sample density for the 1900 Census, even that year includes nearly 50,000 adults for this analysis. The censuses for 1940, 1950, and 1990 also involved sampling by the Bureau of the Census. For 1940 and 1950, the census bureau’s sampling method was based on persons rather than households, with the result that more information is available for “sample-line” persons than for others in the household. The 1990 Census IPUMS includes oversamples for persons residing



in small localities in order to increase sampling precision for these places. We use self-weighted subsamples that have been generated for IPUMS users. The IPUMS samples for other census years are self-weighting. This analysis is restricted to native-born residents of the U.S. for two reasons. First, our interest is on long-term historical trends, a subject which is not informed by the treatment of ethnic immigrant groups who dominate differing time periods, and who are themselves made up of diverse types of immigrants. The subject of immigrants is best left for another paper that examines the composition of the foreign-born (country of origin, age, and sex), and distinguishes streams that arrived in family groups or as individuals. Second, the restriction to the native born facilitates the calculation of real cohort experiences by focusing on a population that is relatively closed, and is thus appropriate to follow across successive time periods.

MEASURES AND ANALYTIC APPROACH The objective of this paper is to describe the coresidential parenting experiences of men, compare these experiences to those of women, and explain differences between the sexes and across time periods. Our analysis focuses on the coresidence of native-born adults (ages 18–74) with children under the age of 15. We identify households that contain persons under the age of 15 for an overall assessment of coresidential parenting. We distinguish two types of coresidential parenting relationships – (a) adults who have an “own child” under 15 years of age in the household, and (b) adults who live in a household with children under age 15, none of whom are own children. These social experiences of coresidential parenting can include those of (1) young adults living at home with younger siblings, (2) grandparents and other relatives whose kin have children under age 15, and (3) boarders, lodgers, and assorted other arrangements that lead to sharing a household with children. We use literacy/education and farm residence as individual characteristics that describe levels of human capital, social resources, and life styles, each of which are likely to affect the likelihood of marriage and the number and timing of children. Literacy is the only measure of educational exposure from 1880 to 1920. In 1940 and thereafter, education was measured by years of schooling completed. To obtain a single measure for the time series, we classified persons who were illiterate (1880–1920) and those with less than five years of schooling (1940–1990) as “low education.” In cases where education information was not available for an adult (because of nonsample line status or nonresponse), a third category was added to the educational variable indicating “not available” (8% of the total).

Men’s Flight From Children in the U.S.: A Historical Perspective


The multivariate analysis is based on self-weighting subsamples of the IPUMS files designed to yield about 100,000 cases in each sample year. No subsampling was done for the IPUMS 1900 file since that includes 46,000 cases. Overall, the multivariate models are based on 908,568 persons. The models estimated are binomial and multinomial logistic regressions for (1) living with children or not, and (2) for the competing outcomes of not living with children aged less than 15 (the reference category), living with own child, and living with other children but no own children. These models assess (a) overall sex differences in the likelihood of these outcomes, (b) sex differences controlling for demographic composition (race, period, age group), and (c) sex differences with additional controls for farm status, education, and marital status.2 We pool the samples over time, and treat census year as an additive covariate. There are fairly substantial changes in the effects of some variables over time, which we note where appropriate.

RESULTS We present our basic findings on the trends in coresidential parenthood in the figures that follow. The figures display lines in which each decennial data point is exactly plotted. We used an exponential smoothing procedure to more accurately portray the trends between census years. This procedure is particularly useful in that it handles ten- and twenty-year gaps between censuses in consistent fashion. We then turn to a multivariate analysis of the determinants of living with children less than 15, whether own or other children.

TRENDS The upper pair of lines in Fig. 1 shows that 42% of men aged 18–74 lived with an own child in 1880. The percentage of men living with an own child declined very slowly to 33% in 1940 and to 28% in 1990. There was a major interruption, however, between 1950 and 1970, when the baby boom increased coresidence with own children temporarily to a higher level than in 1880, approaching 50%. While data for the post-World War II period have been used to suggest a dramatic decline in fatherhood (Eggebeen, 1999), the long-term trend is clearly quite different when we do not choose a starting date that is an historical anomaly. Throughout the period, men were somewhat less likely to live with an own child than were women, although the differences were negligible in 1960 and 1970. Again, taking that period as a starting point shows increased sex


Fig. 1. Percentage of Native-Born Adults Aged 18–74 who Coresided with at Least One Own Child or with an “Other” Child Under Age 15, by Sex of Adult and Year.

Men’s Flight From Children in the U.S.: A Historical Perspective


differences, but these again replicate the magnitude of differences evident in the decades around the beginning of the twentieth century. This measure, however, underestimates the extent to which adults live in parental-like relationships with children. Many of those living with other children contribute to them by providing resources of time, example, and material support as grown siblings and other relatives. This pattern, it turns out, is more characteristic of men than of women. The lower lines in Fig. 1 show the trend in coresidence with other children under age 15. In 1880, 26% of men lived with some other child aged less than 15, making this more than half as common as living with own children. Living with children in this fashion, however, has declined even more rapidly than living with own children. By 1940, only 15% of men were coresidential “parents” by virtue of living in a household with a child that is not their own. Coresidence with other children declined very dramatically between 1950 and 1960. Young persons increasingly left their parental homes and any siblings, normally to marry and have children of their own, and of course, the likelihood that there were younger siblings to live with declined as well. The share of those living with other children fell to below 7%, which was about one quarter of the percentage living with own children, and this has held fairly steady since that time. Thus, an increasing portion of coresidential fatherhood is linked to residence with an own child. Further, although earlier in the period this pattern was particularly characteristic of men, by 1960 men and women had converged, with only about a one percentage-point difference between them in the years between 1960 and 1990. How, then, do these two trends combine? Figure 2 shows the changes from 1880 to 1990 in the percentage of native-born men and women who reside in households where there is at least one child younger than age 15, whether their own or not. Overall, the view is similar, but there are interesting differences. The baby boom of the late 1940s now appears much more limited in its temporal impact than it did when we considered only living with own children, with a major peak only for 1950. The decline is reestablished much more quickly, with adults in 1960 little more likely to live with children under 15 than they were in 1940, as the decline in coresidence with other children accelerated. Ignoring the deviations of the baby boom years, the long-term trend was towards deceleration of the historic decline in coresidential parenthood. The coresidential parenting experience of men declined slowly but steadily between 1880 and 1940, amounting to a decline of about twenty points over sixty years (averaging 0.33% annually). Between 1940 and 1990 the percentage of men


Fig. 2. Percentage of Native-Born Adults Aged 18–74 who Coresided with at Least One Child Under Age 15, by Sex of Adult and Year.

Men’s Flight From Children in the U.S.: A Historical Perspective


18–74 experiencing coresidential parenthood decreased by 13 points (averaging 0.26% annually). Of the 33-point decline in coresidential parenting from 1880 to 1990, only 39% occurred after 1940. This is a different, and we argue more accurate, portrayal of fatherhood in America – a long-term decline in coresidential fatherhood, interrupted only by the aberrations associated with the baby boom. Viewed in this way there is no recent crisis in the coresidential fathering experiences of American men; the situation has been developing for a long time, and applies to both men and women. The combined view also suggests that not only has the long-term trend in the point-prevalence rate of parenting for men been quite similar to that for women, but the differences between the sexes in total coresidential parenthood are considerably less than for living with own children. The differences in the early years are proportionately smaller than for living with own children and the period of near total convergence lasts much longer: from 1940 to 1970. After 1970, a difference in the coresidential parenting experiences of men and women reappeared, but it was only 4 percentage points. Although proportionately, the difference is about 15%, such a widening in the parenting experiences of men and women would not appear to constitute a crisis in the parenting experiences of men. These cross-sectional depictions of trends in the coresidential parenting experience of men point up the error in thinking about fatherhood in America as facing a crisis. The parental experience is a persistent feature of men’s lives throughout American history. But the prevalence of that experience has continued its long downward decline in recent years, as fewer younger and older men live with children. To trace the increased compactness of coresidential fatherhood in the lives of men, we calculated coresidential parenthood for real birth cohorts by age as traced through successive censuses, including only those cohorts for whom census coverage occurred with enough regularity to calculate a lifetime portrait (Fig. 3). For cohorts born before 1900, coresidential parenthood was a conspicuous feature of the entire adult life; for those born afterwards, it was an experience that was increasingly focused in the second decade of adulthood, ages 30–39. Beginning with the cohort of 1910–1919, who entered adulthood between 1928 and 37, the shape of coresidential parenthood over the lifetime shifted dramatically. Much higher peaks became evident at ages 30–39 relative to younger and older ages, and these were peaks that were higher than for the cohorts who preceded them. Whereas the cohorts born in the nineteenth century never had more than 50 to 60% living with children at any age, those born between 1900 and 1940 approached a level of coresidential parenthood of almost 80% during their 30s.


Fig. 3. Percentage of Native-Born Male Adults Aged 18–69 who Coresided with at Least One Child Under Age 15, by Birth Cohort and Age.

Men’s Flight From Children in the U.S.: A Historical Perspective


However, the older cohorts continued their involvement in coresidential parenthood much longer, with levels about as high in their 40s as in their 30s and with relatively little decline as they passed through their 50s. The younger cohorts’ concentrated pattern of coresidential parenthood, in contrast, led to a sharp decline over the middle-adult ages. Only 14% of men born in the 1930s were living with children by the time they reached their 50s, compared with nearly 50% of men born in the 1870s. Considering only the levels for all ages obscures the new life course of parenthood. These comparisons certainly suggest considerable change in the cumulative exposure of men to coresidential parenting, but even the most recent of these changes are simply a continuation of long-term trends, not a sudden, dramatic new crisis. As with all true cohort analyses, the processes examined reflect past history. But the findings here are consistent with the period changes described above, supporting and informing our discussion of the lives of men, children, and families.

MULTIVARIATE ANALYSES These broad characterizations of coresidential parenthood have portrayed the parenthood experience of American men from 1880 to 1990, both in crosssectional and cohort terms, and in contrast with the experience of women. But they have not controlled for demographic differences in composition that may affect levels or trends in coresidential parenthood. Nor have they provided analytic explanations of male and female differences. To accomplish this, we estimated the likelihood of living with children under age 15, distinguishing in some models between living with own and other children, in order to observe the effects of gender over time, as well as of other social and economic factors likely to influence parenthood. Our first model examines the odds of being a coresidential “parent” to any child, whether own or other (Table 1). The relative odds that men live with children, on average, pooling across census years, was 0.948 (Model I), indicating that, over the entire period, men were only about 5% less likely to be living with a child than were women. Taking into account changes over the time period and in the age structure of the adult population (Model II) sharpens this difference considerably to 0.917, given the decline in the sex ratio in recent decades and women’s older age distribution. Taking into account race, farm status, and education increases the gender difference slightly more (Model III). Controlling these factors, men are only 90% as likely to live with children as are women, although nearly 20% of this difference reflects gender differences in marital status (Model IV).



Table 1. Logistic Regression Coefficients (Odds Ratios) for Living with Children: United States, 1880–1990. Variables

Model I

Sex (base = Female) Male 0.948 Age Group (base = 30–39) 18–29 40–49 50–59 60–74 Year (base = 1880–1900) 1920–1940 1950–1960 1970–1990 Race (base = White) Black Farm Status (base = Non-farm) Farm Education (base = Low educ.) Not Available Better Education Marital Status (base = Never-married and others) Married and spouse present

Model II

Model III

Model IV




0.464 0.467 0.148 0.070

0.456 0.457 0.140 0.066

0.663 0.425 0.129 0.067

0.538 0.726 0.371

0.613 0.714 0.504

0.577 0.601 0.453





1.505 0.732

1.522 0.715 3.498

Note: All values are significant at p < 0.0001.

There has been substantial variation over time in the effects of gender, however, as we saw at the bivariate level in Fig. 1. Controlling for the other factors in the model does relatively little to change the patterns observed there (results not presented). In the early part of the twentieth century (1920 and 1940), men were as equally likely as women to be living with children under age 15, closing the gap apparent during the previous period; by 1990, men’s relative odds had dropped to 0.775. The other results of the model are unsurprising. The likelihood of living with children declines with age over the adult years, and has declined with time, with the exception of the two decades of the baby boom. Our examination of these patterns over time (not presented) reinforces the compression of parenthood effects shown in Fig. 3. Black men and women are considerably more likely to live with children, as are those living on farms and the married. The racial effect represents a sizable increase since 1940, before which differences were small to nonexistent. The

Men’s Flight From Children in the U.S.: A Historical Perspective


effect for farm residence, in contrast, has attenuated substantially over time (data not presented). Those with more education are less likely to be living with children under age 15 than are those with less, a pattern which has been very stable over time, despite changes in measurement and educational level. Those for whom educational information was not available were considerably more likely to be living with children. As most of these cases are from the 1950 census, however, which did not obtain this information for household members who were not in the special sample, this result is likely to mirror the effects of the baby boom years. Our next analysis distinguishes between parenting through residence with own children and with a child that is not an own child (Table 2). We show odds ratios for the results for living with own children relative to not living with any children (the first column in each model) and for those living with other children (again relative to not living with children) for the same four models shown in Table 1. These results are based on a multinomial logistic regression that controls for the other category of “parental” coresidence. We also examined differences over time in these relationships; those we found to be parallel to the results discussed above, and will not be repeated here. Again, we see that, for the period as a whole (Model I), men were about 10% less likely to live with own children (0.9) but nearly 13% more likely to live with other children. Introducing the other controls has almost no effect on the sex difference in living with own children (although in Model III, before controlling for marital status, men are only 84% as likely to live with own children). Once these factors are controlled, however, the gender difference in living with other children is reduced considerably, with men about 4% more likely to live with such children than women. The pattern of effects for living with own children closely parallels that for living with any children, but this is not the case for living with other children. There is a sharp and continuous decline over historical time for this category, unlike the baby boom induced pattern of fall, rise, and then fall for own children. The relative odds of living with other children in the most recent decades (1970–1990) are only 25% of those of the period 1880–1900. The effects of race, farm residence, and education are also sharper for living with other children than for living own children, and, of course, there are opposite effects of being married, with the married much more likely to live with own children, and less likely to live with other children. The only somewhat puzzling result is that for age. Living with own children relative to living with no children is sharply age graded, with younger adults much more likely to do so than older adults. When we consider living with

Model I


Model II

Model III


Table 2. Multinomial Logistic Regression Coefficients (Odds Ratios) Predicting Living with Own and Other Children: United States, 1880–1990. Model IV

Own child Non-own child Own child Non-own child Own child Non-own child Own child Non-own child vs vs vs vs vs vs vs vs No child No child No child No child No child No child No child No child







0.302 0.453 0.109 0.017

3.009 0.687 0.773 0.960*

0.299 0.442 0.103 0.016

2.981 0.677 0.746 0.934

0.441 0.337 0.070 0.012

2.525 0.837 1.058** 1.183

0.560 0.891 0.487

0.523 0.469 0.186

0.641 0.937 0.664

0.589 0.353 0.245

0.559 0.738 0.620

0.622 0.389 0.248









1.228 0.695

2.818 0.857

1.172 0.678

2.784 0.824



Note: * p < 0.05, ** p < 0.0005, and all the rest are significant at p < 0.0001.


Sex (base = Female) Male 0.898 1.125 Age Group (base = 30–39) 18–29 40–49 50–59 60–74 Year (base = 1880–1900) 1920–1940 1950–1960 1970–1990 Race (base = White) Black Farm Status (base = Non-farm) Farm Education (base = Low educ.) Not Available Better Education Marital Status (base = Never-married and others) Married and spouse present

Men’s Flight From Children in the U.S.: A Historical Perspective


other children, in contrast, while those aged 20–29 are more than twice as likely to do so as older adults, there is little difference among those of other ages, with only a slight increase at “grandparental” ages (50 + ). We also examined these models separately by sex (data not presented), but these analyses, like those just presented, do little to change the conclusions that can be drawn from the simple examination of Figs 1 and 2. The male process of coresidential parenting differs little from that of females. Parenting one’s own child is more compactly age-graded among women than men, reflecting the greater likelihood that men may have more than one wife and the greater age range over which men can father children. Marriage is similarly linked to residence with an own child, although because there are a larger number of unmarried women with children, the effect is much greater for men than for women.

CONCLUSION This paper presented an analysis of change over the period 1880 to 1990 in the proportions living with children for both men and women. We distinguished trends in living with own versus non-own children and controlled for other factors likely to have changed over time. The most important finding of this study is that the decline in living with children is really much more gradual than studies using data for recent decades reveal, and has been nearly equally characteristic of men and women. Our results agree with recent research that women were certainly more likely to live with children than were men in 1990 (King, 2000), but that has been the case throughout most of the period. What may have made this seem more extraordinary to contemporary eyes is that, with the decline in the number of adults living in households containing children, women’s single parenthood had become more visible. The fact that the current difference has expanded from the near equality of the baby boom years may also be a contributing factor. Again, the baby boom presents a false basis for comparison. Perhaps the most dramatic element in men’s “flight from children,” however, is the decline in living with non-own children. The rise in separate living and the decline in fertility that leaves few late adolescents with younger siblings have together removed what had been a substantial number of men from children’s lives. This, too, is a pattern that is characteristic of women, as well. Other changes underway during these 110 years actually contributed little to our understanding of the decline in parenthood. The move off the farm clearly contributed, as has the decline in the proportions married. The increase in education, however, is not so clearly linked, although the crudeness of our



measures of education for the entire period suggests that we need to look in greater detail at the recent decline, when we can use more detailed measures of education. We also need to expand this analysis. The parenthood patterns of the foreign born play a predominant role both at the beginning and the end of the century and need to be untangled. Further, for the years 1940, 1950, and 1990, reasonable data on stepparenthood are available on the IPUMS. As this is a role that is particularly important – and problematic – for men, it is unfortunate that we do not have better census data in the most recent period for studying this growing phenomenon.

NOTES 1. Unfortunately, there is no consistent way to distinguish own from stepchildren in U.S. censuses prior to 1990. The census relationship category was “child of head” and most subsumed their stepchildren under this category. The IPUMS has generated variables indicating particularly egregious cases likely to be stepchildren, based on age, or sometimes differences in last names, but differences in individual censuses are great. 2. There is a wide-spread feeling that patterns of custody in the case of divorce have changed dramatically over this period (e.g. LaRossa, 1997), with divorced men routinely given custody in the nineteenth century. We examined this question by looking at changes in the proportions of divorced men living with children over time (data not presented). At no time were divorced men more likely to be living with children than divorced women, with coresidential parenthood always at least twice as high among women than among men, although the overall gender difference might reflect the tendency of never-married mothers to report themselves as divorced in the census. However, there was a massive increase in the gender difference during the middle of the twentieth century, when it reached a peak of nearly 6 : 1. In the past 40 years, however, there has been a sharp decline, with divorced women again about twice as likely as divorced men to be coresidential parents. This might reflect at least in part the growth in cohabitation with partners who have children among divorced men.

ACKNOWLEDGMENTS Paper presented at the annual meetings of the Population Association of American, Los Angeles, CA, March 25, 2000. Support for this research was provided by NICHD Grant HD35527-01, “Race and Parental Roles of Men and Women, 1880–1990,” by the NICHD Network on Family and Child Well-Being and by an NICHD Population Center grant to Brown University. We thank Steven Ruggles for his tireless effort to create the Census IPUMS files that made this work feasible and for his suggestions about our analysis of these data; we also thank Zewdu Woubalem for his careful work on the tabulations.

Men’s Flight From Children in the U.S.: A Historical Perspective


REFERENCES Bumpass, L., & Raley, R. K. (1995). Redefining single-parent families: Cohabitation and changing family reality. Demography, 32, 97–109. Bumpass, L., Raley, R. K., & Sweet, J. (1995). The changing character of stepfamilies: Implications of cohabitation and nonmarital childbearing. Demography, 32, 425–436. Cherlin, A. (1992). Marriage, Divorce, and Remarriage (rev. and enlarged ed.). Cambridge, MA: Harvard University Press. Eggebeen, D., & Uhlenberg, P. (1986). Changes in the organization of men’s lives: 1960–1980. Family Relations, 34, 251–257. Eggebeen, D. (1999). The changing course of fatherhood: Men’s experience with children in demographic perspective. Paper presented at the annual meeting of the Population Association of America, New York, NY. Juby, H., & LeBourdais, C. (1998). The changing context of fatherhood in Canada: A life course analysis. Population Studies, I, 163–175. King, R. B. (2000). Time spent in parenthood status among adults in the United States. Demography, 36, 377–385. Kobrin, F. (1976). The fall in household size and the rise of the primary individual in the United States. Demography, 13, 127–138. Kramarow E. (1995). Living alone among the elderly in the United States: Historical perspectives on household change. Demography, 32, 335–352. LaRossa, R. (1997). The Modernization of Fatherhood. Chicago, IL: University of Chicago Press. Magnuson, D. L., & King, M. L. (1995). Comparability of the Public Use Microdata Samples: Enumeration Procedures. Historical Methods, 28, 27–32. Michael, R., Fuchs, V., & Scott, S.. (1980). Changes in the propensity to live alone: 1950–1976. Demography, 17, 39–56. Pampel, F. (1983). Changes in the propensity to live alone: Evidence from consecutive crosssectional surveys, 1960–1976. Demography, 20, 433–447. Robinson, J. G. (1988). Perspectives on the completeness of coverage of population of the United States decennial censuses. Paper presented at the annual meeting of the Population Association of America, New Orleans. Ruggles, S. (1996). Living arrangements of the elderly in the United States. In: T. K. Hareven (Ed.), Aging and Intergenerational Relations: Historical and Cross-Cultural Perspectives (pp. 254–263). Berlin, Germany: de Gruyter. Sandefur, G., & Tuma, N. B. (1987). Social and economic trends among the aged in the United States, 1940–1985 (Discussion Paper No. 849-87). Madison, WI: Institute for Research on Poverty. Santi, L. (1990). Household headship among unmarried persons in the United States, 1970–1985. Demography, 27, 219–232. Ruggles, S., & Sobek, M. (1997). Integrated Public Use Microdata Series: Version 2.0. Minneapolis: University of Minnesota, Historical Census Projects.

This Page Intentionally Left Blank



Sandra L. Hofferth and John F. Sandberg ABSTRACT The purpose of this paper is to examine changes between 1981 and 1997 in how a representative sample of American children spends their time on a weekly basis, focusing on overall differences in time use. We also examine how the time of specific children varies depending on the age and gender of the child, presence of and employment status of parents, the number of children in the family, and the level of parental education. Data come from the Time Use Longitudinal Panel Study, 1975–1981 and the 1997 Child Development Supplement to the Panel Study of Income Dynamics. Results show a pattern of increased time in structured activities such as school, day care, sports, and art activities, and reduced time in unstructured play, television viewing, visiting, and passive leisure. While a few of these changes are related to increased maternal employment, most tend to be related to demographic characteristics such as increased education and reduced family size.

INTRODUCTION In the past twenty years, American families have experienced major demographic and economic changes. Such changes include increased maternal employment, increased rearing of children by single parents, the completion of more years of schooling by mothers and fathers, and declines in family size. At Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 193–229. Copyright © 2001 by Elsevier Science Ltd. All rights of reproduction in any form reserved. ISBN: 0-7623-0776-5




the same time there has been increased concern about children’s development of the skills needed to successfully negotiate childhood and the transition to adulthood. The ways children spend their time are an indicator of the values parents place on the attainment of various skills and contexts for learning which may have long-term consequences for their development (Larson & Verma, 1999). These values are likely to have changed over time along with the demographic characteristics. In spite of the great interest of educational and health institutions in the well-being of children and the demographic changes described above, almost no previous research has been conducted to see exactly how American children’s lives have changed over the past decade and a half (Robinson & Godbey, 1997). An understanding of how children spend their time at present, changes over time, and variations by family characteristics is a critical first step in studying demographic change and child well-being. Demographic Context Among the most important of the demographic changes is that of increased maternal employment. In 1997 two-thirds of the mothers of preschool children were working, compared with 47% in 1980. Because of increases in maternal employment, the enrollment of children in early education and care programs has risen (Hofferth, 1995). Most studies have found that employed mothers spend less time caring for children than those not employed (Nock & Kingston, 1988; Robinson & Godbey, 1997). However, the differential may be declining. A study using U.S. data from the 1920s to the 1980s reports that parental time caring for children rose rather than declined over the period, in spite of increased maternal employment (Bryant & Zick, 1996). Another study using recent data shows an increase in the time employed mothers spent in child care as a primary activity between 1965 and 1985 (compared with a decrease for nonemployed mothers) (Robinson & Godbey, 1997). Changes in the ways both parents allocate time and their balance of decision-making power may have important impacts on children (Nock & Kingston, 1988; Bryant & Zick, 1996; McLanahan, 1988; McLanahan & Bumpass, 1988). Increased maternal employment has also led to a shift from home production of goods and services for a family’s own use to their purchase (Jacobs, Shipp & Brown, 1989). Between 1965 and 1985, the shopping time of both employed and nonemployed women increased, with the largest increase (26%) among employed women (Robinson & Godbey, 1997). This is consistent with the increased purchase of goods and services outside the home. Between 1992 and 1997 alone, service industries created more than half of all new jobs (U.S. Bureau of the Census, 2000).

Changes in American Children’s Time, 1981–1997


A second important demographic trend is the rise in divorce. Today it is estimated that two-thirds of first marriages will end in divorce, more than twice as many as two decades ago (Martin & Bumpass, 1989). Almost one-quarter of white and more than half of black children are currently born to unmarried mothers. These two changes lead to a very high proportion of children – 60% according to recent estimates – who are expected to spend at least part of their childhood with only one parent (Hernandez, 1993). Parental divorce and single parenthood have been identified as risk factors for socioemotional problems, school failure, school drop-out, early parenthood, and becoming a female family head (Hetherington, 1989; McLanahan, 1985). In 1997 28% of children were living with only one parent (U.S. Bureau of the Census, 1998). Many are concerned about the ability of single mothers and fathers to adequately supervise their children and keep them focused on their schooling. Single mothers have been found to spend three fewer hours per week in child care than married mothers (Robinson & Godbey, 1997). Thus we might expect differences in how children in two-parent and single parent families spend their time. In contrast to the sociodemographic changes discussed so far, increased parental schooling levels should raise parental investments in their children. Besides monetary advantages, education affects the values, knowledge, experience, time allocation, and aspirations that parents bring to childrearing. Previous research has shown that high-income mothers invest more heavily in their children by spending more time with them in educational activities (Leibowitz, 1974a, b, 1977). About one-fifth of mothers had completed some college in the 1950s, compared with two-fifths in the 1980s. Fewer than one in ten black mothers and fathers had completed some college in the 1950s. In the 1980s that percentage had climbed to one out of three black fathers and one out of four black mothers (Hernandez, 1993). These changes suggest that a larger proportion of children may be better off today than several decades ago. This trend may be offset by the increased concentration of immigrant children in American schools. In contrast to the native-born, immigrants often have lower educational levels. They bring diverse cultural habits and languages and may have different preferences for their own and their children’s activities. Finally, reduced fertility might influence children in a positive way. The number of children born to the average woman has declined substantially. Each woman born after 1945 is expected to bear about two children by the time she completes childbearing (U.S. Bureau of the Census, 1984; U.S. Bureau of the Census, 1988). In contrast, the average woman born between 1930 and 1939 had about three children. Children in small families consistently perform better academically than those in larger families; it has been argued that this is due to



the greater time and resources that can be devoted to the former (Blake, 1989; Zajonc & Markus, 1975). One explanation Robinson and Godbey (1997) offer for finding decreased parental time for nonemployed mothers is smaller family size. When results are adjusted for family size, parental time per child actually increases over time. The aforementioned economic and demographic changes have also had profound effects on institutions that are charged with caring for children – child care providers, preschool programs, and schools. As the time children spend out of the home increases, how out-of-home time is spent will be crucial to child well-being. If they play less at home, do they play more somewhere else? While this paper does not describe activities in out-of-home settings, we gain at least an idea of the relative distribution of time in the home and in other outof-home environments. In other research we will examine the activities occurring in those other environments. We discuss the implications of trends in activities at home in the context of all children’s environments. This paper also does not focus upon the implications of time allocations for children’s well-being. Although most research has focused on cognitive scores, recent research (Eccles & Barber, 1998) has found that determining how well children do is not a simple matter of examining time spent in learning and achievement test scores. In fact, children who do well are those who are involved in a variety of activities during their school years and who are able, consequently, to develop competencies and skills in those areas (Larson & Verma, 1999). Hofferth and Sandberg (2001) find reading to be associated with higher achievement test scores and sports participation with both less aggressive behavior and higher test scores. How Might Children’s Activities Change? The major demographic changes discussed above include increased maternal employment, increased parental education, the reduced fraction of children living in two-parent families, and reduced family size. How might we expect changes in maternal employment to affect children’s activities? First, given that children need care, we expect an increase in the time children spend in school and in child care. The latter includes preschool programs, day care, and babysitters for young children and before- and after-school programs for school-age children. Second, the lives of young children may be as tightly scheduled as the lives of their working parents. Thus, we may see an increase in structured activities. By structured activities we mean activities that generally are scheduled at a certain time and place on a regular basis, such as team sports, music and dance.1 These may come at the expense of unstructured

Changes in American Children’s Time, 1981–1997


activities such as free play and outdoor activities. As mothers spend more time in the work force, other activities are bound to decline. The time women spend in household work has declined significantly over the past several decades (Gershuny & Robinson, 1988). While children and other family members may contribute more time to offset this decline, housework time may simply decline overall. Maternal employment time may increase at the expense of informal family time such as time spent eating meals, with fast-food and carry-out substituting for home-cooked meals. Other discretionary activities, such as church attendance and visiting, and time spent reading with or helping children do homework may also decline. Even time children spend sleeping may suffer if parents’ schedules become more important than children’s. One study (Bianchi & Robinson, 1997) failed to find an association between children’s time spent reading, studying or doing household work and maternal employment net of other factors, but further examination is needed. Given the increased schooling of parents and greater encouragement for children’s reading by more highly educated parents, today’s children may spend more time reading and studying. Bianchi and Robinson (1997) found children spent more time reading or being read to in households with more educated parents relative to those in households with less educated parents. On the other hand, if children spend less time at home and reading tends to be a home-based activity (Hofferth & Jankuniene, 2001), even private reading time may suffer. It is unclear how television watching may have changed over time. Greater parental education may lead to less television viewing (Robinson & Godbey, 1997). Bianchi and Robinson (1997) found an inverse relationship between parental education and children’s television viewing, suggesting that television viewing may decline as average levels of schooling rise. If parents substitute educational programs for noneducational ones, however, it may lead to more viewing. The positive effects of increased parental schooling may be offset by the increased proportion of children living in single parent families. With less total parental time available for their children, we expect a number of activities to suffer, including mealtimes and free play. Single mothers spend less time in meal preparation and household cleaning than married mothers (Zick & Allen, 1996). Under these circumstances, whether children contribute more or whether everyone will do less is not clear. Television watching, an inexpensive entertainment option, may increase. The increase in nontraditional families, including single working parent and dual earner families may be associated with declines in traditional family activities such as church attendance and participation in church-related activities. Previous research (Bianchi & Robinson, 1997) failed to find a relationship between the number of parents



and the time children spend reading, watching television, studying, and doing household work. While representative of California, that study was not nationally representative. Activities of a sample of California children may differ from those in other parts of the U.S. either because of climatic differences or because of cultural differences in the population. As family sizes decline, on the one hand children should reap the advantages of more parental time, which may translate into increased time spent in educational and leisure time activities. On the other hand, having fewer children also means that the availability of siblings to play with and help with homework declines. The resources to raise a large family need to be spread among more children; consequently, we expect that children in large families will be less likely to participate in costly activities and more likely to participate in free or low-cost activities. There will be more household work in a large family and each child may be expected to do more. Consistent with this expectation, Bianchi and Robinson (1997) found that children in large families spent more time in household work than children in small families. Significance and Focus of the Paper The purpose of this paper is, first, to examine how American children spent their time in 1981 and 1997, focusing on overall differences in time use by child age and gender in a set of 18 key activities of children between ages 3 and 12. This research focuses upon children’s activities during the school year, and does not represent the activities of children during the summer months. Second, we examine how the time of American children varies depending on the presence of and employment status of parents, the number of children in the family, and the level of parental education. How these factors affect the time children spend in educational activities such as studying and reading, in family activities such as household chores, or in leisure activities such as watching television, playing sports, and visiting with others will help us understand how demographic changes might have produced the changes in activities we described earlier. In addition, if we find overall differences among children of different characteristics in how they use time, differences that relate to later health and achievement, we gain important policy and programmatic levers to use in improving such outcomes. Finally, we discuss the implications of our findings for parents, policy makers, and for research. Parents have little to go on in making important decisions regarding their daily activities. Policy-makers likewise have been operating in the dark regarding the educational activities of children.

Changes in American Children’s Time, 1981–1997


METHODS AND DATA The measurement of children’s time is an issue that a number of researchers have wrestled with. In this section we describe measurement issues, then describe our two sources of national information on children’s time. Measurement of Children’s Time There are several ways of assessing how much time and in what activities children engage. The most accurate way to collect such data would be through observation. However, such methods are costly, intrusive, and limited in the amount of a day that can be covered. Another accurate way to collect information is by time sampling, in which respondents write down the activity they are engaged in whenever a beeper sounds. This methodology is also costly and intrusive. The most common method in survey research is to ask parents directly how much time they spend in certain activities, such as reading to their child. While simple and widely used, this method is known to be biased. First, it is subject to social desirability bias. Parents will report more time spent on desirable activities (such as reading with the child) than on less desirable ones. Second, there is no baseline against which to check consistency, validity, or reliability. Times have been shown to be quite inaccurately reported in some cases (Juster & Stafford, 1985; Marini & Shelton, 1993). In contrast, substantial methodological work has established the validity and reliability of data collected in time-diary form (Juster & Stafford, 1985). The present study is based upon complete 24-hour time diaries collected from parents and their 3–12 year old children in 1981 and 1997. Two Data Sets Data for this study on children’s time use come from 24 hour time diaries used in two surveys of the American population; the 1997 Child Development Supplement to the Panel Study of Income Dynamics, and from a small 1981 follow-up to the 1976–77 Study of Time Use in Social and Economic Accounts2, The Time Use Longitudinal Panel Study, 1975–1981,3 both collected by the University of Michigan using a similar methodology. The Time Use Longitudinal Panel Study, 1975–1981 Prior to 1997, the only nationally representative data on U.S. children’s time use had been collected in 1981 by the University of Michigan from a sample of 620 families, with diaries from 620 respondents, 376 spouses, and 492



children aged 3–17 years old as part of the Time Use Longitudinal Panel Study, 1975–1981 (Timmer, Eccles & O’Brien, 1985). Eligible households for this study were the 920 heads and wives interviewed in at least three out of four waves in the 1975–76 Study of Time Use in Economic and Social Accounts. While adults were interviewed in 1981 up to four times, from most children only two time diaries, one for a school day and one a non-school-day, were obtained. Child diaries were obtained during the school year only. Many studies continue to utilize estimates based on these data for lack of a recent alternative (Task Force on Youth Development and Community Programs, 1992). Until 1997 no national time diary data were available to document changes since 1981, though results of a California study have been reported (Bianchi & Robinson, 1997). For this paper, we use the 1981 data to build a child level file with time diary measures and demographic characteristics. Individual child records are weighted by the product of 1981 sample attrition weight (which adjusts the sample for loss of respondents from the 1975–1976 study), a post stratification factor (by race and education of the head of the household) derived from comparison to the March 1981 Current Population Survey4 to make the sample more comparable to the U.S. population in 1981, and a sub-selection weight that adjusts for the probability that a child in a given household was sampled, and also for non-response of sampled children. The Child Development Supplement to the Panel Study of Income Dynamics The study sample was drawn from the 1997 Child Development Supplement (CDS) to the Panel Study of Income Dynamics (PSID). The PSID is a 30-year longitudinal survey of a representative sample of U.S. men, women, children, and the families in which they reside. In 1997, the PSID added a refresher sample of immigrants to the U.S. (since 1968) so that the sample represents the U.S. population in 1997. When weights are used, the PSID has been found to be representative of U.S. individuals and their families (Fitzgerald, Gottschalk & Moffitt, 1998a). With funding from the National Institute of Child Health and Human Development (NICHD), data were collected in 1997 on up to two randomly selected 0–12-year-old children of PSID respondents both from the primary caregivers and from the children themselves. The CDS survey period began in March 1997 and, with a break from mid-June through August, ended on December 6, 1997. As in the 1981 study, child interviews took place only during the school year. Interviews were completed with 2,380 households containing 3,563 children. The response rate was 90% for those families regularly interviewed in the core PSID and 84% for those contacted the first time this year for the immigrant refresher to the sample, yielding a combined

Changes in American Children’s Time, 1981–1997


response rate for both groups of 88%. Post-stratification weights based upon the 1997 Current Population Survey are used to make the data nationally representative. The individual level child file used in this analysis is weighted by the product of the core PSID family weight, a post-stratification factor (by race and education of household head) based on comparison to the 1997 Current Population Survey, and again, a sub-selection weight that adjusts for the probability that a child in a given household was sampled and also for nonresponse of sampled children. Time Diaries As did the 1981 Study of Time Use in Social and Economic Accounts, the Child Development Supplement collected a complete time diary for one weekday and one weekend day for each child age 3–12 in the family. The time diary, which was interviewer-administered to parent and child, asked several questions about the child’s flow of activities over a 24-hour period beginning at midnight of the designated day. These questions ask the primary activity that was going on at that time, when it began and ended, and whether any other activity was taking place. In the coding process, children’s activities are classified into ten general activity categories (paid work, household activities, child care, obtaining goods and services, personal needs and care, education, organizational activities, entertainment/social activities, sports, hobbies, active leisure, passive leisure), and further subdivided into three-digit subcategories (such as parent reading to a child) that can be recombined in a variety of ways to characterize children’s activities. For comparison purposes, the primary activities of children 3 to 12 were classified into the 18 major categories used by Timmer and colleagues in the early 1980s (Timmer et al., 1985). These categories were expanded to separate shopping from household work and to separate day care from school. Youth groups were also distinguished from the broader “visiting” category. Time spent traveling for the purposes of engaging in a specific activity is included in that category. Secondary activities are not measured. For example, time spent doing housework with the television on where housework was the primary activity is not counted as time “watching television.” Thus, some activities that are often secondary may be underestimated.5 Given that many activities are occasional, we would not expect all children to engage in most of these on a daily basis. However, we want to abstract from this to describe the activities of American children in general. Since not all children do every activity each day, the total time children spend in an activity is a function of the proportion who engage in the activity and the time those participating spend in it. An estimate of weekly time is computed by



multiplying weekday time (including those who do not participate and have zero time) by five and weekend day time by two, after removing children who do not have both a weekend and weekday diary.6 Selecting children ages 3 to 12 with two diaries, sample sizes were reduced to 2,148 cases; missing data on some of the demographic variables further reduced the sample to 2,132. Time use data are by nature quite sensitive to errors resulting in extreme or unusually large values for a particular activity; an error in reporting not only has ramifications for estimates of time spent in that particular activity, but in all other activities in the day as well. Such problems are less troublesome in our 1997 data due to the large sample size, but potentially quite problematic for the 1981 data. In addition, imputing daily values to a hypothetical week as we do here exacerbates these problems. Though care was taken during the data coding and cleaning stage to assure that all activities were temporally contiguous and that they summed to 24 hours for a given day, there are some outlying values in major activities that deserve special attention. Time use estimates were analyzed for both years for extreme values, and where the authors deemed there to be a substantial probability of data error, that case was removed from the forthcoming analyses completely.7 After removing these cases, the analytic sample sizes for 1981 and 1997 were 222 and 2,119, respectively. Robinson and Godbey (1997) distinguish among contracted time (work, school), committed time (household and child care obligations), personal time (eating, sleeping, personal care), and free time (everything else). We generally use this model with some small changes because we are concerned with children, not adults. Since they have to be in school but don’t have to work, we treat school rather than work as children’s “contracted” time. Personal care time is time spent eating, sleeping, and caring for their personal needs. Few children have “committed” time; we include household work along with free time. Young children simply accompany their parents in this activity. Free or discretionary time, therefore, consists of household work and shopping, studying, church attendance, youth groups, visiting, sports, outdoors activities, hobbies, art activities, which includes dance and music lessons, playing TV watching, reading, household conversations, and other passive leisure (which includes going to movies and sports events as a spectator). Measurement of the Demographic Variables The demographic variables used to analyze the impact of family factors on the time of children 3–12 include age of child, gender of child, number of parents (one or two), employment of wife (not employed vs. employed), education of

Changes in American Children’s Time, 1981–1997


mother (no college vs. some college), and number of children (one or two vs. three or more).

RESULTS Number of Activities The results presented first are based upon a total of 2,119 children between age 3 and age 12 whose parents had completed time diaries for them (or with them) for two days in the previous week in 1997 and 222 children with two diaries in 1981. The larger the number and variety of activities the more valid the data. That is, diaries with more detail are assumed to represent children’s days better than those with fewer activities. In addition, the degree of detail should be comparable between 1981 and 1997. We found that children were reported to have 23 activities on an average weekday and 24 activities on an average weekend day in 1997 compared with 28 activities on an average weekday and 25 activities on an average weekend day in 1981. Thus the earlier survey documented more activities. However, the number of different (nonrepeated) activities reported, the “variety of activities,” averaged 13 on a weekday or on a weekend in both years. Therefore, the two years seem reasonably comparable. In addition, the amount of time that was not ascertained (1%) is comparable across the two years. An Overview of Changes in Children’s Time between 1981 and 1997 Table 1 lists detailed categories and the percentage of children engaging in each primary activity in their estimated week. From this table it is clear that all children sleep, eat, and engage in or receive personal care (bathing, etc). Almost all play and watch television but few engage in market work or hobbies. There are some significant changes: Household work showed both increases and decreases. The proportion who do the basic household tasks such as cooking and cleaning has declined 11.5%, from 78 to 69%, while the proportion who shop has increased 21%, from 43 to 52%. There were declines in participation in many activities. The proportion studying declined 15%, from 54 to 46%; attending church declined 30%, from 37 to 26%; participating in youth groups declined 34%, from 38 to 25%; visiting declined 31%, from 75 to 52%; and participating in other passive leisure declined 24%, from 66 to 50%. Increases in participation occurred in some cases. The proportion


Table 1. Percentage of Children Participating in 21 Daily Activities, 1981–1997, by Age (3–12-year-olds). 1981 61 0% 54% 39% 100% 100% 100% 53% 26% 35% 36% 66% 41% 18% 0% 20% 100% 98% 25% 28% 69% 2%

* p < 0.05, ** p < 0.01, *** p < 0.001

665 0% 63% 60%*** 100% 100% 100% 51% 17%* 24% 21%*** 50%* 74%*** 19% 2% 35%** 98% 98% 53%*** 37% 53%** 24%***

1981 60 3% 79% 36% 98% 100% 100% 88% 34% 37% 33% 69% 69% 13% 2% 20% 93% 96% 37% 41% 53% 3%

Age 6–8 1997 602 2% 65%** 50%** 100% 100% 100% 93% 55%*** 24%** 25% 48%*** 75% 14% 2% 24% 91% 96% 43% 31% 45% 12%**

1981 101 12% 92% 50% 100% 100% 100% 88% 82% 37% 42% 83% 65% 21% 2% 17% 90% 97% 32% 37% 73% 2%

Age 9–12 1997 851 3%*** 78%*** 47% 100% 100% 100% 91% 62%*** 28%** 27%*** 56%*** 76%** 17% 3% 21% 88% 94% 34% 28%* 51%*** 4%

1981 222 6% 78% 43% 99% 100% 100% 78% 54% 37% 38% 75% 59% 18% 2% 19% 93% 97% 32% 36% 66% 2%

All Ages 1997 2119 2%*** 69%*** 52%** 100%* 100% 100% 79% 46%** 26%*** 25%*** 52%*** 75%*** 17% 3% 26%** 92% 96% 42%*** 32% 50%*** 13%***


N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare

Age 3–5 1997

Changes in American Children’s Time, 1981–1997


participating in sports increased 21%, from 59 to 75%; participation in art activities increased 37%, from 19 to 26%; reports of reading increased 31%, from 32 to 42%; and day care attendance increased five-fold, from 2 to 13%. Weekly Hours of Nondiscretionary Time Participation is only part of the picture; increases or declines in time spent may outweigh declines in participation. To examine the total change in time due to changes in both participation and time spent in activities, we focus on hours from our weekly estimates. Table 2 shows the average weekly hours and minutes 3 to 12-year-old children spent in major activities, from the 1997 and 1981 data. Consistent with the increase in maternal employment over the period, the time spent in school increased about 2 hours per week, or 25 minutes per school day, from 24 hours and 45 minutes to 26 hours and 48 minutes in an average week. Day care time increased from about 14 minutes to almost 3 hours per week over the period, due both to increased participation and to greater time spent. Personal care time also increased over the period, from just over 6 hours a week to 8 hours per week. The increase in personal care time, which besides bathing includes preparation to go places, packing up, and getting dressed, is consistent with a larger proportion of time spent away from home. The more time spent away from home, the more time it takes to get ready to go. Time spent sleeping increased about 1 hour and time spent eating decreased about 1 hour. While the increase in time spent sleeping is not consistent with an increased time pressure, the decline in time spent eating is. Other research suggests that families are reducing the frequency with which they sit down together to share a family meal (Kinney, Dunn & Hofferth, 2000). With many activities for children conflicting with the dinner hour, families are often unable to eat together at home. They may eat on the run or in shifts, which cuts down on meal time. Consistent with this story, we show later on that time spent eating has dropped for children with employed but not with nonemployed mothers. Subtracting time spent eating, sleeping, in personal care, school, and child care from the 168 total weekly hours in 1981 and then in 1997, our first finding is that children’s free time as a proportion of total weekly time declined from 1981 to 1997, from 56.5 hours in 1981 to 49 hours in 1997, a decline of approximately 12% (Fig. 1). If you remove sleep and make it relative to nonsleep hours (Larson & Verma, 1999), it amounts to a decline from 58 to 52% of the child’s day. The decline in free time (7.5 hours per week) is largely due to increased time spent in school and child care (Table 2). But it is also due to increased personal care time.


Table 2. Weekly Time Children Spent in 21 Activities, 1981–1997, by Age (3–12-year-olds). 1981 61 0:00 2:09 2:35 6:18 9:43 77:19 14:30 0:25 2:30 0:49 2:58 1:31 0:13 0:00 0:28 25:50 15:14 0:29 0:37 2:59 0:10 1:12 99%

* p < 0.05, ** p < 0.01, *** p < 0.001

665 0:00 2:20 3:44* 8:32*** 9:24 76:11 12:05 0:36 1:03*** 0:21** 3:04 4:08*** 0:37* 0:05 1:12*** 17:21*** 13:52 1:24*** 0:48 2:35 7:30 1:07 99%

1981 60 0:27 2:49 0:59 6:13 9:08 70:04 27:52 0:52 1:36 0:41 3:40 6:01 0:28 0:03 0:21 14:58 15:55 0:59 1:07 1:58 0:12 1:37 99%

Age 6–8 1997 602 0:06 2:07 2:38*** 7:53*** 8:05** 70:49 32:46** 2:08*** 1:15 0:42 2:48 5:13 0:30 0:04 0:45*** 11:10** 12:54* 1:09 0:30* 1:33 1:33*** 1:23 99%

1981 101 0:43 5:18 1:57 6:21 8:13 65:36 29:02 3:22 2:24 0:56 3:48 4:51 0:46 0:03 0:22 7:24 20:01 1:03 0:53 3:24 0:18 1:14 99%

Age 9–12 1997 851 0:11* 3:42*** 2:24 7:53*** 7:23** 67:24** 34:03*** 3:41 1:20** 0:49 2:40** 6:33** 0:36 0:07 0:54** 8:54* 13:36*** 1:14 0:27* 2:19* 0:24 1:24 99%

1981 222 0:27 3:46 1:52 6:18 8:52 70:01 24:45 1:53 2:13 0:50 3:32 4:15 0:32 0:02 0:23 14:30 17:35 0:53 0:53 2:53 0:14 1:20 99%

All Ages 1997 2119 0:06** 2:49*** 2:53*** 8:05*** 8:13*** 71:07* 26:48** 2:16* 1:13*** 0:38** 2:50** 5:25 0:35 0:05 0:57*** 12:12*** 13:29*** 1:16** 0:35*** 2:11*** 2:57*** 1:18 99%


N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for

Age 3–5 1997

Changes in American Children’s Time, 1981–1997


Fig. 1. Percentage of Weekly Time Spent in Nondiscretionary and Discretionary Activities.

Discretionary Time If there were no other changes in how time is spent, because of the decline in discretionary time of 12%, we would expect an average decline of 12% in



discretionary activities. In fact, there were other shifts in the use of free time from 1981 to 1997 which differ from the overall change in discretionary time (Fig. 2). Consistent with our hypotheses, the time children spent playing declined 16% between 1981 and 1997, only slightly more than the decline in free time overall. Participation itself did not change; almost all children play. Television viewing as a primary activity declined by 23%, more than the decline in free time. This was not due to a decline in the proportion who watch; almost all children watched television during both 1981 and 1997. The time spent attending church, participating in youth groups, passive leisure, and other household conversations also declined significantly and by more than discretionary time. These are due partially to declines in participation. Of the household work categories, housework declined, while shopping increased. These are partially but not totally a function of changes in participation. The time spent in hobbies, organized sports, and arts activities increased. Studying and reading both increased over time. Changes in studying are due to increases in time spent, since participation actually declined. Changes in reading are due both to increased participation and to increased time. As a result, we see the composition of children’s free time changing. Less time is now spent in unstructured activities such as playing and passive activities such as television viewing and more time is spent in structured activities such as art and sports and in educational activities such as studying and reading (Fig. 3, Fig. 4). Our overall hypothesis that activities requiring substantial parental input would decline is supported. Church attendance and time in youth groups, visiting, and other household conversations declined. The press for time and lowered home production are reflected in lower children’s time in household work and greater time spent shopping. Support is provided for our hypothesis of an increase in structured activities such as the arts (lessons) and sports and a decline in unstructured play. Support is also shown for the hypothesis that there would be increases in activities such as studying and reading. However, we cannot tell from this table whether these are a result of increased maternal employment or some other demographic change between 1981 and 1997. Maternal Employment as Reason for Change To what extent changes in time use are due to changes in maternal employment can be determined by examining whether similar changes occurred among nonworking and working mothers between 1981 and 1997. In fact, only a few of the above-described changes in time use are linked to changes in maternal employment. The decline in time spent eating is related to employment; this decline across the period did not occur among children of mothers who were

Changes in American Children’s Time, 1981–1997 209

Fig. 2. Changes in Weekly Hours by Age from 1981 to 1997 (All Ages).



Fig. 3. Percentage of Discretionary Weekly Time in Activities, All Ages, 1981.

not employed and time spent eating meals was lower among children of employed compared with nonemployed mothers in 1997 (Table 3). As anticipated, the increase in daycare time is related to employment; however, the increase in time spent in school is not. There was no change among children of employed mothers in school time; the change occurred among children of nonemployed mothers. Thus the increase in time spent in school is likely to be due to a preference for greater use of schools and school activities by families of nonworking mothers as well as by families of working mothers. While these data show a significant difference in the time children 3–12 spent in school in 1997 by maternal employment status, the difference is considerably smaller than it was in 1981. Other data show that, in fact, the use of preschool programs has converged among working and nonworking mothers over the past several decades (Hofferth, Shauman, Henke & West, 1998).

Changes in American Children’s Time, 1981–1997


Fig. 4. Percentage of Discretionary Weekly Time in Activities, All Ages, 1997.

Most other changes are not linked to maternal employment. Time spent in personal care increased among both groups and there is no difference in personal care between children of employed and nonemployed mothers in 1997. The decline in household work, the increase in shopping, the decline in time spent in church, and the increase in art activities occurred among children of nonworking as well as working mothers. Increases in sports participation were largest among children of nonemployed mothers as well, equalizing the sports participation of children of working and nonworking mothers in 1997. In sum, changes occurred for children of both employed and nonemployed mothers over this period. Described in cultural terms, these may be related to changing preferences for activities for children of all ages, not just as a form of



Table 3. Weekly Time Children (Ages 3–12) Spent in 21 Activities, 1981–1997, by Maternal Employment Status. Not Employed

1981 N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for





95 766 126 1267 0:14 0:03** 0:36 0:07*** 3:47 2:44** 3:44 2:54** 1:54 3:10** 1:49 2:41** 6:41 8:04*** 6:00 8:07*** 8:52 8:36 8:52 8:01*** 71:44 72:22 68:42 70:31** 20:35 25:37*** 27:52 27:12 1:46 2:14 1:57 2:18 2:28 1:20*** 2:00 1:09*** 0:37 0:32 0:59 0:39 3:44 2:29** 3:22 3:02 2:37 5:11*** 5:29 5:31 0:25 0:29 0:36 0:37 0:05 0:04 0:00 0:06 0:23 1:03** 0:23 0:54*** 17:19 13:21*** 12:22 11:28 18:09 14:36*** 17:09 12:53*** 0:51 1:20** 0:54 1:13 1:08 0:35*** 0:40 0:32 2:33 2:09 3:08 2:12** 0:01 0:33 0:24 4:30*** 1:55 1:17 0:52 1:12 99% 99% 99% 99% 168:00:00 168:00:00 168:00:00 168:00:00

Not Employed vs. Employed 1981 1997

** * ** ***

*** *** **

** *** ** ***

*** ***

* ***

* p < 0.05, ** p < 0.01, *** p < 0.001

child care but to achieve other goals families have for their children (Kinney et al., 2000). Because of these converging trends, the time use of children of working mothers was similar to the time use of children of nonworking mothers in 1997. There were a few differences, however. Compared with children of nonemployed mothers, in 1997, children of employed mothers spent less time shopping, sleeping, eating, playing, and watching TV, and more time in school and day care.

Changes in American Children’s Time, 1981–1997


Differences by Family Structure Increased single parenthood is the second major demographic trend that could potentially have influenced children’s time use over the past decade and a half. Most of the trends in children’s time are similar for those living with one parent and those living with two parents (Table 4). Trends for children in single parent families were not always statistically significant because of the small number of children living with one parent in 1981. One interesting difference is that the

Table 4. Weekly Time Children (Ages 3–12) Spent in 21 Activities, 1981–1997, by Number of Parents.

Two Parents 1981 N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for


Single Parent 1981


195 1570 27 463 0:27 0:05*** 0:18 0:09 4:11 2:52*** 0:42 2:42*** 1:57 3:08*** 1:11 1:57 6:36 8:00*** 4:06 8:26*** 9:05 8:19*** 7:20 7:55 70:01 71:17** 69:56 70:59 22:56 26:21*** 37:54 27:28*** 1:57 2:18 1:21 2:12 2:00 1:17*** 3:42 1:00*** 0:43 0:40 1:36 0:22*** 3:45 2:49** 1:54 2:50 3:49 5:19*** 7:26 5:41 0:35 0:35 0:01 0:32 0:02 0:06 0:00 0:01 0:26 1:00*** 0:00 0:47* 15:24 12:35*** 07:53 10:48 17:12 12:56*** 20:25 15:33 0:59 1:25** 0:03 0:44* 0:59 0:38*** 0:01 0:18* 3:09 2:22** 0:52 1:32 0:07 2:35*** 1:03 4:29 1:30 1:11 0:04 1:24 99% 99% 100% 99% 168:00:00 168:00:00 168:00:00 168:00:00

* p < 0.05, ** p < 0.01, *** p < 0.001

Two Parents vs. Single Parent 1981 1997

*** *** ***

*** * **

*** ** ** * *** ** ** *** ** ** ** ** **

** ***

* *** *** *** *** *** ***



time children of single parents spent in school declined (from a previously high level), while that of two parents rose, making the overall levels of enrollment similar in 1997. Another interesting difference is that time in sports rose for children of two parents but not for children of single parents. Because the latter was already at a high level, sports time converged for the two groups between 1981 and 1997. Finally, the decline in church time occurred for children in single parent and in two-parent families; their levels were closer in 1997 than in 1981. There are, however, some differences in levels of time spent in activities by family structure. Children living with one parent spend more time in day care due to the greater chance their mother is employed. As a result, their time at home is reduced as are almost all other activities. Compared with children living with two parents, children living with one parent spend less time eating, attending church, in youth groups, in hobbies, art activities, playing, reading, in household conversations, and in other passive leisure. One exception is that they spend more time watching television. There is no difference in the time children spend in household work, though they spend less time shopping. Presumably they have less money to shop or, alternatively, shopping on the part of parents becomes more intensive and less leisurely as other constraints increase. Our hypothesis that children spend more time assisting single mothers with household work was not supported. Maternal Education and Changes in Children’s Time Increased maternal education is the third key change of interest to have occurred over the past decade and a half. The data suggest the diffusion of cultural norms and values regarding expenditures of time that were formerly concentrated among families with an better-educated mother. Most trends were similar for children of mothers with no college compared to those with some college in terms of declines in time spent in housework, and increases in time spent shopping, in personal care, and in day care (Table 5). Also similar for both groups are increases in active leisure (sports) and declines in passive leisure (passive leisure, TV, household conversations). In general, these changes made the activities of children of less-educated mothers more similar to those of children of educated mothers. Even though more differences by maternal education in levels of activity time are observed in 1997 than they were in 1981, these are most likely due to the large sample size in 1997 compared with 1981. The major differences in children’s activities in 1981 by maternal education were participation in youth groups, art activities, television viewing, and reading. In 1981, children of mothers with some college spent more time in art and reading, and less time in TV than children of mothers with

Changes in American Children’s Time, 1981–1997


Table 5. Weekly Time Children (Ages 3–12) Spent in 21 Activities, 1981–1997, by Maternal Education.

No College 1981 N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for


Some College 1981


163 1048 59 980 0:30 0:02*** 0:15 0:10 3:36 2:46** 4:12 2:52*** 1:49 2:44*** 1:58 3:01* 6:18 8:11*** 6:17 8:01*** 8:50 8:09** 8:57 8:18 69:50 71:36** 70:31 70:45 24:29 27:01* 25:26 26:13 1:45 2:06 2:12 2:28 2:18 1:08*** 1:57 1:20 0:59 0:27*** 0:23 0:46 3:25 2:47 3:51 2:52 4:17 5:06 4:10 5:44* 0:25 0:29 0:50 0:39 0:02 0:03 0:01 0:07 0:15 0:45*** 0:44 1:10 14:40 11:55*** 14:02 12:27 18:30 15:15*** 15:02 11:40*** 0:44 0:57 1:17 1:35 0:47 0:29*** 1:07 0:38*** 2:37 1:48*** 3:37 2:35* 0:11 2:35*** 0:22 3:30** 1:35 1:30 0:38 0:57 99% 99% 100% 99% 168:00:00 168:00:00 168:00:00 168:00:00

No College vs. Some College 1981 1997


** ** **



** * * ***

* *

*** *** *** *** **

* p < 0.05, ** p < 0.01, *** p < 0.001

no college. In 1997, these same differences held and, in addition, children of mothers with some college also spent more time in market work, sports, outdoor activities, hobbies, household conversations, studying, other passive leisure, and day care. Such differences represent differences in values, certainly, but also may reflect differences in the ability to afford such activities. Passive leisure, which includes going to paid entertainment such as movies, for example, declined for both groups of children; however, levels of passive



leisure remained higher for children of college-educated compared with noncollege-educated mothers. One interesting shift in children’s activities is that, in 1981, children of mothers with some college were significantly less likely to be enrolled in youth groups. In 1997, in contrast, children with mothers with some college were more likely to be enrolled in youth groups than children of mothers with no college education. Participation in youth groups increased (though only marginally) for those with some college while declining for those with no college. These results support the argument that better-educated mothers make a greater investment in children’s learning through encouraging reading and studying. They also support the hypothesis that better-educated mothers make a greater investment in children’s structured activities such as sports and arts, but this extends to include a variety of other skill-based activities such as hobbies and youth groups. Additionally, we see a convergence in activities. With the sole exception of television viewing, which is still much lower for children of college-educated mothers, differences in activities between children of college and non-college educated mothers are small. Differences by Family Size The fourth demographic trend of interest here is the decline in family size. Again, the same overall trends hold for large and small-sized families (Table 6). There are, however, several interesting differences between small and large families both in 1981 and in 1997. Children do more household work in large than in small families and they spend less time shopping. It is more efficient due to economies of scale to produce goods and services at home than to purchase them when the family is large. This is the only group, however, in which we see children do more household work. Children spent more time watching television in large compared with small families in 1997. Television serves as an inexpensive form of entertainment. Because of the large number of alternative forms today, family size affects television viewing more than it did in 1981. Compared with children in small families, children in large families spend less time eating meals, less time sleeping, less time in day care and more time in school. The smaller amount of time children from large families spend in day care is consistent with other data on day care use (Hofferth, Brayfield, Deich & Holcomb, 1991); in large families, caring for children at home is more efficient than paying for out-of-home care. However, the increased time in school may reflect the use of free or low-cost before and after-school activities which may or may not be considered “school.” Children in large families spend less time visiting and they spend less time in conversation with adults, both of

Changes in American Children’s Time, 1981–1997


Table 6. Weekly Time Children (Ages 3–12) Spent in 21 Activities, 1981–1997, by Family Size. < 3 Children 1981 N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for


3 + Children 1981


82 1188 140 844 0:36 0:05*** 0:20 0:07* 3:02 2:40 4:11 3:04*** 2:35 3:04 1:25 2:35*** 6:25 8:13*** 6:14 7:57*** 8:58 8:21 8:49 8:04** 70:46 71:33 69:34 70:45 23:51 24:51 25:15 29:04*** 2:13 2:11 1:40 2:24** 2:08 1:12*** 2:15 1:16*** 0:59 0:36 0:44 0:36 3:32 3:04 3:31 2:29** 3:40 5:25** 4:35 5:22 0:26 0:35 0:34 0:33 0:01 0:05 0:02 0:05 0:33 0:59* 0:17 0:55*** 15:30 12:20*** 13:54 11:58** 17:22 13:08*** 17:43 14:06*** 0:54 1:19*** 0:51 1:11 0:25 0:37 1:08 0:29*** 2:00 2:13 3:24 2:07*** 0:32 4:18 0:03 1:13** 1:20 1:01 1:19 1:32 99% 99% 99% 99% 168:00:00 168:00:00 168:00:00 168:00:00

< 3 vs. 3 + Children 1981 1997

* **

** ** * ** ***


* ** *** ** *



* p < 0.05, ** p < 0.01, *** p < 0.001

which are probably due to the fact that they have playmates among their siblings. Age Differences in Trends Over Time So far we have focused upon differences between 1981 and 1997 across all age groups. However, we know that how children spend time changes as they grow older. One issue is the extent to which our overall findings on changes in time



hold for children of some or all age groups. This is particularly important to know, since some activities are age-specific. For example, we find that household work declined overall, but particularly for the 6–8 and 9–12-yearolds, who were more likely to do such work (Figures 5–7). Shopping, in contrast, increased more for the 3–5 and 6–8 year-olds than for the 9–12 yearolds. The increase reflects primarily the increased proportion of young children who accompany their parents shopping, rather than increased time spent doing so. Age differences in the proportion who shop have disappeared since 1981 (Table 1). While there is evidence of increased studying and reading over all children, significant increases in studying occurred only among 6–8-year-olds and significant increases in reading only among 3–5 year-olds. The main reason for the increase in studying among 6–8-year-olds was an increase in the proportion who did some studying at all, from one-third to more than one-half. The fact that significant increases in reading occurred among 3–5 year-olds probably reflects parents’ increasing concern with preparing children for school. Increased enrollment in day care centers and preschools may also be associated with children doing more reading at early ages. There is certainly no evidence that 6–12-year-olds are doing any more reading than in 1981. The reason for the increased time 3–5-year-olds spend reading is the increased proportion reading or being read to, which doubled between 1981 and 1997 (Table 1). Only about one-third of 6–12-year-olds read for pleasure in 1981 and 1997, with no significant change over time. While the overall increase in time spent in sports was not significant, we see that sports participation increased dramatically among both 3–5 and 9–12-yearolds. The major reason for the increase among the 3–5-year-olds was the increased proportion engaging in sports, which almost doubled over the period. There was an increase in participation among 9–12-year-olds, but it was smaller. The decline in time spent in church attendance, youth groups, and visiting is a function primarily of the smaller proportion of children who participate. The declines are consistent across all age groups. Art activities increased among children of all ages. The increase in time spent in art activities among 3–5-year-olds reflects increased participation, whereas the increased time by older children reflects increased time spent. Household conversations and other passive leisure declined primarily for the 9–12-year-olds. While playing declined overall, playing actually increased among 9–12-year-olds. This may reflected increased used of video and computer games among this age group.

Changes in American Children’s Time, 1981–1997 219

Fig. 5. Changes in Weekly Hours by Age from 1981 to 1997 (3–5-Year-Olds).


Fig. 6. Changes in Weekly Hours by Age from 1981 to 1997 (6–8-Year-Olds).

Changes in American Children’s Time, 1981–1997 221

Fig. 7. Changes in Weekly Hours by Age from 1981 to 1997 (9–12-Year-Olds).

Participation in 1981 Girls

107 8% 77% 42% 100% 100% 100% 75% 41% 34% 37% 76% 61% 25% 0% 15% 94% 97% 34% 29% 61% 3%

114 4% 80% 44% 99% 100% 100% 82% 65% 38% 39% 73% 58% 11% 3% 22% 93% 96% 29% 42% 71% 2%

* p < 0.05, ** p < 0.01, *** p < 0.001

Gender Diff.


*** *




1097 1% 66% 48% 100% 100% 100% 79% 46% 26% 24% 49% 80% 18% 2% 21% 93% 96% 41% 32% 49% 13%

1022 2% 73% 56% 100% 100% 100% 80% 45% 25% 25% 54% 69% 16% 3% 32% 91% 95% 44% 31% 51% 12%

Gender Diff.

*** ***

** ***

Trend in Participation 1981–1997 Boys


*** ** ** **

* *** *** *** *

*** *** *** *** **




*** ** *** ***

** ***


N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare


Participation in 1997


Table 7. Proportion of Children (Ages 3–12) Participating in 21 Weekly Activities and Weekly Hours, 1981–1997, by Gender.

Time Spent in 1981 Girls

107 0:43 2:52 1:53 6:05 8:45 70:39 22:58 1:24 1:45 0:43 3:40 5:08 0:36 0:00 0:25 16:19 18:18 0:49 0:42 2:13 0:24 1:27 99% 168:00:00

114 0:10 4:35 1:49 6:30 8:59 69:25 26:24 2:19 2:38 0:55 3:24 3:25 0:27 0:04 0:21 12:47 16:54 0:56 1:02 3:30 0:04 1:13 99% 168:00:00

* p < 0.05, ** p < 0.01, *** p < 0.001

Gender Diff.

* ***

* ** *

* * **




1097 0:03 2:26 2:34 7:36 8:14 70:53 26:18 2:21 1:11 0:38 2:39 6:40 0:36 0:04 0:39 12:52 13:35 1:11 0:33 2:15 3:28 1:02 99% 168:00:00

1022 0:08 3:13 3:13 8:36 8:11 71:22 27:18 2:11 1:14 0:37 3:01 4:03 0:32 0:06 1:15 11:27 13:22 1:20 0:35 2:06 2:23 1:35 99% 168:00:00

Gender Diff.

* *** *** ***


*** ***




*** * *** ** ** ** ** **

*** *** ***


* **

*** *** *** *


*** * *** *** *** ***


N Market work Household work Shopping Personal care Eating Sleeping School Studying Church Youth groups Visiting Sports Outdoors Hobbies Art activities Playing Television Reading Household conversations Other passive leisure Daycare Not Ascertained (NA) % of time accounted for


Time Spent in 1997

Trend in Time Spent 1981–1997

Changes in American Children’s Time, 1981–1997

Table 7. Continued.



Gender Differences Some sex-stereotyped gender differences in children’s activities are quite evident and have not declined over time. Girls reported more household work in both years and more shopping in 1997 than did boys and they spent more time at it (Table 7). Girls also spent more time in personal care than did boys in 1997. Boys are more likely to be involved in sports and they also spend more time in such activities. Girls, in contrast, are more likely to participate in and spend more time in less strenuous activities such as art and music in 1997. Girls also spent less time than boys in free play in both years. Some of the trends have led to greater similarity in the activities of boys and girls in 1997. In particular, while girls were more likely than boys to study in 1981, the two were similar in 1997 – about 45–46% studied. This is because the proportion of girls who studied actually declined between 1981 and 1997 while the proportion of boys who studied stayed the same. Increased studying time occurred primarily among boys, as a result. This increase may be due to the use of computers for homework and Internet research. Boys’ spend significantly more time using computers than girls (not shown). There was a convergence in school time, church attendance, and other passive leisure as well. In 1981 girls spent more time in school. Boys increased their time at school more than girls between 1981 and 1997; the difference was no longer significant in 1997. Girls were more active than boys in church activities in 1981; by 1997 the difference in time was not significant. A third area in which boys and girls are converging is in other passive leisure. In 1981 girls spent more time in passive leisure than boys. Girls’ use declined between 1981 and 1997 such that the two did not differ significantly on passive leisure in 1997. Areas of nonconvergence include household work, shopping, personal care, sports, art, play, and day care. For example, while participation in sports increased for both boys and girls, the level was greater for boys to begin with and the gap did not diminish. Differences widened for the arts, where girls increased both participation and time while boys did not. Trends in activities for boys and girls are similar, for the most part, with both spending more time shopping, in personal care, and in day care between 1981 and 1997.

SUMMARY AND CONCLUSIONS Between 1981 and 1997 two major changes in children’s time occurred. First, the proportion of time that is nondiscretionary, that is, which is taken up by school or day care, personal care, eating, and sleeping increased significantly. This increase occurred across the board, for every group of children. A portion

Changes in American Children’s Time, 1981–1997


of this change, particularly the increase in day care time and the decline in time spent eating, was due to maternal employment. However, other changes represent changed values. For example, children of nonemployed mothers spent more time in school in 1997 compared with 1981. Children of both employed and nonemployed mothers spent more time in personal care in 1997 than in 1981. As a result of these changes, the proportion of time that is discretionary declined. Children’s time use changed in other important ways between 1981 and 1997. One major change is the increase in structured activities such as sports and art activities and the decline in unstructured play and television viewing. Many of these changes may be due to overall changes in societal preferences. The decline in household work, the increase in shopping, the decline in time spent in church, the decline in play and television viewing and the increase in art activities occurred among children of nonworking as well as working mothers. Increases in sports participation were largest among children of nonemployed mothers as well, thus not linked directly to work. Maternal work contributed only to declines in meal time and to increased day care. The contribution of cultural values is seen in differences by maternal education. Children of mothers with some college education spent more time in art activities, reading, sports, outdoor activities, hobbies, youth groups, household conversations, studying, and other passive leisure and less time watching television than children of mothers without any college. As the educational levels of the population rise, we expect continued increases in structured and educational activities. Family structure and family size are important correlates of how children spend their time. Children in single-parent families are disadvantaged in that they spend less time shopping, attending church, in youth groups, in hobbies, playing, reading, in household conversations, and more time in other passive leisure, day care, and television viewing than children in two-parent families. Children in large families also have a different experience of childhood than children from smaller families, spending more time in household work, less time shopping, and more time watching television. What are the implications of these changes for families and children? First, families are a lot busier today; with two parents working and children’s structured activities increasing, to accomplish everything requires very tight scheduling. Declining time spent eating meals reflects these pressures and constraints. Families also may spend more time running errands to purchase goods at stores rather than making things at home. This is reflected in the increased time children spend shopping, the largest chunk of which is



explained as the time children spend accompanying parents on errands and shopping trips, and the declining time spent in traditional household work. What is surprising is that television time has not increased; in fact, it has declined for all groups of children, and it has declined more than the overall decline in free time. Television is a passive activity that does not contribute to children’s achievement test scores (Hofferth & Sandberg, 2001) so this can be interpreted as a positive trend. On the mixed news side, the total time spent reading, only about one hour per week, has increased significantly over the period, but the base level was small and that change is mostly among children 3–5 years of age. Other research finds reading to be the activity that is most strongly associated with better reading comprehension (Hofferth & Sandberg, 2001) and, consequently, with success in school. While studying increased significantly between 1981 and 1997, the time spent studying is still small, only about two hours per week, and the increase was concentrated only among 6–8-year-olds. In addition, studying is not necessarily associated with doing better in school because children who study more include those are having school problems. While we found that children spend less time playing, this mostly reflects decreased time spent at home; the decline in play time is about the same as the overall decline in free time. It should be pointed out, however, that this estimate may be biased because we focused solely on home time. Children may be playing in their preschool programs and have some free time at school. Thus, this study provides only a partial picture of children’s time. Current research is examining what children do at school as well as at home. There is one cautionary note. We found that children spend less time eating meals. This is not the only time children and parents spend talking, but since time spent just sitting and talking as the main activity (household conversations) also declined significantly between 1981 and 1997, there may a basis for concern that shared family activities are declining. This study presented a picture of children’s time in 1997 and trends since 1981, but it is only the first look. A second wave of data will be collected from these same children and their families in the year 2002. The question of the relationship of time to child behavior and well-being that could not be addressed in this paper may be better addressed in the future with longitudinal data.

NOTES 1. While ideally we would like to be able to measure the regularity of schedules to more precisely define “scheduled” and “unscheduled,” we did not ask parents about

Changes in American Children’s Time, 1981–1997


scheduling and we have only one weekday and one weekend time diary, not enough to establish regularity precisely. While activities may have both structured and unstructured aspects, for the most part, as we have defined them, sports are structured and play and outdoor activities are unstructured. 2. Time Use in Economic and Social Accounts, 1975–1976; data originally collected by F. Thomas Juster, Paul Courant, Greg J. Duncan, John P. Robinson and Frank P. Stafford of the Survey Research Center, Institute for Social Research, The University of Michigan. Data furnished by the Inter-university Consortium for Political and Social Research, ICPSR study #7580. Neither the collectors of the original data nor the Consortium bear any of the responsibility for the analyses or interpretations presented here. 3. Time Use Longitudinal Panel Study, 1975–1981; data originally collected by F. Thomas Juster, Martha S. Hill, Frank P. Stafford, and Jacquelynne Eccles Parson of the Survey Research Center, Institute for Social Research, The University of Michigan. Data furnished by the Inter-university Consortium for Political and Social Research, ICPSR study #9054. Neither the collectors of the original data nor the Consortium bear any of the responsibility for the analyses or interpretations presented here. 4. The data from the Current Population Survey, March 1981: After-tax money income estimates were originally collected by the U.S. Dept. of Commerce, Bureau of the Census, and were made available for these analyses by the Inter- university Consortium for Political and Social Research, ICPSR study #8269. Neither the collectors of the original data nor the Consortium bear any of the responsibility for the analyses or interpretations presented here. 5. The specific activities that make up each of the 21 categories are available from the authors. 6. Two children who, in 1997, spent the entire week visiting were also excluded. 7. The most obvious examples of such likely errors were children whose weekly hours of sleep fell below 21 hours, or above 112. Another obvious example would be the 3 year old in 1981 whose imputed weekly market work time was 58 hours. A complete list of cases removed from the analyses and rational for these decisions is available from the second author on request.

ACKNOWLEDGMENTS Funding for this research was provided by grant #U01HD37563 from the National Institute of Child Health and Human Development and by the Center for the Ethnography of Everyday Life, an Alfred P. Sloan Center for the Study of Working Families.

REFERENCES Bianchi, S., & Robinson, J. (1997). What did you do today? Children’s use of time, family composition, and the acquisition of social capital. Journal of Marriage and the Family, 59, 332–344. Blake, J. (1989). Family Size and Achievement. Berkeley, CA: University of California Press.



Bryant, W. K., & Zick, C. (1996). An Examination of Parent-Child Shared Time. Journal of Marriage and the Family, 58, 227–237. Eccles, J., & Barber, B. (1998). Student Council, Volunteering, Basketball, or Marching Band: What Kind of Extracurricular Involvement Matters? Journal of Adolescent Research. Fitzgerald, J., Gottschalk, P., & Moffitt, R. (1998a). An Analysis of Sample Attrition in Panel Data: The Michigan Panel Study of Income Dynamics. Journal of Human Resources, 33(2), 251–299. Gershuny, J., & Robinson, J. P. (1988). Historical Changes in the Household Division of Labor. Demography, 25(4), 537–552. Hernandez, D. J. (1993). America’s Children: Resources from Family, Government and the Economy. New York: Russell Sage. Hetherington, E. M. (1989). Coping with family transition: Winner, losers, and survivors. Child Development, 60, 1–14. Hofferth, S. L., & Sandberg, J. F. (2001). How American Children Use their Time. Journal of Marriage and the Family, 63(3), 295–308. Hofferth, S. (1995). Caring for Children at the Poverty Line. Children and Youth Services Review, 17(1–3), 61–90. Hofferth, S., Brayfield, A., Deich, S., & Holcomb, P. (1991). National Child Care Survey 1990. Washington, D.C.: The Urban Institute. Hofferth, S., & Jankuniene, Z. (2001). Life After School. Educational Leadership, 58(7), 19–23. Hofferth, S., Shauman, K., Henke, R., & West, J. (1998). Characteristics of Children’s Early Care and Education Programs. Washington, D.C.: National Center for Educational Statistics. Jacobs, E., Shipp, S., & Brown, G. (1989). Families of Working Wives Spending More on Services and Nondurables. Monthly Labor Review, 112(2), 15–23. Juster, F., & Stafford, F. P. (1985). Time, Goods, and Well-Being. Ann Arbor, MI: Institute for Social Research. Kinney, D. A., Dunn, J. S., & Hofferth, S. L. (2000). Family Strategies for Managing the Time Crunch. Paper presented at Conference on. Work and Family: Expanding the Horizons, San Francisco, CA. Larson, R., & Verma, S. (1999). How Children and Adolescents Spend Time Across the World: Work, Play, and Developmental Opportunities. Psychological Bulletin, 125(6), 701–736. Leibowitz, A. (1974a). Education and Home Production. American Economic Review, 64(2), 243–250. Leibowitz, A. (1974b). Home investments in children. Journal of Political Economy, 82(2, Part II), S111-S131. Leibowitz, A. (1977). Parental Inputs and Children’s Achievement. Journal of Human Resources, 12, 242–251. Marini, M. M., & Shelton, B. A. (1993). Measuring Household Work: Recent experience in the United States. Social Science Research, 22, 361–382. Martin, T. C., & Bumpass, L. L. (1989). Recent trends in marital disruption. Demography, 26, 37–51. McLanahan, S. (1985). Family structure and the reproduction of poverty. American Journal of Sociology, 90, 873–901. Nock, S., & Kingston, P. (1988). Time with children: The impact of couples’ work-time commitments. Social Forces, 67, 59–85. Robinson, J. P., & Godbey, G. (1997). Tiime for Life: The Surprising Ways Americans Use Their Time. University Park, PA: Pennsylvania State University.

Changes in American Children’s Time, 1981–1997


Task Force on Youth Development and Community Programs (1992). A Matter of Time: Risk and Opportunity in the Nonschool Hours. New York: Carnegie Corporation. Timmer, S. G., Eccles, J., & O’Brien, K. (1985). How Children Use Time. In: F. S. Juster & F. P. (Eds), Time, Goods, and Well-Being (pp. 353–382). Ann Arbor, MI: Institute for Social Research. U.S. Bureau of the Census (1984). Childspacing among Birth cohorts of American women: 1905–1959. Current Population Reports, Series P-20 (385). U.S. Bureau of the Census (1988). Fertility of American Women: June 1987. Current Population Reports, P-20 (427). U.S. Bureau of the Census (1998). Family Composition Begins to Stabilize in the 1990s. Available:–88.html (Accessed 29 May 98). U.S. Bureau of the Census (2000). Service Industries New Economy’s Biggest Generator of Jobs: Mississippi leads States, Census Bureau Reports. Available: (Accessed 7/7/2000). Zajonc, R. B., & Markus, G. B. (1975). Birth order and intellectual development. Psychological Review, 82, 74–88. Zick, C., & Allen, C. (1996). The Impact of Parents’ Marital Status on the Time Adolescents Spend in Productive Activities. Family relations, 45, 65–71.

This Page Intentionally Left Blank


This Page Intentionally Left Blank



Monica Kirkpatrick Johnson, Sabrina Oesterle and Jeylan T. Mortimer ABSTRACT In this chapter, we consider the connections between macro-level changes in work and family structures in the U.S. and the micro-level attitudes of adolescents regarding future work and family lives. Using nationally representative data, we first examine whether adolescents’ orientations toward work and family changed in the last two decades. With this national picture in mind, we then examine data from a community panel study of adolescents to assess whether adolescents anticipate conflict between work and family roles, how they respond to such concerns, and what difficulties they have in integrating work and family during early adulthood. We find that adolescents remain highly interested in work and family roles, and have rising aspirations for educational attainment and extrinsic work rewards. However, the age at which they expect to marry has risen, and support for both cohabitation and childbearing outside of

* An earlier version of this paper was presented at the 1999 Meetings of the American Sociological Association, Chicago.

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 233–261. Copyright © 2001 by Elsevier Science Ltd. All rights of reproduction in any form reserved. ISBN: 0-7623-0776-5




marriage has grown over this period. Although a minority of high school seniors, mostly females, anticipate that their future family roles will interfere with their career plans, there is evidence that young people are planning the timing of their investments in education, work, and family roles in order to avoid difficulties. As adolescents enter young adulthood, they become much more likely to anticipate that their family roles will interfere with their career plans, and among those who have made family transitions, the incidence of work-family conflict is high.

INTRODUCTION Adolescence is a stage in the life course in which plans and goals for the future come to the forefront. Key components of these future orientations are aspirations for education, occupation, and family activities (Nurmi, 1991). High school seniors in particular, as they are on the threshold of adulthood, face questions about who they will be as adults and the steps to be taken in order to meet their goals. Adolescents’ plans and orientations toward work and family are clearly linked to adult outcomes. As students leave high school, they must make decisions regarding investments in post-secondary education and labor market participation. Plans and orientations toward work and family, along with the actual opportunities and constraints encountered along the way, play an important part in shaping the work and family trajectories of young adults (Gerson, 1985; Johnson, 1999; Sassler & Schoen, 1999). The social psychological model of attainment postulates that socioeconomic background influences post-secondary educational attainment, occupational success, and earnings largely indirectly, mediated by adolescents’ plans and aspirations (Sewell & Hauser, 1975). Family formation at this life stage also has implications for later attainment, especially for females (Marini, 1984; Marini et al., 1989). Continued monitoring of youth’s plans and aspirations provides clues about the social and economic futures of the next generation of adults. The institutions of work and family, and the linkages between them, have undergone much change during the past half-century, creating a new context for adolescent planning and decision-making. The way in which adolescents respond to these contextual changes provides important information about future work and family structures. The so-called traditional, normative work-family model, as it developed in the U.S. by mid-century, provided both men and women with clearly defined roles and thus a relatively predictable future in terms of work and family arrangements. The responsibilities of earning a living to provide financial

Adolescents’ Anticipations of Work-Family Conflict


support to the family, on the one hand, and unpaid family and household labor, on the other, were divided between husband and wife, respectively. Although some married women participated in paid labor, especially working class and minority women (Goldin, 1990), men’s principal contribution was economic and women’s was caretaking. Major changes in the structure of work-family arrangements since that time have created a new context within which adolescents plan their futures. This new context is characterized by increasing uncertainty and greater potential conflict between family and work roles. One of the most consequential changes has been the sharp increase in women’s participation in the labor market (Bianchi & Spain, 1996). Women’s raised aspirations for rewards outside the family have been fueled by the women’s movement and rising levels of education. Although high aspirations for education and occupation may be a major motivation for women’s participation in the labor force, current economic trends suggest that the economic well-being of the family will increasingly depend on women’s contribution to the family’s livelihood. Contemporary young people in the U.S. are starting their work and family lives in times of a healthy and growing economy and historically low unemployment rates. However, they are also facing a future of vast and increasing income inequality. The level of income inequality in the U.S. is now the highest among industrialized societies (Mischel, Bernstein & Schmitt, 1999). The bottom 60% of families has lower real incomes, adjusted for inflation, in 1999 than in 1977; the top fifth of households now earns 50% of the total income share (Shapiro & Greenstein, 1999). This situation has been attributed in part to an increase in contingent and nonstandard employment and declining real wages, especially for men. Since the mid-1960s, the share of male workers in long-term jobs (10 or more years with the same firm) has declined by almost 10% (Mischel et al., 1999). An increase in contingent employment, such as on-call work, temporary work, day labor, contracting, and part-time work, fosters job instability and insecurity. Typically, contingent work settings are also of lower quality than regular fulltime employment, as they pay lower wages and are less likely to provide benefits. In 1995, 25% of male and 34% of female workers were in such nonstandard, less than full-time employment (Kalleberg et al., 1997). Since the mid-1970s, median incomes dropped especially for young men between the ages of 25 to 34 (–19.6%) but also for middle-aged men between the ages of 35 to 44 years (–15.1%) (Mischel et al., 1999). The combination of these economic changes has increasingly undermined men’s ability to be the family’s sole breadwinner (Wallace, 1998). Women’s contribution to the share of family incomes has continuously grown over the



last 20 years, particularly for families at the bottom and middle of the income distribution (Mischel et al., 1999). Increasingly, women are also single mothers, either solely responsible, or predominantly responsible, for the economic welfare of their families. These trends have profoundly changed family experiences. The hours families devote to paid labor have risen, particularly among married-couple families with children. Families with children worked 326 hours more in 1996 than in 1970 (husband and wives combined), which is about four work weeks more per year and person (husband or wife) than in the late 1970s (Mischel et al., 1999). Publication of The Overworked American (Schor, 1991) and The Time Bind (Hochschild, 1997) have focused public attention on the amount of time families devote to paid labor. Moreover, with the shift to employment in the service sector, the demand for laborers to work evening, night, and weekend shifts has grown substantially, creating additional difficulty in coordinating childcare and maintaining satisfactory marriages (Presser, 1998). These economic and employment shifts have important implications for women. Although most contemporary marriages are dual-worker families in which the husband and wife share the economic provider role (Ahlburg & De Vita, 1992), they generally do not share the homemaker role to the same extent. Women retain major responsibility for childcare and housework and, therefore, face a double burden (Coltrane, 2000). Despite women’s growing financial contributions to their families, men have been resistant to taking on a greater share of the family and household labor (Hochschild, 1989; Gager, 1995). This arrangement clearly creates a potential for conflict between work and family roles and responsibilities, especially for females, but increasingly for males as well. Although these changes in work-family arrangements have been well documented (e.g. Bianchi & Spain, 1986; Gerson, 1998), research is needed on young people’s reactions to this changing context. The potential for conflict between work and family roles is clear considering the macro-level trends. We do not know, however, whether contemporary young people indeed expect difficulties with respect to these arrangements in their own lives and how they may behave in response to such anticipation. In other words, what is needed is an examination of the links between macro-level changes in work and family structures and micro-level processes, including adolescents’ plans for, and orientations toward, their future work and family lives. Because the responsibility for family and household labor has not shifted as much as paid work roles, we might expect that contemporary young women are more aware than young men of the potential for conflict between work and family, and are more reactive to such potential. The stalled revolution at home

Adolescents’ Anticipations of Work-Family Conflict


(Hochschild, 1989) may lead women to be more strategic when it comes to planning their families and careers. Women may also increasingly feel the need to take responsibility for their own and their children’s lives, not only because their economic security has been weakened by males’ increasing inability to financially support a family, but also because sharply rising divorce rates have contributed to greater uncertainty about the stability of marriage. Women may also do more planning for their future family lives because they have more efficacy with respect to family matters, which encourages planfulness or strategic thinking (Nurmi, 1997). For both young men and young women, changes in the linkages of work and family may have led to a retreat from marriage. Given the greater economic independence of women and greater uncertainty about marriage, one might expect that marriage has become less necessary and less desirable for young women today (see Oppenheimer, 1994, 1997 for a review and critique of this hypothesis). For young men, one might expect that they would also have less interest in family. The family has become less dependent on men’s success in the provider role, and women are less likely to be dedicated to supporting their husbands’ careers and more likely to expect greater contributions from their husbands to household labor. Declining rates of marriage, especially at younger ages (Ahlburg & De Vita, 1992), and delay of childbirth (Bianchi & Spain, 1996) can be viewed as signs of withdrawal from the institution of family. Alternatively, they can be interpreted as indicators of planfulness in a new work-family context – strategic sequencing of education and entry into work and family roles. Men and women may be delaying family formation in order to avoid the strain of meeting the demands of career and family development simultaneously. The rising age of first marriages may account for much of the apparent decline in marriage rates (Oppenheimer, 1994). Females, in particular, may be anticipating greater conflict between the demands of employment and family roles and may do more thinking and strategic planning about the timing of family formation in relation to their educational and employment goals. Consistent with a sequencing strategy, young people could turn to alternative family arrangements, such as cohabitation, because they provide more flexibility and require less commitment. Early commitment to another person and accommodation to that person’s employment opportunities and life goals could curtail geographic mobility and generally interfere with a young person’s establishment in the labor market. Cohabitation has been found to be more attractive to young people today, especially when economic resources are low (Seltzer, 2000).



In the context of dramatic changes in families, employment, and the nexus of employment and family life, this chapter considers young people’s reactions to these changes. We seek to create a better understanding of the link between macro-level changes in work and family structures and micro-level plans and orientations. We examine both historical trends in adolescents’ plans and orientations at the national level and current dynamics in a cohort of Minnesota youth as they make the transition to adulthood. Using U.S. national data on high school seniors, drawn from the Monitoring the Future study, we first assess whether young people’s orientations toward work and family have changed over the last two decades. In order to investigate whether adolescents anticipate conflict between work and family roles, and how they respond to such concerns, we turn to data from the Youth Development Study, a panel study of adolescents in St. Paul, Minnesota. Because this study follows adolescents as they enter early adult employment and family roles, we are also able to examine the difficulties young people have in integrating work and family during this life stage.

NATIONAL TRENDS IN ADOLESCENTS’ ORIENTATIONS TOWARD WORK AND FAMILY Data and Measures The data on historical trends in orientations toward family and work are derived from the Monitoring the Future study, a repeated cross-sectional survey of U.S. high school seniors.1 Data were collected using a multistage cluster sampling technique that included 125 public and private schools in order to obtain a nationally representative sample of high school seniors. The study began in 1975; surveys have been administered to about 16,000 students annually. Due to the broad range of topics covered by the Monitoring the Future survey, the questionnaire has been split into six forms. Except for a core set of background questions for the total sample, most other items are answered by one of the sub-samples of about 2,500 students. The sample size for the percentages reported below is, thus, about 2,500 each year. The Monitoring the Future study provides measures that illustrate young people’s orientations toward family by the desire to get married, the preferred timing of marriage, and the desire to have children. Seniors’ desire to marry was measured by the question, Which do you think you are most likely to choose in the long run? (getting married; I have no idea; not getting married; am already married). The respondents were also asked to rate the importance of Having a good marriage and family life (1 = not important; 4 = extremely

Adolescents’ Anticipations of Work-Family Conflict


important). The preferred timing of marriage was measured by the question, If it were just up to you, what would be the ideal time for you to get married? (1 = Within the next year or so; 2 = Two or three years from now; 3 = Four or five years from now; 4 = Over five years from now; and 5 = I don’t want to marry). The desire for children was gauged by the question, If you did get married (or are married), how likely is it that you would want to have children? Response categories ranged from 1 = Very unlikely to 5 = Very likely with 3 = Uncertain. The Monitoring the Future survey also addressed the degree of acceptance of various family arrangements. Respondents’ views on cohabitation were assessed by the question, It is usually a good idea for a couple to live together before getting married in order to find out whether they really get along (1 = Disagree; 5 = Agree). To assess attitudes toward bearing children outside of marriage, respondents were asked to complete the following sentence: A man and a woman who decide to have and raise a child out of wedlock are . . . . The five response categories were: experimenting with a worthwhile alternative lifestyle; doing their own thing and not affecting anyone else; living in a way that could be destructive to society; violating a basic principle of human morality; and none of the above. Orientations toward work include the importance attached to work, work values, and preferences about dividing employment and childcare arrangements in the seniors’ families of procreation. Respondents indicated the importance of work by agreeing or disagreeing with the statement, I expect my work to be a very central part of my life. Work values were measured by ratings of a range of job characteristics on a four-point scale of importance; 1 = Not important and 4 = Very important. Extrinsic work values, for example, referenced job characteristics like A job that has high status and prestige, A job where the chances for advancement and promotion are good, and A job which provides you with a chance to earn a good deal of money. Finally, respondents’ preferences for allocating employment and childcare roles were measured by the question, Imagine you are married and have one or more pre-school children. How would you feel about each of the following working arrangements (1 = Not at all acceptable; 4 = Desirable)? The arrangements included: Husband works full-time, wife doesn’t work; Husband works fulltime, wife works about half-time; Both work full-time; Both work about half-time; Husband works about half-time, wife works full-time; and Husband doesn’t work, wife works full-time. The Monitoring the Future survey also includes a battery of questions regarding educational aspirations. Plans to obtain a bachelor’s degree were indicated by answers to the question, How likely is it that you will do each of



the following things after high school? – Graduate from college (four-year program). The scale of responses ranged from 1 = Definitely not to 4 = Definitely will. National Trends in Future Orientations Consistent with national trends in the age of first marriages, young people are increasingly planning to marry at later ages. Three-quarters of high school seniors in the Monitoring the Future study indicated in 1995 that they wanted to wait four or more years to get married, up from two-thirds in 1976. However, these national data clearly show that youth continue to be interested in marrying as well as having children. The proportion of high school seniors who think that they will most likely get married is quite high and has increased slightly over the last two decades. In 1976, 69% of boys and 80% of girls thought they would likely get married in the future. These proportions grew to 77% for boys and 83% for girls in 1995. Somewhat less than two-thirds of the students in the Monitoring the Future study indicated in 1995 that they wanted to have children after getting married. This proportion has been stable since the mid-1970s. Similarly, there are no signs of withdrawal of commitment to the family as work-family structures have changed in recent decades. The importance young people attach to the family has not diminished, and in fact has risen slightly since the mid-1970s. The proportion of respondents indicating that a good marriage and family life was quite or extremely important is presented in Fig. 1 for this time period. In 1976, the majority (88%) of high school seniors thought that having a good marriage and family life was quite or extremely important. Yet an even larger share (92%) did so in 1995. Consistent with their plans to marry and have children, family life continues to be valued highly by both males and females. Both in terms of the importance they attach to family life and their anticipations of marrying, females show a somewhat stronger orientation to the family than males. Although young Americans’ desire to have families has remained strong over the past two decades, they have become increasingly accepting of both cohabitation before marriage and out-of-wedlock childbearing. Agreement with the statement, It is a good idea to live together before marriage, increased 19% between 1976 and 1995. Just over 73% agreed with this statement in 1995. In 1976, 43% agreed that raising a child out of wedlock was a worthwhile alternative lifestyle or a private matter that does not affect anyone else. In 1995, as shown by agreement with these statements, 52% accepted the

Adolescents’ Anticipations of Work-Family Conflict


Fig. 1. Importance of Good Marriage and Family Life.

idea of rearing a child outside of marriage. Thus, young people are growing more tolerant of alternatives to traditional, and particularly legalistic family arrangements. Turning to adolescents’ orientations toward work, most students expect work to be a very central part of their lives. The proportion of respondents who agree or mostly agree that work is very central to life (see Fig. 2) has declined somewhat since the mid-1980s. In 1995, two-thirds of the respondents expected work to be very central in their lives, down from three-quarters in 1976. The importance of work decreased more steeply for males than for females between 1985 and 1995; as a result, in the 1990s slightly more females than males expected that work would be a very central part of their lives. The shift in the males’ responses is particularly interesting within the context of a changing male work experience. Eroding wages and diminishing job security may make the employment role a less attractive one to consider a very central part of one’s life. Despite some slight downward shift in the importance of the work role, adolescents do not expect any less in terms of the rewards they seek from employment. In fact, Marini, Fan, Finley and Beutel (1996), who examined the Monitoring the Future work values data, report that the importance young people attach to extrinsic rewards of work, such as pay and prestige, and the



Fig. 2. Work is Central to Life.

importance attached to job security increased slightly in the last few decades. This shift was greater for females than males. Whereas young men once valued the extrinsic rewards of work more highly than women, young females have caught up with men in their extrinsic work values, and particularly in their evaluation of advancement opportunities (Marini et al., 1996). About two-thirds of males and females rated having good chances for advancement as very important in 1995. Males still attached slightly greater importance to the chance to earn a great deal of money; 62% of males and 51% of females rated this job reward as very important in 1995. Consistent with these increasing expectations about the extrinsic rewards from work, young peoples’ educational aspirations have also increased over the last decades. Education is one of the most important means by which greater pay, prestige, and advancement opportunities are achieved. Trends in the proportion of respondents who plan to graduate from a four-year college or university are shown in Fig. 3. Compared to the mid-1970s, when about 50% of the respondents expected that they would probably or definitely graduate from a four-year college, about three-quarters of respondents in 1995 planned to obtain a college degree. Whereas a somewhat smaller share of women than men planned to get a college degree in 1976, a greater proportion of females (81%) than males (74%) have college plans in 1995.

Adolescents’ Anticipations of Work-Family Conflict


Fig. 3. Educational Aspiration.

Notwithstanding the importance of paid work to females, both males and females prefer work-family arrangements in which mothers leave the labor force or significantly reduce their hours of employment when there are small children in the family. Both male and female students in the Monitoring the Future study increasingly reject the traditional work-family arrangement during these early child rearing years (husband works full-time, wife does not work), however, in favor of wives’ part-time employment (Fig. 4). By 1995, males appear equally likely to accept work-family arrangements in which the wife does not work or the wife works part time. Females, in contrast, show a clear preference for remaining in the labor force part-time over no employment at all. National employment data indicate that many contemporary young men and women both work full time even when they have pre-school children. Most seniors, however, do not regard an arrangement where both parents work full time as desirable or acceptable. Nevertheless, support for this work-family arrangement more than doubled over the 20-year period of the study. There is also little support for arrangements in which fathers reduce their paid work (not shown). Taken together, the national data suggest that adolescents continue to place great emphasis on both family and work roles. Both males and females remain highly interested in marriage and parenting. They have simultaneously become more accepting of cohabitation and out of wedlock childbearing, though,


Fig. 4. Preferred Working Arrangements in Future Family with Preschool Children.

Adolescents’ Anticipations of Work-Family Conflict


perhaps reflecting a desire for flexibility rather than defined pathways of family formation. It is also apparent that young people increasingly expect to marry at later ages. Such a delay would certainly facilitate meeting their high educational aspirations and expectations regarding occupational rewards. In the 1990s, females are more likely than males to view work as an important part of their adult lives. It is important to note that over time, as females have become more interested in extrinsic work features (reaching parity with males), they have remained more interested than males in the social and altruistic rewards of work. For example, in 1995 approximately 59% of females and 33% of males rated the opportunity to help others as very important; 37% of females and 27% of males rated the opportunity to work with a lot of other people as very important. But despite these various manifestations of the salience of work, females are also more committed than males to their future family roles. They are more likely to plan to marry and attach greater importance to having a good marriage and family life. For contemporary young females, with high aspirations and expectations in terms of both work and family, there is great potential for conflict. The Monitoring the Future data do not contain further information regarding the potential conflict adolescents perceive between work and family life. Adolescents may expect little difficulty in combining work and family roles, or may not have given the matter much thought. Given their strong interest in family and their high expectations and aspirations for attainment in their work, however, it may also be that young people, and especially young women, are concerned about how they will navigate work and family roles. Anticipating problems, they may plan their futures in ways that attempt to minimize workfamily conflict. It is possible, for example, that the Monitoring the Future respondents’ desire to delay family formation is in response to the perceived need to sequence investments in education, work, and family formation. In addition, these cross-sectional data on high school seniors are unable to address questions about the actual experiences of work-family conflict for young people. Seniors in high school have not yet faced the actual demands of full-time employment, and very few have become parents. The demands of work and family increase as one moves through the young adult years. The transition to adulthood is a unique phase in the life course for resolving work and family conflicts (Aronson, 1999). Those men and women who marry or become parents relatively early may face special difficulty in dealing with work-family conflict, as they often have fewer resources and are attempting to launch family and employment roles simultaneously. To address these possibilities, we turn to a panel study of contemporary adolescents.



THE YOUTH DEVELOPMENT STUDY Data and Measures We utilize data from the first nine waves of the Youth Development Study (YDS), a prospective longitudinal study of adolescents in St. Paul, Minnesota. The study design appears in Table 1. The study began in 1988 with a panel of 1,000 adolescents, chosen randomly from a list of enrolled ninth graders in the St. Paul Public Schools.2 Questionnaires were completed by participants in their school classrooms during each of their four years in high school (1988–1991); they were mailed to students who were not present at school on the two days of administration or who had left the school district. Data for the fifth through ninth waves of the study were collected with questionnaires obtained by mail. The first eight waves of the study were conducted each spring, 1988–1995; the ninth wave was conducted two years later in 1997. Panel retention through the fourth wave was 93%; retention through the ninth wave was nearly 79%. During the first and fourth years of the study, the parents of the respondents were also surveyed by mail, providing, among other things, direct measures of socioeconomic background. Although we compare aggregate responses from YDS fourth-wave respondents, when most were seniors in high school, with Monitoring the Future data, it should be recognized that the data sets are not fully comparable. Aside from one being a national and the other a local study, they utilize different sets of measures and the YDS, but not the Monitoring the Future study, includes youth who do not attend high school (about 10% of YDS respondents in Wave 4). A range of orientations to the future, including family and achievement plans, was measured while the participants were in high school. In the domain of family, respondents were asked about expectations for marriage, the age at which they anticipated marriage, and how important they expected marriage and parenthood would be to them as adults. In the domains of education and work, respondents indicated their educational aspirations and plans, and how important they expected their future occupations or careers to be. Respondents were also asked about their plans for employment after entering family roles, and whether they planned to remove themselves from the labor force for any length of time to take care of children. In the years following high school (1992–1995; 1997), respondents were mailed life history calendars (Freedman et al., 1988), on which they indicated activities and family status changes during the previous year, including school attendance, part and full-time work, marriage, parenthood, and living arrangements in monthly units. We measured marriage and parenthood status

Administration in School Full Survey

Administration Wave Year Grade Level Age Retention Rate*

1 1988 9 14–15 –

2 1989 10 15–16 96.2%

3 1990 11 16–17 95.4%

4 1991 12 17–18 92.8%

Short Survey

Mail Survey Full Survey

Short Survey

5 1992

6 1993

7 1994

8 1995

9 1997

18–19 81.3%

19–20 77.7%

20–21 79.6%

21–22 77.6%

23–24 78.6%

Adolescents’ Anticipations of Work-Family Conflict

Table 1. Youth Development Study Research Design.

* 1010 consented to participate in Fall, 1987.




by whether or not respondents had entered into marriage or parenthood by the Wave 8 (1995) and by the Wave 9 (1997) data collections. Anticipated work-family conflict was measured in the senior year of high school (Wave 4) by the question, Are you concerned that any of the following things might interfere with your future career plans? (circle all that apply.) Marriage and Children were listed among the options. In Wave 8, anticipated interference and interference already experienced were measured by the question, Which of the following problems have interfered with your career plans so far? Are you concerned that any might interfere in the future? (circle all that apply.) Again, Marriage and Children were listed among the options for both questions. Whether marriage and children had already interfered with career plans was also asked in Wave 9, but concerns about this type of interference in the future were not. It is important to note that these measures tap only a limited portion of the work-family conflict domain, and reference conflict in only one direction – family roles affecting career plans. In the next section, we examine whether adolescents anticipate work-family conflict in their futures and what plans they have to deal with conflicts that arise in the simultaneous pursuit of work and family roles. We then consider how anticipations of work-family conflict change as adolescents leave high school, taking on adult roles in the labor force and forming their own families. Finally, we examine the prevalence of actual interference between family and career. We draw on selected prior findings from the Youth Development Study and incorporate new analyses directly relevant to the anticipation and experience of work-family conflict. Results Anticipated Conflict and Plans for Dealing With It The work and family plans and orientations of YDS adolescents in Wave 4, when most were seniors in high school, are quite consistent with the national trends observed in the Monitoring the Future data, and indicate great potential for future work-family conflict. The YDS data further show that adolescents perceive this potential conflict, and that females in particular are planning their futures with this in mind. As seniors in high school, YDS respondents considered both work and family roles as central to their lives, and had high aspirations for educational and occupational attainment, much like seniors in the Monitoring the Future study. YDS participants’ senior year aspirations and orientations related to work and family are summarized in Table 2. Consistent with the Monitoring the Future data, we find that young people in our panel have strong family

Adolescents’ Anticipations of Work-Family Conflict


Table 2. YDS Adolescents’ Work and Family Plans and Aspirations in the Senior Year of High School (Wave 4). Females


Importance of marriage (% very or extremely important) Importance of parenthood (% very or extremely important) Expect to marry (% yes) Anticipated marriage age (years)

88.7 91.7 85.9 24.5

82.7** 78.6*** 76.6** 25.7***

Aspirations for a 4-year college degree (%) Importance attached to career (% very or extremely important) Extrinsic work values (mean) Expect to work after having children (% yes) Expect that spouse will work after having children (% yes)

70.8 93.3 17.6 96.6 90.3

68.0 90.8 + 17.0** 97.6 46.3***

* p < 0.05; ** p < 0.01; *** p < 0.001 for gender difference + A comparison of means by gender (vs. proportion shown here) indicates a statistically significant difference ( p < 0.05).

orientations. Approximately 86% of YDS respondents indicated that marriage and their relationship with their spouse would be very or extremely important to them in the future; 86% also rated future parenthood as this important. As noted earlier, 92% of respondents in the Monitoring the Future study felt family life was highly important, using a similar measure. Most YDS respondents also plan to marry (86% of females, 77% of males). In fact, the proportions of males and females that think they will marry are nearly identical across the two studies. Our local data are also consistent with national trends in that females continue to demonstrate a stronger orientation toward family than males, despite their interest and commitment to paid work. YDS females were more likely to plan to marry; they expected to marry at a younger age by about a year; and they attached greater importance to both marriage and to parenthood than did their male age peers. Only about 5% of females did not plan to have children, whereas this was true for about 13% of males. YDS adolescents also indicate strong interests in education and work. About 70% of the sample aspired to graduate from a four-year college or university. Consistent with these aspirations, males and females in the study placed a great deal of importance on their future work roles. For example, just over 90% of YDS respondents report that their occupation or career will be very or extremely important to them in the future. As noted earlier, a somewhat smaller



proportion of Monitoring the Future respondents thought work would be very central to their lives. Also consistent with the national trends observed in the Monitoring the Future study, YDS females anticipated that their occupation/career would be even more important to them as adults than did YDS males. And whereas the Monitoring the Future data indicated that young women had caught up with males in terms of their concern with the extrinsic rewards of working, YDS females expressed stronger extrinsic reward values than did YDS males. YDS females almost universally expected to work, both after marriage and after having children. High expectations with regard to achievement in paid work and family formation potentially set the stage for later work-family conflict. Moreover, previously reported results from the YDS show that for both males and females anticipated involvement in work and in family go together (Johnson & Mortimer, 2000). These spheres are not seen as being mutually exclusive or constituting a zero-sum game, such that involvement in one sphere necessarily requires pulling back from the other. In fact, those who attached greater importance to family also attached greater salience to career, had higher achievement aspirations, and placed greater importance on the intrinsic and extrinsic rewards of work. Given the high level of anticipated involvement in both employment and family roles, and the fact that they are positively interrelated, there is great potential for these roles to conflict. The findings also indicate, however, that females were more likely than males to plan in anticipation of their need to combine work and family roles (Johnson & Mortimer, 2000). Females with higher educational and occupational aspirations and stronger intrinsic job values planned later marriages. Thus, while they may not have anticipated lesser involvement in family life once formed, greater anticipated involvements in work and education were associated with an anticipated postponement of family life. In contrast, these aspirations did not affect males’ desired marital timing. Moreover, females were more likely to anticipate that their family roles would interfere with their careers than males; twice as many females as males anticipated that family roles would interfere with their career plans when they were seniors in high school (see Figs 5 and 6).3 Females also expected to forego labor force participation, at least for a period of time, to take care of infants or young children. Over two-thirds of the young women expected to be out of the labor force a year or longer after the birth of a child. Few males in our panel expected to remove themselves from the labor force for the purpose of taking care of young children. In fact, substantial proportions of males remained uncertain about, or did not anticipate that their future wives would be in the

Adolescents’ Anticipations of Work-Family Conflict


Fig. 5. Percent Anticipating Interference of Marriage with Career, Waves 4 and 8.

labor force after having children. Only 46% of males expected that their wives would be employed after having children, and 44% responded that they did not know whether or not they expected their wives would work after having children. These expectations regarding employment and childcare arrangements are consistent with the general observations offered by the Monitoring the Future study. The most preferred working arrangements in the Monitoring the Future study were those in which the wife was not employed or was employed part time, with the husband employed full time in either case. Adolescents’ plans

Fig. 6. Percent Anticipating Inferference of Children with Career, Waves 4 and 8.



with respect to combining work and family clearly involve women removing themselves from the labor force for a period of time or limiting their work hours to care for young children. This may be the reason YDS females anticipated that their family roles would interfere with their career plans. A major source of conflict for these young women will be reconciling their expectations about paid work and its rewards with their desires to limit their employment while their children are young. Adolescents’ plans for managing work and family roles reveal their concerns with temporal issues. As seniors, both males and females who planned later marriages anticipated less interference between their family roles and their career plans. This suggests that adolescents expect that delaying family formation would facilitate combining employment and family roles.4 Females who attached greater importance to their future careers, and those who were more interested in the extrinsic rewards of work – income, prestige, security and advancement – also expected to return to the labor market more quickly after having children, devoting less time to full-time child care. Young, careeroriented women may be to some extent aware of what is required of them to achieve their goals, and the negative consequences of labor market withdrawal. As a whole, Youth Development Study participants reflect national trends in that they place great importance on both work and family roles in adulthood. Also consistent with national trends, young women in the Youth Development Study attach greater importance to both of these roles than do young men. By studying this cohort, it is also apparent that young people, especially young women, anticipate difficulty in combining these roles. Two major orientations regarding the resolution of this anticipated conflict are evident. First, both males and females anticipate that women will take time off from work, or reduce their hours of employment, in order to meet their desired goals for family life. Second, young people are planning to marry at later ages in order to realize their goals in both the work and family domains. Change in Anticipated Work-Family Conflict during the Transition to Adulthood As noted earlier, some adolescents anticipated conflict between their future work and family roles. Figures 5 and 6 show the proportions of respondents who anticipated that marriage and having children, respectively, would interfere with their career plans. During their senior year in high school, about one-fifth of females and one-tenth of males anticipated that parenting would interfere with their career plans; about half as many expected that marriage would interfere with their career plans. Thus, consistent with females’

Adolescents’ Anticipations of Work-Family Conflict


continued greater responsibility for children, YDS females at this age were clearly more concerned about children as a source of work-family conflict than were males ( p < 0.001). They were also more concerned about marriage as a potential source of conflict with their career plans ( p < 0.05), perhaps anticipating restriction with respect to geographic mobility needed to pursue their work-related goals, or anticipating needing to follow a husband for his work. The high degree of commitment seniors in high school demonstrated with respect to their future work and family roles was accompanied by a level of concern about how these roles would be combined. Females, who placed greater importance on both roles than did males, and who clearly expected (and were expected by their male peers) to make employment-related sacrifices to care for pre-school children, were the most concerned. The overall proportion of adolescents that reported anticipated conflict was quite low, however. Combining responses across the two family roles, only 23.3% of females and 10.6% of males in Wave 4 anticipated that marriage and/or parenthood would interfere in the future with their career plans. Following high school, more participants came to expect that family roles would interfere with their career plans. As the young people approached actual career and family transitions, more anticipated difficulties. Combining responses across the two family roles, 30.8% of females and 26.4% of males in Wave 8, four years after high school, anticipated that marriage and/or parenthood would interfere in the future with their career plans. This upward trend in anticipating work-family conflict was greater for males than females. By Wave 8 there were no longer significant gender differences in anticipated interference. In other words, males, who were less concerned at an earlier age, caught up with females at least by the time they were about 21 or 22 years old. This is consistent with previous findings from the Youth Development Study, noted in the last section, indicating that females were more planful with respect to combining their future work and family roles as seniors in high school. Females’ educational and occupational aspirations were tied to their expectations for the age at which they would marry (Johnson & Mortimer, 2000). Females face the issues surrounding work and family roles at an earlier stage in life than their male counterparts. The anticipation of conflict runs across major demographic categories. Whether or not adolescents had a parent who attended college, had a family income above the sample median, or came from a two-parent family made no difference in the prevalence of anticipating that family roles would interfere with career plans. Among females, whites were more likely to anticipate



interference than non-whites ( p < 0.05). This racial difference was similar among males, but did not achieve statistical significance. Actual Work-Family Conflict During the Young Adult Years Many respondents had begun to form families during the period immediately following high school. Approximately 14% of females and 10% of males had married by Wave 8 ( p < 0.10), four years after most had graduated from high school. Approximately 33% of females and 14% of males had become parents by this time ( p < 0.001). Two years later, in Wave 9, about 23% of both females and males had married; 40.5% of females and 26.6% of males had become parents ( p < 0.001). Consistent with national data, many more young adults in the Youth Development Study were parents than had married (nearly one-third of births in the U.S. are to single women (National Center for Health Statistics, 1995)). As indicated in our earlier analysis of the Monitoring the Future data, having children without being married is becoming more acceptable to young people. Cohabitation is also more acceptable. By Wave 9, six years after high school, 56% of females and 45% of males in the study had spent at least some period of time cohabiting with a romantic partner ( p < 0.01). Prior analysis of YDS data revealed that relatively early marriage limited young women’s human capital investment in post-secondary education and part-time work, but had no significant implications for young men’s corresponding investments (Mortimer & Johnson, 1999). Young women who married in the first four years following high school achieved fewer months of post-secondary schooling and part-time work than those who had not married by this time. In contrast, early parenting was constraining to both men’s and women’s human capital development through schooling (and part-time work), and, for women only, through full-time labor force participation (Mortimer & Johnson, 1999). In this way, early family formation interfered with educational and occupational attainment. Females’ greater anticipation that family roles would interfere with their career plans is consistent with what transpired when they entered family roles in early adulthood. Young women made greater sacrifices in terms of their investments in post-secondary education and labor force participation. With respect to early parenting, males also made sacrifices in terms of their educational attainment. Additional insight into the ways in which young adults navigate the conflicting demands of work and family roles is gained by examining interruptions in employment with the birth of a child. Over half of those who had become parents by the Wave 8 survey reported that either they themselves or their partners/spouses had taken time off from work with the birth of a child.5 Of those females who reported taking time off, 81% remained out of the labor

Adolescents’ Anticipations of Work-Family Conflict


Table 3. Proportion of YDS Participants Reporting Work-Family Conflict. Females


Experienced interference from marriage (of those married), Wave 8 Experienced interference from marriage (of those married), Wave 9

7.1 33.3

7.2 31.9

Experienced interference from children (of parents), Wave 8 Experienced interference from children (of parents), Wave 9

40.7 77.3

18.8* 53.0*

* p < 0.05; ** p < 0.01; *** p < 0.001 for gender difference

force for at least a month or reduced their work schedule for some longer period. When their partners or spouses took time off from work, 90% took off less than one month. For males, 85% of those who reported taking time off from work remained out of the labor force for less than one month. In fact, just over half of the males reported taking off less than one week of work. Their partners or spouses, on the other hand, did interrupt their own employment. Just over three-quarters of the males’ partners or spouses who took time off from work remained out of the labor force for at least one month. Respondents also reported in Waves 8 and 9 whether their family roles had already interfered with their career plans. The proportions of young men and women who reported such conflicts are presented in Table 3. For those who had entered marriage and/or parenthood, the actual experience of work-family conflict was widespread, particularly by Wave 9. In Wave 8, about 7% of married males and females reported that marriage had interfered with their career plans. By Wave 9, about 30% of married males and females reported that marriage had interfered with their career plans. Males and females appear to be equally likely to have experienced at least some interference when it came to their marital roles. Our measure distinguishes only between the experience of some or no interference, without capturing the extent of the difficulty. Other findings suggest, however, that females in fact experienced greater difficulties. As noted earlier, marriage limited young women’s investments in postsecondary schooling and part-time work, which supports educational investments, but did not limit men’s similar investments. Clearly, children are a source of greater interference with career plans than marriage. Consistent with dominant patterns of child rearing, and women’s earlier plans to interrupt their employment to care for children, females were much more likely than males to report that parenting had interfered with their career plans. An unplanned pregnancy, in particular, can be disrupting and very consequential for a career trajectory. Of those who had become parents by



Wave 8, 18.8% of males and 40.7% of females had experienced interference with their career plans. Of those who had become parents by Wave 9, two years later, 53% of males and 77.3% of females reported that they had experienced interference. It is striking how prevalent this experience was; a majority of both males and females reported that their parental role interfered with their career plans. The proportion of females experiencing this form of difficulty was especially great. Thus, the level of reported work-family conflict among those who have made family transitions was quite high. Moreover, the number of respondents who experienced interference represents over one-fourth of the entire panel (including those who had not married or become parents and were thus not eligible for this form of work-family conflict). Neither race nor whether adolescents had a parent who attended college, had a family income above the sample median, or came from a two-parent family made a difference in the likelihood of experiencing interference of family roles with career plans.6 It is apparent that the experience of work-family conflict in early adulthood was more prevalent than the anticipation of conflict when respondents were seniors in high school. Aronson’s (1999) in-depth interviews with a selection of YDS women in 1996–1997 also indicate that the experience of work-family conflict is greater than the anticipation. Young women who had become mothers, many of whom were not married and had become pregnant as teenagers, experienced great difficulty balancing the conflicting demands of work and caring for their children. They received very little support from the fathers of their children or from their husbands/partners. Women who had not become mothers, in striking contrast, were quite optimistic about how they would balance work and family, and expected to have spouses that would share family responsibilities. Aronson suggests that this latter group of women may have unrealistic views about the ease of combining work and family roles in the future, and that they may later experience difficulties similar to those women who have already become mothers. She also suggests, however, that nonmothers may be better planners. For example, they may delay child bearing until it can be combined well with career demands, and may marry men who will be more involved in family labor.

DISCUSSION AND CONCLUSION Adolescents’ plans and orientations with respect to adult work and family roles have demonstrated both stability and change in response to major changes in work and family structures in the U.S.. During a time of growing uncertainty and greater potential role-conflict in adult work and family life, national trends

Adolescents’ Anticipations of Work-Family Conflict


in adolescents’ orientations toward work and family indicate that young people continue to place great emphasis on both family and work roles. Males and females remain highly interested in marriage and parenting. At the same time, there is a growing acceptance of non-traditional family arrangements and a growing desire to marry at later ages. High aspirations and expectations for work and family on the part of both males and females have great potential for conflict in the future. In the 1990s, females are more committed to their family roles than males and more likely to view work as an important part of their adult lives. Because of their greater commitment to these roles, and their own (as well as their male peers’) preference for having mothers care for pre-school children while fathers work full time, there is greater potential for females to experience conflict as they navigate these roles in early adulthood. The findings from the YDS panel study indicate that young people are somewhat concerned with combining work and family roles, and some do anticipate conflict. In this context, the timing of investments in education, work, and family formation becomes an important part of adolescents’ planning. The results indicate that young people may believe postponing family formation will facilitate the achievement of career and family goals. In the years immediately following high school, more and more young people come to expect that family roles will interfere with their career plans. As more young people face decisions about investments in work, career, and family formation, concern about work-family conflict grows. While still in high school, YDS females were more likely to anticipate work-family conflict. Males caught up, however, in the years after high school. Females begin thinking about the complex issues involved in managing both work and family roles in adulthood at an earlier age and, as mentioned above, demonstrate more strategic planning in this realm. Work-family conflict is an important issue facing members of this cohort. As they move into adult family roles, many report that these roles have indeed interfered with their career plans. The prevalence of work-family conflict is considerably more common than was anticipated in adolescence. Perhaps much of this unexpected difficulty is due to unanticipated pregnancy. Many who did not anticipate having problems combining work and family roles did not anticipate becoming a parent at these ages either. That experienced difficulty exceeds anticipated difficulty also reflects the somewhat unrealistic expectations adolescents generally hold for their futures (Gottfredson, 1981; Shu & Marini, 1997). Our findings do not provide information on the nature of the interference respondents anticipate or experience as they make the transition to adulthood.



Future research is needed to examine how the nature of work-family conflict changes over the life course. The difficulties in combining work and family life, as well as the resources available to deal effectively with such difficulties, likely vary with the timing of family formation and the stages of work careers. The anticipation of work-family conflict is becoming an important feature of growing up. While maintaining high expectations for their futures, young people are also thinking of the difficulties involved in meeting the demands of both work and family roles. As seniors in high school, only a small group of young people (primarily females) was specifically concerned that their future family roles might interfere with their careers. In the years that immediately follow, as young people move into adult roles, more become concerned. While those anticipating this kind of difficulty remain a minority, it should be noted again that this represents only one type of difficulty young people may encounter in balancing their work and family lives. Moreover, the pervasive extent of work-family conflict among those making family transitions during this stage in the life course is striking. Because of data limitations, we cannot directly compare young people today with those of earlier cohorts. However the widespread experience of conflict is certainly compatible with Buchmann’s (1989) characterization of the early life course as becoming increasingly uncertain and blurred with respect to age-graded expectations and roles. Further research on the transition to adulthood at different historical times is required to find out whether this life course stage is becoming more difficult and complex, and whether young people increasingly navigate this process in an individualized manner.

NOTES 1. The Monitoring the Future data were obtained from the following published sources: Bachman et al., 1980, 1981, 1986, 1993; Johnston et al., 1997; and Schulenberg et al., 1995. 2. Consent to participate was obtained from 64% of all eligible invitees. A probit analysis of the decision to participate was conducted to test whether participants selected themselves on the basis of high socioeconomic status. This analysis used information from the 1980 Census reported at the tract level, the most recent Census data available when data collection began, to characterize the neighborhoods of all eligible families. The contextual dimensions constituted the independent variables in the equation, along with age, gender, and the school in which the child was enrolled; the decision to participate was the dependent variable. This analysis is reported elsewhere (Finch, Shanahan, Mortimer & Ryu, 1991); importantly, no socio-economic contextual variables (for example, race, family composition, median household income, parents’ education and occupational level) proved to be statistically significant predictors of

Adolescents’ Anticipations of Work-Family Conflict


participation in the study. Thus, the sample is representative of the general population of ninth graders attending the St. Paul public schools. 3. The direction of work-family conflict is important in understanding adolescents’ anticipations. A study examining the anticipation that work would conflict with family life found male college students to be more concerned than females (Livingston et al., 1996). Together, the findings of the two studies clearly reflect continued expectations of greater female responsibility for household labor and greater male responsibility for paid labor. 4. Alternatively, confidence could lead to both delay in family formation and less anticipated interference, rendering this relationship spurious. 5. Those parents who were not in the labor force at the time of the child’s birth are not considered to have taken time off, but are included in the denominator for these proportions. 6. We also examined whether anticipating interference as a high school senior influenced the likelihood of family formation in the first few years after high school. Such anticipation might be protective, leading adolescents to postpone family formation. Alternatively, it might be predictive, in that anticipating interference is simply an accurate prediction of one’s future. Neither of these hypotheses received support in the current study.

ACKNOWLEDGMENT This research was supported by a grant, Work Experience and Mental Health: A Panel Study of Youth, from the National Institute of Mental Health (M42843).

REFERENCES Ahlburg, D. A., & De Vita, C. J. (1992). New realities of the American family. Population Bulletin, 47, 1–44. Aronson, P. (1999). The balancing act: Young women’s expectations and experiences of work and family. In: T. Parcel (Ed.), Research in the Sociology of Work (Vol. 7, pp. 55–83). Greenwich CT: JAI Press. Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1980). Monitoring the Future: Questionnaire responses from the nation’s high school seniors, 1976. Ann Arbor, MI: Institute for Social Research, University of Michigan. Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1981). Monitoring the Future: Questionnaire responses from the nation’s high school seniors, 1980. Ann Arbor, MI: Institute for Social Research, University of Michigan. Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1986). Monitoring the Future: Questionnaire responses from the nation’s high school seniors, 1985. Ann Arbor, MI: Institute for Social Research, University of Michigan. Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1993). Monitoring the Future: Questionnaire responses from the nation’s high school seniors, 1990. Ann Arbor, MI: Institute for Social Research, University of Michigan. Bianchi, S. M., & Spain, D. (1986). American women in transition. New York: Russell Sage Foundation.



Bianchi, S. M., & Spain, D. (1996). Women, work, and family in America. Population Bulletin, 51(3). Washington, D.C.: Population Reference, Inc., December 1996. Buchmann, M. (1989). The script of life in modern society: Entry into adulthood in a changing world. Chicago: University of Chicago Press. Coltrane, S. (2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family, 62, 1208–1233. Finch, M. D., Shanahan, M., Mortimer, J. T., & Ryu, S. (1991). Work experience and control orientation in adolescence. American Sociological Review, 56, 597–611. Freedman, D., Thornton, A., Camburn, D., Alwin, D., & Young-Demarco, L. (1988). The life history calendar: A technique for collecting retrospective data. In: C. C. Clogg (Ed.), Sociological Methodology (Vol. 18). San Francisco, CA: Jossey-Bass. Gager, C. T. (1995). Is love stronger than justice?: Perceptions of fairness in the division of household labor. Unpublished Doctoral Dissertation, University of Pennsylvania. Gerson, K. (1985). Hard choices: How women decide about work, career, and motherhood. Berkeley: University of California Press. Gerson, K. (1998). Gender and the future of the family: Implications for the postindustrial workplace. In: D. Vannoy & P. J. Dubeck (Eds), Challenges for Work and Family in the Twenty-first Century (pp. 11–21). New York: Walter de Gruyter. Goldin, C. (1990). Understanding the gender gap: An economic history of American women. New York: Oxford University Press. Gottfredson, L. S. (1981). Circumscription and compromise: A developmental theory of occupational aspirations. Journal of Counseling Psychology Monograph, 28, 545–79. Hochschild, A. (1989). The second shift. New York: Basic Books. Hochschild, A. (1997). The time bind: When work becomes home and home becomes work. New York: Metropolitan Books. Johnson, M. K. (1999). Change in job values during the transition to adulthood. Paper presented at the Annual Meeting of the American Sociological Association, Chicago, August. Johnson, M. K., & Mortimer, J. T. (2000). Work-family orientations and attainments in the early life course. In: T. L. Parcel & D. B. Cornfield (Eds), Work and family: Research informing policy (pp. 215–48). Sage. Johnston, L. D., Bachman, J. G., & O’Malley, P. M. (1997). Monitoring the Future: Questionnaire responses from the nation’s high school seniors, 1995. Ann Arbor, MI: Institute for Social Research, University of Michigan. Kalleberg, A. L., Rasell, E., Hudsin, K., Webster, D., Reskin, B. F., Cassirer, N., & Appelbaum, E. (1997). Nonstandard work, substandard work: Flexible work arrangements in the U.S. Washington, D.C.: Economic Policy Institute. Livingston, M. M., Burley, K., & Springer, T. P. (1996). The importance of being feminine: Gender, sex role, occupational and marital role commitment, and their relationship to anticipated work-family conflict. Journal of Social Behavior and Personality, 11, 179–192. Marini, M. M. (1984). Women’s educational attainment and the timing of entry into parenthood. American Sociological Review, 49, 491–511. Marini, M. M., Shin, H., & Raymond, J. (1989). Socioeconomic consequences of the process of transition to adulthood. Social Science Research, 18, 89–135. Marini, M. M., Fan, P., Finley, E., & Beutel, A. M. (1996). Gender and job values. Sociology of Education, 69, 49–65.

Adolescents’ Anticipations of Work-Family Conflict


Mischel, L., Bernstein, J., & Schmitt, J. (1999). The state of working America 1998–99. Ithaca, NY: Cornell University Press. Mortimer, J. T., & Johnson, M. K. (1999). Adolescent part-time work and post-secondary transition pathways: A longitudinal study of youth in St. Paul, Minnesota. In: W. Heinz (Ed.), From Education to Work: Cross National Perspectives (pp. 111–148). New York: Cambridge University Press. National Center for Health Statistics (1995). Vital statistics of the United States, 1991. Vol I, Natality. Washington, D.C.: Public Health Service. Nurmi, J. (1991). How do adolescents see their future? A review of the development of future orientation and planning. Developmental Review, 11, 1–59. Nurmi, J. (1997). Self-definition and mental health during adolescence and young adulthood. In: J. Schulenberg, J. Maggs & K. Hurrelmann (Eds), Health Risks and Developmental Transitions During Adolescence (pp. 395–419). Cambridge University Press. Oppenheimer, V. K. (1994). Women’s rising employment and the future of the family in industrial societies. Population and Development Review, 20, 293–342. Oppenheimer, V. K. (1997). Women’s employment and the gain to marriage: The specialization and trading model. Annual Review of Sociology, 23, 431–53. Presser, H. B. (1998). Toward a 24 hour economy: The U.S. experience and implications for the family. In: D. Vannoy & P. J. Dubeck (Eds), Challenges for Work and Family in the Twentyfirst Century (pp. 39–47). New York: Walter de Gruyter. Sassler, S., & Schoen, R. (1999). The effect of attitudes and economic activity on marriage. Journal of Marriage and the Family, 61, 147–59. Schor, J. (1991). The overworked American: The unexpected decline in leisure. New York: Basic Books. Schulenberg, J., Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (1995). American adolescents’ views on family and work: Historical trends from 1976–1992. In: P. Noack, M. Hofer & J. Youniss (Eds), Psychological Responses to Social Change: Human Development in Changing Environments. New York: Walter de Gruyter. Seltzer, J. A. (2000). Families formed outside marriage. Journal of Marriage and the Family, 62, 1247–1268. Shapiro, I., & Greenstein, R. (1999). The widening income gulf. Washington, D.C.: Center on Budget and Policy Priorities. Shu, X., & Marini, M. M. (1997). Occupational aspirations and opportunities: Coming to terms with life’s limitations. Paper presented at the 1997 Annual Meetings of the American Sociological Association, Toronto. Sewell, W. H., & Hauser, R. M. (1975). Education, occupation, and earnings: Achievement in the early career. New York: Academic Press. Wallace, M. (1998). Downsizing the American dream: Work and family at century’s end. In: D. Vannoy & P. J. Dubeck (Eds), Challenges for Work and Family in the Twenty-first Century (pp. 23–38). New York: Walter de Gruyter.

This Page Intentionally Left Blank



Tama Leventhal and Jeanne Brooks-Gunn ABSTRACT Given growing interest in the neighborhoods in which children live and interact, this article examines current findings and provides a developmental framework for the coming century. Our approach moves away from viewing neighborhood contexts as a form of “social address” to considering the processes through which neighborhood effects might operate on child well-being. Also emanating from this concern, the framework regards neighborhood influences on child development as well as neighborhood contexts per se as dynamic rather than static. Three models are identified: Institutional resources – the quality, availability, accessibility, and affordability of different resources in the community; Relationships – parental characteristics, support networks, and behavior and home environment attributes; and Norms/Collective efficacy – the extent of community formal and informal institutions present to monitor the behavior of residents as well as the presence of physical risk to residents. Building on these models, examples of research incorporating

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 263–301. 2001 by Elsevier Science Ltd. ISBN: 0-7623-0776-5




neighborhood change are provided using data on children followed over the first 8 years of life, including an investigation of the dynamics of neighborhood conditions (duration and timing) and an exploration of residential mobility patterns. Residing more years in an affluent neighborhood by age 5 was positively associated with children’s IQ scores, and neighborhood affluence at birth and age 5 and neighborhood poverty at age 5 were associated with children’s IQ and achievement scores at age 8. In terms of mobility patterns, a majority of families remained in similar neighborhoods over time with the biggest shift in families’ circumstances occurring when children were young; most changes in neighborhood economic conditions centered around movements into and out of middle-income neighborhoods. Finally, the last section points to the importance of considering the role of neighborhoods in children’s lives in the twenty-first century and identifies the research designs and methodologies scholars can employ to incorporate this developmental framework into their research.

While the corpus of research that has emerged over the last decade tells us what neighborhood structural dimensions (e.g. income/SES) are associated with particular outcomes (see Leventhal & Brooks-Gunn, 2000, for a review of these findings), it does not shed light on the processes through which neighborhood effects may operate on child and adolescent well-being. In essence, we have not progressed much further from where we were in 1990, at the time of the now classic review by Jencks and Mayer (1990). Their review identified five theoretical models for linking individual outcomes with neighborhood influences: (a) “neighborhood institutional resource models” which suggest that neighborhood resources may affect children through access to resources that promote healthy development, such as libraries, parks, police protection, and medical facilities; (b) “collective socialization models” which posit that neighborhood influences affect children via community social organization, including the presence of adult role models, supervision, and monitoring; (c) “contagion (or epidemic) models” which highlight problem behavior and hypothesize that the negative behavior of neighbors or peers strongly affects the behavior of others; (d) “competition models” which purport that neighbors (or peers) compete for scarce community resources; and (e) “relative deprivation models” which suggest that neighborhoods affect individuals through their assessments of their situations relative to neighbors or peers. Despite calls for more theoretically driven work, contemporary neighborhood research continues to view neighborhood contexts as another form of “social address”

Changing Neighborhoods and Child Well-Being


(similar to the social address of SES of a family) with little specification of actual processes. Also emanating from this concern, most neighborhood researchers have examined neighborhood influences on child development as well as neighborhood contexts per se as static. A majority of families move from neighborhood to neighborhood (Wood, Halfon, Scarlata, Newacheck & Nessim, 1993). In addition, moving in and of itself can be stressful for children if it requires changing schools, making new friends, and adjusting to a new neighborhood (even if there is no shift in neighborhood resources); however, there are likely to be developmental age sensitivities to the effects of moving with potentially larger consequences for adolescents than children (Haveman, Wolfe & Spaulding, 1991; McLanahan & Sandefur, 1994). The existing neighborhood research has also not acknowledged that the economic and social structure of neighborhoods also changes (even if families do not move; see Korbin & Coulton, 1997, for an example of different neighborhood trajectories in the Cleveland, Ohio, area). In this article, we take stock of where the current field of neighborhood research is at as well as where it will be going in the next century. We address the concerns outlined, in turn, and whenever possible, provide links across them. Specifically, the first section draws upon the existing literature on neighborhood effects on child and adolescent well-being as well as other relevant research to reformulate the theoretical models outlined by Jencks and Mayer (1990) to aid empirical explorations of the processes through which neighborhood influences may operate on development. Building on the theoretical models, the following section provides an exemplar for exploring the dynamics of neighborhood conditions, modeled after the literature on the dynamics of family income. Finally, the last section points to the importance of considering the role of neighborhoods in children’s lives in the twenty-first century and how scholars can incorporate the developmental framework laid out here into their research (see Brooks-Gunn, Duncan, Leventhal & Aber, 1997; Duncan & Raudenbush, 1999; Leventhal & Brooks-Gunn, 2000; Raudenbush & Sampson, 1999, for methodological reviews of neighborhood effects).

WHERE WE ARE GOING: MECHANISMS OF NEIGHBORHOOD EFFECTS ON CHILD AND ADOLESCENT OUTCOMES Before addressing the discussion of theoretical models, it is important to ascertain: “What is a neighborhood?” Several approaches have been used to



define the neighborhood unit of analysis. The most common strategy is to rely upon information from the U.S. Decennial Census, available from the census forms the population fills out on the first of April during the first year of every decade. A neighborhood is frequently defined in terms of a census tract; tract boundaries are identified with the advice of local communities working under Census Bureau guidelines and, generally, reflect prominent physical and social features that delineate neighborhoods (e.g. major streets, railroads, ethnic divisions). Census tracts contain approximately 3,000 to 8,000 individuals. Other bureaucratically defined units available from administrative data sources and, typically employed in city-specific studies, include health districts, police districts, and school districts; these units often overlap and are used in conjunction with census data. Alternatively, ethnographic descriptions of neighborhoods suggest that study participants perceive boundaries differently. Likewise, when participant reports of neighborhood conditions are collected, neighborhood boundaries are frequently not specified; however, residents’ reports of neighborhood boundaries have been found to closely resemble census tracts (Sampson, 1997). Unless participant reports are obtained from a separate sample, as in the case of a community survey, neighborhood measures may be confounded with other measures reported by study participants. We address measurement issues further in the last section. In addition, most neighborhood research has centered around urban rather than rural areas; while much of this discussion may apply, rural context may have their own unique set of issues that are not fully addressed here (see Crockett, Shanahan & JacksonNewsom, 2000, for a review). Explorations of the pathways/processes of neighborhood influences also use several strategies. One approach has been to investigate neighborhood structural dimensions to see whether each has a positive or negative effect on child well-being and to hypothesize how neighborhood influences may be operating using the five models identified by Jencks and Mayer (1990). A second strategy has been to examine threshold or non-linear effects of neighborhoods on child and adolescent well-being using a contagion framework (Crane, 1991; Jencks & Mayer, 1990). A third approach has been to examine the association between child and adolescent well-being and neighborhood structural and behavioral characteristics thought to be associated with neighborhood social organization (Coulton, Korbin, Su & Chow, 1995; Elliott, Wilson, Huizinga, Sampson, Elliot & Rankin, 1996; Sampson, Raudenbush & Earls, 1997; Sampson & Groves, 1989). Several of these researchers have even explored the mechanisms through which neighborhood organization is fostered or hindered and the consequences for child development (Elliott et al., 1996; Sampson, 1997; Sampson & Groves, 1989). A fourth

Changing Neighborhoods and Child Well-Being


ecological strategy focuses on the multiple contexts that shape children and youth including families, schools, and peers as well as neighborhoods (Aber, Gephart, Brooks-Gunn, Connell & Spencer, 1997). These different contextual influences have been conceptualized as both mediators (e.g. neighborhood effects may be transmitted through parental behavior or school characteristics) and moderators (e.g. neighborhood effects may vary depending on peer group attributes). In general, all of these approaches hypothesize that neighborhood effects are most likely indirect, operating through more proximal behaviors, although accompanying empirical explorations of these underlying mechanisms of neighborhood influences have been impeded because theoretical and empirical specifics are not coherently outlined (i.e. by outcome, age of child, and specific mediators or moderators). The limited empirical work investigating mediated models reveals that, in fact, neighborhood effects are largely indirect, operating through community- and family-level processes; approximately half of the variance in neighborhood effects on child and adolescent outcomes appears to be accounted by these processes (Elliott et al., 1996; Klebanov, Brooks-Gunn, Chase-Lansdale & Gordon, 1997; Sampson, 1997). Building on the work of Jencks and Mayer (1990) and the research that has followed, we lay out a framework by specifying: (a) more theoretically driven models; (b) the level at which each mechanism might operate (individual, family, school, peer, community); (c) the processes through which the mechanisms may operate; and (d) models that may be most relevant for studying particular outcomes and developmental age groupings (i.e. young childhood, late childhood, young adolescence, and late adolescence). Moving beyond the current approach, we also incorporate a dynamic component into these models, addressing issues of neighborhood influences over time, movement of individuals across neighborhoods, and change within neighborhoods. The models include: Institutional Resources – the quality, availability, accessibility, and affordability of different resources in the community; Relationships – Parental characteristics, support networks, and behavior and home environment attributes; and Norms/Collective Efficacy – the extent of community formal and informal institutions present to supervise and monitor the behavior of residents (especially youth) as well as the presence of physical risk to residents. According to Jencks and Mayer’s (1990) framework, the institutional resource model combines resource and competition models because the hypothesized mechanism of community-level resources overlaps. Relative deprivation may also be subsumed under this model with individuallevel mechanisms at play. Contagion and collective socialization models are



combined under the norms and collective efficacy frame because community institutions and peer group behavior are the primary vehicles of neighborhood influences for both models; these models are also difficult to disentangle empirically with both focusing on negative behaviors. Finally, the relationship model includes elements of collective socialization but is limited to familylevel mechanisms of regulation. The revised models are complementary rather than conflicting with the utility of each model depending largely on the outcome of interest. Institutional resource mechanisms may be most relevant for studying achievement outcomes, and norms/collective efficacy models most salient for investigations of problem behaviors. Relationship mechanisms may be most useful for examining how parents actually socialize their children or how parents’ own relationships influence parental socialization. Relationship mechanisms may be particularly relevant for young children due to the primacy of the family during the early years; whereas, during adolescence, collective efficacy models may be most salient because of the prominence of the peer group during this period and the need for community supervision and monitoring. In terms of institutional resource models, throughout the childhood years, parents or parental figures manage and supervise children’s lives, so institutional resources are likely to have more indirect effects on children through familial processes. Institutional Resources The quality, availability, accessibility, and affordability of different types of institutional resources in the community, including learning, social and recreational activities, child care, schools, health care services, and employment opportunities, are relevant to child and adolescent development. While there are a few studies exploring the association between these community resources and child well-being, the question of whether exposure to institutional resources over time influences development in a linear or nonlinear fashion has not been raised; this issue is explored in the following section. However, the literature on divorce and single-parenthood points to how a change (or discontinuity) in neighborhood conditions may be associated with child development. Specifically, residential mobility was found to account, in large part, for the negative consequences of parental divorce and separation on adolescent outcomes (McLanahan & Sandefur, 1994). The researchers hypothesize that the decline in family economic resources that accompanies parental divorce and separation is associated with moves to neighborhoods with lower quality resources (although, this premise was not directly tested). Now,

Changing Neighborhoods and Child Well-Being


we examine the evidence linking neighborhood resources to child development. The availability of learning activities in the community, such as libraries, family resource centers, literacy programs, and museums for families to draw upon for their children’s learning stimulation may promote school readiness and achievement-related outcomes. A study of young children in the Infant Health and Development Program (IHDP) found that learning experiences outside of the home did not mediate neighborhood effects on children’s cognition; learning experiences inside the home mediated neighborhood effects on child outcomes, which may be linked to relationship mechanisms (Klebanov, Brooks-Gunn, McCarton & McCormick, 1998). Additional research with older children who have more direct access to community resources is merited. The presence of social and recreational activities, such as parks, sports programs, art and theater programs, and community centers, may foster children’s physical and social development. Quantitative and qualitative research suggests that when neighborhoods have few learning, recreational, or social activities available within the immediate community, parents will use alternative strategies to obtain resources for their children, such as drawing upon resources from surrounding communities (Elder, Eccles, Ardelt & Lord, 1995; Jarrett, 1997). For young children, the availability, affordability, and quality of child care available in the community is another important resource through which neighborhood effects may be transmitted. As documented by a growing body of research, the attributes of available child care and early intervention programs have implications for children’s cognitive, socioemotional, and physical development as well as parenting outcomes (Brooks-Gunn, Berlin & Fuligni, 2000; Benasich, Brooks-Gunn & Clewell, 1992; Campbell & Ramey, 1994; Hayes, Palmer & Zaslow, 1990; Lee, Brooks-Gunn, Schnur & Liaw, 1990; McKey, Condelli, Granson, Barrett, McConkey & Plantz, 1985; Reynolds, 1994; Yoshikawa, 1994). The results of community-level analyses revealed that the supply was low and the quality of child care poor in lowincome neighborhoods, suggesting that child care may be a resource for which residents in these neighborhoods compete (Fuller, Coonerty, Kipnis & Choong, 1997). Schools are an important institutional resource for older children and adolescents and are perhaps a more proximal environment than neighborhoods. Neighborhood resources may shape school characteristics, such as quality, climate, and demographics, and in turn, influence outcomes (Jencks & Mayer, 1990). A recent study based on experimental data found that children and youth who moved from public housing in high-poverty neighborhoods to low-poverty



neighborhoods and to low- to middle-income neighborhoods were more likely to attend schools with higher pass rates on achievement tests than their peers who stayed in high-poverty neighborhoods, and discrepancies in school environments were due in large part to the presence of more resources and more advantaged student populations (Ludwig & Ladd, 1998). Limited evidence suggests that school characteristics, such as students’ attitudes, may mediate neighborhood effects on older children’s and adolescents’ risk behaviors (e.g. Ennett, Flewelling, Lindrooth & Norton, 1997). Access to health care services is another community resource through which neighborhood effects may operate. A majority of the research focusing on links between neighborhood structural characteristics and health outcomes (e.g. low birth weight, injury, maltreatment) has not examined the extent of health and social services present in the community. In related work, residing in low- and middle-income neighborhoods was associated with more emergency room visits than residing in affluent neighborhoods; families living in middle-income neighborhoods reported more doctor visits than families in low- and highincome neighborhoods (effects were found controlling for family characteristics). In addition, among children from lower income families, those who resided in higher income neighborhoods had fewer doctors’ visits than their counterparts who resided in lower income neighborhoods (Brooks-Gunn, McCormick, Klebanov & McCarton, 1998). These results indicate that access to different types of health services may vary by neighborhood SES. Opportunities for employment in the community are relevant for adolescents and may operate at two levels. The first, actual employment opportunities available in the community, is a community-level mediator, and the second, adolescents’ expectations about the opportunities for employment, is an individual-level mediator. This latter dimension roughly corresponds with Jencks and Mayer’s (1990) models of collective socialization and relative deprivation in terms of individuals weighing their own options relative to their family, neighbors, and peers in the surrounding community. In essence, adults in the neighborhood may act as role models for children and youth. While studies with large survey samples highlight the negative developmental consequences of extensive adolescent employment, such as problem behavior and substance use, researchers focusing on low-income samples emphasize the benefits of adolescent employment, such as economic gains and adult monitoring that may lead to increased school engagement and decreased problem behavior (Bachman & Schulenberg, 1993; Gleason & Cain, 1997; Leventhal, Graber & Brooks-Gunn, in press; Mortimer, Finch, Ryu, Shanahan & Call, 1996; Newman, 1999; Steinberg, Fegley & Dornbusch, 1993; Sullivan, 1989, 1996). Depending on the neighborhood context, opportunities for

Changing Neighborhoods and Child Well-Being


employment may hinder access to resources or provide access to resources. Employment opportunities for minority youth in low-income neighborhoods may be restricted, and these youth may encounter residential and racial segregation in the local labor market (Freeman, 1991; Iceland, 1997; Kirschenman & Neckerman, 1991; Newman, 1999; Osterman, 1991; Wilson, 1997). Low-income, minority youth in the Gautreaux Program who moved from public housing to the suburbs were more likely to be employed, have jobs with benefits, and earn higher wages than their peers who remained in the city (Kaufman & Rosenbaum, 1992; Popkin, Rosenbaum & Meaden, 1993). At the individual-level, expectations about the employment opportunities available to adolescents, undoubtedly, are altered by neighborhood contexts. How children and youth, especially those from low-income communities, reconcile discrepancies between their expectations and their daily observations of adults around them may affect educational attainment, teenage childbearing, criminal involvement, and substance use (Billy, Brewster & Grady, 1994; Ogbu, 1991; Paulter & Lewko, 1987; Willis, 1977). Relationships There are several ways in which parental relationships may intervene between neighborhood influences and child and adolescent well-being, including parental characteristics and support networks available to parents, parental behavior, and the home environment. The literature on economic hardship and unemployment suggests that low-income and child outcome links are mediated by feelings of financial strain and depression that in turn affect parenting behavior (Conger, Ge, Elder, Lorenz & Simons, 1994; McLoyd, 1990). Parents’ relationships moderate or mediate the impact of depression (mental and physical health) and are important for parental active coping and sense of efficacy as a parent. Extrapolating from this theoretical model, neighborhood poverty could potentially alter parental well-being and subsequent child outcomes vis-à-vis parenting behavior. One study, however, found that neighborhood poverty and affluence (both compared to neighborhood middleincome) were not associated with mothers’ of emotional distress or coping skills (but researchers did not examine links between neighborhood conditions, relationships and child outcomes; Klebanov, Brooks-Gunn & Duncan, 1994). Among low-income mothers who moved out of public housing in high-poverty neighborhoods into low-poverty neighborhoods, superior mental and physical health was reported compared with mothers who remained in high-poverty neighborhoods (Katz, Kling & Liebman, in press; Leventhal & Brooks-Gunn, in press). Yet another study based on families with adolescents who lived in



disadvantaged neighborhoods found that parental efficacy mediated the use of family management strategies among African American parents but not European American parents (Elder et al., 1995). Parents’ access to support networks and social connections have been hypothesized to mediate the effect of neighborhood economic resources on child and adolescent development (Cook, Shagle & Degirmencioglu, 1997). Again, based on models of economic hardship, social support available to parents might mitigate the stress associated with living in a dangerous and impoverished neighborhood, and consequently, reduce the adverse effects of parental stress on child outcomes (Conger et al., 1994; Elder et al., 1995; McLoyd, 1990). In addition, friends and relatives within the neighborhood might be relied upon for child care when parents are unavailable to monitor their children (Logan & Spitze, 1994). Residential mobility has the potential to disrupt parents’ and children’s social networks leading to more unfavorable child outcomes (McLanahan & Sandefur, 1994). However, among minority, low-income families in the Yonkers Project and Gautreaux Program, no differences in social support were found between families who moved to middle-income neighborhoods and families who stayed in low-income urban neighborhoods (Briggs, 1997a, b; Rosenbaum. Popkin, Kaufman & Rusin, 1991). Finally, links between neighborhood SES and parental social support have been found such that social support was lower in high- and low-income neighborhoods compared with middle-income neighborhoods (Klebanov et al., 1994). Parental behavior thought to be associated with neighborhood of residence and subsequent child well-being includes warmth, harshness, and supervision and monitoring. Parental warmth involves emotional responsivity and support displayed by parents towards children, while harshness entails critical and unresponsive behavior towards children. Supervision and monitoring include the extent to which parents oversee and regulate their children’s activities. Living in a poor neighborhood (compared with a middle-income neighborhood) was found to be associated with lower maternal warmth with preschoolers (Klebanov et al., 1994). In addition, a negative association between community disadvantage (assessed via Census) and adolescent problem behavior was found to be accounted for by the overall quality of parenting (monitoring, warmth/support, inductive reasoning, harsh discipline, hostility, and communication), assessed via videotaped parent-child interactions (Simons, Johnson, Beaman, Conger & Whitbeck, 1996). Again, drawing from the models of economic hardship, neighborhood conditions, particularly poverty and danger, may increase parental stress and anxiety and lead to more harsh parenting (Conger et al., 1994; McLoyd, 1990).

Changing Neighborhoods and Child Well-Being


There is some empirical work linking neighborhood conditions to harsh parenting. Specifically, two experimental studies found that parents who moved to more affluent neighborhoods used less harsh parenting practices than parents who remained in poorer neighborhoods (Briggs, 1997a; Leventhal & BrooksGunn, in press); in one of the studies superior mental health was also reported among mothers who moved. Another study found that parental reports of neighborhood danger were positively associated with the use of harsh control and verbal aggression (Earls, McGuire & Shay, 1994). Parental supervision and monitoring could mediate neighborhood effects by increasing or decreasing children’s exposure to the surrounding neighborhood; however, the subsequent effect of parental supervision and monitoring may depend on the neighborhood context (Gonzales, Cauce, Friedman & Mason, 1996; Lamborn, Dornbusch & Steinberg, 1996; Pettit, Bates, Dodge & Meece, 1999). Parents in dangerous and impoverished neighborhoods may employ more restrictive parenting practices with the goal of limiting exposure to harmful community influences, according to ethnographers (Anderson, 1991; Burton, 1990; Burton & Jarrett, in press; Furstenberg, 1993; Jarrett, 1997). Empirical data support this premise. In a quasi-experimental study of mobility, low-income parents who moved to middle-income neighborhoods used less restrictive monitoring practices with youth than parents who stayed in the lowincome neighborhoods (Briggs, 1997a). Another study explicitly investigating mediated effects found that the positive association between neighborhood low-SES and adolescent pregnancy was mediated by parental monitoring of early dating behavior (Hogan & Kitagawa, 1985). Several aspects of the home environment are hypothesized to be associated with indirect neighborhood effects on children and adolescents, including the provision of learning experiences, the physical environment, the presence of routines and structure, and the level of violence. The provision of learning experiences in the home entails, for example: reading to or with the child; the presence of books and reading materials; the presence of a work space for the child; and age-appropriate toys/games (Bradley, 1995; Caldwell & Bradley, 1984). The beneficial effect of living in a high-SES neighborhood on children’s cognition, verbal ability, and behavior problem scores was mediated by the quality of the home learning environment in two samples of young children (controlling for family characteristics; Klebanov et al., 1997). In a subsequent study (Klebanov et al., 1998), the provision of learning experiences inside the home, not outside of the home as might have been expected, was found to mediate neighborhood effects on child outcomes (i.e. mothers in poor/ dangerous neighborhoods were not less likely to take their children on educationally-enriching outing). What parents do with their children on a one-



on-one basis (stimulating activities in the home) seems to be critical for their children’s cognitive growth rather than mere exposure to outside-the-home activities. Another recent study also found that the negative effect of neighborhood risk (assessed via participant report) on young children’s teacher-reported social competence and reading achievement was partially accounted for by the quality of the home environment (controlling for child and family characteristics; Greenberg, Cole, Lengua & Pinderhughes, 1999). Characteristics of the physical environment include safety, cleanliness, space allocation, lighting, and decor (Bradley, 1995; Caldwell & Bradley, 1984). The quality of the physical environment of the home is likely to have the largest effect on children’s physical health. Living in a neighborhood with a high proportion of low-income residents is associated with lower quality physical home environments than living in a neighborhood with a high proportion of middle-income residents, controlling for family SES (Klebanov et al., 1994). Residing in a low-income neighborhood, in turn, has been found to be associated with an increased risk of child injury, likely due in part to unsafe play areas inside the home (Durkin, Davidson, Kuhn, O’Connor & Barlow, 1994). Along these same lines, a randomized study of mobility found that children who moved out of public housing in high-poverty neighborhoods had fewer injuries and asthma attacks in the past six months than children who remained in high-poverty neighborhoods (Katz et al., in press). Routines and structure established by parents, such as regular mealtimes and bedtimes, are thought to be central to children’s development (Bradley, 1995; Boyce, Jensen, James & Peacock, 1983). Routines are important to models of community socialization as proposed by Wilson (1987). At the theoretical level, Wilson (1987, 1991) has argued that routines are lacking in many disadvantaged neighborhoods because of the decline in social structure. However, indirect models assessing links between neighborhood effects and child outcomes via routines have not been empirically investigated. One study based on a randomized mobility experiment found that among families who lived in public housing in high-poverty neighborhoods, parents who moved into private housing in low-poverty neighborhoods provided more structure and routines in their children’s lives than parents who remained in poor neighborhoods (Leventhal & Brooks-Gunn, in press). Alternatively, researchers investigating residential mobility suggest that moving (above and beyond neighborhood quality) is likely to alter practices within the home with potentially negative consequences for children and youth, at least in the short-term (Leventhal, Brooks-Gunn & Klebanov, 2000). Children’s exposure to violence in the home as either a witness or a victim is another mechanism through which neighborhood effects may be transmitted

Changing Neighborhoods and Child Well-Being


to children and adolescents (especially mental and physical health outcomes). A well-known study of a poor urban community found that children were exposed to high levels of violence within both the neighborhood and the home (Martinez & Richters, 1993; Richters & Martinez, 1993). Researchers from the Project on Human Development in Chicago Neighborhoods (PHDCN) are beginning to explore the intersection of violence in the home and violence in the community (Earls, personal communication; see Kindlon, Wright, Raudenbush & Earls, 1996; Selner-O’Hagan, Kindlon, Buka, Raudenbush & Earls, 1998, for methodological work underway). Norms/Collective Efficacy Social disorganizational theory hypothesizes that community structural characteristics (low-SES, ethnic heterogeneity, residential instability, singleparenthood) determine the extent of formal and informal institutions that are present to supervise and monitor the behavior of residents, especially youth. Besides monitoring residents’ activities, community institutions regulate the proliferation of physical risk to resident, particularly children and youth, including violence and victimization, access to harmful and illegal substances, and other threats to physical well-being. Mechanisms include supervision, monitoring, and collective norms as well as the processes through which they are established. Neighborhood social organization is thought to be disrupted when internal neighborhood structure undergoes change, for instance, when an influx of new immigrants enters a neighborhood or a substantial loss of employment opportunities occurs (Coulton & Pandey, 1992; Wilson, 1987, 1997). Under the collective efficacy frame, the social connections that exist among residents entail mutual trust, shared values, and a willingness to intervene on behalf of the community (these types of relations are much more diffuse than friendship networks and close personal ties [see Sampson, 1999, for more detailed distinction]). Recently, Sampson and colleagues (1997) have termed this sociostructural attribute “collective efficacy” based on the work of Bandura (1986). More specifically, collective efficacy (assessed via a community survey) describes the extent of social connections in the community and the degree to which residents monitor the behavior of others in accordance with socially accepted practices and the overall goal of supervising children and maintaining public order (i.e. a combined measure of social control and social cohesion). In other words, collective efficacy is the ability of neighborhood residents to supervise the activities of children and youth. Low rates of residential mobility in the neighborhood have been shown to foster collective



efficacy (Elliott et al., 1996; Sampson et al., 1997). At both the community- and individual-levels, mechanisms of social control are linked to rates of problem behavior (Elliott et al., 1996; Sampson, 1997; Sampson et al., 1997). According to social disorganization theory, peers are the central mechanism through which community socialization influences youth, and these influences are thought to be primarily adverse because they are enhanced when community institutions and norms fail to regulate peer group behavior (Sampson, 1992; Sampson & Groves, 1989; Shaw & McKay, 1942). Developmentally, peer influences are hypothesized to be most prominent during adolescence, but peer influences begin earlier. Young children had the largest number of peer interactions as well as exposure to aggressive peers in their neighborhoods, compared with other settings, such as child care, family, and organized play groups (Sinclair, Pettit, Harrist, Dodge & Bates, 1994). Aggressive play may be more tolerable when it is unstructured and unsupervised. Among early adolescents, the close friends of antisocial boys were likely to live in the same neighborhood and to have more unstructured and unsupervised activities (Dishion, Andrews & Crosby, 1995). Quasi-experimental data also revealed that among youth who moved from low-income to middle-income neighborhoods, those with high levels of problem behavior maintained greater contact with peers in the old, low-income neighborhoods (Briggs, 1997a). Do peers mediate or moderate neighborhood effects on adolescent development? Two studies report that in socially disadvantaged neighborhoods, there are fewer formal and informal institutions to monitor peer group activities, and this lack of organized activities is positively associated with delinquency, problem behavior, and negative peer group affiliation and negatively associated with prosocial competence (Elliott et al., 1996; Sampson & Groves, 1989). Peer deviance has also been found to mediate the negative effect of neighborhood disadvantage on adolescent emotional well-being (Simons et al., 1996). Peer support, on the other hand, has been shown to moderate neighborhood influences on adolescent antisocial behavior, substance use, and school achievement (Dubow, Edwards & Ippolito, 1997; Gonzales et al., 1996). Specifically, in high-risk neighborhoods, peer support appears to have more adverse effects, and in low-risk neighborhoods, peer support appears to have more beneficial effects. The number of community institutions is also thought to be associated with the presence of risk to residents (exposure to danger, violence, crime, and illegal or harmful substance); when institutions are lacking, the threat to residents, especially children and youth, is greater. Across two experimental housing programs in which families relocated from public housing to less poor

Changing Neighborhoods and Child Well-Being


neighborhoods, parents reported safety for their children as their primary reason from wanting to move (Briggs, 1997a; HUD, 1996). In a follow-up study, children who moved to low-poverty neighborhoods through one of the programs were less likely to be the victim of a personal crime than their peers who did not move (Katz et al., in press). Neighborhoods that parents perceive as dangerous are viewed as lower quality for child rearing than less dangerous neighborhoods (Earls et al., 1994). One study examined the indirect effects of perceived danger on adolescent mental health and found that youth in low-SES neighborhood perceived greater danger than their peers in high-SES neighborhoods, and perceived neighborhood danger was negatively associated with their emotional well-being (Aneshensel & Sucoff, 1996). Across a number of studies examining aggregate- or neighborhood-level outcomes, findings indicate that neighborhoods that are comprised of many poor residents, are not residentially stable, and have high concentrations of racial and ethnic minorities are likely to expose residents to greater physical risks including relatively high rates of crime, juvenile delinquency, low weight births, infant mortality, morbidity, and child physical abuse (Chow & Coulton, 1992; Collins & David, 1990; Coulton et al., 1995; Daly & Wilson, 1997; Drake & Pandey, 1996; see also O’Campo, Xue, Wang & Caughy, 1997). Access to illegal and harmful substances varies across neighborhoods as a function of the presence of community institutions to supervise the behavior of residents. In predominantly middle-SES neighborhoods, African-American young adolescents were more likely to be sold cigarettes than EuropeanAmerican adolescents, and these youth were more likely to be sold cigarettes in neighborhoods with a high proportion of African-American residents (Landrine, Klonoff & Alcaraz, 1997). In the Yonkers Project where youth moved from low- to middle-income neighborhoods, youth who stayed in lowincome neighborhoods had greater access to alcohol, assessed via number of liquor stores in community, than youth who moved to middle-income neighborhoods (Briggs, 1997b). High levels of drug activity in the neighborhood have also been linked to increased school rates of cigarette use (Ennett et al., 1997).

WHERE WE ARE GOING: INCORPORATING MODELS OF CHANGE INTO NEIGHBORHOOD RESEARCH Our goal here is to integrate an understanding of the dynamics of neighborhood conditions into our developmental framework for understanding how neighborhood effects may be transmitted to children and youth by recognizing the fact that families move across neighborhoods and that neighborhoods change.



Undoubtedly, these situations are likely to influence development if they are associated with parental relationships, institutional resources, and collective efficacy, as reviewed earlier. As an exemplar, we focus on neighborhood income (poverty and affluence) in this section. Neighborhood income was selected because this dimension has been most frequently studied and other structural characteristics, such as racial/ethnic diversity and residential instability, tend to be highly correlated with income/SES measures (Duncan & Aber, 1997). Separate indicators of affluence and poverty are used because the presence of poor and affluent neighbors may have differential effects on child and adolescent well-being (Brooks-Gunn, Duncan, Klebanov & Sealand, 1993; Jencks & Mayer, 1990). Unlike the research on family income, studies on neighborhood income have not considered the dynamic role of neighborhood poverty and affluence (timing, duration, depth) on child well-being. Based on research using data from a single time point, neighborhood affluence (compared to neighborhood middle-income) has demonstrated consistent associations with children’s and adolescents’ achievement-related outcomes, and neighborhood poverty (compared to neighborhood middle-income) has demonstrated somewhat consistent associations with children’s and adolescents’ behavioral and socioemotional outcomes (see Leventhal & Brooks-Gunn, 2000b, for a review). However, it is possible that this pattern of findings may change by studying the effects of neighborhood income over time (Brooks-Gunn et al., 1997). Incorporating a developmental perspective will also point to potential processes through which neighborhood effects may be operating on child well-being. One developmentally-oriented approach to examining the effects of neighborhood income on child well-being is to parallel the framework for studying the dynamics of family income by looking at the duration and timing of neighborhood poverty and affluence (Duncan & Brooks-Gunn, 1997; Duncan, Brooks-Gunn & Klebanov, 1994). Does the number of years a child lives in a poor (or affluent) neighborhood matter for development? Neighborhood poverty (rather than affluence) may have a cumulative negative effect on child achievement that does not exhibit itself by examining neighborhood income from a single time point, or neighborhood affluence may have a cumulative positive effect over time on behavioral outcomes (see Hughes, Furstenberg & McDonald, 1998, for a rare example of work examining neighborhood conditions at multiple time points). Does neighborhood affluence matter more for achievement during young childhood and neighborhood poverty matter more during late adolescence? We do not know. In general, two separate bodies of work have examined residential mobility. The first has focused on patterns of mobility with some exploration of the

Changing Neighborhoods and Child Well-Being


causes of mobility (Long, 1988; South & Crowder, 1998), and the second has considered the association between residential mobility, including the frequency of moves, and child development (Duncan et al., 1998; Haveman et al., 1991; Wood et al., 1993). Our framework attempts to bridge these bodies of literature while also incorporating research on neighborhood effects. Specifically, our strategy for studying neighborhood change entails longitudinal, developmental, and contextual components. In the remainder of this section, we illustrate examples of neighborhood research that incorporate this approach including an examination of the duration and timing of neighborhood income effects on child development and an exploration of patterns of residential mobility. These examples draw upon data from the Infant Health and Development Program (IHDP; Gross, Spiker & Haynes, 1997; Infant Health and Development Program, 1990). The IHDP was an eight-site randomized clinical trial designed to evaluate the efficacy of comprehensive early intervention services during the first three years of life in reducing the developmental and health problems of low birth weight (LBW; ≤ 2500 g), premature ( ≤ 37 weeks) infants (N = 985). The sample was stratified at each site on a number of key variables. One-third of the infants were randomly assigned to receive a variety of services intended to enhance child competence – high quality child care, family support services, and pediatric follow-up care. The control group received only pediatric follow-up care (Gross, Spiker & Haynes, 1997; IHDP, 1990). The children have been followed annually from birth through age 8. Children’s cognition was assessed by means of full-scale IQ tests when they were 5 and 8 years of age and their reading and math achievement was assessed via the Woodcock-Johnson Tests of Achievement-Revised when they were 8 years-old (N = 813 European-American and African-American children). Neighborhood income measures were constructed by matching family addresses to census neighborhood identifiers at the censustract level. Addresses were matched from the time of the children’s birth (1980 Census data) and when the children were 5 and 8 years of age (1990 Census data). Indicators of both neighborhood poverty and affluence were constructed based on previous work with this sample (Brooks-Gunn et al., 1993; Duncan et al., 1994; Brooks-Gunn et al., 1998; see Table 1). The measures used for neighborhood poverty and affluence are somewhat different across the 1980 and 1990 Censuses, but are more equivalent when adjusted for the Consumer Price Index (CPI). As indicated in Table 1, two indicators of neighborhood affluence are used at ages 5 and 8 – affluence and high affluence – to have some comparability in measures across the ages, while also recognizing that families’ circumstances tended to improve over time and that the real dollar amounts (inflation adjusted) across the Censuses are not necessarily equivalent.



Table 1. Description of Neighborhood Income Measures for Families in the Infant Health and Development Program. M (SD) Children at Birth Poverty (fraction families < $10,000) Affluence (fraction families > $30,000) Children at Age 5 Poverty (fraction families < $15,000) Affluence (fraction families between $35,000–$50,000) High Affluence (fraction families > $50,000) Children at Age 8 Poverty (fraction families < $15,000) Affluence (fraction families between $35,000–$50,000) High Affluence (fraction families > $50,000)

0.31 (0.18) 0.18 (0.14) 0.24 (0.18) 0.33 (0.01) 0.27 (0.20) 0.22 (0.18) 0.34 (0.01) 0.28 (0.20)

Note: At birth, neighborhood measures based on 1980 U.S. Census, and at ages 5 and 8, measures based on 1990 U.S. Census. N = 813.

When compared to two nationally representative samples of families with children (Panel Study of Income Dynamics and National Longitudinal Study of Youth-Child Supplement), families in the IHDP are somewhat more economically and socially disadvantaged, but a range of socioeconomic groups is represented (Brooks-Gunn et al., 1993; Chase-Lansdale et al., 1997; Smith, Brooks-Gunn & Klebanov, 1997). The descriptive statistics presented in Tables 1 and 2 provide insight into the dynamic nature of neighborhood income. Specifically, over time, families tended to improved their circumstances by moving to neighborhoods with fewer poor residents (fraction of families with incomes less than $10,000 at birth and fraction of families with incomes less than $15,000 at ages 5 and 8) and more affluent residents (fraction of families with incomes greater than $30,000 at birth and fraction of families with incomes between $35,000 and $50,000 at ages 5 and 8; see Table 1). When looking at the duration (or length) of residence in a poor or affluent neighborhood, the findings in Table 2 reveal that by age 8, most children had, on average, lived at least one year in a poor as well as an affluent neighborhood (these findings are largely attributable to racial differences in neighborhood residence; findings not presented). Now, we examine links among the dynamics of neighborhood conditions and children’s development.

Changing Neighborhoods and Child Well-Being


Table 2. Description of Duration of Neighborhood Poverty and Affluence Measures for Families in the Infant Health and Development Program. M (SD)/Percentages Duration of Neighborhood Poverty1 Birth – Age 5 # years live poor neighborhood Continuously live poor neighborhood Transiently live poor neighborhood Never live poor neighborhood Birth – Age 8 # years live poor neighborhood Continuously live poor neighborhood Transiently live poor neighborhood Never live poor neighborhood Duration of Neighborhood Affluence2 Birth – Age 5 # years live affluent neighborhood Continuously live affluent neighborhood Transiently live affluent neighborhood Never live affluent neighborhood Birth – Age 8 # years live affluent neighborhood Continuously live affluent neighborhood Transiently live affluent neighborhood Never live affluent neighborhood

0.66 (0.80) 20.8% 24.3% 54.9% 0.97 (1.14) 15.9% 33.1% 51.1% 0.58 (0.76) 16.3% 24.9% 58.8% 0.98 (1.15) 15.6% 32.9% 51.5%

Note: N = 813. 1 At birth, a poor neighborhood was defined as a neighborhood in which 40% or more of families had incomes less than $10,000, and at ages 5 and 8, a poor neighborhood was defined as a neighborhood in which 30% or more of families had incomes less than $15,000 (all based on 1980 U.S. Census). 2 At birth, an affluent neighborhood was defined a neighborhood in which 30% or more of families had incomes greater than $30,000, and at ages 5 and 8, an affluent neighborhood was defined as a neighborhood in which 50% or more of families had incomes greater than $35,000 (all based on 1990 U.S. Census).

STUDYING NEIGHBORHOOD EFFECTS ON CHILD WELL-BEING OVER TIME The data on neighborhood income in the IHDP (birth, age 5, and age 8) permit an investigation of the association between the duration of neighborhood poverty and affluence and children’s cognition and achievement as outlined above. To explore this issue, Ordinary Least Squares (OLS) Multiple Regression analyses were conducted to examine the association between the duration of neighborhood poverty and affluence across early and late childhood (birth to age 5 and birth to age 8) and children’s IQ and achievement scores. All



analyses controlled for family income, maternal age, race, verbal ability, education, and marital status, and child gender, birth weight, neonatal health status, intervention status, and study site. Separate analyses were conducted for neighborhood poverty and affluence. The first set of analyses examined whether the number of years a child lived in a poor or affluent neighborhood was associated with children’s IQ scores at ages 5 and 8 and children’s achievement at age 8 (i.e. linear effects; see Table 3). Living more years in an affluent neighborhood by age 5 was positively associated with children’s IQ scores. Since IQ and achievement tests are standardized with a mean of 100 and a standard deviation of about 15, the unstandardized coefficients represent IQ points. Accordingly, children who lived in affluent neighborhoods had test scores that were approximately 2.2 points higher for each year they resided in an affluent neighborhood. The next set of analyses examined whether living persistently or transiently compared to never living in an affluent neighborhood was associated with child outcomes (i.e. non-linear effects; see Table 3). The results indicate that living

Table 3. Duration of Neighborhood Affluence Associated with Children’s Cognitive and Achievement Outcomes in the Infant Health and Development Program.

Linear Effects # Years live in affluent neighborhood

Non-Linear Effects Continuously live affluent neighborhood

Transiently live affluent neighborhood

Never live affluent neighborhood

Age 5 IQ

Age 8 IQ

Age 8 Read Acht

Age 8 Math Acht

2.21* (1.07) [0.10]

0.89 (0.69) [0.06]

0.22 (0.87) [0.01]

–1.09 (0.87) [–0.07]

4.02 + (2.19) [0.09] 3.08* (1.53) [0.08] omitted

3.16 (2.16) [0.07] 1.29 (1.35) [0.04] omitted

–2.94 (2.72) [–0.04] 2.81 (1.71) [0.07] omitted

–4.92 + (2.73) [–0.08] –0.69 (2.72) [–0.02] omitted

Note: Unstandardized coefficients with standard errors in parentheses and standardized coefficients in brackets. All coefficients derived from OLS regression analyses that control for maternal age, race, verbal ability (missing data on mother’s verbal ability), education, marital status, child gender, birth weight, neonatal health, intervention status, and site. All models significant at p < 0.001. N = 813; + p < 10; * p < 0.05

Changing Neighborhoods and Child Well-Being


transiently (i.e. 1 year) in an affluent neighborhood by age 5 compared to never living an affluent neighborhood was associated with an increase in IQ scores of 3 points. Across analyses, there were no significant effects for the duration of neighborhood poverty. Thus, findings indicate that residing in an affluent neighborhood at birth or at age 5 benefited children’s cognition by age 5. The lack of findings at age 8, but significant effects found at age 5 suggests that the observed neighborhood effects could really be a proxy for a school effect, signifying entry into higher quality schools among children in more affluent neighborhoods. The beneficial effect of neighborhood affluence at birth, on the other hand, most likely operates through familial processes (see Klebanov et al., 1997). We also explored whether neighborhood income during young childhood (birth and age 5) or late childhood (age 8) was more highly associated with child well-being at age 8. OLS multiple regression analyses were conducted to examine the association between the timing of neighborhood poverty and affluence and children’s IQ and achievement scores (all analyses controlled for maternal, family, and child characteristics). A single model was run for each outcome that included measures of neighborhood income (poverty or affluence) at birth and ages 5 and 8 (see Table 4). Living in a poor neighborhood at age 5, as opposed to a non-poor neighborhood, was negatively associated with children’s IQ and math achievement scores at age 8. For neighborhood affluence, living in an affluent neighborhood at birth compared to a non-affluent neighborhood was positively associated with children’s IQ scores at age 8. Neighborhood affluence at age 5 was also positively associated with children’s IQ, reading achievement, and math achievement scores in late childhood; on average, children in affluent neighborhoods scored about 2.2 points higher than their peers in non-affluent neighborhoods. Together, these findings suggest that neighborhood conditions at age 5 are more strongly associated with late childhood outcomes than neighborhood conditions at birth and at age 8. Specifically, neighborhood affluence at age 5 (the time of school entry) may be most salient for cognitive and achievement outcomes. In addition, neighbors with incomes in the thirty thousand dollar range (absolute amount) appeared to hold particular significance for children’s cognitive and school achievement over time. Families with incomes in this range may represent some type of threshold with respect to beneficial neighborhood effects for children’s cognitive and achievement outcomes.2 Alternatively, this finding may reveal something about the quality of schools in such neighborhoods – that is, children living in more affluent neighborhoods may attend higher quality schools than their peers in less



Table 4. Timing of Neighborhood Poverty and Affluence Associated with Children’s Cognitive and Achievement Outcomes at Age 8 in the Infant Health and Development Program.

Timing of Neighborhood Poverty Birth Fraction of families < $10,000 Fraction of families > $10,000 Age 5 Fraction of families < $15,000 Fraction of families > $15,000 Age 8 Fraction of families < $15,000 Fraction of families > $15,000 Timing of Neighborhood Affluence Birth Fraction of families > $30,000 Fraction of families < $30,000 Age 5 Fraction of families $35,000–$50,000 Fraction of families > $50,000 Fraction of families < $35,000 Age 8 Fraction of families $35,000–$50,000 Fraction of families > $50,000 Fraction of families < $35,000


Reading Achievement

Math Achievement

–1.86 (4.14) [–0.03] omitted

–3.97 (5.26) [–0.04] omitted

0.15 (5.24) [0.00] omitted

–8.82* (4.46) [–0.09] omitted

–5.64 (5.67) [–0.05] omitted

–12.60* (5.66) [–0.12] omitted

4.01 (4.41) [0.04] omitted

0.11 (5.56) [0.00] omitted

5.93 (5.60) [0.06] omitted

10.28* (5.28) [0.09] omitted

3.07 (6.67) [0.02] omitted

5.06 (6.66) [0.04] omitted

16.55* (6.76) [0.09] 0.01 (5.11) [0.00] omitted

21.06* (8.61) [0.11] –6.94 (6.49) [–0.07] omitted

27.38*** (8.63) [0.15] –0.15 (6.50) [0.00] omitted

–4.11 (6.50) [–0.02] –1.74 (4.97) [–0.02] omitted

–5.49 (8.21) [–0.03] 4.32 (6.30) [0.05] omitted

–10.08 (8.24) [–0.06] –5.46 (6.32) [–0.06] omitted

Note: Unstandardized coefficients with standard errors in parentheses and standardized coefficients in brackets. All coefficients derived from OLS regression analyses that control for maternal age, race, verbal ability (missing data on mother’s verbal ability), education, marital status, child gender, birth weight, neonatal health, intervention status, site, and missing income data. All models significant at p < 0.001. N = 813; * p < 0.05; ** p < 0.01; and *** p < 0.001.

Changing Neighborhoods and Child Well-Being


affluent neighborhoods which accounts for their enhanced cognition and achievement. The magnitude of the neighborhood effects reported here are consistent with other studies using data from a single time point. Similar to the current findings, a previous review of studies examining the association between neighborhood structural dimensions (measured via Census data and controlling for a host of family-level characteristics) and child and adolescent outcomes revealed that neighborhood effects were small to moderate, accounting for about five% of the variance in child outcomes (Leventhal & Brooks-Gunn, 2000). In contrast, the effect size for family-level variables, such as income, are more substantial, ranging from moderate to larger, depending on the age at which income is assessed and the outcome under investigation (Duncan & Brooks-Gunn, 1997). Studying Neighborhood Mobility and Change Over Time We also explored patterns of residential mobility vis-à-vis neighborhood income on moves over two age periods (birth to age 5 and age 5 to age 8) among families in the IHDP (see Table 5). The mobility patterns displayed fall under three categories: an improvement in neighborhood economic conditions, a decline in neighborhood economic conditions, and no change in neighborhood economic conditions (due in part to families who do not move). A majority of families remained in similar neighborhoods over time with the biggest shift in families’ circumstances occurring when their children were young. Not surprisingly, most changes in neighborhood economic conditions centered around movements into and out of middle-income neighborhoods with most moves being to a neighborhood at the next income level, as opposed to larger changes in neighborhood income (e.g. poor to middle-income versus poor to affluent/high-affluent). Most families tended to improve their neighborhood surroundings by moving to relatively more advantaged neighborhoods. This finding parallels the data on family income which suggest that, on average, family incomes tend to rise (rather than fall) over time (Duncan, Yeung, Brooks-Gunn & Smith, 1998). Analyses of the data, as expected, revealed that the direction of neighborhood change was associated with family income (results available from authors upon request). Looking at the overall trajectories of family mobility for the eight years, a majority of families, 50%, remained in similar neighborhoods; 30% of families were on upward trajectories (i.e. moved upward at least one time and remained in similar neighborhoods otherwise); 11% of families were on downward trajectories (i.e. moved downward at least one time and remained in similar



Table 5. Mobility Patterns Among Families in the Infant Health and Development Program1.

Improvement in Neighborhood Conditions Poor to Middle-Income Poor to Affluent/High-Affluent Middle-Income to Affluent/High-Affluent Affluent to High-Affluent Decline in Neighborhood Conditions Middle-Income to Poor Affluent to Poor/Middle-Income High-Affluent to Poor/Middle-Income/Affluent No Change, Moved No Change, Did Not Move

Birth to Age 5

Age 5 to Age 8

27.1 12.1 3.5 8.1 3.4 10.1 5.9 4.3 — 32.2 30.6

13.6 5.1 1.7 5.1 1.5 10.4 5.2 2.7 2.5 8.2 67.7

Note: Table presents percentages. N = 813. 1 At birth, a poor neighborhood was defined as less than 10% of families with incomes above $30,000; a middle-income neighborhood was defined as 10%–30% of families with incomes above $30,000; and an affluent neighborhood was defined as greater than 30% of families with incomes above $30,000 (all based on 1980 U.S. Census data). At ages 5 and 8, a poor neighborhood was defined as less than 10% of families with incomes above $50,000; a middle-income neighborhood was defined as 10% – 30% of families with incomes above $50,000; an affluent neighborhood was defined as 30% – 50% of families with incomes above $30,000; and a high-affluent neighborhood was defined as greater than 50% of families with incomes above $50,000 (all based on 1990 U.S. Census data).

neighborhoods otherwise), and 10% of families were on oscillating trajectories (i.e. moved upward at one time and downward at the other). As evident in Table 5, the cells for particular types of moving patterns (rather than general nature of moves) are small for this sample, making links to child development problematic. However, these analyses are provided as an exemplar for studying mobility patterns and the variability in children’s neighborhood conditions over time.

A FRAMEWORK FOR STUDYING THE NEIGHBORHOODS IN WHICH CHILDREN LIVE In this section, we forecast how the next generation of neighborhood research will be conducted in the twenty-first century – that is, we consider the major research initiatives that are required to address the research gaps of this past decade. In doing so, we turn to the methodological corollaries of the theoretical

Changing Neighborhoods and Child Well-Being


issues raised throughout this article. We highlight appropriate research designs as well as approaches to measuring neighborhood contexts that tap process and change. Finally, we discuss the problem of endogeneity in the study of neighborhood change. Longitudinal Designs for Studying Neighborhood Effects on Development Over Time The greatest limitation to date is the absence of longitudinal data on child wellbeing and neighborhood contexts. Conducting analyses on the dynamics of neighborhood conditions (i.e. duration and timing of neighborhood poverty and affluence), residential mobility patterns, and neighborhood change requires multiple assessments of neighborhood conditions over time (and hopefully, multiple assessments of child outcomes over time). We suggest several longitudinal research designs that are most amenable to addressing these topics, including: (a) national or multisite large studies; (b) neighborhoodbased designs; and (c) experimental or quasi-experimental designs. The first approach involves the use of national or multisite studies to examine models of neighborhood change and child development. These studies, generally, have a large range of neighborhood (and family) types and permit estimations of neighborhood effects based on few children/families per neighborhood. Such a distribution is optimal because it minimizes multicollinearity among the neighborhood dimensions under investigation (Duncan, Connell & Klebanov, 1997; Duncan & Raudenbush, 1999). To date, a majority of the neighborhood research (which is cross-sectional) has employed this approach with data sets, such as the Panel Study of Income Dynamics (PSID; Hill, 1991); the National Longitudinal Study of Youth-Child Supplement (NLSY-CS; Baker & Mott, 1989); and the Infant Health and Development Program (IHDP; Gross & Spiker, 1997; IHDP, 1990). The advantage of this strategy is that the data sets are readily available, have followed families over extended periods of time (with plans for future data collection), and include some family process data that are typically available in smaller studies (see also Brooks-Gunn, Berlin, Leventhal & Fuligni, 2000, for a discussion of large studies available for public use in the coming years). As in the examples provided, these studies are ideal for exploring the dynamics of neighborhood conditions. Given the large sample sizes of national studies, these data sets also permit an examination of mobility patterns with links to child well-being. A major limitation of these studies, however, is the restriction of neighborhood measures to Census data.



The second strategy recommended is neighborhood-based designs that incorporate neighborhoods in the study design. Specifically, the sampling frame includes certain types of neighborhoods as well as a range of neighborhood types intended to be representative of the target population of neighborhoods under investigation. Sampling is also conducted to ensure a certain number of children per neighborhood (e.g. approximately 15 to 30 study participants per neighborhood; Duncan & Raudenbush, 1999) to permit multilevel analyses, such as Hierarchical Linear Modeling (HLM; Bryk & Raudenbush, 1992). These analytic techniques account for the nested nature of neighborhood-based designs (i.e. children are nested within neighborhoods), and yield estimates of the variability of outcomes both within and across neighborhoods, providing more reliable estimates of neighborhood effects on child well-being. A recent example of this approach is the Project on Human Development in Chicago Neighborhoods (PHDCN; Sampson et al., 1997). In this study, cluster analyses were performed on Chicago Census data to define neighborhood clusters (groups of 2 to 3 tracts that are internally homogenous with respect to SES and racial/ethnic makeup and geographically contiguous). Then, Census data were used to define two stratification variables – SES (3 levels) and racial/ethnic composition (7 levels) – which were cross-classified, and a stratified probability sample of 80 neighborhood clusters was selected for the longitudinal component of the study. Approximately 105 children per neighborhood cluster were interviewed for the longitudinal study (see Earls & Buka, 1997, for further details). In general, researchers utilizing neighborhoodbased designs have found that neighborhoods are internally quite heterogeneous, and more variability exists within neighborhoods than across neighborhoods (Cook et al., 1997; Elliott et al., 1996; Furstenberg et al., 1999). Further, in studies of pathways of neighborhood influences, the ability to examine processes within and across neighborhoods is critical to understanding how neighborhood effects operate on child well-being. For example, Cook and colleagues (1997) suggest that some family processes may vary more across families than neighborhoods and consequently, act as moderators (rather than mediators) of neighborhood influences. In terms of residential mobility patterns and neighborhood change, researchers employing this strategy are beginning to follow study participants as they move across neighborhoods, while also sampling new families as they move into target neighborhoods, to see how families select themselves into and out of neighborhoods (Sastry, GhoshDastidar, Adam & Pebley, 2000). The final strategy recommended for incorporating models of neighborhood change is experimental or quasi-experimental designs in which families are randomly assigned to reside in particular types of neighborhoods. Although this

Changing Neighborhoods and Child Well-Being


approach may appear infeasible, housing policies, such as housing mobility programs, offer the opportunity to explore how a change in neighborhood context affects children and youth. Housing mobility programs generally involve relocating residents from one neighborhood, such as families living in public housing in low-income neighborhoods, to housing in other neighborhoods, typically less poor neighborhoods. Programs cannot serve all eligible or interested families, so assignment is often random, and housing re-location is often quasi-random based on housing availability. A recent example of this strategy is the Moving to Opportunity for Fair Housing Demonstration (MTO) sponsored by the U.S. Department of Housing and Urban Development in five of the nation’s largest cities. The program randomly assigned housing project residents (low income, predominantly minority families with children) to: an experimental group who received housing vouchers and special assistance and counseling to move with the requirement that they move to low-poverty neighborhoods; a control group who received housing vouchers but no special assistance and no stipulation as to where they had to move; or a second control group who did not receive a voucher or special assistance and thus, remained in public housing. In addition to providing more reliable estimates of true neighborhood effects by minimizing selection as a problem, this approach is well-suited for investigations of the potential mechanisms through which neighborhood effects are transmitted. These studies also permit an exploration of the consequences of differential exposure to neighborhood contexts. Measuring Neighborhood Contexts Over Time To study neighborhood processes and change, researchers must move beyond Census data to alternative data sources because the Census is only conducted every ten years and is limited to sociodemographic information on neighborhood contexts (Leventhal & Brooks-Gunn, 2000). These other data sources that twenty-first century neighborhood researchers will need to consider include: (a) alternative administrative data, (b) systematic social observations, (c) community surveys, (d) neighborhood expert surveys, and (e) qualitative data. These methods are typically employed in conjunction with Census data. Measurement issues are fundamental to the study of neighborhoods because how neighborhoods are measured has implications for assessing the magnitude of neighborhood effects on child development as well as evaluating neighborhood change. Alternative administrative data sources, like the Census, are based on bureaucratically defined units and are available from city, state, and federal agencies. Examples of such data are vital statistics from health departments,



crime reports from police departments, school records from education departments, and child abuse and neglect records from human and social service departments. Like the Census, many of these data sources are readily available. Administrative data are typically collected annually, in contrast to the Census, making them well-suited for modeling change (Coulton et al., 1995). In addition, administrative data are useful for testing theoretical models. For example, data on crime could be used to examine the collective efficacy model (e.g. Sampson et al., 1997); data on abuse and neglect could be used to explore the relationships model (Coulton, Korbin & Su, 1999; Korbin & Coulton, 1997); and data on schools could be used to investigate institutional resource models (Jencks & Mayer, 1990). Another promising strategy for tapping neighborhood processes is systematic social observations or windshield surveys which entail having trained observers use a structured format to characterize neighborhoods (BarnesMcGuire & Reiss, 1993; Kohen, Brooks-Gunn, Leventhal & Hertzman, 1999; Raudenbush & Sampson, 1999; Reiss, 1971; Spencer, McDermott, Burton & Kochman, 1997; Taylor, Gottfredson & Brower, 1984). There are several means of data collection – videotaping, rater checklists, and audio-taping. Evidence from early studies employing systematic social observations indicates that it may be particularly useful for assessing social organizational features of neighborhoods discussed under collective efficacy models (Perkins & Taylor, 1996; Raudenbush & Sampson, 1999; Spencer et al., 1997). Of the various data collection means, audio-taping may be the most economical and unbiased method of gathering process data. The third method for assessing neighborhood processes and change is community surveys in which non-study participants are interviewed about their neighborhoods, yielding measures of neighborhood contexts that are independent from measures obtained by study participants (Sampson, 1997; Sampson et al., 1997). Community surveys typically employ neighborhood-based designs (with an independent sample; O’Brien, 1990; see also Sampson, 1997). Again, this strategy is just coming on line, but initial findings suggest that community surveys also reliably assess neighborhood social organization (Perkins & Taylor, 1996; Sampson, 1997; Sampson et al., 1997). Community surveys, however, are expensive to conduct. The next approach for obtaining processes-oriented data is neighborhood expert surveys which require interviewing key community leaders, such as prominent religious, political, business, and social leaders in the community, about their neighborhoods (Crain, personal communication; Earls & Buka, 1997). This technique may be useful for ascertaining information on the

Changing Neighborhoods and Child Well-Being


political and social organization of neighborhoods and how they change over time. The final method for gathering data on neighborhood processes and change is through ethnographic interviews (Burton & Jarrett, 2000). Recently, several large-scale studies with neighborhood-based designs have incorporated qualitative substudies to supplement the quantitative data collected in other components of the study (Brooks-Gunn et al., 2000). Qualitative data permit a rich exploration of mechanisms of neighborhood influence, particularly family relationships and collective efficacy. They also provide a more nuanced examination of factors associated with neighborhood trajectories, particularly neighborhood resilience, than quantitative approaches (Korbin & Coulton, 1997). Endogeneity in the Study of Neighborhood Change Our discussion of neighborhood change entails two distinct processes, one at the family-level and the other at the neighborhood-level. Specifically, families move, selecting themselves into and out of neighborhoods, and neighborhoods themselves may change (as a function in part of mobility patterns). Both processes may create biases of endogeneity in models that examine neighborhood change and child development because some omitted variable at either the family- or neighborhood-level, respectively, may account for any observed effects. It is likely that family mobility is a greater threat to internal validity because it has the potential to bias processes at both levels. For instance, a family may respond to negative change in a neighborhood (neighborhood change) by moving out of the neighborhood (negative selection). Addressing these issues is difficult and can best be handled through study design. As discussed under neighborhood designs, one approach is to follow families over time as they move across neighborhoods. Likewise, a strategy for tracking neighborhood change is to examine neighborhoods over time and chart families who move into and out of them, as well as to monitor accompanying changes in social organization. A new study of Los Angeles neighborhoods, L.A. FANS, has incorporated both of these components. PHDCN is also following participants over time in addition to conducting a repeated, cross-sectional community survey of Chicago neighborhoods (see Brooks-Gunn et al., 2000, for overviews of these studies). In addition, analytic techniques can be used to address concerns about omitted variable bias including instrumental variables, sibling analyses, and behavioral genetics models (Leventhal & Brooks-Gunn, 2000). An instrumental variable approach entails using an unbiased variable (instrument) to estimate neighborhood



measures to reduce the correlation between neighborhood and outcome measures that may be due to unobserved characteristics and then substituting the estimated neighborhood measure in subsequent analyses for the unadjusted neighborhood measures (e.g. Foster & McLanahan, 1996); sibling analyses take advantage of variability in neighborhood conditions over time by comparing the effects of neighborhood characteristics on siblings or first cousins, in essence, holding unmeasured family characteristics constant (e.g. Aaronson, 1997); and behavioral genetic models compare individuals of varying genetic relatedness to estimate the contribution of genetic and environmental factors and further partial environmental influences into familywide (e.g. neighborhood) and child-specific components (e.g. Caspi, Taylor, Moffitt & Plomin, 2000). Most of these analytic approaches to addressing endogeneity, however, have not incorporated neighborhood change into their applications; sibling models may be most amenable to studying change because they capitalize on discontinuities over time (across individuals within a family).

CONCLUSION What did we learn about the role of neighborhoods in children’s lives during the last quarter of this century and where will neighborhood research go in the next century? During the past decade, links between child and adolescent wellbeing and neighborhood contexts suggest continuing this line of research (Leventhal & Brooks-Gunn, 2000). However, a more developmentally-oriented framework for studying the neighborhoods in which children live is required. First, to understand neighborhood effects on child well-being, it is necessary to study the processes through which neighborhoods influence children and youth. To this end, we provided three models (institutional resources, relationships, and norms/collective efficacy) to facilitate theoretical and empirical explorations of these processes. Second, the dynamic nature of neighborhood conditions are important to examine. Exemplars were provided for neighborhood income, modeled after the research on the dynamics of family income. Third, how neighborhood change brought about by either residential mobility or internal neighborhood processes affects child and family well-being is critical. Patterns of mobility were explored as an illustration of such work. Finally, we laid out the methodology necessary to undertake research on neighborhood change. We hope this article will stimulate more developmental research on the neighborhoods in which children live. At the very least, we seek to raise awareness of these issues as the field of neighborhood research continues to

Changing Neighborhoods and Child Well-Being


grow and become a prominent research and policy agenda of the twenty-first century. Incorporating a more developmental frame when studying neighborhoods has implications for not only how we as researchers think about neighborhood contexts, but also the policies and programs which shape child and family well-being.

NOTES 1. Addresses when the children were age 3 were also matched to 1980 Census data; however, since there were substantial missing data at age 3 (n = 202; 25%) compared to the other ages, these data were not examined. Families with missing data at age 3 consisted primarily of families who had moved away from the IHDP catchment sites (Klebanov, personal communication). It also should be noted that previous studies examining neighborhood effects in the IHDP at ages 3 and 5 employed measures of neighborhood income assessed at a single time point (birth). 2. In terms of 1979 U.S. dollars, families with incomes in the $30,000 range had an income-to-needs ratio (income adjusted for poverty threshold based on household size) of five or higher (based on family size of 3 to 4), and in 1989 U.S. dollars, families with income in the $35,000 to $50,000 range had an income-to-needs ratio of about four (based on family size of 3 to 4). Thus, while the absolute dollar amounts have changed, the relative standing of families with incomes in the $30,000 range remained fairly stable.

ACKNOWLEDGMENTS The authors would like to thank the U.S. Department of Housing and Urban Development (HUD), the Russell Sage Foundation, and the National Institute of Child Health and Human Development for their support. We are also grateful to the National Science Foundation, the NICHD Research Network on Child and Family Well-Being, and the Columbia University Public Policy Consortium. Additional support was provided by the Social Science Research Council’s Working Group on Communities and Neighborhoods, Family Processes, and Individual Development. In addition, we thank Pamela Klebanov for her advice on data analyses. We would also like to thank Sandy Hofferth for comments on an earlier draft of this manuscript.

REFERENCES Aaronson, D. (1997). Sibling Estimates of Neighborhood Effects. In: J. Brooks-Gunn, G. J. Duncan, & J. L. Aber (Eds), Neighborhood Poverty: Policy Implications in Studying Neighborhoods (Vol. 2, pp. 80–93). New York: Russell Sage Foundation Press. Aber, J. L., Gephart, M., Brooks-Gunn, J., Connell, J., & Spencer, M. B. (1997). Neighborhood, family, and individual processes as they influence child and adolescent outcomes. In: J.



Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Context and Consequences for Children (Vol. 1, pp. 44–61). New York: Russell Sage Foundation Press. Anderson, E. (1991). Neighborhood effects on teenage pregnancy. In: C. Jencks & P. E. Peterson (Eds), The Urban Underclass (pp. 375–398). Washington, D.C.: The Brookings Institution. Aneshensel, C. S., & Sucoff, C. A. (1996). The neighborhood context and adolescent mental health. Journal of Health and Social Behavior, 37, 293–310. Bachman, J. G., & Schulenberg, J. (1993). How part-time work intensity relates to drug use, problem behavior, time use, and satisfaction among high school seniors: Are these consequences or merely correlates? Developmental Psychology, 29(2), 220–235. Baker, P. C., & Mott, F. L. (1989). NLSY child handbook 1989: A guide and resource document for the National Longitudinal Survey of Youth 1986 Child Data. Columbus, OH: Center for Human Resources Research, Ohio State University. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Barnes-McGuire, J., & Reiss, A. (1993). Systematic social observation Manual: Project on human behavior in Chicago Neighborhoods (Report prepared for Foundation for Child Development). Cambridge, Harvard School of Public Health: Author. Benasich, A. A., Brooks-Gunn, J., & Clewell, B. C. (1992). How do mothers benefit from early intervention programs? Journal of Applied Developmental Psychology, 13, 311–362. Billy, O. J., Brewster, K. L., & Grady, W. R. (1994). Contextual effects on the sexual behavior of adolescent women. Journal of Marriage and the Family, 56, 387–404. Boyce, W. T., Jensen, E. W., James, S. A., & Peacock, J. L. (1983). The Family Routines Inventory: Theoretical origins. Social Sciences and Medicine, 17(4), 193–200. Bradley, R. H. (1995). Environment and parenting. In: M. Bornstein (Ed.), Handbook of Parenting (pp. 235–261). Hillsdale, NJ: Erlbaum. Briggs, X. S. (1997a). Moving up versus moving out: Neighborhood effects in housing mobility programs. Housing Policy Debate, 8(1), 195–234. Briggs, X. S. (Ed.) (1997b). Yonkers Revisited: The early impacts of scattered-site public housing on families and neighborhoods. Report submitted to the Ford Foundation. New York, Teachers College: Author. Bronfenbrenner, U. (1986). Ecology of the family as context for human development: Research perspectives. Developmental Psychology, 22(6), 723–742. Brooks-Gunn, J., Berlin, L. J., & Fuligni, A. S. (2000). Early childhood intervention programs: What about the family? In: J. P. Shonkoff & S. J. Meisels (Eds), Handbook on Early Childhood Intervention (2nd ed., pp. 549–588). New York: Cambridge University Press. Brooks-Gunn, J., Berlin, L. J., Leventhal, T., & Fuligni, A. S. (2000). Depending on the kindness of strangers: Current national data initiatives and developmental research. Child Development [Special Issue on “New Directions for Child Development in the TwentyFirst Century”], 71(1), 257–268. Brooks-Gunn, J., Duncan, G., J., & Aber, J. L. (Eds) (1997a). Neighborhood poverty: Context and consequences for children (Vol. 1). New York: Russell Sage Foundation Press. Brooks-Gunn, J., Duncan, G. J., Klebanov, P. K., & Sealand, N. (1993). Do neighborhoods influence child and adolescent development? American Journal of Sociology, 99(2), 353–395. Brooks-Gunn, J., Duncan, G. J., Leventhal, T., & Aber, J. L. (1997). Lessons learned and future directions for research on the neighborhoods in which children live. In: J. Brooks-Gunn, G.

Changing Neighborhoods and Child Well-Being


J. Duncan, & J. L. Aber (Eds), Neighborhood Poverty: Context and Consequences for Children (Vol. 1, pp. 279–297). New York: Russell Sage Foundation Press. Brooks-Gunn, J., McCormick, M. C., Klebanov, P. K., & McCarton, C. (1998). Young children’s health care use: Effects of family and neighborhood poverty. Journal of Pediatrics, 132, 971–975. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: Sage Publications. Burton, L. M. (1990). Teenage childbearing as an alternative life-course strategy in multigenerational black families. Human Nature, 1(2), 123–143. Burton, L. M., & Jarrett, R. L. (2000). In the mix, yet on the margins: The place of families in urban neighborhood and child development research. Journal of Marriage and the Family, 62(4), 1114–1135. Caldwell, B. M., & Bradley, R. H. (1984). Home Observation for Measurement of the Environment. Little Rock, AK: University of Arkansas Press. Campbell, F. A., & Ramey, C. T. (1994). Effects of early intervention on intellectual and academic achievement: A follow-up study of children from low-income families. Child Development, 65, 684–698. Caspi, A., Taylor, A., Moffitt, T. E., & Plomin, R. (2000). Neighborhood deprivation affects children’s mental health: Environmental risk identified in a genetic design. Psychological Science, 11(4), 338–342. Collins, J. W., & David, R. J. (1990). The differential effect of traditional risk factors on infant birthweight among blacks and whites in Chicago. American Journal of Public Health, 80(6), 679–681. Conger, R. D., Ge, X., Elder, G. H., Jr., Lorenz, F. O., & Simons, R. L. (1994). Economic stress, coercive family process, and developmental problems of adolescents. Child Development, 65(2), 541–561. Cook, T. D., Shagle, S. C., & Degirmencioglu, S. M. (1997). Capturing social process for testing mediational models of neighborhood effects. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Policy Implications in Studying Neighborhoods (Vol. 2, pp. 94–119). New York: Russell Sage Foundation Press. Coulton, C. J. (1997). Potential and problems in developing community-level indicators of children’s well-being. In: R. M. Hauser, B. V. Brown & W. R. Prosser (Eds), Indicators of Children’s Well-Being (pp. 372–391). New York: Russell Sage Foundation Press. Coulton, C. J., Korbin, J. E., & Su, M. (1991). Neighborhood and child maltreatment: A multilevel study. Child Abuse and Neglect, 23(11), 1019–1040. Coulton, C. J., Korbin, J. E., Su, M., & Chow, J. (1995). Community level factors and child maltreatment rates. Child Development, 66(5), 1262–1276. Coulton, C. J., & Pandey, S. (1992). Geographic concentration of poverty and risk to children in urban neighborhoods. American Behavioral Scientist. Special issue: The impact of poverty on children, 35(3), 238–257. Crane, J. (1991). The epidemic theory of ghettos and neighborhood effects on dropping out and teenage childbearing. American Journal of Sociology, 96(5), 1126–1159. Crockett, L. J., Shanahan, M. J., & Jackson-Newsom, J. (2000). Rural youth: Ecological and life course perspectives. In: R. Montemayer (Ed.), Advances in adolescent development: Vol. 9. Adolescent experiences: Cultural and economic diversity in adolescent development. Thousand Oaks, CA: Sage. Daly, M., & Wilson, M. I. (1997). Life expectancy, economic inequality, homicide, and reproductive timing in Chicago neighborhoods. British Medical Journal, 314, 1271–1274.



Dishion, T. J., Andrews, D. W., & Crosby, L. (1995). Antisocial boys and their friends in early adolescence: Relationship characteristics, quality, and interactional process. Child Development, 66, 139–151. Dornbusch, S. M., Ritter, L. P., & Steinberg, L. (1991). Community influences on the relation of family status to adolescent school performance: Differences between African Americans and non-Hispanic whites. American Journal of Education, 38(4), 543–567. Drake, B., & Pandey, S. (1996). Understanding the relationship between neighborhood poverty and specific types of child maltreatment. Child Abuse and Neglect, 20, 1003–1018. Dubow, E. F., Edwards, S., & Ippolito, M. F. (1997). Life stressors, neighborhood disadvantages, and resources: A focus on inner-city children’s adjustment. Journal of Clinical Child Psychology, 26(2), 130–144. Duncan, G. J., & Aber, J. L. (1997). Neighborhood models and measures. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Context and Consequences for Children (Vol. 1, pp. 62–78). New York: Russell Sage Foundation Press. Duncan, G. J., & Brooks-Gunn, J. (Eds) (1997). Consequences of growing up poor. New York: Russell Sage Foundation Press. Duncan, G. J., Brooks-Gunn, J., & Klebanov, P. K. (1994). Economic deprivation and earlychildhood development. Child Development, 65(2), 296–318. Duncan, G. J., Connell, J., & Klebanov, P. K. (1997). Conceptual and methodological issues in estimating causal effects of neighborhoods and family conditions on individual development. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Context and Consequences for Children (Vol. 1, pp. 219–250). New York: Russell Sage Foundation Press. Duncan, G. J., & Raudenbush, S. W. (1999). Assessing the effects of context in studies of children and youth development. Educational Psychologist, 34(1), 29–41. Duncan, G. J., Yeung, W. J., Brooks-Gunn, J., & Smith, J. R. (1998). How much does childhood poverty affect the life chances of children? American Sociological Review, 63, 406–423. Durkin, M. S., Davidson, L. L., Kuhn, L., O’Connor, P., & Barlow, B. (1994). Low-income neighborhoods and the risk of severe pediatric injury: A small-area analysis in Northern Manhattan. American Journal of Public Health, 84(4), 587–592. Earls F., & Buka S. L. (1997). Project on Human Development in Chicago Neighborhoods: Technical Report. Rockville, MD: National Institute of Justice. Earls, F., McGuire, J., & Shay, S. (1994). Evaluating a community intervention to reduce the risk of child abuse: Methodological strategies in conducting neighborhood surveys. Child Abuse and Neglect, 18(5), 473–485. Elder, G. H., Eccles, J. S., Ardelt, M., & Lord, S. (1995). Inner-city parents under economic pressure: Perspectives on the strategies of parenting. Journal of Marriage and the Family, 57, 771–784. Elliott, D., Wilson, W. J., Huizinga, D., Sampson, R., Elliott, A., & Rankin, B. (1996). The effects of neighborhood disadvantage on adolescent development. Journal of Research in Crime and Delinquency, 33(4), 389–426. Ennett, S. T., Flewelling, R. L., Lindrooth, R. C., & Norton, E. C. (1997). School and neighborhood characteristics associated with school rates of alcohol, cigarette, and marijuana use. Journal of Health and Social Behavior, 38, 55–71. Foster, E. M., & McLanahan, S. (1996). An illustration of the use of instrumental variables: Do neighborhood conditions affect a young person’s chance of finishing high school? Psychological Methods, 1(3), 249–260.

Changing Neighborhoods and Child Well-Being


Freeman, R. B. (1991). Employment and earnings of disadvantaged young men in a labor shortage economy. In: C. Jencks & P. E. Peterson (Eds), The Urban Underclass (pp. 103–121). Washington, D.C.: The Brookings Institution. Fuller, B., Coonerty, C., Kipnis, F., & Choong, Y. (1997). An unfair head start: California families face gaps in preschool and child care availability. Report published by Berkeley-Stanford PACE Center, Yale University, and the California Child Care Resource and Referral Network: Growing Up in Poverty Project. Furstenberg, F. F., Jr. (1993). How families manage risk and opportunity in dangerous neighborhoods. In: W. J. Wilson (Ed.), Sociology and the Public Agenda (pp. 231–238). Newbury Park, CA: Sage Publications. Furstenberg, F. F., Jr., Cook, T. D., Eccles, J., Elder, G., & Sameroff, A. (Eds) (1999). Managing to make it: Urban families and adolescent success. Chicago, IL: University of Chicago Press. Gleason, P. M., & Cain, G. G. (1997). Earnings of black and white youth and their relation to poverty. Madison, WI: University of Wisconsin, Institute for Research on Poverty (Discussion Paper No. 1138–97). Gonzales, N. A., Cauce, A., Friedman, R. J., & Mason, C. A. (1996). Family, peer, and neighborhood influences on academic achievement among African-American adolescents: One-year prospective effects. American Journal of Community Psychology, 24(3), 365–387. Greenberg, M., T., Lengua, L. J., Coie, J. D., & Pinderhughes, E. E. (1999). Predicting developmental outcomes at school entry using a multiple-risk model: Four American communities. Developmental Psychology, 33(35), 403–417. Gross, R. T., Spiker, D., & Haynes, C. W. (Eds) (1997). The Infant Health and Development Program for low birth weight premature infants. Stanford, CA: Stanford University Press. Haveman, R., Wolfe, B., & Spaulding, J. (1991). Childhood events and circumstances influencing high school completion. Demography, 28(1), 133–157. Hayes, C. D., Palmer, J. L., & Zaslow, M. E. (1990). Who cares for America’s children? Child care policy for the 1990s. Washington, D.C.: National Academy Press. Hill, M. S. (1991). The Panel Study of Income Dynamics: A user’s guide. Thousand Oaks, CA: Sage Publications. Hogan, D. P., & Kitagawa, E. M. (1985). The impact of social status, family structure, and neighborhood on the fertility of black adolescents. American Journal of Sociology, 90(4), 825–855. Hughes, M. E., Furstenberg, F. F., Jr., & McDonald, K. B. (1998). Neighborhoods, mobility, and the successful development of at-risk youth. Paper presented at the Meeting of the Population Association of America, Chicago, IL. Iceland, J. (1997). Urban labor markets and individual transitions out of poverty. Demography, 34(3), 429–441. The Infant Health and Development Program (1990). Enhancing the outcomes of low birthweight, premature infants: A multisite randomized trial. Journal of the American Medical Association, 263(22), 3035–3042. Jarrett, R. L. (1997). Bringing families back in: Neighborhoods’ effects on child development. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Policy Implications in Studying Neighborhoods (Vol. 2, pp. 48–64). New York: Russell Sage Foundation Press.



Jencks, C., & Mayer, S. (1990). The social consequences of growing up in a poor neighborhood. In: L. E. Lynn & M. F. H. McGeary (Eds), Inner-City Poverty in the United States (pp. 111– 186). Washington, D.C.: National Academy Press. Katz, L. F., Kling, J., & Liebman, J. (in press). Moving to Opportunity in Boston: Early results of a randomized mobility experiment. Quarterly Journal of Economics. Kaufman, J., & Rosenbaum, J. (1992). The education and employment of low-income black youth in white suburbs. Educational Evaluation and Policy Analysis, 14(3), 229–240. Kindlon, D. J., Wright, B. D., Raudenbush, S. W., & Earls, F. (1996). The measurement of children’s exposure to violence: A Rasch analysis. International Journal of Methods in Psychiatric Research, 6, 187–194. Kirschenman, J., & Neckerman, K. M. (1991). “We’d love to hire them, but . . .”: The meaning of race for employers. In: C. Jencks & P. E. Peterson (Eds), The Urban Underclass (pp. 203– 232). Washington, D.C.: The Brookings Institution. Klebanov, P. K., Brooks-Gunn, J., Chase-Lansdale, P. L., & Gordon, R. (1997). Are neighborhood effects on young children mediated by features of the home environment? In: J. BrooksGunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Context and Consequences for Children (Vol. 1, pp. 119–145). New York: Russell Sage Foundation Press. Klebanov, P. K., Brooks-Gunn, J., & Duncan, G. J. (1994). Does neighborhood and family poverty affect mothers’ parenting, mental health, and social support? Journal of Marriage and the Family, 56(2), 441–455. Klebanov, P. K., Brooks-Gunn, J., McCarton, C. M., & McCormick, M. C. (1998). The contribution of neighborhood and family income upon developmental test scores over the first three years of life. Child Development, 69(5), 1420–1436. Kohen, D. E., Brooks-Gunn, J., Leventhal, T., & Hertzman, C. (1999). Neighborhood income and social and physical disorder in Canada: Associations with young children’s competencies. Manuscript in preparation. Korbin, J., & Coulton, C. J. (1997). Understanding the neighborhood context for children and families: Combining epidemiological and ethnographic approaches. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty: Policy Implications in Studying Neighborhoods (Vol. 2, pp. 65–79). New York: Russell Sage Foundation Press. Lamborn, S. D., Dornbusch, S. M., & Steinberg, L. (1996). Ethnicity and community context as moderators of the relations between family decision making and adolescent adjustment. Child Development, 67, 283–301. Landrine, H., Klonoff, E. A., & Alcaraz, R. (1997). Racial discrimination in minors’ access to tobacco. Journal of Black Psychology, 23(2), 135–147. Lee, V., Brooks-Gunn, J., Schnur, E., & Liaw, F. (1990). Are Head Start effects sustained? A longitudinal comparison of disadvantaged children attending Head Start, no preschool, and other preschool programs. Child Development, 61, 495–507. Leventhal, T., & Brooks-Gunn, J. (in press). Moving to Opportunity: What about the kids? In: J. Goering (Ed.), Choosing a Better Life? How public housing residents selected a HUD experiment to improve their lives and those of their children. New York: Russell Sage Foundation. Leventhal, T., & Brooks-Gunn, J. (2000). The neighborhoods they live in: Effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin, 126(2), 309–337. Leventhal, T., Brooks-Gunn, J., & Klebanov, P. K. (2000). Type, number, and timing of household events of household events over the first eight years of life: Associations with children’s

Changing Neighborhoods and Child Well-Being


cognition and achievement. New York: Center for Children and Families, Teachers College. Leventhal, T., Graber, J. A., & Brooks-Gunn, J. (in press). Adolescent transitions into young adulthood: Antecedents, correlates, and consequences of adolescent employment. Journal of Research on Adolescence. Logan, J. R., & Spitze, G. D. (1994). Family neighbors. American Journal of Sociology, 100(2), 453–476. Long, L. (1988). Migration and residential mobility in the United States. New York: Russell Sage Foundation. Ludwig, J., & Ladd, H. L. (1997). The effects of MTO on educational opportunities in Baltimore: Early evidence. Paper presented at the First Moving to Opportunity Conference sponsored by HUD, Washington, D.C. Martinez, J. E., & Richters, P. (1993). The NIMH Community Violence Project: II. Children’s distress symptoms associated with violence exposure. Psychiatry, 56, 22–35. McKey, R. H., Condelli, L., Granson, H., Barrett, B., McConkey, C., & Plantz, M. (1985). The impact of Head Start on children, families and communities (Final report of Head Start Evaluation, Synthesis and Utilization). Washington, DC: Administration for Children, Youth, and Families, United States Department of Health and Human Services. McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts and what helps. Cambridge, MA: Harvard University Press. McLoyd, V. C. (1990). The impact of economic hardship on Black families and development. Child Development, 61, 311–346. Mortimer, J. T., Finch, M. D., Ryu, S., Shanahan, M. J., & Call, K. T. (1996). The effects of work intensity on adolescent mental health, achievement, and behavioral adjustment: New evidence from a prospective study. Child Development, 67, 1243–1261. Newman, K. (1999). No shame in my game: The working poor in the inner city. New York: Knopf and the Russell Sage Foundation. O’Brien, R. (1990). Estimating the reliability of aggregate-level variables based on individuallevel characteristics. Sociological Methods and Research, 18, 473–504. Ogbu, J. U. (1991). Minority coping responses and school experience. Journal of Psychohistory, 18(4), 433–456. Osterman, P. (1991). Gains from growth? The impact of full employment on poverty in Boston. In: C. Jencks & P. E. Peterson (Eds), The Urban Underclass (pp. 122–134). Washington, D.C.: The Brookings Institution. Paulter, K. J., & Lewko, J. H. (1987) Children’s and adolescents’ views of the work world in times of economic uncertainty. In: J. H. Lewko (Ed.), How Children and Adolescence View the World of Work (pp. 21–31). San Francisco: Jossey-Bass. Perkins, D. D., & Taylor, R. B. (1996). Ecological assessments of community disorder: Their relationship to fear of crime and theoretical implications. American Journal of Community Psychology, 24(1), 63–107. Pettit, G. S., Bates, J. E., Dodge, K. A., & Meece, D. W. (1999). The impact of after-school peer contact on early adolescent externalizing problems is moderated by parental monitoring, perceived neighborhood safety, and prior adjustment. Child Development, 70(3), 768–778. Popkin, S. Rosenbaum, J., & Meaden, P. (1993). Labor market experiences of low-income black women in middle-class suburbs: Evidence from a survey of Gautreaux program participants. Journal of Policy Analysis and Management, 12, 556–573.



Raudenbush, S. W., & Sampson, R. J. (1999). “Econometrics”: Toward a science of assessing ecological settings, with application of HLM. Sociological Methodology, 29, 1–41. Reiss, A. (1971). Systematic observations of natural social phenomena. In: H. Costner (Ed.), Sociological Methodology (pp. 3–33). San Francisco: Jossey-Bass. Reynolds, A. J. (1994). Effects of a preschool plus follow-on intervention for children at risk. Developmental Psychology, 30(6), 787–804. Richters, P., & Martinez, J. E. (1993). The NIMH Community Violence Project: I. Children’s as victims of and witnesses to violence. Psychiatry, 56, 7–21. Rosenbaum, J. E., & Popkin, S. J. (1991). Employment and earnings of low-income blacks who move to middle-class suburbs. In: C. Jencks & P. E. Peterson (Eds), The Urban Underclass (pp. 342–356). Washington D.C.: The Brookings Institution. Rosenbaum, J. E., Popkin, S. J., Kaufman, J. E., & Rusin, J. (1991). Social integration of lowincome black adults in middle-class white suburbs. Social Problems, 38(4), 448–461. Sampson, R. J. (1992). Family management and child development: Insights from social disorganization theory. In: J. McCord (Ed.), Facts, Frameworks, and Forecasts: Advances in Criminological Theory (Vol. 3, pp. 63–93). New Brunswick, NJ: Transaction Books. Sampson, R. J. (1997). Collective regulation of adolescent misbehavior: Validation results from eighty Chicago neighborhoods. Journal of Adolescent Research, 12(2), 227–244. Sampson, R. J. (1999). What “community” supplies. In: R. Ferguson & W. T. Dickens (Eds), Urban Problems and Community Development (pp. 241–292). Washington, D.C.: Brookings Institution Press. Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing socialdisorganization theory. American Journal of Sociology, 94(4), 774–780. Sampson, R. J., Raudenbush, S. W., & Earls F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924. Sastry, N., Ghosh-Dastindar, B., Adams, J., & Pebley, A. (2000). The design of a multilevel, longitudinal survey of children, families, and communities: The Los Angeles Family and Neighborhood Survey. Santa Monica: RAND. Selner-O’Hagan, M. B., Kindlon, D. J., Buka, S. L., Raudenbush, S. W., & Earls, F. J. (1998). Assessing exposure to violence in urban youth. Journal of Child Psychology and Psychiatry and Allied Disciplines, 39(2), 215–224. Shaw, C., & McKay, H. (1942). Juvenile delinquency and urban areas. Chicago, IL: University of Chicago Press. Simons, R. I., Johnson, C., Beaman, J. J., Conger, R. D., & Whitbeck, L. B. (1996). Parents and peer group as mediators of the effect of community structure on adolescent behavior. American Journal of Community Psychology, 24(1), 145–171. Sinclair, J. J., Pettit, G. S., Harrist, A. W., Dodge, K. A., & Bates, J. E. (1994). Encounters with aggressive peers in early childhood: Frequency, age differences, and correlates of risk behaviour problems. International Journal of Behavioral Development, 17(4), 675–696. Smith, J. R., Brooks-Gunn, J., & Klebanov, P. K. (1997). The consequences of living in poverty for young children’s cognitive and verbal ability and early school achievement. In: G. J. Duncan & J. Brooks-Gunn (Eds), Consequences of Growing Up Poor (pp. 132–189). New York: Russell Sage Foundation Press. South, S., & Crowder, K. (1998). Leaving the ‘hood: Residential mobility between black, white, and integrated neighborhoods. American Sociological Review, 63, 17–26. Spencer, M. B., McDermott, P. A., Burton, L. M., & Kochman, T. J. (1997). An alternative approach to assessing neighborhood effects on early adolescent achievement and problem behavior. In: J. Brooks-Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty:

Changing Neighborhoods and Child Well-Being


Policy Implications in Studying Neighborhoods (Vol. 2, pp. 164–181). New York: Russell Sage Foundation Press. Stack, C. B. (1974). All our kin. New York: Harper & Row. Steinberg, L., Fegley, S., & Dornbusch, S. M. (1993). Negative impact of part-time work on adolescent adjustment: Evidence from a longitudinal study. Developmental Psychology, 29(2), 171–180. Sullivan, M. L. (1989). Getting paid: Youth crime and work in the inner city. Ithaca, NY: Cornell University Press. Sullivan, M. L. (1996). Developmental transitions in poor youth: Delinquency and crime. In: J. A. Graber, J. Brooks-Gunn & A. C. Petersen (Eds), Transitions Through Adolescence: Interpersonal Domains and Context (pp. 141–164). Hillsdale, NJ: Erlbaum. Taylor, R. B., Gottfredson, S., & Brower, S. (1994). Block crime and fear: Defensible space, local social ties, and territorial functioning. Journal of Research in Crime and Delinquency, 21(4), 303–331. Willis, P. (1977). Learning to labor: How working-class kids get working-class jobs. New York: Columbia University Press. Wilson, W. J. (1987). The truly disadvantaged: The innercity, the underclass, and public policy. Chicago, IL: University of Chicago Press. Wilson, W. J. (1991). Studying inner-city social dislocations: The challenge of public agenda research. American Sociological Review, 56(1), 1–14. Wilson, W. J. (1997). When work disappears. New York: Alfred J. Knopf. Yoshikawa, H. (1994). Prevention as cumulative protection: Effects of early family support and education on chronic delinquency and its risks. Psychological Bulletin, 115(1), 28–54. Wood, D., Halfon, N., Scarlata, D., Newacheck, P., & Nessim, S. (1993). Impact of family relocation on children’s growth, development, school function, and behavior. Journal of the American Medical Association, 270(11), 1334–1338.

This Page Intentionally Left Blank



Penelope L. Maza* INTRODUCTION The child welfare system in the U.S. reflects many of the conflicts inherent in trying to establish the acceptable relationships among federal, state, and local levels of government, and children and their parents. These relationships have been the focus of debate and advocacy since the nineteenth century when the Women’s Christian Temperance Union sought government intervention in domestic violence, including both violence against women and children by alcohol-abusing husbands and fathers. Throughout the twentieth century concerns about these relationships were raised by a mix of social movements and activists including the Progressives of the 1920s, the socialists of the 1930s, the social activists of the 1960s, and the Christian conservatives of the 1980s and 1990s. This paper will explore how the child welfare system in the U.S. has historically operated within the context of this debate with particular

* The interpretations, recommendations and conclusions in this article are those of the author and do not necessarily reflect or represent the view of the Children’s Bureau; Administration for Children and Families; Administration on Children; Youth and Families, or the Department of Health and Human Services.

Children at the Millennium: Where Have We Come From, Where Are We Going? Volume 6, pages 303–323. Copyright © 2001 by Elsevier Science Ltd. All rights of reproduction in any form reserved. ISBN: 0-7623-0776-5




emphasis on the role of the federal government. It will also address current issues in these relationships and identify the pressing issues being faced in the twenty-first century. For the purposes of this paper, the public child welfare system encompasses child protective services, foster care, aftercare services, services to prevent removal from the home and other services which promote permanency for children in safe and stable living environments. This paper will use the term “foster care” to refer to care provided on a 24-hour basis under the responsibility of the public child welfare agency. The form of this care can include, but is not limited to, a foster family home with a relative or nonrelative, group home, or institution. This paper could also have included a broader array of services such as those delivered to unmarried teenage parents, juvenile justice programs, family support services, day care, and education. Although these services are critical for children, the focus of this paper is the foster care system, its precedents and outcomes, because this is the part of the child welfare system that has consistently been the subject of considerable controversy.

CONTEXT Historically, children have been seen as parental property, and in earlier days, as paternal property. The conception of children as property complicates the ability of government to intervene in their protection. As attention to the issue of child maltreatment has increased, the distinction between acceptable parental discipline practices and maltreatment has become more defined, although it still varies across the country and among child protection agencies and workers. The developing definition of maltreatment has become a major force in the child welfare system since maltreatment is now the primary reason four out of five children and their families receive services from the system (U.S. Department of Health and Human Services [DHHS], 1997). It is clear, however, that the definition of maltreatment is intimately tied to the circumstances associated with poverty since, as discussed below, the historical roots of the child welfare system lay in keeping children safe and/or with their families in the face of poverty. Even in recent years, those in the lower income strata have a higher likelihood of maltreating their children than those in higher income categories. The Third National Incidence Study of Child Abuse and Neglect, which included both reported and unreported child maltreatment, found significant differences in the incidence of all types of maltreatment except emotional neglect and fatal injuries for children from families with

Confronting the Dilemmas of Child Welfare: Past, Present and Future


different incomes. For example, the incidence rate per 1,000 children was 47.0 for children from families making less than $15,000 per year compared to 2.1 for children from families making more than $30,000 per year (DHHS, 1996). Slightly more than one-fourth (28%) of the primary caretakers of children receiving services from child welfare agencies in 1994 were regularly working either full or part-time, and for almost half (45%) of the primary caretakers government programs were their main source of support (DHHS, 1997). There has been debate about the reasons why poor children are overrepresented in the child welfare system, particularly among those reported for maltreatment. One reason given for this overrepresentation is that poor families are more likely to be subjected to the scrutiny of government agencies, such as the welfare department and the courts, than families with higher incomes. The alternative explanation is that maltreatment is, in fact, more prevalent among the poor because poverty increases stress and that poverty itself is a danger to children (Pelton, 1978). The second explanation has far more support in the empirical literature (Berrick, 1998). African-American children have also been historically overrepresented in the child welfare system. In 1961, African-American children constituted only about one-fourth (24%) of the children receiving services from public child welfare agencies (Jeter, 1963). This increased to an estimated 29% in 1977 and to 41% in 1994 of those children who received services from the public child welfare agency. In 1994, African-American children were twice as likely as white children (56% vs. 28%) to be placed in foster care (DHHS, 1997). By FY 1998, African-American children constituted 44% of the children in foster care, 53% of the children waiting to be adopted, and 46% of the children adopted from the public child system (DHHS, 2000d). This over-representation of African-American children exists even though there is no difference between whites and African Americans in the incidence of maltreatment (DHHS, 1996). African-American children who enter the child welfare system differ from white children in a number of ways. AfricanAmerican children enter the system at a younger age; they are more likely to come from economically disadvantaged backgrounds, and, overall, have more identified problems than white children. However, even when these factors are held constant, African-American children are more likely to be placed in foster care than white children (DHHS, 1997). The role of the federal government in the child welfare system has evolved over time. The U.S. Children’s Bureau has provided leadership in child welfare since early in the twentieth century, although much of its early emphasis was on child labor (U.S. Department of Health, Education and Welfare [DHEW], 1962). As will be discussed below, since the 1930s, the U.S. Children’s Bureau



has played a variety of roles in the child welfare field. Other national organizations such as the Child Welfare League of America and the Children’s Defense Fund have also worked to change various aspects of the child welfare system. However, the primary responsibility for the delivery of services to this population has been at the state level and, in some states, at the county level. During the first half or more of the twentieth century, most of the funding for foster care and other services for children and their families came from state and local government and private sources (Jeter, 1963). Over time, the role of federal, state and county funds has increased. More recently, private agencies are obtaining most of their funding from the public sector (Malm & Maza, 1988). Although each state child welfare program is structured somewhat differently, two major structures are dominant. These are based on the extent of local/county vs. state funding and management control (Orr, 1999). The most frequently used structure is the state-administered, county-supervised program in which the state provides most of the funding and the employees of the county-level agencies are state employees. In contrast, a significant subset of states, including New York, Pennsylvania, Colorado, and California, have a state-administered, county-supervised system where counties provide most of the funding and the employees work for the counties. Of course, there are states with mixed structures, such as Ohio, in which counties vary as to their relationship to the state or where cities operate as if they are counties, such as Baltimore and New York City. Overlaying these relationships is federal funding from a variety of programs, most importantly Title IV-E of the Social Security Act. This provision provides over $6.5 billion in federal matching payments towards the costs associated with some foster care as well as other programs. Because of the complex leadership and oversight structure in the states and the over 3,000 counties and local jurisdictions, one might expect little similarity in how these systems operate. However, research has identified patterns of foster care service delivery which can been seen in the states, regardless of their policy context (Wulczyn, Brunner & Goerge, 1999). This suggests that, even with this complexity, these systems adjust to changes in similar ways, particularly to changes in federal requirements. The federal government, beginning in about 1935, provided some funds to support various kinds of family services and foster care. As discussed below, in 1961, the AFDC program was amended by establishing the Title IV-A foster care program which provided federal matching funds to states for the costs associated with the out-of-home placement of certain children. This increased the financial commitment of the federal government to child welfare,

Confronting the Dilemmas of Child Welfare: Past, Present and Future


particularly to foster care, and set the stage for a major increase in the financial role of the federal government in child welfare, which was statutorily prescribed in 1980. However, the role the federal government has played in child welfare as it relates to this financial commitment has varied considerably based on the prevailing political philosophy of the administration in power. At various times the federal government’s relationship to the states has varied from neglect to control through the use of financial sanctions if states were deemed not to have met program requirements. In recent years, the federal government has attempted to establish egalitarian partnerships with the states. The goals of the child welfare system are inherently conflicting. While expected to cast a wide net to identify maltreated children, it is also expected to decrease the rate of false-positive identifications of maltreatment. Furthermore, the system is simultaneously criticized for removing too many children from their homes while keeping too many children in their own homes under risky circumstances. Moreover, there are conflicting beliefs about the effects of removal on child development. It is simultaneously believed that both foster care and maintaining the child in an abusive home impede a child’s development. In addition, there are conflicting definitions of a “good” outcome. For example, there is great emphasis today on increasing the number of children adopted from the public foster care system but, in many ways, adoption may signify a major failure of the system to treat families’ problems and address families’ needs. Furthermore, while family reunification is generally seen as a positive outcome, a substantial proportion of reunifications fail and the child is returned to care (Wulczyn et al., 1999). An exit to emancipation is not seen as a good outcome because permanency was not achieved for those children, although a substantial proportion of those children return to their parents’ home on their own after leaving care (Westat Inc., 1991). Finally, some outcomes that are defined as permanent are less than permanent (e.g. reunification). Even adoption has a failure rate. The disruption (return of the child to foster care prior to finalization) rate may be as high as 10–13% (Stolley, 1993; Goerge, Howard, Yu & Radomsky, 1995). Returning an adopted child to foster care after finalization (adoption displacement) is not unusual since children may have to be in foster care to receive certain services.

THE PAST The current public child welfare system is an amalgamation of three different service populations that were identified in the nineteenth century. The first



category of service population was the poor. The primary service for the poor in the nineteenth century was the almshouse. Almshouses became defined as negative places for children to live (DHHS, 1995; Schene, 1998). By the late nineteenth century, orphanages, which began as a way to house orphans created by the Civil War, were seen as a viable alternative to the almshouse. As these facilities gained popularity, they began to offer some services, such as education and health care, which had not been offered in the almshouses. Many of the children living in orphanages were not true orphans; that is, they had at least one living parent. Many of these parents placed their children in these facilities because they felt they provided a better environment for them than living, usually, with a single parent (DHHS, 1995). By the 1900s, orphanages began to go out of style because of the belief that children should be with families and, by mid-century, foster family care became the preferred form of foster care. Institutional care was primarily used for older children or for children who needed special treatment. Other factors that led to orphanages going out of favor were cost and reports of abuse (Schene, 1998). With the enactment of Title IV-A of the Social Security Act in 1935, Aid to Families with Dependent Children (AFDC), came the hope that poverty alone would not be a reason for children to live apart from their parents. Until the enactment of the Temporary Assistance for Needy Families (TANF) program in 1996, AFDC served as the major program that kept children with their families when the primary family problem was financial. By the 1970s, foster family care was the dominant form of care. However, it was based on a number of assumptions, the foremost of which was that women would stay at home to be foster mothers. This feature broke down during the 1980s as more women worked outside of their homes and the nonworking foster parents aged out of the system (DHHS, 1993). In some states relatives were sought and became foster parents. Relative care rose from negligible in 1977 to 29% in 1999 (DHHS, 1997, 2000d). By the late 1990s, relative care was being encouraged by legislation (Public Law 105-89) while at the same time the children in care with relatives were experiencing longer stays in placement than children in other forms of care (Berrick, 1998; Wulczyn et al., 1999). The second group is the child maltreatment population. This group arrived late on the scene. In fact, the first child maltreatment report was investigated by an organization based on the model of the animal welfare movement. In 1874, the New York Society for the Prevention of Cruelty to Children was formed, which focused on rescuing children from maltreatment. For the first quarter of the twentieth century, child protection was dominated by the private agencies.

Confronting the Dilemmas of Child Welfare: Past, Present and Future


Approximately three hundred Societies for the Prevention of Cruelty to Children (SPCC) were established under the auspices of the American Humane Association. The approaches of these organizations varied from emphasizing rescuing children from neglectful and abusive environments to helping families provide caring environments for their children. By the 1930s and 1940s many of the functions performed by these agencies were taken over by a variety of public and voluntary organizations (Schene, 1998). As discussed below, there was increased emphasis on child maltreatment beginning in the 1960s, including the passage of mandatory reporting laws in the states. Consequently, the number of reports to public child welfare agencies of child abuse and neglect increased from 9,563 in 1967 to 2.8 million in 1998, although the rate of victimization appears to be declining (Waldfogel, 1998; DHHS, 2000a). This increase in reports has resulted in the commitment of a substantial proportion of public child welfare agency resources to the receipt, screening and investigation of reports of abuse and neglect. The final population group is the adoption population. This is the one that has probably gone through the most change. In the nineteenth century, adoption was rare and, in fact, consisted of mostly poor and neglected children from big East Coast cities being sent out west on the orphan trains by private child welfare agencies. Potential parents at each stop would view the available children at the railroad station and select the ones they wanted (Schene, 1998). There was virtually no review of the potential parents or monitoring of the adoption. In contrast, by the twentieth century, adoption became a service primarily for infertile couples who wanted infants. The criteria for adopting became more rigid and the process more closely monitored. Most of the babies were found among unmarried pregnant women who did not have an option of abortion and who had been disgraced for being pregnant out-of-wedlock. As the plight of children in the foster care system became more visible, the public child welfare agencies began to redefine the “adoptable” population. By the late 1960s through the 1970s, many children were spending long periods of time in foster care and service providers began to believe that at least some of these children could be adopted. The philosophical thrust was permanency planning which was demonstrated through a project in Oregon and was eventually disseminated throughout the country through funding from the Children’s Bureau (DHHS, 1997). The adoption focus for public agencies then became finding permanent homes for children who could not safely be reunited with their families. Most, but not all, private agencies continued to concentrate on the adoption of infants.



MAJOR LEGISLATIVE POLICY AND PRACTICE INITIATIVES Until 1961 the role of the federal government in child welfare services was limited to the modestly funded Title IV-B program of the Social Security Act, which could be used to pay for various services including foster care. The U.S. Children’s Bureau’s role was primarily to provide technical assistance to states and communities and to collect and disseminate statistics on various child service populations including children in foster care and children adopted through all means including private and independent adoptions. In 1961, the Congress amended the Title IV-A of the Social Security Act to allow federal funds to support payment for court-ordered foster care placements for children who would have been eligible to receive Aid to Families with Dependent Children (AFDC). The purpose of this legislation was to end the financial disincentive for states to remove children from unsafe homes and place them in foster care. It was believed that because the federal government was paying part of the cost for the support of AFDC eligible children when they were living at home, unless a similar proportion of their foster care costs were covered, the states would be reluctant to remove them from their homes to protect them from maltreatment. As part of implementing the Title IV-A foster care program, the then Department of Health, Education and Welfare (DHEW) issued the Flemming rule requiring that states either provide AFDC while working to improve conditions in the home or provide other living arrangements for these children. The Title IV-A foster care program differed in important respects from the Title IV-B program. First, it was an entitlement program. This meant that there was no limit on the amount of funds that could be spent. In contrast, Title IV-B was limited by appropriations. Second, the Title IV-A program was bureaucratically located with the income assistance program, AFDC, rather than in the U.S. Children’s Bureau in conjunction with the Title IV-B program where it could have been tied to broader policy directions for child welfare. To further complicate the situation, in 1969 DHEW mandated the separation of financial and social services within Title IV-A. This covered both the AFDC program and the foster care program. Thus, the opportunity to identify family problems early, which AFDC had provided was now lost to the paperwork of determining eligibility for AFDC (DHHS, 1997). The 1960s saw a substantial increase in the federal emphasis on child abuse and neglect. In 1962, C. Henry Kempe’s article identifying the battered child

Confronting the Dilemmas of Child Welfare: Past, Present and Future


syndrome (Kempe, Silverman, Steele, Droegemueller & Silver, 1962) provided the impetus for a variety of activities to address the issue. In 1963, the U.S. Children’s Bureau developed the first model statute for the reporting of child maltreatment. In 1974, the Congress passed the Child Abuse Prevention and Treatment Act (CAPTA) which expanded the spectrum of reportable behaviors and expanded the range of professionals who were required to report maltreatment (DHHS, 1997; Schene, 1998; Waldfogel, 1998). These two emphases – Title IV-A and CAPTA – were very successful in emphasizing the protection of children via removal and foster care placement. In fact, so many children had entered foster care and remained there that the number of children in care had increased from an estimated 177,000 in public foster care in 1961 (Jeter, 1963) to an estimated 520,000 in 1977 (DHHS, 1997). By the mid-1970s, concern was being expressed about the number of children who were spending their lives in foster care. This became known as “foster care drift.” A number of initiatives were undertaken which focused on permanency planning, the major emphasis for foster care for the balance of the century. A variety of research projects and demonstrations were undertaken which showed that children in foster care could be returned home or adopted and did not have to spend their childhoods in the custody of the state (Schene, 1998). The adoption movement was particularly strong. This movement argued that all children in care, even older, handicapped, and minority children, and children who were members of sibling groups which needed to be adopted together, were adoptable if they were unable to return home to their parents or a relative. In 1978, the Adoption Opportunities Program was added to the CAPTA legislation. This program provided for a national adoption exchange which would list children available for adoption and persons seeking to adopt, technical assistance to states, and demonstrations of methodologies to increase the adoption of children from the public foster care system. It should be noted that these two emphases, maltreatment at the front end of the system and adoption at the back end of the system, tended to ignore such issues as foster care quality and other aspects of the foster care experience. By 1980, there was concern about the rising number of children in foster care and the treatment they were receiving. This concern culminated in the passage of Public Law 96-272, the Adoption Assistance and Child Welfare Act of 1980. The proponents of this legislation believed that it would correct the problems in the system which had increased entries and decreased exits from the public foster care system. The programs this statute created and amended were housed in and administered by the U.S. Children’s Bureau and changed



the role of the federal government in child welfare from primarily a funder of foster care and a provider of technical assistance to a monitor of program protections for children in the system. For the first time federal legislation required states to make “reasonable efforts” to prevent children from entering foster care and to reunify children who had been placed. The legislation eliminated the Title IV-A foster care program and replaced it with the Title IV-E foster care and adoption assistance program. Not only did this program encompass the old Title IV-A foster care program, but also added uncapped entitlement funds for administrative costs and training. In addition, it provided adoption assistance, i.e. federal matching funds, for certain children who were being adopted from the foster care system so that their adoptive families could meet their special needs. There were a number of requirements that the states had to meet to receive these funds. The most well-known requirements, however, were added to the Title-IV-B program that would have an increased authorization through the addition of Section 427. To receive Section 427 funds, states had to conduct an inventory of all children who had been in care six months or more, establish a statewide information system which tracked children in foster care, establish a case review system which reviewed the cases of children in foster care every six months, and have a dispositional hearing in the courts after 18 months and periodically thereafter. In addition, states were to have established reunification programs and preplacement preventive services program to reunify children in care with families and prevent them entering foster care in the first place. The major goals of this legislation were to reduce the time children spent in foster care and the number of children in care. Both of these objectives appear to have been achieved in the early years after the passage of the legislation. Between 1977 and 1982, the median duration of care decreased from 31 to 21 months. By 1982 the number of children in foster care was reported to have dropped from 520,000 (DHHS, 1997) in 1977 to 274,000 (American Public Welfare Association [APWA], 1993) and to as low and 242,000 (Maximus Inc., 1983). Although there probably was a real decline in the number of children in foster care during this period, the extent of the decline may have been exaggerated. A different methodology was used to develop the 1977 and the two 1982 estimates. The 1977 estimate was developed from a weighted national random sample of case records of children receiving services from public child welfare agencies. These records were abstracted by research staff who classified the case as either an in-home services case or a foster care case (DHHS, 1997). In contrast, the first estimate for 1982 was developed using

Confronting the Dilemmas of Child Welfare: Past, Present and Future


reports by states of aggregate data to the American Public Human Services Association (formerly the American Public Welfare Association). Most of these reports were not contemporaneous with the end of 1982, nor were they necessarily covering the same 12-month period (APWA, 1993). The second estimate for 1982 was developed by collecting a limited number of variables from a national random sample of case records identified by the child welfare agency as being in foster care. Since local agency staff abstracted the case records, the determination of whether or not the child was in foster care was made by the caseworkers, not the researchers (Maximus Inc., 1983). These methodological differences should be assessed within the context of the new requirement of Section 427 of the Social Security Act for the establishment of statewide information systems which were to contain information on all the children in foster care. It is possible that during the development and implementation of these new systems, cases that should have been closed previously (e.g. child had returned home and was being monitored in that setting, or child had been discharged from care) were officially closed when the data were entered into the new systems. Whatever the role of each of these factors – real decline, methodological differences, or new statewide information systems – the decline in the number of children in foster care was short-lived. It stopped in 1983 (Gershenson, 1985) and since then it has continued to increase to an estimated 560,000 by the end of FY 1998 (DHHS, 2000d). The duration of care has remained relatively constant at about 18–24 months (Stolley, 1993; DHHS, 2000d). Although in some respects the legislation may have been effective, in many others it was not. For example, preplacement preventive services were not widely established and specific reunification programs still do not exit. Although case reviews and dispositional hearings occurred, in many places they were perfunctory since the courts received no additional funding to handle the massive increased workload. Finally, the federal government was equivocal in its monitoring and enforcement of the legislation. The states had different definitions of such activities as case reviews, dispositional hearings, and what constituted foster care, and the content and format of case plans. However, under the Reagan Administration of the early 1980s, the U.S. Children’s Bureau could not issue guidance and never published final regulations for the Section 427 program. At the beginning of the Reagan Administration, the philosophy was laissez faire with respect to the states because the administration believed that the federal government should have no role in paying for or monitoring social services; by the end of the second term, that philosophy had changed. The role of the U.S. Children’s Bureau became that of an enforcer.



The Bureau conducted compliance reviews of state programs with inconsistent criteria. If the states failed the reviews, the federal government would recover already awarded funds. The reviews were eventually suspended until a final regulation could be published. The legislation was eventually amended, substantially changing the nature of the Section 427 program. By the end of the 1980s, the number of children in foster care was increasing at alarming rates, primarily due to the increasing use of cocaine by women and the imposition of policies in large states that required the reporting of in utero drug exposure as prima facia evidence of child abuse (Maza, 1999a). Although research suggested that the problem was not so much excessive entry of children into the system, but was mostly caused by the inability of agencies to return children home (Wulczyn et al., 1999). In 1993, with Clinton Administration support, the Congress passed an amendment to Title IV-B of the Social Security Act, which established Subpart 2 of Title IV-B to support family preservation and support services. The legislation authorized $1 billion in new funding over a five-year period and was seen as an attempt to substitute for the underfunding of preventive services that occurred during the 1980s. Family support encompasses services designed primarily for families that did not have contact with the child welfare system. The statute defines family support services by the goals they are designed to achieve – “promote the wellbeing of children and families,” “increase the strengths and stability of families,” increase parents’ confidence and competence,” “afford children a stable and supportive family environment,” and “enhance child development.” Services can include respite care, early developmental screening, mentoring, and home visiting (U.S. Government Accounting Office, 1995). Providers and advocates of family support services also include the ways in which the programs work with families as being integral to their qualifying as family support. In general, they are designed to address issues faced by all families, not to address specific problems or replace professional services, and to be delivered in an egalitarian manner (Abt Inc., in press). These services take a variety of forms and, almost by definition, are not designed to prevent child abuse or other problems generally dealt with by the child welfare system. In addition, with the exception of a few home visiting programs, very few had been subject to a rigorous evaluation. However, they were supported by the community-based service providers and the political community as being “good” for families. In contrast, family preservation services are services primarily designed to prevent foster care placement of children. The most visible form of this service is Homebuilders, which was promoted by various foundations. This approach

Confronting the Dilemmas of Child Welfare: Past, Present and Future


uses short-term in-home intervention to prevent foster care. The service model had been tested primarily with adolescents and its approach was to stabilize the family situation to prevent placement. By the late 1980s and early 1990s, Homebuilders was being promoted as a means to prevent the placement of infants and young children from drug-affected households. By the late 1990s, family preservation’s prominence as a solution to the enormous problems faced by the child welfare system had faded. The evaluation findings about the program’s effectiveness to prevent foster care placement were either equivocal or negative (McCrosky & Meezan, 1998; Shuerman, Rzepnicki & Littell, 1994; Littell & Shuerman, 1995), and there were increasing concerns about the safety of the children in the programs (Murphy, 1994; Gelles, 1996). Consequently, the Family Preservation and Support Program was changed to Safe and Stable Families in the reauthorization by Public Law 105-89, the Adoption and Safe Families Act of 1997 (ASFA). This legislation also permitted expenditure of funds for a wider range of services. The message was clear: Safety concerns were to dominate. The primary focus of this legislation, however, was on permanency. The statute contains a number of provisions that emphasize adoption including a provision that requires that most children in the foster care system, after a period of time, be reviewed for appropriateness for termination of parental rights. It also established the first data-based outcome incentive program in child welfare, the Adoption Incentive Program, which rewards states for increasing their number of adoptions of children from the public foster care system (Maza, 2000). Finally, the focus on outcomes was also emphasized in the legislation by requiring the Department of Health and Human Services (DHHS) to develop outcome measures for state child welfare systems. The DHHS is required to publish an annual report on state performance in relation to these measures. The first report was published in FY 2000 (DHHS, 2000d). A second report is required which will propose ways to link funding to these outcomes. Prior to the publication of the report, DHHS issued regulations governing child and family services state plan reviews. A component of this results-oriented review uses performance measures to determine whether the state child welfare program is in conformity with the legislative requirements of Title IV-B of the Social Security Act. Failure to either meet the national standards for performance on these measures or, for those states which fail to meet the standards in their first review and demonstrate improvement in subsequent reviews, will result in the federal government imposing financial penalties on the state child welfare agency (DHHS, 2000e).



CURRENT ISSUES IN CHILD WELFARE The current child welfare system operates on a mix of belief and fact. At the front end of the system are the family support and family preservation emphases, both of which have had little support from research as effective tools to prevent placement or maltreatment. In addition, although there appears to be a leveling off of child maltreatment reports (DHHS, 2000a), a recent study showed that required reporters are not reporting a significant proportion of cases where the children are experiencing harm (DHHS, 1996). This gatekeeping for the system by child protective services is accomplished by screening out cases at the phone call stage, not substantiating cases at the investigation stage by making it excessively burdensome to do so, and by reclassifying some types of abuse so the cases fall outside the responsibility of the agency. These strategies have been used because, as many suspect, the system reached the limit of the number of cases it could handle with its resources sometime in the late 1980s (DHHS, 1997). As discussed below, findings from an ongoing research project may inform this safety vs. family preservation policy debate. The gaps in knowledge in child welfare are too numerous to enumerate here. While the system continues to be in crisis, in general, federal funds for research on the child welfare system have been reduced, particularly for foster care research. Although some local studies are being supported with child abuse and neglect research and demonstration funds, the only current national study focused directly on child welfare is the National Survey of Child and Adolescent Well-Being (NSCAW), which was requested by the Congress in 1996. This six-year longitudinal study of a national random sample of almost 7,000 children is annually collecting data directly from children, parents, and other caregivers. In addition to administrative data, data are also being collected from caseworkers, teachers, and state agency administrators. Because this study is following children after their first contact with the system, it includes children who never have another contact with the system, children who receive supervision from the child welfare agency but do not enter foster care, children who have short foster care experiences, children who have prolonged foster care experiences and children who may have experienced all of the above. A primary goal of this study is to attempt to answer the safety vs. in-home services question by looking at outcomes such as child development. (Although the final findings from this study will not be available for a number of years, they should have a major impact on child welfare policy and practice when they are available [Research Triangle Institute, 1999]).

Confronting the Dilemmas of Child Welfare: Past, Present and Future


Throughout the twentieth century, the foster care population has changed from being primarily a population protected from the ravages of poverty to one being protected from abusive and neglectful parents. Poor and AfricanAmerican children and families have consistently been over-represented. It has also changed from placing children in mostly large institutions, to utilizing foster family homes, and eventually to an extensive use of relative homes. It is also a system that kept children indefinitely but which now is supposed to move children faster to permanency, particularly adoption and reunification. Finally, for those children for whom permanency does not occur, those being emancipated are to be provided the training and supports they need to live on their own as a new emphasis on independent living is being made throughout the 1980s and in the late 1990s. Even with these new emphases, the number of children in foster care is now greater than what it was in the late 1970s and appears to be increasing. The structure of federal funding for child welfare has been the subject of considerable criticism and has been identified as one of the reasons for the increase in the number of children in foster care. The argument is that the substantial difference in the federal funding structure for foster care – an uncapped entitlement in contrast to general child welfare services less than $1 billion – may provide an incentive to agencies to place children in care and keep them there for a longer periods of time than necessary (Wulczyn, 2000). It is true that for a considerable number of years Title IV-B funding has been virtually flat, while Title IV-E funding had increased dramatically (Courtney, 1998). However, research has found that children eligible for Title IV-E matching funds are about as likely to receive in-home services as they are to be in foster care (DHHS, 1997). In order to alleviate this perceived incentive to place children in foster care, states have been experimenting with managed care models of service delivery. Under the Section 1130 of the Social Security Act, DHHS was given authority to waive certain requirements of Title IV-E of the Social Security Act. As of September 1999, there are 24 demonstration waiver projects in 21 states. Six states (Colorado, Connecticut, Maryland, Michigan, Texas, and Washington) are experimenting with managed care/capitated payment systems, which at the beginning of the year, give private agencies under contract with the states the amount of Title IV-E funds they are expected to spend during the year. These states and the contracted private agencies will have flexibility in how the funds are spent (foster care vs. other services) and can keep the savings realized. An additional six states (Florida, Indiana, New York, North Carolina, Ohio, and Oregon) are experimenting with a subset of counties. These demonstrations are required to have rigorous evaluations using experimental designs with,



preferably, random assignment (DHHS, 2000b). Findings will not be available until the early twenty-first century.

THE FUTURE The historically intimate relationship between the Aid to Families with the Dependent Children (AFDC) and the child welfare system was broken in 1996 when the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) was enacted and established the Temporary Assistance for Needy Families (TANF) program. This new program gave states more flexibility in a number of ways, but set a national time limit for the receipt of welfare payments for parents at five years. This time limit will be reached by most states in 2002. The impact of the loss of this “safety net” on the child welfare system has yet to be seen. It is possible that there will be some increase in child maltreatment reporting and/or the number of children in foster care (Kroll, 1996) given the historical and well-documented relationship between child maltreatment and poverty (Needell, Cuccaro-Alamin & Bookhart, 1999). Based on recent research, even a small failure rate in TANF could result in a substantial increase in the number of children in care. For example, a study conducted on data from Illinois, California, and North Carolina found that although less than 3% of the AFDC population in those states in the late 1990s entered foster care, they accounted for over 60% of the entries into foster care (Office of the Assistant Secretary for Planning and Evaluation, 2000). Another study conducted with data from 10 counties in California on AFDC child entrants in 1990 found that within five years, approximately 1 in 30 entered the foster care system (Needell et al., 1999). Welfare reform is also hypothesized to exacerbate the already rising costs of foster care, which remains an uncapped entitlement, while AFDC does not. States may convert relative foster homes previously supported by child-only AFDC to Title IV-E (Courtney, 1995). The Administration for Children and Families (ACF) has funded a number of evaluations of various aspects of welfare reform programs (Administration for Child and Families [ACF] 2000). Other studies that are focusing on these issues have, thus far, found no detrimental effects on child welfare that can be directly attributed to welfare reform (Sengupta, 2000). However, findings reported from these and other studies after implementation of the five-year limit are more likely to identify effects, if there are any. The legislation passed in 1993 that established the waiver demonstration program in child welfare was amended by ASFA in 1997 to permit the granting of additional waiver demonstrations. As discussed above, many of the waivers

Confronting the Dilemmas of Child Welfare: Past, Present and Future


are used to create flexibility in the expenditure of Title IV-E foster care funds, particularly to use them for services. In their applications, states must assure cost neutrality, i.e. the costs for the waiver program over its five years cannot exceed what would have been spent under the traditional program and states are required to conduct rigorous evaluations (DHHS, 2000b). Most of these projects are still in their infancy, but their findings will become public early in the twenty-first century. In addition, a number of other proposals for flexible funding are being discussed. In brief, one would maintain the Title IV-E entitlement and accountability, but would permit states to transfer “saved” Title IV-E funds to Title IV-B to be used for services (American Public Human Services Association, 1998). A second proposal would provide states with a fixed amount of funds to use as they see fit for this population. The final proposal would be to revise the waiver demonstration program by eliminating some of the restrictions such as time limits and the evaluation requirement (Johnson, 2000). Whether or not any of these specific proposals are adopted is not certain; however, it is likely that there will be continuing discussion of and experimentation with different funding structures. Related to waivers, particularly managed care waivers, and to other aspects of ASFA, is the focus on outcomes. The data on foster care and adoption are improving significantly with the development and implementation of the Adoption and Foster Care Analysis and Reporting System. Through this mandatory reporting system, states semi-annually electronically submit individual records on each child who experienced foster care during the reporting period and each child adopted. States are financially penalized if they fail to submit data or fail to submit data that meet pre-established quality standards (Maza, 1999b). In addition to providing reports on children in foster care and children being adopted, the data are being used for a variety of purposes, including the child welfare outcomes annual report, child and family services review outcomes, determining awards under the Adoption Incentive Program, and sampling for the Title IV-E monitoring reviews. This focus on outcomes and accountability could improve services or it could drive the system into a “bean counting” exercise where states focus their activities on passing the monitoring reviews, as anecdotal information has suggested was done in the 1980s. In addition, with child welfare performance measurement at the national level still in its infancy, there will likely be unintended consequences from the selected measures. Some of the measures being developed try to avoid unintended consequences of promoting certain kinds of outcomes. However, it is still likely that there will a myriad of unanticipated outcomes as states try to work on the outcomes specified by the federal



government. This will be especially true as these outcomes become more intimately linked to funding. There is increased interest in relative care as evidenced by the ASFArequired report to the Congress on kinship foster care (DHHS, 2000c). Relative care is another one of the dilemmas of child welfare. Although care provided by the extended family is seen as positive for children, the children in relative care stay in care longer than children in care with non-relative foster parents, and relative foster parents are less likely to receive training and supportive services than non-relative foster parents (DHHS, 1998, 2000c). Relative care accounts for much of the increase in foster care in large states since the late 1980s (Wulczyn et al., 1999). As presented earlier, relative foster care was negligible in 1977 (DHHS, 1997) but accounted for 29% of foster care placements in 1998 (DHHS, 2000d). Some states are attempting to reduce the unnecessary use of relative care in the foster care system. Specifically, Illinois is making distinctions among children with a protective need who are in preexisting relative placements, children who have been removed from the home of parents or relatives because of abuse or neglect, and children with no protective need. The last category will not be served by the child welfare system but will be referred to income assistance. In Illinois, children in the first category were formerly placed in formal foster care with the relative with whom they had been living. Under the new program, these children and their relative caretakers will be provided family support services and/or in-home services for a period of time unless other information is developed which indicates threats to the child’s safety (Illinois Department of Children and Family Services, 1995). In addition, under the waiver demonstration program, eight states (California, Delaware, Illinois, Maryland, Montana, New Mexico, North Carolina, and Oregon) are experimenting with assisted guardianship for relative foster parents who do not choose to adopt the child, but want a more permanent relationship with the child without the excessive oversight that formal foster care requires (DHHS, 2000b). As with adoption assistance, assisted guardianship provides a monthly maintenance payment to the guardian that is less than or equal to the foster care payment. Since these children are in safe permanent placements with relatives and will be there for the foreseeable future, this assistance removes the financial disincentive for permanency – losing the foster care payment. Finally, there will be the backlashes. Of particular concern is an adoption backlash. The number and rate of adoptions is increasing so rapidly (Maza, 2000) that practitioners fear that there will be increases in disruptions, displacements, and dissolutions (relinquishment of child back to foster care with severing of parental rights). If these become more prevalent, it is likely

Confronting the Dilemmas of Child Welfare: Past, Present and Future


that the push toward adoption may be tempered. In addition, because of the rapidity with which these adoptions are to be achieved, it is likely that birth parents will seek relief from the courts, particularly by claiming that they did not receive the needed services prior to the commencement of termination of parental rights proceedings. These likely scenarios may end up slowing the march toward adoption as a preferred outcome for children who cannot return to their families.

CONCLUSION This article has reviewed the history of the child welfare system and has discussed the many internal contradictory goals of the system. It has specifically examined the issues of child safety, permanency, and the roles of the state and federal government. It has concluded that many of the issues that evolved in the twentieth century will be revisited, albeit in different forms, in the twenty-first century. Whatever the outcome of these debates and trends, one can be sure that the foster care system, which serves particularly vulnerable children, will remain controversial and subject to criticism well into the next century.

REFERENCES Abt Associates Inc. (in press). National evaluation of family support programs. Abt Associates, Cambridge, MA. Administration for Children and Families (2000). Ongoing research and evaluation projects: welfare reform and social services research. rd&e.htm. American Public Human Services Association (1998). Linking child welfare outcomes with federal funding: background issues. Unpublished paper. American Public Welfare Association (1993). VCIS child welfare data 1982. Berrick, J. D. (1998). When children cannot remain home: foster family care and kinship care. The Future of Children, 8(1), 72–87. Courtney, M. E. (1995). The foster care crisis and welfare reform. Public Welfare, (Summer), 27–33. Courtney, M. E. (1998). The costs of child protection in the context of welfare reform. The Future of Children, 8(1), 88–103. Gelles, R. J. (1996). The book of David: how preserving families can cost children’s lives. New York: Basic Books. Gershenson, C. P. (1985). 1983 trend of children in foster care. (Child Welfare Research Notes No. 11 [April]). Washington, D.C.: Administration on Children, Youth and Families, Goerge, R., Howard, E., Yu, D., & Radomsky, S. (1995). Adoption disruption and displacement: the Illinois child welfare system, 1976–1994. Working paper. Chapin Hall Center for Children, University of Chicago, Chicago, IL. Illinois Department of Child and Family Services (1995). Illinois HMR reform plan.



Jeter, H. (1963). Children, problems and services in child welfare programs. Washington, D.C.: U.S. Department of Health and Human Services. Johnson, N. L. (2000). Promoting flexible funding in the child protection program. Unpublished paper. Kempe, C. H., Silverman, F. N., Steele, B. F., Droegemueller, W., & Silver, H. K. (1962). The battered-child syndrome. Journal of the American Medical Association, 181, 17–24. Kroll, J. (1996). Devolution and how it impacts on children. Adoptalk, (Winter), 1–2. Littell, J. H., & Schuerman, J. R. (1995). A synthesis of research on family preservation and family reunification programs. Rockville, MD: Westat, Inc. Maximus Inc. (1983). Child welfare indicator survey. McLean, VA. Malm, K., & Maza, P. L. (1988). Sources of agency income. Washington, D.C.: Child Welfare League of America. Maza, P. L. (1999a). Boarder babies and placement in foster care. Clinics in Perinatology, 26(1), 201–211. Maza, P. L. (1999b). The Adoption and Foster Care Analysis and Reporting System. Adoption Factbook III. Washington, D.C.: National Council for Adoption,. Maza, P. L. (2000a). Recent data on the number of adoptions of foster children. Adoption Quarterly, 3(2), 71–81. Maza, P. L. (2000b). Using administrative data to reward agency performance: the case of the federal adoption incentive program. Child Welfare, 74(5), 444–456. McCrosky, J., & Meezan, W. (1998). Family-centered services: approaches and effectiveness. The Future of Children, 8(1), 54–71. Murphy, P. T. (1994). Comment. In: M. R. Ventrell (Ed.), Excellence in Children’s Law (pp. 251– 256). Denver, CO: National Association of Counsel for Children. Needell, B., Cuccaro-Alamin, S., & Bookhart, A. L. (1999). Transitions from AFD.C. to child welfare in California. Children and Youth Services Review, 21(9/10), 815–841. Office of the Assistant Secretary for Planning and Evaluation (2000). Dynamics of children’s movement among the AFD.C., Medicaid and foster care programs prior to welfare reform: 1995–1996. Washington, D.C.: U.S. Department of Health and Human Services. Orr, S. (1999). Child protection at the crossroads: child abuse, child protection, and recommendations for reform. Washington, D.C.: Reason Public Policy Institute. Pelton, L. H. (1978). Child abuse and neglect: the myth of classlessness. American Journal of Orthopsychiatry, 48(4), 608–617. Research Triangle Institute (1999). National Survey of Child and Adolescent Well-Being: Annual Report. North Carolina: Research Triangle Park, NC Research Triangle Institute. Schene, P. (1998). Past, present, and future roles of child protection. The Future of Children, 8(1), 23–38. Schuerman, J. R., Rzepnicki, T. L., & Littell, J. (1994). Putting families first: an experiment in family preservation. New York: Aldine De Gruyter. Sengupta, S. (2000). No rise in child abuse seen in welfare shift. The New York Times on the Web. Stolley, K. S. (1993). Statistics on adoption in the United States. The Future of Children, 3(1), 26–42. U.S. Government Accounting Office (1995). Child welfare: opportunities to further enhance family preservation and support activities (GAO/HEHS-95–112). Washington, D.C.: U.S. Government Accounting Office. U.S. Department of Health, Education and Welfare [DHEW] (1962). Five Decades of Action for Children. Washington, D.C.: U.S. Government Printing Office.

Confronting the Dilemmas of Child Welfare: Past, Present and Future


U.S. Department of Health and Human Services [DHHS] (1993). A national survey of current and former foster parents. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (1995). Orphanage background materials. Unpublished Work. U.S. Department of Health and Human Services [DHHS] (1996). The Third National Incidence Study of Child Abuse and Neglect. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (1997). National study of protective, preventive and reunification services delivered to children and their families. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (1998). Children placed in foster care with relatives: a multi-state study. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (2000a). Child Maltreatment 1998: Reports from the states to the national child abuse and neglect data system. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (2000b). Child welfare demonstrations for fiscal year 2000 and 2001 (Log No. ACYF-CB-IM-200–01). Washington, D.C.: U.S. Government Printing Office U.S. Department of Health and Human Services [DHHS] (2000c). Report to the Congress on kinship foster care. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (2000d). Child welfare outcomes: 1998 annual report. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health and Human Services [DHHS] (2000e). 45 CFR Parts 1355, 1356 and 1357 Title IV-E foster care eligibility reviews and child and family services state plan reviews: final rule. Federal Register, (January 25), 4020–4093. Waldfogel, J. (1998). The future of child protection: How to break the cycle of abuse and neglect. Cambridge, MA: Harvard University Press. Westat, Inc. (1991). A national evaluation of Title IV-E foster care independent living programs for youth. Washington, D.C.: U.S. Department of Health and Human Services. Wulczyn, F. W. (2000). Federal fiscal reform in child welfare services. Children and Youth Services Review, 22(2), 131–159. Wulczyn, F. H., Brunner K., & Goerge, R. (1999). An update from the multistate foster care data archives. Chicago, IL: Chapin Hall Center for Children, University of Chicago.

This Page Intentionally Left Blank