Focus on Single-Parent Families: Past, Present, and Future : Past, Present, and Future 9780313379512, 9780313379505

Focus on Single-Parent Families: Past, Present, and Future brings together in one volume a range of cutting-edge researc

252 42 946KB

English Pages 310 Year 2010

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Focus on Single-Parent Families: Past, Present, and Future : Past, Present, and Future
 9780313379512, 9780313379505

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

FOCUS ON SINGLE-PARENT FAMILIES

Past, Present, and Future

ANNICE D. YARBER AND PAUL M. SHARP

Focus on

Single-Parent Families

This page intentionally left blank

Focus on

Single-Parent Families PAST, PRESENT, AND FUTURE

Editors Annice D. Yarber Paul M. Sharp

Copyright © 2010 by Annice D. Yarber and Paul M. Sharp All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except for the inclusion of brief quotations in a review, without prior permission in writing from the publisher. Library of Congress Cataloging-in-Publication Data is available at www.loc.gov. ISBN: 978-0-313-37950-5 EISBN: 978-0-313-37951-2 14 13 12 11 10

1 2 3 4 5

This book is also available on the World Wide Web as an eBook. Visit www.abc-clio.com for details. Praeger An Imprint of ABC-CLIO, LLC ABC-CLIO, LLC 130 Cremona Drive, P.O. Box 1911 Santa Barbara, California 93116-1911 This book is printed on acid-free paper Manufactured in the United States of America

Contents

Preface Acknowledgments Part One: Historical Trends in Family Life and Structure 1.

What’s Happening to the Family: Demographic and Institutional Changes Larry L. Bumpass

Part Two: Social Diversity and Single-Parent Families 2.

3.

4.

5.

6.

vii ix 1 3

23

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers J. Brian Brown and Daniel T. Lichter

27

The Single-Father Family: Demographic, Economic, and Public Transfer Use Characteristics Brett V. Brown

38

Children Raised in Fatherless Families from Infancy: A Follow-Up of Children of Lesbian and Single Heterosexual Mothers at Early Adolescence Fiona MacCallum and Susan Golombok Family Structure and Children’s Success: A Comparison of Widowed and Divorced Single-Mother Families Timothy J. Biblarz and Greg Gottainer Influences on Delinquency: The Role of the Single-Parent Family Amy L. Anderson

42

60 73

Contents

vi

7.

Parental Incarceration: Implications of Child Welfare Elizabeth I. Johnson and Jane Waldfogel

8.

Grandmothers Raising Grandchildren Catherine Goodman and Merril Silverstein

Part Three: Single-Parent Families: The Lived Experience 9.

Expenditures on Children and Financial Difficulties Mark Lino

86 100

119 123

10. The Psychological Well-Being of Single Parents Anna-Marie Cunningham and Chris Knoester

136

11. Single Parenthood and the Law Lynda H. Walters and Carla Abshire

158

12. The Impact of Parents’ Marital Status on the Time Adolescents Spend in Productive Activities Cathleen Zick and Corrine Allen

183

13. The Effects of Neighborhood and Mother-Teen Relationships on Adolescent Sexual Activity H. Harrington Cleveland and Michael Gilson

194

14. Religious Involvement as a Source of Support? Ann Sorenson, Carl F. Grindstaff, and R. Jay Turner

210

Part Four: Social Problems and Policy

221

15. Rethinking the “Pathology of Matriarchy”: Family Structure, Educational Attainment, and Socioeconomic Success Timothy J. Biblarz and Adrian E. Raftery

223

16. Child Care Policy Reform and Employment of Single Mothers Jay Bainbridge, Marcia K. Meyers, and Jane Waldfogel 17. Explaining Trends in Child Support: Economic, Demographic, and Policy Effects Anne C. Case, I-Fen Lin, and Sara S. McLanahan

245

262

18. Effective Child Support Policy for Low-Income Single Parents Maureen R. Waller and Robert Plotnick

271

Index

295

Preface

We had several goals in creating Focus on Single-Parent Families: Past, Present, and Future. First, recognizing that single-parent families are now a notable and apparently a permanent part of American family life, we wanted to produce an anthology of superior primary source materials covering single-parent families. We believe a book of readings on SingleParent Families that offers academic rigor, addresses the intricate nature of family within the dynamics of the larger society, and provides policy trends would be a valuable contribution to the field. Our second goal was to encourage students to look beyond the obvious and to understand the complexity and diversity of single-parent families. To assist in this process, we present a broad and thorough sampling of scholarship on issues related to single-parent families. We believe that professors and students, and ultimately society, would benefit from a more concentrated coverage of sociological principles and ideas regarding single-parent families. Although most of this scholarship is sociological, we have also drawn on the work of other social sciences. Our third goal was to create pedagogical materials; that is, a set of core readings around which to build a course. Introduction to each of the topical sections and 19 articles allows the book to be used as a supplement or stand-alone text in upper-division undergraduate and graduate level courses dealing with family, social problems, and public policy. We present the material in a way that students find stimulating, clear, and highly readable. At no time in our history have so many adults and children been members of single-parent families. Part One, Historical Trends in Family Life and Structure, is dedicated to the historical, demographic, and economic trends that have come to shape our modern society and the growth of such families.

viii

Preface

Part Two, Social Diversity and Single-Parent Families, takes a look at the diversity of single-parent families once thought to just involve mother-child families. The diversity of the country is found in all families and includes many nationalities, ethnic and racial categories, socioeconomic categories, and an array of combinations that seems to increase with each passing decade. Part Three, Single-Parent Families: The Lived Experience, continues the look at diversity of this family type and adds what has come to be called “the lived experience” meaning that variations in economic conditions, religion, race, family size, and many other factors contribute to the experiences of living in a single-parent family much to the same degree that others experience life in a two-parent family. Single-parent families in their diverse forms appear to have become a permanent fixture on the landscape of American family life. The articles in Part Four, Social Problems and Policy, explore family structure as a “social problem” as well as social policies needed to assist single-parent families in their “lived experience.”

Acknowledgments

We extend our thanks to Debora Carvalko and the staff at Praeger Publishers for giving us the opportunity to publish this work. Clearly, Debora saw the importance of our work and showed both vision and support throughout the project. We extend our thanks for the support we received from Auburn University Montgomery as well as our colleagues who offered suggestions and feedback.

This page intentionally left blank

Part One

Historical Trends in Family Life and Structure

Once a rare family type, single-parent families have become a notable and permanent feature of family life in the United States. Single-parent families, at least historically, had been most often formed as a result of widowhood, with many single-parent families being formed as a result of the death of the mother during childbirth or the death of the father in relation to occupational accidents or war. Though most often poor, these families were seen as having been created through unforeseen misfortune, not necessarily by individual choice. Over the course of the twentieth century, virtually every aspect of modern society changed rapidly as did the structure of the family. With family structure changing at such a swift rate, one growing and extremely important family structure has emerged in a variety of forms: the singleparent family. Historical, demographic, and economic trends play a significant role in the changing family structure. All of the following selections provide a historical look at social and cultural factors that influenced the rise in single-parent families. In his presidential address to the Population Association of America in 1990, Larry L. Bumpass asked “What’s happening to the family?” The trends in various family structures from marriage to fertility formed the basis of his poignant remarks concluding, in part, that profamilial normative pressures have eroded in all areas of the life course, leading to many of these heretofore uncommon family modalities. Although postmodernity and the tremendous speed with which change occurs underlie much of the patterns, Bumpass, as a demographer, nonetheless sets the stage for our look at the single-parent family in America. The major trends covered by Bumpass are each uniquely significant, yet when combined they are

2

Historical Trends in Family Life and Structure

powerful forces that have changed, and indeed are still changing, the most primary of social institutions—the family. As you read this selection, imagine the major changes in family life that laid the groundwork for the emergence of many nontraditional forms of family and family life. First, marital disruption—perhaps no other single factor has caused such profound alterations in family life than divorce. Indeed, family instability dominates core aspects of family for the majority of the population and especially children. Cohabitation and marriage, Bumpass argues, constitute factors contributing to changes in family life. These intimate lifestyles are numerous but here involve persons whom will likely marry. Nonmarital childbearing steeply increased beginning in the late 1960s and most of these births have been unplanned and have often been to cohabitating couples. In addition, the surge in teen childbearing is an extremely important component in unplanned fertility. The changing role and complexity of work is no small influence regarding childbearing and parenthood either, regardless of relationship type. Intergenerational relationships often involve the ascribed status of parents, grandparents, and a plethora of other relations and social support networks that create interdependency, thus linking generations together. Apparently, family and family relationships play a smaller and smaller role in our lives. Bumpass provides significant insight into the increasing individualism not just in the United States but in other Western nations, and the impact on family structure and processes.

Chapter 1

What’s Happening to the Family: Demographic and Institutional Changes Larry L. Bumpass*

As I argued several years ago (Bumpass, 1986), the perspectives of demographic transition theory are relevant to contemporary family change (Davis, 1967; Freedman, 1975). Because fertility cannot be isolated theoretically from the institutional context in which it is embedded, theories about fertility decline are intrinsically also theories about changes in the family as an institution. Hence the question, “What’s happening to the family?” is one that demographers have been asking for a long time, and our perspective has been both historical and dynamic. The institution of the family is not seen as a fixed form against which we can judge current behavior. Rather, it is the collective representation of our changing family experience, as that experience interacts with its environment. Normative expectations play a major role in structuring family patterns, but they also tend to lag behind changing behavior, accommodating in time to behavioral changes. Duncan and Schnore’s (1959) POET formulation draws attention to the reciprocal nature of causation and the interplay between social organization and technological change as technology redefines environment. Certainly, the evolution of our technological culture has radically altered the social context of family relationships. I believe that the theoretical perspectives of a half-century ago were essentially correct, from Ogburn’s emphasis on declining family functions (Ogburn & Tibbitts, 1933) to Lorimer’s description of increasing individualism at the expense of moral obligation (Lorimer, Winston, & Kiser, 1940). Goode’s (1963) World Revolution and Family Patterns was on target * L. L. Bumpass. What’s Happening to the Family? Interactions Between Demographic and Institutional Change. Demography, 1990, 27(4). Reprinted with permission from the Population Association of America.

4

Historical Trends in Family Life and Structure

with respect to many of the linkages between industrial economies and the reduction and reshaping of family roles, though I doubt that the conjugal family system should be seen as the end point of this process. At the same time, the changing structural and economic bases are not the whole story. Lesthaeghe (1983; Lesthaeghe & Surkyn, 1988; Lesthaeghe & Wilson, 1985) argued that individuation and secularization are independent causal forces in family change. These cultural factors exist in symbiotic relationship with modern economies, but they have deep historical roots of their own (Bellah, Madsen, Swidler, Sullivan, & Tipton, 1985; Shorter, 1973; Stone, 1982). When needs and interests conflict, as they do in any collectivity, how much weight must the individual give to the interests of others, in contrast to his or her own interests, and how large a circle of others must be considered? The shrinkage of this circle from a larger community toward the individual alone is nowhere better illustrated than by the level of marital disruption, to which I will return momentarily. I am aware that the stance taken here may appear to run counter to historical work discrediting various links between modernization and family change. However, I think the differences are more apparent than real. This perspective does not require close temporal association between economic or structural shifts and family changes. Underlying ecological contexts of organizational change may create tensions, much like those at a fault line, the effects of which are not seen until the confluence of forces overcomes inertia. The precipitating factors may be economic, cultural, or political events, or they may be no more than the accumulated weight of past change. The most dramatic recent example of this is the collapse in the total fertility rate in Spain, over a single decade, from one of the highest in Europe to one of the lowest (Munoz-Perez, 1987; Sardon, 1986). Our confidence that we understood low fertility was shattered by the baby boom, so we retreated and tried to comprehend this embarrassing disconfirmation. As Ryder put it, demographers were “victims of a failure of nerve: they should have stuck to their theoretical guns” (1979, p. 359). Although the baby boom is long over, the demographic transition in the West most likely is not. We tend to think of the transition in the relative value of childbearing as a past event for today’s low-fertility societies and implicitly assume that we have reached a final equilibrium. However, we have no theoretical basis for this assumption. Indeed, a growing number of authors are suggesting that the factors supporting marriage and parenthood are likely to erode further under the pressure of competing opportunities (e.g., Bernard, 1972; Butz & Ward, 1979; Davis, 1983; Demeny, 1986; Huber, 1980; Keller, 1971; Popenoe, 1988; Ryder, 1979; Westoff, 1978; Westoff & Ryder, 1977). Our consideration of family change profits greatly from research on the baby boom and the subsequent fertility decline. We recognize, as

What’s Happening to the Family: Demographic and Institutional Changes

5

emphasized by both Blake (1972) and Davis (1972), that a broad array of pronatalist influences persist, particularly in the different socialization of males and females and in discrimination against women in the workplace. We also see that some pronatalist forces may occur on a cyclical basis, as emphasized by Easterlin (1978) and Butz and Ward (1979). However, we too often consider one explanatory scheme at a time. We must expand our view to include simultaneously both profamilial and antifamilial factors and to regard family and fertility behavior as a dynamic outcome of these opposing forces. This puts the question in historical perspective, drawing our attention to the trend line around which fluctuations may occur; and it raises the question of which forces are likely to strengthen and which to weaken in the future. In spite of the great inertia of pronatalist forces, subreplacement fertility is the norm in Western industrial societies, and the processes underlying the long-term decline seem far from exhausted. Changes in one family domain may contribute to further changes in that domain or in others. We have experienced major alterations in marriage, divorce, childbearing, women’s employment, parenting, and adult intergenerational relationships—and these have mutually reinforcing effects as well as impacts on normative expectations about family life. I now turn to these issues. The following discussion draws heavily on the 1987–1988 National Survey of Families and Households (Sweet, Bumpass, & Call, 1988), and results not otherwise referenced are based on our tabulations with these data.

MARITAL DISRUPTION The deep historical, cultural, and structural roots of changing family behavior are illustrated well by the trend in cohort divorce rates (Cherlin, 1981; Preston & McDonald, 1979). Although annual rates of divorce have fluctuated around the trend line, the underlying rate of increase in the level of lifetime divorce has been virtually constant for more than 100 years, generating the accelerating curve from 7 percent for marriages in 1860 to the current expectation of well over one-half. It is too soon to conclude that the plateau in the divorce rate of the last decade represents the end of this trend—we must remember that there was a similar 15-year plateau before the takeoff of the late 1960s. In any event, the current level of marital disruption is very high. Castro Martin and Bumpass (1989) recently estimated that almost twothirds of recent first marriages would be likely to disrupt if current levels persist. Further work leads us to suspect that 60 percent may be closer to the mark. However, the exact level of marital disruption is

6

Historical Trends in Family Life and Structure

much less important than the social fact that the majority of recent first marriages will not last a lifetime. The pervasiveness of this experience argues against hypotheses tied to specific locations in the social structure, certainly for the underlying trend. Though there are important differentials—in particular, by race, education, and age at marriage (Castro Martin & Bumpass, 1989)—no group is immune. For example, rates of the early 1980s imply that more than half of first marriages are likely to disrupt among whites, women who attended college, and those who married after the age of 23. Further, the strength of the forces affecting marital stability is illustrated by the experience of Catholics and others who disapprove of divorce. Despite the strong position of the Catholic Church and its buttressing through religious sanction and exhortation, Catholics are no less likely to divorce than are non-Catholics (Bumpass, Castro Martin, & Sweet, in press). Similarly, Thornton (1989) found in his longitudinal analyses that disapproval of divorce is unrelated to subsequent divorce rates, even though attitudes tend to accommodate to having been divorced. This high level of marital instability dominates core aspects of family life for the majority of the population and challenges family norms. The most important impact is on the family contexts of children. About half of today’s young children in the United States will spend some time in a single-parent family, most as a consequence of divorce (Castro Martin & Bumpass, 1989). Furthermore, this is not simply a transitional phase between first and second families: the majority will remain in a motheronly family for the remainder of their childhood (Bumpass & Sweet, 1989a). We have heard about the problems this creates for children so often that we are in danger of being numbed to them. In addition to psychological distress (Heatherington, Camara, & Featherman, 1983; Wallerstein & Blakeslee, 1989; Wallerstein & Kelly, 1980), consequences include a marked increase in poverty for women and children (Garfinkel & Mclanahan, 1986; Mclanahan, Astone, & Marks, in press), effects on parenting practices and adult time available for children (Heatherington et al., 1983; Thomson, Mclanahan, & Curtin, 1990), and substantial negative impacts on children’s educational attainment and their own family and fertility histories (Mclanahan 1985; Mclanahan & Bumpass, 1988). Divorce illustrates the force of secular individualism more dearly than changes in any other family domain. Although there are obviously many cases, such as abusive families, in which divorce is essential for the wellbeing of children, the interests of parents and children in the preservation of a marriage often differ (e.g., Mclanahan, 1989; Wallerstein & Kelly, 1980). Despite the widely recognized impacts on children, we increasingly acknowledge parents’ self-interest as a legitimate criterion for the

What’s Happening to the Family: Demographic and Institutional Changes

7

resolution of this conflict. The proportion saying that a couple should not stay together for the sake of the children increased from 51 percent to 81 percent between 1962 and 1982 (Thornton & Freedman, 1983). As Mclanahan (1989) pointed out, children are virtually powerless in this conflict of interest. Traditional notions of the family cannot long survive such massive disconfirmation. Of course, many families are stable. More than onethird of first marriages are likely to remain intact for life, and more than one-half of the children who begin life in a two-parent family will have intact families throughout childhood. Nonetheless, widespread awareness of how common disruption is leaves little basis, a priori, for confidence in marital survival. Insecurity about the viability of marital relationships may lead to reduced investments in some marriages, which, in turn, lower the prospects that they will last (Becker, 1981). Hence there may be feedback from demographic behavior to the institution of marriage itself. A major case in point is the attachment of young women to the labor force. This long-term trend may be reinforced by the high level of marital disruption. Because marriage provides such a shaky lifetime guarantee, adolescent girls may increasingly recognize that they must be able to support themselves if necessary. In turn, increases in women’s economic independence may reduce the probability that couples will stay in bad marriages (Goode, 1963; Hannan, Tuma, & Groeneveld, 1978) or, perhaps, even stick out the tough times in good marriages. Of course, few actually plan for divorce; prenuptial agreements are news because they are rare and violate our normative expectations. Stable marriage remains the ideal: three-quarters of people surveyed agree that a couple should remain together for a lifetime except under extreme circumstances, and only 12 percent explicitly disagree. Nonetheless, norms against divorce have weakened greatly: only one-third disapprove of divorce even for couples who have a preschool child.

COHABITATION AND MARRIAGE Although the disruption of marriages has the most profound implications for changing family life, the formation of unions has also undergone revolutionary change (Bumpass & Sweet, 1989a). The proportion of first marriages that were preceded by cohabitation increased from 8 percent for marriages in the late 1960s to 49 percent among those in 1985–1986. The accelerating increase in cohabitation has not likely attenuated. Consequently, it is likely that more than half of persons in their 30s have lived in a cohabiting relationship, and more than half of recent marriages were preceded by cohabitation.

8

Historical Trends in Family Life and Structure

How has it happened that what was once morally reprehensible has become the majority experience in just two decades? We are still working to understand the meaning and implications of this trend, but I would guess that two major factors are at issue. The first is the erosion of normative objections. “Shacking up” was offensive, after all, not because couples were sharing cooking and the laundry, but because they were sharing a bed. The revolution in the sexual experience of unmarried persons over the same period has seriously weakened this basis for disapproving of cohabitation. Only one-fifth of young adults now disapprove of premarital sex—even for 18-year-olds—and only one-sixth explicitly disapprove of cohabitation under any circumstances. All else being equal, it seems likely that sharing a household with an intimate friend is more rewarding than maintaining the same relationship at a distance. So with reduced normative constraints, we should expect increases in cohabitation, even if nothing else changes. Something else has changed, though; in particular, we are much less confident about marital stability. As the fragility of marital relationships is increasingly recognized, couples may sense a need to “try out” marriage by living together before making a long-term commitment. In a series of questions about reasons for cohabitation, that “couples can make sure they are compatible before getting married” was indicated far more than any of the other options offered. Of singles who expect to cohabit in the future, more than 80 percent say that they think this is an important reason for cohabitation. Consistent with this view, at least one partner expects marriage in 90 percent of cohabitations. What are the implications of cohabitation for family life more generally? Cohabitation is clearly symptomatic of our changing values about what is unique to marriage. Nonetheless, because the changes are at the boundary of marriage, I believe that cohabitation has not changed the meaning of marriage much for the married population. There are, however, several implications worth noting: 1. First, it is the meaning of “single” that has changed markedly with cohabitation. The reduced number of years spent married during young adulthood should not be viewed as simply a period of unattached living. Most of the rapid decline in marriage has been offset by increasing cohabitation; young persons are setting up joint housekeeping almost as early in their lives as before marriage rates declined. It is also important that almost half of the cohabiting couples have children present (Bumpass, Sweet, & Cherlin, 1989). 2. Associated with this, the event of marriage is a less specific marker of other transitions. Sex, living arrangements, and parenting depend less on marriage. This, of course, is a nuisance for demographic modeling, which assumes such linkages. We have to rethink how we ought to

What’s Happening to the Family: Demographic and Institutional Changes

3.

4.

5.

6.

9

model the dependencies between marriage and related demographic events (e.g., Rindfuss & Parnell, 1989). At the same time, the effect on most duration-dependent measures is probably not great because the average duration of cohabitation is very short (median of 1.5 years). There is not likely to be a single answer to whether unions should be clocked from cohabitation or marriage. Since marriage is not a necessary condition for sharing a household, it may be delayed simply until a more convenient time for the ceremony or until the couple is more certain of their commitment. Hence for some couples “marriage” essentially began when they started living together, whereas others avoid an unstable marriage by splitting before they reach the altar. It seems most likely that these “premarital divorces,” as Sweet calls them, have helped keep the divorce rate from going even higher. If many couples are using cohabitation to test their relationship, and if 40 percent split up without marrying, then we expect those who do marry to have more stable marriages than would have been the case in the absence of cohabitation. Despite all the attention given the results on the higher divorce rate of marriages preceded by cohabitation (Bumpass & Sweet, 1989b), this finding is not contrary evidence on this point. Cohabitation is selective and includes a lower proportion of couples who hold traditional family attitudes (Booth & Johnson, 1988; Bumpass et al., 1989) and a higher proportion of those who are uncertain about their relationship. Further, the data suggest that not all cohabitations may be part of the marriage process. Some may be better characterized as relationships of convenience in which marriage is not at issue. Although three-quarters of cohabitors expect to marry their partner, one-quarter do not. The other partner anticipates marriage in most of these latter cases, and the process may be moving either toward or away from marriage. In 10 percent of cohabiting couples, however, neither partner expects marriage. The planned follow-up to the National Survey of Families and Households (NSFH) will allow us to evaluate the extent to which cohabitors with different marriage expectations subsequently marry. At present, there is some relevant preliminary evidence based on responses to a question on the perceived chances of eventual separation. Cohabitations in which both partners expect marriage resemble recently married couples more than they do other cohabitors. Finally, both marital instability and cohabitation are part of, and contribute to, the reduction in the perceived necessity of marriage. If marriage is not a prerequisite for many of the benefits traditionally associated with it, and if it is only a weak guarantee of those benefits, awareness of these facts may be eroding marriage’s normative prescription. Only one-third of young adults under 25 agree that “it is

10

Historical Trends in Family Life and Structure

better to be married than to go through life single,” and one-quarter disagree (Bumpass et al., 1989). As Thornton (1989) recently emphasized, marriage has become much more discretionary as an adult role. Now, virtually all persons plan to marry, and almost 90 percent are likely to do so if current age-specific rates persist. Nonetheless, both marriage and remarriage rates have continued their downward slide; and for the most recent experience in the NSFH, these declines were not fully offset by cohabitation. Marriage is not going out of style, but it is being progressively delayed. It seems likely that the proportion never marrying, and especially the proportion not remarrying after divorce, will continue to increase.

CHILDBEARING AND PARENTING Parenthood is also losing its ascribed status (Bumpass, 1973). Indeed, an item similar to the one on marriage, whether “it is better to have children than to go through life being childless,” gets similar responses. This is exactly what we would expect from both the theoretical perspective on long-term fertility decline and the progressive legitimization of individual self-interest. As we all know, the total fertility rate has been stable in the United States over most of the two decades when marriage behavior has been changing so radically. However, this stability conceals a great deal of change. The proportion of children born to an unmarried mother has doubled since 1970 to more than one-quarter in 1987. One aspect of this trend is that nonmarital childbearing is now becoming part of the life experience of a significant proportion of women. Life table estimates suggest that 17 percent of white women and 70 percent of black women will have a child while unmarried if recent levels persist. The rapid increase in the prevalence of nonmarital childbearing is evidence of, and contributes to, the erosion of norms against behavior traditionally described by such terms as “illegitimacy” and “bastard.” It is illuminating on this point that one-third of our unmarried respondents under 35 agreed to the item, “It would be all right for me to have a child while unmarried if I planned on marrying,” and that only half explicitly disagreed. This is not to say that unmarried childbearing has become desirable for many, only that the normative constraints against it have weakened greatly. Why might this be so? I think the answer again lies largely in the level of marital instability. If marriage assures neither a stable two-parent family for the child nor lifetime economic security for the woman, the importance of marrying to “legitimize” a birth is much less compelling

What’s Happening to the Family: Demographic and Institutional Changes

11

(O’Connell & Moore, 1980). Further, as with cohabitation, it is decreasingly relevant that the birth provides evidence of nonmarital sex. More than two-thirds of nonmarital births are reported by their mothers as unplanned. Hence I argue that the increase has been driven far more by changing orientations toward marriage, and the reduced stigma of nonmarital childbearing, than by desires to have children out of wedlock. Falling rates of marriage and remarriage and increased rates of marital disruption have simply lengthened the unmarried period of risk for unplanned childbearing. It is an interesting twist in all of this that one-quarter of all out-ofwedlock births are to cohabiting couples and hence constitute two-parent unmarried families. But in the present context, it is best to regard the vast majority of these as simply unplanned childbearing that happens to occur during cohabitation, rather than as a form of planned “marital” fertility. Most of these result from unplanned pregnancies, and one-third of the parents do not marry each other (Bumpass & Sweet, 1989a). The unusually high birth rate among teenagers is one important component of unplanned fertility (Trussell, 1988). In the current “stable” fertility setting, almost two-thirds of teenage births are now premarital, and 80 percent are either premarital or premaritally conceived (U.S. Bureau of the Census, 1988). Overall, nearly two-fifths of births in the early 1980s were either timing or number failures (Pratt & Horn, 1985). Thus there is a substantial component of recent fertility that could be drastically reduced in the future. I think we will continue to move toward a “perfect” fertility-control society of the sort that Westoff and I reflected on 20 years ago (Bumpass & Westoff, 1970). RU486 (the French abortion pill) and future such developments are likely to move us a bit farther along that path. To the extent that we reduce unplanned fertility, I expect a substantial effect on completed family size—in part because delayed fertility tends to lead to births foregone (Rindfuss & Bumpass, 1978). I expect further declines in intended fertility for other reasons as well. First, once again, is our high level of marital instability. As Weiss and Willis (1985) pointed out, the very substantial prospect of single parenting is likely to be increasingly recognized as among the potential costs of motherhood. For men, the argument runs that the potential benefits are likely reduced, given that after divorce most children stay with their mothers and have very reduced contact with their fathers. The second reason for expecting further fertility decline is the difficulty of juggling family and work obligations. One aspect of this is the theme developed so well by Presser (1989). With the increasingly equal involvement of women in the labor force, arranging for child care has become something of a nightmare, which recent legislation has only begun to remedy.

12

Historical Trends in Family Life and Structure

Quite apart from child care, though, complex work schedules often compete with potential family time. For example, questions on work schedules in the NSFH revealed that among two-earner couples with children at home, three-fifths have some work hours between 5:00 and 10:00 in the evening and half have weekend hours; about one-sixth work most of the hours during these periods. The employment of mothers of young children is very instructive with respect to our relative values. When there is a conflict between our values concerning the parenting needs of children and economic roles, our priorities are demonstrated by our behavior. The importance of a mother’s full-time care for an infant has been built into our value system and reinforced by major sources of advice on child care. Three-quarters of U.S. women in 1970 said that they thought that a mother’s working was harmful to a preschool child (Mason & Bumpass, 1975). Yet in the years since, the employment of mothers of young children has doubled to more than one-half. Only one-third now agree that a mother’s working is harmful, but more instructive, almost another third are uncertain and neither agree nor disagree. Even among mothers of preschoolers who are working fulltime, only about half disagree that such work is harmful for children. It is important that I not be misunderstood on this point. I am not asserting that mothers’ working is harmful; the evidence seems mixed at best. My point is that this major change in parenting behavior has occurred despite the belief that it is or may be harmful. This change is often cast as being driven by economic necessity. Surely many mothers of young children may work against their preferences because they have to. We must also remember, though, that economic need is a highly amorphous concept, always seeming to outstrip what we have. It seems more appropriate to recognize women’s employment as yet another process with deep historical roots, driven in large part by the extension to women of individual opportunities that have long been available to many. As Davis (1984) emphasized, it has been a continuous process over the century, with the increases among mothers of young children the last to occur because of our strong cultural values against such employment. Thus even in the context of stable marriage, when parenting competes with other adult interests and roles, self-interest is a legitimate basis for decision. By “self-interest” I do not intend to denote selfishness in a narrow hedonistic sense but, rather, concern with what one ought to do for oneself: for example, the need for a woman to be capable of financial independence. There is also a clear normative component to individuation that is illustrated in the declining legitimacy of “housewife” as a full-time adult identification for women or, indeed, in expected lifestyles and consumption patterns. Our low birth rate and increasing nonparenthood (Bloom & Trussell, 1984) are the outcomes of the struggle between parental roles as prescribed

What’s Happening to the Family: Demographic and Institutional Changes

13

by our culture and other adult commitments, in which family roles often lose out. Low fertility exists throughout Western societies in the face of persisting gender inequality. I think we will move inexorably toward true equality, and the full implications of that for fertility are yet to be seen.

MARITAL RELATIONSHIPS Turning to marital relationships, increasing women’s employment may be altering the mate selection process (Oppenheimer, 1988), the nature of marriage bargains (Becker, 1981), and marriage relationships and marital stability (Cherlin, 1981; Hannan et al., 1978). Indeed several authors (Bennett, Bloom, & Craig, 1989; Schoen & Owens, 1990; Thornton, 1989a) have suggested that what men and women want out of marriage may be diverging, making it harder for bargains to be struck in the marketplace and contributing to lower marriage rates. Husbands and wives may have increasingly different expectations about the division of household labor, and some may even disagree about whether the wife should be employed. We find that one-third of husbands in recent marriages prefer that their wife not be employed. Nonetheless, males are not the only source of disagreement on this issue. One-sixth of wives would rather not be working, including 10 percent of those employed full time; and of women who prefer to work, the vast majority want less than full-time employment. The difference in the amount of hours that men and women say they would prefer to work undoubtedly reflects both differential socialization and the realities for women, of having to combine household and labor market obligations. In the context of the attention given wives’ employment, we find surprisingly little relationship in the aggregate between female employment and measures of marital quality. On the other hand, effects do appear when the joint preferences of spouses are taken into account. Furthermore, which combinations affect marital quality depends on whether you are talking to the husband or the wife. For example, in marriages in which the wife wants to work and is working despite her husband’s preference that she stay home, trouble in the marriage is associated with her employment when the husband is the respondent but not when the wife is. On the other hand, when the wife does not want to work but her husband wants her to, her employment is associated with trouble in the marriage only when she is the respondent. In other words, when there is a discrepancy between spouse preferences, if the wife works in violation of the traditional power relationship, husbands see it as disruptive of the marriage; whereas, if the wife does not work, in deference to the traditional power relationship, she sees it as disruptive.

14

Historical Trends in Family Life and Structure

Thus, there are important issues of spouse disagreement over the wife’s employment. At the same time, we must remember that employment is not a point of conflict in the majority of marriages. About threequarters agree about the wife’s employment—most that she should be working. Among these couples, the wife’s employment has no effect one way or the other on marital quality. We find that employment beyond the hours regarded as normal in our society seems related to marital quality. Trouble in the marriage is reported more often if the wife is working more than 45 hours or if either spouse works on a job where the shift varies (White & Keith, 1990). A major issue in the effect of wives’ employment on couples’ relationships is the extra task load that working women carry because of persisting traditional expectations. The good news is that male household efforts are responsive to the level of the wife’s employment. In our data the actual hours contributed by the husband (and not just the proportion of total effort) increase by 30 percent between couples in which the wife is not working and those in which she is working full time (Dutchin-Eglash, 1988); the time fathers spend in care and feeding of young children similarly increases with the mother’s hours of employment. The bad news, though it is little “news,” is that husbands carry only about one-third of the household task load even if their wives are working full time. Husbands and wives agree that household task allocation is unfair to the wife. What is striking is the marked difference between male and female respondents in the association between the perceived fairness of household task allocation and reports that the marriage may be in trouble. There is only a modest association among males, but females who regard the division as unfair are much more likely to report trouble in their marriage. Further, this effect increases with the number of hours that a wife works. Wives working full time are twice as likely to report having trouble in their marriage if they regard the division of labor as unfair to them (40 percent compared with 18 percent among those who regard it as fair). Hence the “second shift” (Hochschild, 1989) of working women affects the potential stability of many marriages—at least as seen by women. For instance, among persons with separations in the last 10 years, women are more than twice as likely as men to report that they wanted the separation (Mclanahan, 1989). In the midst of all this disruption and changing role behavior, why hasn’t the family disappeared? Why does anyone bother to form families with children? Although the answer seems obvious because of our profamilial cultural values, it is worth reiterating that I view family relationships as being in a dynamic tension, with many profamilial factors involved in this balance. Once again, I am drawn back to the classroom in Ann Arbor, where Freedman passionately argued that the relational aspects of family were

What’s Happening to the Family: Demographic and Institutional Changes

15

not available in the marketplace. Intimacy with a partner in the context of commitment provides unique aspects of meaning and social support. The data make it clear that parenting creates many strains, but raising children also provides many rewards of more intimate and holistic relationships than are to be found in work roles. The status our society ascribes to marital and parenting roles is eroding in comparison with those of the workplace, but the majority of us continue to find family roles intrinsically and uniquely rewarding.

INTERGENERATIONAL RELATIONSHIPS A part of this is clearly found in intergenerational relationships in midlife and later years, as these provide a major source of social connectedness and support—perhaps all the more important (at least in relative terms) because of the greater volatility of marital relationships. The ascribed aspect of this relationship makes parents and children more permanent members of one’s “convoy” of social support (Kahn & Antonucci, 1981) over the life course. Friends and even spouses may come and go, but it is harder to trade in one’s parents—and for women at least—one’s children. Most parents and children live near one another and continue to interact on a regular basis throughout life (Hoyert, 1990; Shanas, 1982). Help with household tasks, baby-sitting, emotional support, and financial exchanges (and most important, the potential for these if needed) weave a fabric of interdependency linking the generations together. We may not fully appreciate the importance of some aspects of intergenerational support because they are specific to certain life stages. Our data on family help with first-home purchases illustrate this point nicely. Among those who have purchased a home since 1980, about one-quarter reported that they had help from relatives (mostly parents). For those who received such help, about half were gifts and half loans, and the median amount was almost $5,000. Similarly, although independent living on the part of the elderly has become increasingly common—reflecting both our values and improved economic ability to implement them (Michael, Fuchs, & Scott, 1980)— coresidence is far from rare even today. Of the elderly with living children, one-sixth presently share a household with an adult child, and one-quarter of the elderly in ill health do so. From the middle generation’s point of view, over a quarter of persons aged 55–64 report that a parent has lived with them since they have had their own household. We miss much of this in cross-section because the period of coresidence tends to be short—a period of reduced independence preceding death or institutionalization.

16

Historical Trends in Family Life and Structure

At the same time that these critical intergenerational linkages persist, they too are likely being altered. The proportion of the elderly seeing a child at least once a week declined by 25 percent between 1962 and 1984, and there has been a marked decline in the functional interactions of help with household repairs by men and help with housework by women. For example, the proportion of women over 65 with living children, who reported that they had helped a child with housework during the last month declined from 52 percent, to 39 percent (Shanas, 1982), to 8 percent over the 1962–1987 period. Many factors may be involved, including the marked increase in employment of their daughters and the high level of marital instability. The full effect of marital instability on intergenerational relations will not be seen until the cohorts that have experienced the high instability of the last two decades approach retirement. Given that most fathers become rather marginal in the lives of their children within a few years of marital disruption (Furstenberg & Nord, 1985), what does this imply about future intergenerational relationships as these fathers enter old age? Among persons with both parents living, those separated from their fathers during childhood are much less likely to see him regularly: 50 percent see their fathers more than once a year, compared with about 90 percent among those whose parents are still married at interview. As we would expect, the differences are not as dramatic for mothers; but even so, the proportion having little contact with their mothers is twice as high among those whose parents are no longer living together. Lower levels of exchange are also associated with divorce and remarriage in the middle generation.

CONCLUSION Thus, it seems that profamilial normative pressures have eroded in all areas of the life course. The nature of these value changes was dramatically illustrated in a recent piece by Easterlin and Crimmins (1988). They noted that there has been almost a doubling since 1970 in the proportion of young people who set financial success as a major life goal, with an associated decline in the proportion concerned with finding meaning in life. Not coincidentally, there was also a marked decline in the proportion who think parents should spend more time with their children. Reporters often want me to say whether recent family change is “good” or “bad.” I have to respond that there is no single answer because we hold values that are incompatible. From the perspective of adult individual freedom, much is being gained. The steady erosion of the ideological baggage of patriarchy is good for both men and women. The increasing opportunity to choose rewarding occupations and rewarding intimate relationships is not something many of us would wish to reverse. At the same time, we continue to care about the well-being of children, about

What’s Happening to the Family: Demographic and Institutional Changes

17

those without adequate social and economic support, and about the human capital of the next generation. As Popenoe (1988) argued, asking whether the family is disappearing misses the point. What is at issue is not the persistence of the institution of the family but, rather, the nature of family patterns in the relevant future—and the opportunities and costs of those patterns. Understanding the long-term character of institutional change should direct social policy toward the amelioration of negative consequences, rather than toward attempts to reverse the tide. The underlying causes are not in events or policies of the last several decades, and it is most unlikely that social policy can significantly alter the course of these trends. I agree with Teitlebaum and Winter (1985) that the arguments against outweigh those in favor of pronatalist policies in Western societies. Nonetheless, with continuing subreplacement fertility, I expect to see an increase in pronatalist concern—motivated by variations on the 1930s themes of “national decline.” Experience to date suggests that even very expensive programs are unlikely to have much effect. On the other hand, attempts in this direction could inadvertently improve the well-being of families to the extent that they include factors such as family allowances, parental leave, and child care. I am cautiously optimistic about the prospects for constructive legislation, though I arrive at this optimism through a perverse form of cynicism. Although I clearly prefer that policies be implemented out of a concern for children and the less privileged, our record in this regard is checkered at best. On the other hand, our society may be willing to commit major resources to prevent further deterioration of our economic competitiveness; and we may have the foresight to understand the importance for this goal of our investment in children. Adequate health insurance, quality day care, and a family income floor for all children seem essential ingredients for bolstering the human capital of the generation that must pay the taxes when we are old. In addition, we just may be able to achieve adequate provision for the long-term care of the elderly through the conjunction of the self-interest and political power of the baby boom. We tend to focus on the consequences for health care when the baby boom reaches older ages. As recognized in a recent series of articles in the New York Times, however, this generation’s influence will first be felt as these individuals seek help with the long-term care of their own parents. The baby boom represents an increased voting block in middle age, when a higher proportion than ever before will have surviving parents and parents living to be very old. So what is happening to the family? Family relationships occupy an important but ever shrinking space in our lives (Espenshade, 1985). This is the continuation of a long-term process and is not confined to one country. Trends in cohabitation, marriage, fertility, and marital disruption are

18

Historical Trends in Family Life and Structure

widely shared across Western industrial societies. To my mind, major causes include the individualizing tendency of participation in our economy and cultural values of individualization that both facilitate this participation and are reinforced by it. There is no reason to think that these processes are exhausted or are likely to reverse. Hence the demographic transition in the West is likely to continue for the foreseeable future, with implications for family change all across the life course.

REFERENCES Bachrach, C. A., Horn, M. C., Mosher, W. D., & Shimizu, I. M. (1985). National Survey of Family Growth, cycle III: Sample design, weighting, and variance estimation. Vital Health Statistics, 2(98). Hyattsville, MD, National Center for Health Statistics. Becker, G. S. (1981). A treatise on the family. Cambridge, MA: Harvard University Press. Bellah, R. N., Madsen, R., Swidler, A., Sullivan, W. M., & Tipton, S. M. (1985). Habits of the heart: Individualism and commitment in American life. Berkeley: University of California Press. Bennett, N. G. (1990). Out-of-wedlock childbearing, marriage, and divorce in the United States. Paper presented at the Conference on Demographic Perspectives on the American Family: Patterns and Prospects, Albany, NY. Bennett, N. G., Bloom, D. E., & Craig, P. H. (1989). The divergence of black and white marriage patterns. American Journal of Sociology, 93, 692–722. Bernard, J. (1972). The future of marriage. New York: World Publishing. Blake, J. (1972). Coercive pronatalism and American population policy. In R. Parke & C. F. Westolf (Eds.), Aspects of population growth policy (pp. 85–109) (Vol. 6 of the Commission on Population Growth and the American Future Research Reports). Washington, DC: U.S. Government Printing Office. Bloom, D., & Trussell, J. (1984). What are the determinants of delayed childbearing and permanent childlessness in the U.S.? Demography, 21, 613–623. Booth, A., & Johnson, D. (1988). Premarital cohabitation and marital success. Journal of Family Issues, 9, 255–272. Bourgeois-Pichat, J. (1981). Demographic change in Western Europe. Population and Development Review, 7(1), 19–42. Bumpass, L. L. (1973). Is low fertility here to stay? Family Planning Perspectives, 5, 68–69. ———. (1982). Employment among mothers of young children: Changing behavior and attitudes. CDE working paper 26. Madison: University of Wisconsin, Center for Demography and Ecology. ———. (1986). Marriage and childbearing late in the demographic transition. Paper presented at the Population Studies Center 25th Reunion Symposium, Ann Arbor, MI. Bumpass, L. L., & Castro Martin, T. (1990). Couple characteristics and the perceived stability of unions. NSFH working paper 30. Madison: University of Wisconsin, Center for Demography and Ecology.

What’s Happening to the Family: Demographic and Institutional Changes

19

Bumpass, L. L., Castro Martin, T., & Sweet, J. A. (In press). Background and early marital factors in marital disruption. Journal of Family Issues. Bumpass, L. L., & Sweet, J. A. (1989a). Children’s experience in single-parent families: Implications of cohabitation and marital transitions. Family Planning Perspectives, 21(6), 256–260. ———. (1989b). National estimates of cohabitation: Cohort levels and union stability. Demography, 26, 615–625. Bumpass, L. L., Sweet, J. A., & Cherlin, A. (1989). The role of cohabitation in declining rates of marriage. NSFH working paper 5. Madison: University of Wisconsin, Center for Demography and Ecology. Bumpass, L. L., & Westoff, C. F. (1970). The “perfect contraceptive” population: The extent and implications of unwanted fertility in the U.S., 1960–65. Science, 169, 1,177–1,182. Butz, W. P., & Ward, M. P. (1979). Will U.S. fertility remain low? A new economic interpretation. Population and Development Review, 5, 663–688. Castro Martin, T., & Bumpass, L. L. (1989). Recent trends in marital disruption. Demography, 26, 37–51. Cherlin, A. (1981). Marriage, divorce, and remarriage. Cambridge, MA: Harvard University Press. Davis, K. (1967). Population policy: Will current programs succeed? Science, 158, 730–739. ———. (1972). The American family in relation to demographic change. In C. F. Westoff & R. Parke, Jr. (Eds.), Demographic and social aspects of population growth (Vol. 1 of the Commission on Population Growth and the American Future Research Reports). Washington, DC: U.S. Government Printing Office. ———. (1983). The future of marriage. American Academy of Sciences Bulletin, 36, 15–43. ———. (1984). Wives and work: Consequences of the sex role revolution. Population and Development Review, 10, 397–417. Demeny, P. (1986). Population and the invisible hand. Demography, 23, 473–487. Doubt on trial marriage raised by divorce rates. (1989, June 9). New York Times, p. A1. Duncan, O. D., & Schnore, L. (1959). Cultural, behavioral, and ecological perspectives in the study of social organization. American Journal of Sociology, 65, 132–145. Dutchin-Eglash, S. (1988). Housework and the division of household labor: 1987–88. Unpublished master’s thesis, University of Wisconsin-Madison, Dept. of Sociology. Easterlin, R. A. (1978). The economics and sociology of fertility: A synthesis. In C. Tilly (Ed.), Historical studies of changing fertility (pp. 57–134). Princeton, NJ: Princeton University Press. Easterlin, R. A., & Crimmins, E. M. (1988). Recent social trends: Changes in personal aspirations of American youth. Sociology and Social Research, 72(4), 217–223. Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 91, 310–330. Espenshade, T. (1985). Marriage trends in America: Estimates, implications and underlying causes. Population and Development Review, 11, 193–245.

20

Historical Trends in Family Life and Structure

Farley, R. (1988). After the starting line: Blacks and women in an uphill race. Demography, 25, 477–495. Freedman, D. S., & Thornton, A. (1990). The consumption aspirations of adolescents. Youth and Society, 21, 259–281. Freedman, R. (1975). The sociology of human fertility: An annotated bibliography. New York: Irvington. Furstenberg, F. F., Jr., & Nord, C. W. (1985). Parenting apart patterns of child rearing after marital disruption. Journal of Marriage and the Family, 47, 893–905. Garfinkel, I., & Mclanahan, S. (1986). Female headed families and public policy: A new American dilemma? Washington, DC: Urban Institute. Goode, W. J. (1963). World Revolution and Family Patterns. New York: Free Press. Hannan, M., Tuma, N., & Groeneveld, L. P. (1978). Income and independence effects on marital dissolution: Results from the Seattle and Denver income maintenance experiments. American Journal of Sociology, 84, 611–633. Heatherington, E. M., Camara, K., & Featherman, D. L. (1983). Intellectual functioning and achievement of children in one-parent households. In J. A. Spence (Ed.), Assessing achievement (pp. 208–284). San Francisco: Freeman. Hochschild, A. (1989). The second shift: Working parents and the revolution at home. New York: Viking Penguin. Hoyert, D. L. (1990). Intergenerational exchange of financial and household assistance, NSFH working paper 24. Madison: University of Wisconsin, Center for Demography and Ecology. Huber, J. (1980). Will U.S. fertility decline toward zero? The Sociological Quarterly, 21, 481–492. Kahn, R. L, & Antonucci, T. C. (1981). Convoys of social support a life course approach. In S. B. Kiesler, J. N. Morgan, & V. K. Oppenheimer (Eds.), Aging: Social change. New York: Academic Press. Keller, S. (1971). Does the family have a future? Journal of Comparative Family Studies, (Spring), 1–14. Lesthaeghe, R. (1983). A century of demographic and cultural change in Western Europe: An exploration of underlying dimensions. Population and Development Review, 9, 411–436. Lesthaeghe, R., & Surkyn, J. (1988). Cultural dynamics and economic theories of fertility change. Population and Development Review, 14, 1–45. Lesthaeghe, R., & Wilson, C. (1985). Modes of production, secularization, and the pace of the fertility decline in Western Europe, 1870–1930. In A. Coale & S. Watkins (Eds.), The decline of fertility in Europe (pp. 261–292). Princeton, NJ: Princeton University Press. Lorimer, F., Winston, E., & Kiser, L. K. (1940). Foundations of American population policy. New York: Harper. Mason, K. A., & Bumpass, L. L. (1975). U.S. women’s sex-role ideology, 1970. American Journal of Sociology, 80, 1212–1219. Mclanahan, S. (1985). Family structure and the reproduction of poverty. American Journal of Sociology, 90, 873–901. ———. (1989). The two faces of divorce: Women’s and children’s interests, Report DP 903–89. Madison: University of Wisconsin, Institute for Research on Poverty. McLanahan, S., & Bumpass, L. L. (1988). Intergenerational consequences of family disruption. American Journal of Sociology, 94, 130–152.

What’s Happening to the Family: Demographic and Institutional Changes

21

McLanahan, S. S., Astone, N. M., & Marks, N. F. (In press). The role of mother only families in reproducing poverty. In A. Huston (Ed.), Children in poverty. Cambridge, UK: Cambridge University Press. Menken, J. L. (1985). Age and fertility: How late can you wait? Demography, 22, 469–485. Michael, R. T., Fuchs, V. R., & Scott, S. R. (1980). Changes in the propensity to live alone: 1950–1976. Demography, 17, 39–56. Munoz-Perez, F. (1987). Le déclin de la fécondité dans Ie sud de l’Europe. Population, 6, 911–942. National Center for Health Statistics. (1990). Advance report of final marriage statistics, 1987. Monthly Vital Statistics Report, 38(12) (Suppl.). Hyattsville, MD: Public Health Service. O’Connell, M., & Moore, M. J. (1980). The legitimacy status of first births to U.S. women aged 15–24, 1939–1978. Family Planning Perspectives, 12(1), 16–25. Ogburn, W. F., & Tibbitts, C. (1933). The family and its functions. In Recent social trends, President’s Research Committee on Social Trends (ch. 13). New York: McGraw Hill. Oppenheimer, V. K. (1970). The female labor force in the United States: Demographic and economic factors governing its growth and changing composition. Westport, CT: Greenwood Press. ———. (1988). A theory of marriage timing. American Journal of Sociology, 94, 563– 591. Popenoe, D. (1988). Disturbing the nest: Family change and decline in modern societies. New York: Aldine de Gruyter. Pratt, W. F., & Horn, M. C. (1985). Wanted and unwanted childbearing: United States, 1913–82. Vital and Health Statistics, No. 108. Hyattsville, MD: U.S. Dept. of Health and Human Services, National Center for Health Statistics. Presser, H. B. (1989). Can we make time for children? The economy, work schedules, and child care. Demography, 26, 523–543. Preston, S. H. (1984). Children and the elderly: Divergent paths for America’s dependents. Demography, 21, 435–457. Preston, S. H., & McDonald, J. (1979). The incidence of divorce within cohorts of American marriages contracted since the Civil War. Demography, 16, 1–26. Rindfuss, R., & Bumpass, L. L (1978). Age and the sociology of fertility: How old is too old? In K. Taeuber, L. Bumpass, & J. Sweet (Eds.), Social demography (pp. 43–56). New York: Academic Press. Rindfuss, R. R., & Parnell, A. M. (1989). The varying connection between marital status and childbearing in the United Slates. Population and Development Review, 15, 447–470. Ryder, N. B. (1979). The future of American fertility. Social Problems, 26, 359–369. Sardon, J.-P. (1986). Evolution de la nuptialite et de la divortialite en Europe Depuis la fin des Annees 1960. Population, 3, 463–482. Schoen, R., & Owens, D. (1990). A further look at first unions and first marriages. Paper presented at the conference on Demographic Perspectives on the Family: Patterns and Prospects, Albany, NY. Schumpeter, J. A. (1988). Decomposition. Population and Development Review, 14, 499–506. (Originally published 1942, in Capitalism, socialism and democracy (ch. 14). New York: Harper & Row.)

22

Historical Trends in Family Life and Structure

Shanas, E. (1982). National Survey of the aged. Report prepared for the Administration on Aging. Washington, DC: U.S. Dept. of Health and Human Services. [DHHS Pub. No. (OHDS) 83–20425.] Shorter, E. (1973). Female emancipation, birth control and fertility in European history. American Historical Review, 78, 605-MO. Sober, E. (1984). The nature of selection: Evolutionary theory in philosophical focus. Cambridge, MA: MIT Press. Stone, L. (1982). The historical origins of the modem family. Fifth annual O. Meredith Wilson lecture in history. Salt Lake City: University of Utah, Dept. of History. Sweet, J., Bumpass, L. L., & Call, V. (1988). The design and content of the National Survey of Families and Households. NSFH working paper I. Madison: University of Wisconsin, Center for Demography and Ecology. Teitlebaum, M. S., & Winter, J. M. (1985). The fear of population decline. New York: Academic Press. Thomson, E., Mclanahan, S. S., & Curtin, R. B. (1990). Family structure and parental socialization. NSFH working paper 17. Madison: University of Wisconsin, Center for Demography and Ecology. Thornton, A. (1989a). Changing attitudes towards family issues in the United States. Journal of Marriage and the Family, 51, 873–895. ———. (l989b). Myths about the family: Reaching history sideways—sociological alchemy and other thoughts about family and history. Seminar, Center for Demography and Ecology, University of Wisconsin-Madison. Thornton, A., & Freedman, D. (1983). The changing American family. Population Bulletin, 38, 1–44. Trussell, J. (1988). Teenage pregnancy in the United Slates. Family Planning Perspectives, 20(6), 262–272. Uhlenberg, P., & Eggebeen, D. (1986). The declining well-being of American adolescents. The Public Interest, 82, 25–38. U.S. Bureau of the Census. (1988). Fertility of American women: June 1988. Current Population Reports, Ser. P-20, No. 436. Washington, DC: U.S. Government Printing Office. Waite, L., & Goldscheider, F. (1990). Work in the home: The productive context of family relationships. Paper presented at the conference on Demographic Perspectives on the Family: Patterns and Prospects, Albany, NY. Wallerstein, J. S., & Blakeslee, S. (1989). Second chances: Men, women and children a decade after divorce. New York: Tichnor and Fields. Wallerstein, J. S., & Kelly, J. B. (1980). Surviving the breakup: How children and parents cope with divorce. New York: Basic Books. Weiss, Y., & Willis, R. J. (1985). Children as collective goods in divorce settlements. Journal of Labor Economics, 3, 268–292. Westoff, C. F. (1978). Some speculations on the future of marriage and fertility. Family Planning Perspectives, 10(2), 79–83. Westoff, C. F., & Ryder, N. B. (1977). The predictive validity of reproductive intentions. Demography, 14, 431–454. White, L., & Keith, B. (1990). The effect of shift work on the quality and stability of marital relations. Journal of Marriage and the Family, 52, 453–462.

Part Two

Social Diversity and Single-Parent Families

Part One was dedicated to the historical, demographic, and economic trends that have come to shape our modern society and the growth of single-parent families. At no time in our history have so many adults and children been members of such families. Part Two is a look at the diversity of single-parent families once thought to involve just mother-child families. The diversity of the country is found in all families and includes many nationalities, ethnic and racial categories, socioeconomic categories, and an array of combinations that seems to increase with each passing decade. Among the most important aspects of the diversity issues in singleparent families is not just economics but poverty. As we also saw in Part One, economic stability is perhaps the major key to familial stability over the life course of the family and especially for children. Here, Brown and Lichter delve into the issues of poverty, welfare, and single mothers in rural areas as compared with those in the more often discussed and researched metropolitan areas. A very interesting aspect of the research is the issue of livelihood strategies used by these mothers. The multiple objectives of this research endeavor included cohabitation, coresidence with other adults, employment, and welfare. The single-father family—a growing trend in the United States— accounts for about one-fifth of single-parent families. Historically ignored due to their low incidence, father-child families have significantly increased in both their visibility and viability. Single fathers experience similar difficulties to single mothers such as balancing work and child care, as well as their social lives and interacting with the court system. However, they also experience significant role ambiguity as single fathers raising their children alone. Brown’s research covers demographic trends

24

Social Diversity and Single-Parent Families

that have increased single-father families and especially focuses on public transfer usage. As public assistance programs such as Medicaid/ Medicare, Food Stamps, and reduced educational benefits change due to political reforms, single-father families are quickly becoming more important in policy debates and reform. This is especially true as it relates to cohabiting versus noncohabiting fathers. The research and policy fields will continue to evolve regarding father-only households and the new roles and challenges may well uncover the positive and negative impacts of these families as they continue to grow in number and become more visible to the nation. Alternative lifestyles, combined with single-parenting issues, are MacCallum and Golombok’s focus in their work on the development of children raised since infancy in fatherless families, by either lesbian or single heterosexual women. They bring to the surface the growing phenomenon and acceptance of nontraditional lifestyles and parenting. This research involves a longitudinal follow-up study of children who are now 12 to 13 years of age and who were reared since infancy in three types of families: lesbian-mother families, heterosexual mother-only families, and two-parent heterosexual families. Does family structure affect children’s success? Some researchers have found that family processes—the quality of interaction—to be more important to family well-being than family structure as they compare five different family types: adoptive, two-parent biological, single-mother, stepfather, and stepmother households. Educational attainment of children is significantly altered by various family structural types and thus affects the eventual socioeconomics in intergenerational contexts. Family structure and the child’s success are not based only on family types, but single-parent families have often been the focus for delinquency potential of children. An ongoing debate among scholars, researchers, and the public continues to affect policy issues from education to criminal justice, while providing little clarity. In addition to the debate regarding delinquency and family structure, Johnson and Waldfogel provide some very interesting information on the recent trends in parental incarceration and the implications for child welfare. The increase in mother incarceration has especially affected foster care increases and grandparent caregivers. Children of incarcerated parents are often disadvantaged long before the parent becomes incarcerated, and many of these children suffer from a multitude of problems directly and indirectly associated with parental incarceration. Incarceration of parents can often lead to grandparents raising their grandchildren. Intergenerational poverty and welfare were key issues of the 1996 Personal Responsibility and Work Opportunity Reconciliation Act. Of course, extended families have existed in various forms throughout the nation’s history, but we learned in the previous research,

Social Diversity and Single-Parent Families

25

coresidency might often involve grandparents and other single mothers living together as an economic strategy. Goodman and Silverstein examine the phenomenon of grandmothers raising grandchildren. There are many reasons for grandparents, especially grandmothers, to assume parental duties either as custodial or as coparenting caregivers, and those cultural variations, especially ethnicity, are extremely important in terms of family cohesion.

This page intentionally left blank

Chapter 2

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers J. Brian Brown and Daniel T. Lichter*

Single mothers face unique challenges to lasting economic security in America’s new welfare policy environment. Welfare reform, which culminated in the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), has sought to encourage economic selfsufficiency, placed strict time limits on receipt of cash assistance, and promoted marriage and two-parent families (Blank, 2002; Lerman, 1999). The early results have been encouraging. Welfare case loads declined by more than 50 percent between 1993 and 2001, a figure that exceeded even the most optimistic forecasts. Moreover, poverty rates among single mothers, although still high, are nevertheless at record lows, while a declining share of America’s children are now residing in (often poor) single-parent families (Fields & Casper, 2001; Lichter & Jayakody, 2002; Primus, 2001). Whether recent optimism persists is far from certain as unemployment rates climbed to nearly 6 percent during the post-2000 recessionary period. It also is unclear whether declining caseloads and rising maternal employment rates nationwide since 1993 improved the life circumstances of single mothers, especially those living in nonmetropolitan areas (Lee, Harvey, & Neustrom, 2002; Parisi et al., 2004). Indeed, single mothers face the daily challenge of “making ends meet,” balancing their commitments to work and family, and navigating an increasingly punitive welfare system (e.g., Edin, 2000; McDonald & Armstrong, 2001). This is no less true

* J. B. Brown & D. T. Lichter. Poverty, Welfare, and the Livelihood Strategies of Nonmetropolitan Single Mothers. Rural Sociology, 2004, 69(2), 282–301. Copyright 2004 by Rural Sociological Society (MO). Reproduced with permission of Rural Sociological Society (MO) via Copyright Clearance Center.

28

Social Diversity and Single-Parent Families

in nonmetropolitan areas (Monroe & Tiller, 2001; Struthers & Bokemeier, 2000; Wells, 2002). Unfortunately, we know surprisingly little about the economic and family circumstances of nonmetropolitan single mothers (Snyder & McLaughlin, 2004). In this paper, we identify and evaluate the various livelihood strategies used by single mothers to make ends meet. Unlike previous studies, we identify important similarities and differences in employment, cohabitation, and coresidence with relatives and other adults among single mothers living in metropolitan and nonmetropolitan areas. We then evaluate the extent to which the “livelihood strategies” of single mothers are associated with economic well-being and welfare dependency (e.g., receipt of cash assistance and food stamps). Our study highlights the adaptive strategies used by single mothers in the 1990s and beyond. We accomplish our objectives using data on 2,123 single mothers from the 1995 National Survey of Family Growth (Cycle 5), a nationally representative data set available from the National Center for Health Statistics (Potter et al., 1997). The welfare waiver process in the early 1990s and the passage of PRWORA represented a clear policy response to longstanding concerns about nonworking single mothers; unwed childbearing among poor women; and the intergenerational transmission of poverty and welfare dependence. Indeed, maternal employment is positively associated with material well-being of single-parent families (Smith et al., 2000). In nonmetro areas, poor families are more likely than their metro counterparts to include two working parents, but are less likely to receive cash assistance. The average dollar value of cash assistance is lower in nonmetro areas (Jensen & Chitose, 1997; Lichter & Jensen, 2001).

LIVELIHOOD STRATEGIES OF SINGLE MOTHERS Unfortunately, these facts often feed traditional and increasingly outdated notions about the family life and economic self-sufficiency of single mothers living outside of metropolitan areas (Struthers & Bokemeier, 2000; Wells, 2002). Single mothers, if they find a good job, often have difficulty securing affordable daycare services. Close friends and families often lack the means to provide financial assistance in times of need. The receipt of welfare and food stamps may bring unwanted stigma in nonmetro areas (Rank & Hirschl, 1988). And, for women who wish to marry, the men available to them may lack stable jobs that pay a family wage (Albrecht et al., 2000; McLaughlin et al., 1999). Clearly, given the economic self-sufficiency and family formation goals of welfare reform, it is perhaps more important than ever to reexamine the livelihood strategies and economic well-being of nonmetro women.

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers 29

WORK AMONG SINGLE MOTHERS Work provides an important route to economic self-sufficiency. Since the mid-1990s, however, the “working poor” have replaced the “welfare poor” as the largest share of America’s poor female-headed families (Lichter & Jensen, 2001). The jobs available to single mothers and other low-skilled workers are often unstable, offer few benefits (e.g., health insurance, child care), and pay low wages. A large minority of single mothers who leave public assistance face unemployment. Many find it difficult to maintain stable employment (Sawhill, 2001). Harris (1996) found that women who returned to welfare experienced inadequate social and child-care support. The economic benefits associated with a booming economy in the 1990s were offset by reductions in government cash assistance programs (Porter & Dupree, 2001). Clearly, for many single mothers, finding a job has been an adaptive response to “work-first” welfare reform programs. Unfortunately, we only have a limited understanding of the economic benefits of employment among single mothers in nonmetro areas, and whether employment opportunities and constraints differ from their metro counterparts (Lerman et al., 2001). On a positive note, the share of family income from employment among nonmetro female-headed families increased from 33 percent to 54 percent between 1994 and 1999 (Lichter & Jensen, 2001). Yet, overall annual earnings and wage rates remain lower in nonmetro than metro areas, and nonmetro workers face higher rates of unemployment and underemployment (Conger & Elder, 1994; Slack & Jensen, 2002). The implication is that work-based solutions to poverty are arguably less likely to be a “solution” to poverty and welfare dependence in nonmetro areas, where single mothers have limited educational and job training opportunities, and where jobs that pay a living wage are less available (Jensen & Chitose, 1997; Weber, Duncan, & Whitener, 2002).

COHABITATION AMONG SINGLE MOTHERS For a growing share of young adults, cohabitation is an increasingly important stage in the transition to marriage (Bumpass, Raley, & Sweet, 1995), and it has become a common family context for both childbearing and childrearing. Roughly 40 percent of nonmarital births in the United States are born to cohabiting couples (Bumpass & Lu, 2000). Cohabitation is especially common among low-income mothers. Moving in with a boyfriend may be an adaptation to economic hardship. Indeed, Acs and Nelson (2002) found that children living with a single mother and her cohabiting partner are less likely to be poor than

30

Social Diversity and Single-Parent Families

are children living with only a single mother. Cohabitation can offer financial and nonfinancial benefits for a single mother. Clearly, one financial incentive for a single mother to cohabit is the sharing of housing and utility costs. Cohabiting partners may also share child care responsibilities and augment women’s networks of family and friends and, indirectly, their access to financial assistance or social support. Cohabiting single mothers also may be better able than other single mothers to pursue employment, work more hours, or obtain training necessary for securing a good job. We know surprisingly little about patterns of cohabitation in nonmetro areas (Snyder, Brown, & Condo, 2004). However, there are good reasons to believe that cohabitation, as a livelihood strategy, may be less prevalent among nonmetro single mothers. Traditional family values, gender roles, and other normative constraints may make unmarried cohabitation a less acceptable option for single mothers. Single mothers in nonmetro areas may also be less able than metro single mothers to partner with “economically attractive” men who provide economic support and who subsidize single mothers’ job search activities (McLaughlin, Gardner, & Lichter, 1999). Although cohabitation may be positively associated with economic well-being among single mothers, this association may be weaker in nonmetro areas.

DOUBLING-UP AMONG SINGLE MOTHERS As an adaptation to economic hardship, single mothers may move in with parents, extended kin or other family, or other nonrelatives; some may never have left the parental home. Previous studies have clearly shown that welfare benefit levels are associated with the independent living arrangements of single mothers; single mothers are more likely to form separate households in high—rather than in low—welfare benefit states (Bane & Ellwood, 1986; Moffitt, Reville, & Winkler, 1998). Welfare reform in the 1990s has made cash assistance no longer available to single mothers under age 18 who do not reside with other adults. This restriction may anchor teen unwed mothers to the parental home, but may also have created new economic hardships that cause some teen mothers to give up their children rather than rear their children alone without government cash assistance (Brandon, 2000). Indeed, the number of children living with at least one grandparent (with or without a biological parent) increased over the past decade (Bryson & Casper, 1999). Is “doubling-up” among single mothers more or less common in nonmetro areas, and is it a livelihood strategy that promotes economic well-being? Lichter and Jensen (2001) found that roughly 20 percent of

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers 31

nonmetro single mothers now reside with other kin or in multifamily households. This figure represents a significant increase from the early 1990s. As in the case of cohabitation, nonmetro single mothers may find that coresidence with other family members provides financial and child care assistance.

GOVERNMENT ASSISTANCE As an economic livelihood strategy, the receipt of government cash assistance represents an alternative to work and economic support from family and friends. Although nonmetro families (including femaleheaded families with children) have higher rates of poverty than metro families, they are less likely to receive cash assistance (Lichter & Jensen, 2001). Jensen and Eggebeen (1994) show that single mothers in nonmetro areas are less likely to receive public assistance than single mothers in metro areas and that the dollar value of assistance is often too low to lift them and their children out of poverty. Poor people in nonmetro areas often face greater stigma, misinformation about welfare eligibility, and significant obstacles to receipt, such as limited transportation (Rank & Hirschl, 1988; Weber et al., 2002). The low cash assistance levels in states with substantial nonmetro populations (e.g., Mississippi) are often insufficient to meet even the most basic needs for food, shelter, and clothing (Parisi et al., 2004). Lack of jobs, unreliable transportation, and inadequate child care place nonmetro single mothers at greater risk of being sanctioned off welfare rolls for failing to comply with mandated work requirements (Swenson, White, & Murdock, 2001). The goal was to eliminate any real or perceived incentive to “use” out-of-wedlock childbearing as a mechanism or pathway to leaving the parental home. Whether the receipt of welfare is an adaptive strategy or a manifestation of economic hardship and single parenthood is unclear in the new welfare policy environment. Therefore, in the latter half of our analysis we examine how the livelihood strategies identified above are associated with government assistance among similarly (dis)advantaged metro and nonmetro single mothers. In this paper, our main objective is to identify nonmetro-metro differences in the family and economic livelihood strategies of single mothers and evaluate their relationship to economic well-being. The substantive and policy implications are significant: nonmetro single mothers may be triply disadvantaged. That is, nonmetro single mothers may face higher rates of poverty and greater barriers to welfare receipt; and the various livelihood strategies (identified here) may provide little assurance of escaping poverty and welfare dependence.

32

Social Diversity and Single-Parent Families

DATA AND METHODS We use data from the National Survey of Family Growth (NSFG), a national probability sample of women aged 15 to 45 designed to produce national estimates on pregnancy and reproductive and child health (Potter et al., 1997). Cycle 5 of the NSFG provides detailed retrospective life history information, including family background, marital and nonmarital relationship histories, and fertility experiences for 10,847 women in 1995 (Abma et al., 1997). We restrict our sample to mothers who are single (or separated) at the time of the interview. Although mothers are often involved with nonresident children’s lives, and childless women often raise nonbiological children, our research is restricted to mothers most germane to current welfare policy debates: single women raising their own children under age 18. Our analytic sample consists of 2,123 single mothers aged 18 to 45. Sampling weights are used in the reporting of descriptive statistics to take into account the NSFG sampling design and response bias, and to approximate nationally representative estimates. Our economic outcome of interest in the first part of our analysis is the family income-to-poverty ratio (multiplied by 100), which is defined as the ratio of income of the mother to her official poverty line, as designated by the Office of Management and Budget. The income-to-poverty ratio thus is a family-size adjusted measure of family income (using the income equivalence levels implied by the poverty lines for families of different sizes). Income-to-poverty ratios below 100 indicate family income below the official poverty line. The mean income-to-poverty ratio for single mothers in our sample is 175 (SD 1⁄4 158). Thirty-seven percent are officially poor. Nonmetro women reside outside of Metropolitan Statistical Areas (MSAs), which include core counties containing a central city and the adjacent counties that are functionally integrated with the core. Nonmetro single mothers comprise 18 percent of our sample. We also specify whether nonmetro and metro residents live in the South. We identify four family livelihood strategies. We begin with maternal employment. About 43 percent of all single mothers are employed fulltime, defined here as working 35 hours or more per week. Others are either working part-time (18 percent), not looking for work, or temporarily not working (40 percent). Next, 22 percent of these single mothers are cohabiting with a male partner. One-third of the single mothers in this sample are coresiding or doubling-up with at least one other adult (not counting the cohabiting partner). Finally, the receipt of food stamps and cash assistance can be a livelihood strategy for single mothers, especially for those whose parenting and/or child-rearing activities limit the amount of time spent in paid employment, school, or vocational training. Overall, 35 percent of the sample received cash assistance (AFDC, Aid to

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers 33

Families with Dependent Children) and 45 percent received food stamps. Not surprisingly, single mothers rely heavily on government assistance. We also control for maternal age, distinguishing between single mothers aged 18 to 29 (41.5 percent) and those aged 30 to 45 (58.5 percent). Our measure of marital history indicates that one-third of our sample has never been married, and that the majority are divorced or separated. Children under age 18 represent financial and social demands related to child care that make stable and high-paying employment difficult. Onehalf of these single mothers are raising one child, 30 percent are raising two children, and the remaining 20 percent are raising three or more children. Other children increase financial demands on the household and may affect decisions to work or seek welfare. About 14 percent of the single mothers reside with nonbiological minor children. Child support provides 36 percent of these single mothers with additional income. Education is associated with better paying and stable employment opportunities, reducing the likelihood of welfare dependency and poverty-level wages. Thirty-three percent have not completed high school, 59 percent have completed high school and possibly some higher education, and the remaining 9 percent have college degrees. Regarding race, about one-half of the single mothers are white, 32 percent are black, and other racial/ethnic groups make up the remainder of the sample. To measure socioeconomic background, we examined whether the parents of single mothers dropped out of high school. Slightly less than one-half of single mothers’ parents did not complete high school.

DISCUSSION AND CONCLUSION Welfare caseloads declined rapidly during the 1990s (Lee et al., 2002; Lichter & Jayakody, 2002). Reductions in welfare dependency, economic self-sufficiency through work, and the promotion of two-parent families continue to be major goals of the welfare reform agenda. Whether welfare reform has hurt or helped nonmetro single mothers is unclear (Weber et al., 2002). Indeed, much of the welfare debate and evaluation has centered on inner-city populations and the metro underclass. The economic circumstances of nonmetro single mothers are typically ignored or forgotten in public policy debates (Lichter & Jensen, 2001). Our main objective here has been to inform the welfare debate by examining the various livelihood strategies of single mothers in nonmetro America. How do nonmetro single mothers make ends meet? Are the livelihood strategies available to them—work, cohabitation, and doubling-up—less likely to lift them out of poverty? Our results show that nonmetro mothers have lower family income and higher rates of poverty than their counterparts in metro areas

34

Social Diversity and Single-Parent Families

(cf. Lichter & Jensen, 2001; Weber et al. 2002). This finding is not new, but is being reconfirmed in recent research (see Snyder & McLaughlin, 2004). What is new is that—despite considerably different social circumstances and labor market opportunities—current nonmetro single mothers seem to adopt livelihood strategies that are remarkably similar to metro single mothers. Only in the case of welfare receipt are nonmetro single mothers different from their metro counterparts. Still, slightly more than one-half of nonmetro single mothers receive cash assistance from the government. Our results suggest that nonmetro single mothers have not been immune to economic and family changes over the last several decades, and their strategies to make ends meet are similar to their metro counterparts. The livelihood strategies identified here, however, do not necessarily translate evenly into reductions in poverty or increases in family income for metro and nonmetro single mothers. Indeed, the income-to-poverty ratios of employed nonmetro single mothers were lower than their counterparts in metro areas. This has important implications for “work first” welfare programs that aim to move poor single mothers into the labor force. Work-based welfare reform may be less likely to achieve its goals (e.g., reductions in welfare dependence or poverty) in nonmetro areas. Moreover, although cohabitation and doubling-up may improve family income, it is also true that these livelihood strategies are less strongly associated with improvements in family income in nonmetro areas than in metro areas. Much of the difference seems to be due to the social and demographic risk factors that place cohabiting mothers at greater risk of poverty. Moreover, as an adaptive strategy to economic hardship, nonmetro poor single mothers also are less likely than their metro counterparts to receive food stamps and cash assistance. Welfare is less likely to provide an economic safety net for nonmetro single mothers. This is one reason our examination of livelihood strategies among rural women is especially important from a policy standpoint (see Wells, 2002). Our study of nonmetro single mothers provides a baseline for future research. It addresses the current dearth of research on nonmetro single mothers and the livelihood strategies they use to improve the economic well-being of their families. The evidence reported here concerning women’s economic well-being should not distract us from identifying and evaluating other important dimensions of women’s well-being, such as mental and physical health, self-esteem and self-efficacy, the quality of parent-child relations, and healthy intimate partnering. Moreover, our cross-sectional analyses, while providing useful information on the current livelihood strategies of single mothers, are less suited for making strong causal statements about women’s economic and social adaptations to economic hardship. Lacking panel data, we do not know directly how single mothers respond behaviorally to economic hardship—whether

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers 35

they benefit from becoming employed, entering cohabiting unions, or doubling-up with other adults. Additional research is needed to better understand how changes in livelihood strategies help promote reductions or changes in poverty (or not), and how adaptive strategies affect welfare receipt. Our study thus provides a starting point rather than the final answer. Finally, though state welfare reform has been ongoing since the early 1990s, it also is unclear whether the post-PRWORA period of the late 1990s has ushered in a new or different set of livelihood strategies. Much of the debate over reauthorization of the 1996 welfare reform bill centered on marriage promotion (Lichter, Graefe, & Brown, 2003), yet it is unclear whether welfare reform has created new incentives that have changed the family formation behaviors of women, including women living in nonmetro areas. Job training and work supports (e.g., child care subsidies), as well as demand-side solutions (e.g., nonmetro economic development), will continue to play a large role in addressing the persistent problems of low income nonmetro mothers. Indeed, our study shows that employment opportunities or, more precisely, the lack of employment opportunities, may play an unusually large role in securing (or not) the long-term economic well-being of nonmetro single mothers and their children. REFERENCES Abma, J. C., Chandra, A., Mosher, W. D., Peterson, L. S., & Piccinino, L. J. (1997). Fertility, family planning, and women’s health: New data from the 1995 National Survey of Family Growth. Vital and Health Statistics, 23. National Center for Health Statistics, Hyattsville, MD. Acs, G., & Nelson, S. (2002). What if welfare reform led to a rise in cohabitation? Presented at the Annual Meetings of the Association of Public Policy Analysis and Management, Dallas, TX, Nov. 7–9, 2002. Albrecht, D. E., Albrecht, C. M., & Albrecht, S. L. (2000). Poverty in nonmetropolitan America: Impacts of industrial, employment, and family structure variables. Rural Sociology, 65, 87–103. Bane, M. J., & Ellwood, D. T. (1986). Slipping into and out of poverty: The dynamics of spells. Journal of Human Resources, 21, 1–23. Blank, R. M. 2002. Evaluating welfare reform in the United States. NBER working papers 8983. National Bureau of Economic Research, Inc., Cambridge, MA. Blank, R. M. (2002). Evaluating welfare reform in the United States. NBER working papers 8983. National Bureau of Economic Research, Inc., Cambridge, MA. Brandon, P. D. (2000). Did the AFDC program succeed in keeping mothers and young children living together? Social Service Review, 74, 214–230. Bryson, K. R., & Casper, L. M. (1999). Coresident grandparents and grandchildren. P23–198. U.S. Bureau of the Census. Washington, DC: Current Population Reports, Special Studies.

36

Social Diversity and Single-Parent Families

Bumpass, L. L., & Lu, H. (2000). Trends in cohabitation and implications for children’s family contexts in the United States. Population Studies, 54, 29–41. Bumpass, L. L., Raley, K. R., & Sweet, J. A. (1995). The changing character of stepfamilies: Implications of cohabitation and nonmarital childbearing. Demography, 32, 425–436. Conger, R. D., & Elder, Jr., G. H. (1994). Families in troubled times: Adapting to change in rural America. New York: Aldine de Gruyter. Cotter, D. A. (2002). Poor people in poor places: Local opportunity structures and household poverty. Rural Sociology, 67, 534–555. Edin, K. (2000). What do low income single mothers say about marriage? Social Problems, 47, 112–133. Fields, J., & Casper, L. M. (2001). America’s families and living arrangements: March 2000. P20–537, U.S. Census Bureau. Washington, DC: Current Population Reports. Harris, K. M. (1996). Life after welfare: Women, work, and repeat dependency. American Sociological Review, 61, 407–426. Jensen, L., & Chitose, Y. (1997). Will workfare work? Job availability for welfare recipients in rural and urban America. Population Research and Policy Review, 16, 383–395. Jensen, L., & Eggebeen, D. J. (1994). Nonmetropolitan poor children and reliance on public assistance. Rural Sociology, 59, 45–65. Lee, M. A., Harvey, M., & Neustrom, A. (2002). Local labor markets and caseload decline in Louisiana in the 1990s. Rural Sociology, 67, 556–577. Lerman, R. I. (1999). Retreat or reform? New U.S. strategies for dealing with poverty. Social Policy Review, 11. Lerman, R. I., McKernan, S., & Pindus, N. (2001). Welfare reforms and employment of single mothers: Are rural areas keeping pace? Rural America, 16(3), 22–28. Lichter, D. T., Graefe, D. R., & Brown, J. B. (2003). Is marriage a panacea? Union formation among economically disadvantaged unwed mothers. Social Problems, 50, 60–86. Lichter, D. T., & Jayakody, R. (2002). Welfare reform: How do we measure success? Annual Review of Sociology, 28, 117–141. Lichter, D. T., & Jensen, L. (2001). Poverty and welfare among rural female-headed families before and after PRWORA. Rural America, 16, 28–35. McDonald, K. B., & Armstrong, E. M. (2001). De-romanticizing black intergenerational support: The questionable expectations of welfare reform. Journal of Marriage and Family, 63, 213–223. McLaughlin, D. K., Gardner, E. L., & Lichter, D. T. (1999). Economic restructuring and changing prevalence of female-headed families in America. Rural Sociology, 64, 394–416. Moffitt, R. A., Reville, R., & Winkler, A. E. (1998). Beyond single mothers: Cohabitation and marriage in the AFDC program. Demography, 35, 259–278. Monroe, P. A., & Tiller, V. V. (2001). Commitment to work among welfare-reliant women. Journal of Marriage and Family, 63, 816–828. Parisi, D., McLaughlin, D. K., Grice, S. M., Taquino, M., & Gill, D.A. (2004). TANF participation rates: Do community conditions matter? Rural Sociology, forthcoming.

Poverty, Welfare, and Livelihood Strategies of Nonmetropolitan Single Mothers 37 Porter, K. H., & Dupree, A. (2001). Poverty trends for families headed by single working mothers: 1993–1999. Pub No. 01–144. Center on Budget and Policy Priorities. Washington, DC. Potter, F. J., Iannachione, V. G., Mosher, W. D., Mason, R. E., Kavee, J. D., & Botman, S. L. (1997). National Survey of Family Growth Cycle 5: Design, Estimation, and Inference. Vital and Health Statistics 2. National Center for Health Statistics, Hyattsville, MD. Primus, W. (2001). What next for welfare reform? The Brookings Review, 19(3), 17–19. Rank, M. R., & Hirschl, T. A. (1988). A rural-urban comparison of welfare exits: The importance of population density. Rural Sociology, 53, 190–206. Sawhill, I. (2001). From welfare to work. Brookings Review, 19(3), 4–8. Slack T., & Jensen, L. (2002). Race, ethnicity, and underemployment in nonmetropolitan America: A 30-year profile. Rural Sociology, 67, 208–233. Smith, J. R., Brooks-Gunn, J., Klebanov, P. K., & Lee, K. (2000). Welfare and work: Complementary strategies for low-income women? Journal of Marriage and Family, 62, 808–821. Snyder, A. R., Brown, S. L., & Condo, E. P. (2004). Residential differences in family formation: The significance of cohabitation. Rural Sociology, 2004. Snyder, A. R., & McLaughlin, D. K. (2004). Female-headed families and poverty in rural America. Rural Sociology, 69, 127–149. Struthers, C. B., & Bokemeier, J. L. (2000). Myths and realities of raising children and creating family life in a rural county. Journal of Family Issues, 21, 17–46. Swenson, T., White, S., & Murdock, S. (2001). Time limit and sanction effects on metropolitan and nonmetropolitan welfare caseloads. Unpublished manuscript presented at the Annual Meetings of the Rural Sociological Society, Albuquerque, NM, August 15–19. Weber, B., Duncan, G. J., & Whitener, L. A. (2002). Rural dimensions of welfare reform. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. Wells, B. (2002). Women’s voices: Explaining poverty and plenty in a rural community. Rural Sociology, 67, 234–54.

Chapter 3

The Single-Father Family: Demographic, Economic, and Public Transfer Use Characteristics Brett V. Brown*

For the past several decades, there has been a substantial and continuous increase in families headed by single fathers, both absolutely and as a percentage of all families with children (Bianchi, 1995; Casper & Bryson, 1998; Eggebeen, Snyder, & Manning, 1996; Garasky & Meyer, 1996). The Census Bureau estimates that there were over 2.1 million such families in the United States in 1998, an increase of over 50 percent since 1990. They now represent 18 percent of all single-parent families with children under age 18 (Casper & Bryson, 1998). Until very recently, little was known about single-father families beyond the most basic demographic information offered in government reports. Their rapid growth, however, has prompted scientists and policymakers to begin to ask questions about the social forces driving the trend. How do such families differ from two-parent and single-mother families in their social dynamics, socioeconomic circumstances, and need for social supports? What are the consequences for the children who grow up in such families? The continued and rapid growth in the 1990s of both cohabiting and noncohabiting single-father families suggests that these family forms may well become increasingly common in the coming decade. This is the time to begin considering more systematically the relationship of

* B. V. Brown. The Single-Father Family: Demographic, Economic, and Public Transfer Use Characteristics. Marriage & Family Review, 2000, 29(2), 203–220. Reprinted with permission from Taylor & Francis.

The Single-Father Family

39

social policy to single-father families, and to identify the considerable research needs and the data needed to fulfill them. Basic descriptive information on the demographic and socioeconomic circumstances of single-father families is provided in this article, with particular attention given to the receipt of public transfers. Such information is much needed in an era of dramatic changes in the nation’s social safety net—changes that have taken place in ignorance of the likely consequences for single-father families. Time trend data from 1984 to 1996 are also examined to document how the characteristics of these families have changed during this period of rapid growth. First and foremost, it is clear that a substantial proportion of both noncohabiting and cohabiting single-father families depends on public transfers to enhance their well-being. Receipt of the Earned Income Tax Credit (EITC) is the most widespread, affecting 46 percent of cohabiting and 64 percent of noncohabiting single-father families in 1996. One in five noncohabiting single fathers rely on public health insurance to cover their youngest child, and one in seven rely on food stamps to help feed their families. The percentages are even larger for cohabiting single fathers. Rates of public assistance receipt are relatively modest among noncohabiting single-father families at 7 percent, but play a larger role in cohabiting father households in which 15 percent receive public assistance, and the average annual amount received exceeds $3,700. Most of the public assistance does not come directly to the father in such families, however, but through the father’s partner. The public transfer programs on which noncohabiting and cohabiting single-father families are most reliant (the EITC, Medicaid/Medicare, Food Stamps, and free or reduced-price school lunches) have all experienced significant expansions in recent years (Committee on Ways and Means, 1998), and are still substantially controlled at the federal level. There was talk during the 1996 Congress about reducing most of these programs and giving the states more control over them, though by and large that has not happened. The welfare reform legislation of 1996 had the largest impact on the Aid to Families with Dependent Children (AFDC) program. Thus we may infer that welfare reform will have little impact on the financial well-being of noncohabiting single-father families (who are less likely to receive this kind of public assistance) and a modest impact on cohabiting single-father families (whose partners are moderately likely to receive this kind of public assistance). The socioeconomic characteristics of these father-headed families confirm previous findings that single fathers have less education and considerably less earnings and income than their married counterparts (Eggebeen et al., 1996; Meyer & Garasky, 1993). In addition, the socioeconomic gap between single and married fathers has been getting larger

40

Social Diversity and Single-Parent Families

since 1984; married fathers have had substantial gains in educational outcomes. The living arrangements and employment patterns of single fathers are suggestive of strategies that single fathers may be adopting regarding income maximization, parenting, and child care arrangements. Nearly one in five noncohabiting single fathers live with an adult relative compared with one in twenty married fathers. A parent of the father is present in about two-thirds of those cases. Of particular interest are the one in nine noncohabiting single fathers in 1997 who had not worked at all during the previous year. It may be that many of these fathers have opted to become full-time caretakers for their children. Fathers who are awarded custody of their children may in part be awarded custody because of a willingness or desire to perform such a role. Alternatively, it may be that single fathers who do not have good employment prospects have opted for the caretaker role, perhaps within the context of an extended household. Research on patterns of child care and caretaking within such families would yield important insights into the social dynamics of such extended family situations. The analyses of the socioeconomic characteristics of father-headed households suggests that single fathers have less education and considerably less earnings and income than their married counterparts. In addition, the socioeconomic gap between single and married fathers has been getting larger since 1984; married fathers have had substantial gains in educational attainment and household income, while income and education levels for both types of single-father families have stagnated. The children growing up in single-father families are finding themselves increasingly financially disadvantaged relative to children in two-parent families, though their resources remain high relative to single-mother families (U.S. Department of Health and Human Services, 1998). One important area of research that has received very little attention to date is the nature of the parent-child relationship within single-father families as well as the larger social dynamics that exist in such households. Research has shown that mothers and fathers do act differently in fulfilling their parenting roles in two-parent families and as noncustodial parents (see Lamb, 1997; Marsiglio, 1995). Very little research exists, however, on the nature of parenting in single-father families. It is likely that many such fathers are taking on a wider range of parenting activities than they would undertake if a spouse were present. As single-father families become more common, their example might feed back into our more general notions of what it is to be a father, affecting the parenting practices of married fathers as well.

The Single-Father Family

41

REFERENCES Bianchi, S. M. (1995). The changing demographic and socioeconomic characteristics of single parent families. Marriage and Family Review, 20, 71–97. Bumpass, L. L., & Raley, R. K. (1995). Redefining single-parent families: Cohabitation and changing family reality. Demography, 32, 425–36. Casper, L., & Bryson, K. (1998). Household and family characteristics: March 1998. Current Population Reports, P20–515 and P20–515u. Washington, DC: U.S. Census Bureau. Committee on Ways and Means. (1998). 1998 green book. Washington, DC: U.S. Federal Interagency Forum on Child and Family Statistics. (1998). ———.Nurturing fatherhood: Improving data and research on male fertility, family formation and fatherhood. Washington, DC: Government Printing Office. Eggebeen, D. J., Snyder, A. R., & Manning, W. D. (1996). Children in single-father families in demographic perspective. Journal of Family Issues, 17, 441–465. Garasky, S., & Meyer, D. R. (1996). Reconsidering the increase in father-only families. Demography, 33, 385–393. Hernandez, D. J. (1993). America’s children: Resources from families, the government, and the economy. New York: Russell Sage. Lamb, M. E. (Ed.) (1997). The role of the father in child development, 3rd ed. New York: John Wiley and Sons, Inc. Marsiglio, W. (Ed.) (1995). Fatherhood: Contemporary theory, research, and social policy. Thousand Oaks, CA: Sage Publications. Meyer, D. R., & Garasky, S. (1993). Custodial fathers: Myths, realities, and child support policy. Journal of Marriage and the Family, 55, 73–89. U.S. Department of Health and Human Services. (1998). Trends in the well-being of American children and youth: 1998. Washington, DC: U.S. Department of Health and Human Services/Assistant Secretary for Planning and Evaluation.

Chapter 4

Children Raised in Fatherless Families from Infancy: A FollowUp of Children of Lesbian and Single Heterosexual Mothers at Early Adolescence Fiona MacCallum and Susan Golombok*

It is estimated that in the United Kingdom today, 22 percent of children are being raised in a family without a father present (Office of National Statistics, 2002). Early research on the psychological well-being of children from fatherless families concentrated on situations in which the father had left following separation or divorce, exposing the child to discordant parental relationships as well as to the loss of a once-present parent. These studies found children of single mothers to be at increased risk for cognitive, social, and emotional problems (for reviews, see Biller, 1974; Herzog & Sudia, 1973). However, single parenthood following separation or divorce may involve consequences other than just father absence, specifically those of financial hardship, low socioeconomic status, and lack of social support. Indeed, in an analysis of four representative samples of U.S. single mother families, it was found that the single most important factor contributing to the difficulties experienced by children of lone parents in later life was the lower family income associated with single parenthood (McLanahan & Sandefur, 1994). Studies that controlled for these social and economic factors demonstrated that father absence in

* F. MacCallum & S. Golombok (2004). Children Raised in Fatherless Families from Infancy: A Follow-Up of Children of Lesbian and Single Heterosexual Mothers at Early Adolescence. Journal of Child Psychology and Psychiatry, 45(8), 1407–1419. Copyright 2004, Wiley. Reproduced with permission of Blackwell Publishing Ltd.

Children Raised in Fatherless Families from Infancy

43

itself does not negatively impact the child’s intellectual ability or socioemotional adjustment (Broman, Nichols, & Kennedy, 1975; Crockett, Eggebeen, & Hawkins, 1993; Ferri, 1976). The other confounding issue in the early studies was that the children had been through the experience of parental separation. Children whose parents have divorced or separated show poorer psychological adjustment than children whose fathers have died, in terms of incidence of behavioral problems (Ferri, 1976; Rutter, 1971) and coping with the transition to adulthood (McLanahan & Sandefur, 1994). The most influential factor seems to be the exposure to parental conflict, which was found in a review by Amato (1993) to be the most significant predictor of emotional distress in the children of divorced parents. Divorce also creates adjustment difficulties for mothers, who may have raised levels of depression and anxiety (Hetherington & Stanley-Hagan, 2002). In a recent study in the United Kingdom, the high rate of depression found among single mothers was found to be associated with psychological disorder in children (Dunn et al., 1998). Thus, the negative outcomes seen in some studies of father-absent families cannot necessarily be generalized to those children who are reared by their mother without a male partner from birth or shortly thereafter, and have no experience of marital disruption or family realignment. It is possible, however, that other pressures on these mothers such as social stigma and lack of social support may interfere with their parenting role and leave their children vulnerable to emotional and behavioral problems. From a general population study in the United Kingdom in the 1970s, a small group of families was identified in which the father had never been resident, and it appeared that the children were not showing any adverse effects due to their family situation (Ferri, 1976). Conversely, a series of reports by Weinraub and colleagues, who studied a sample of families in the United States, found that children of these solo mothers had more behavioral problems, lower social competence, and poorer school performance than children from two-parent families (Weinraub, Horvath, & Gringlas, 2002). However, these negative outcomes were found to be associated with the low maternal social support and heightened maternal stress experienced by some of the solo mothers, rather than directly related to single parenthood. It is important to remember that single mothers will vary with respect to their age, social and economic status, and educational level, and it is the complex combination of these and other factors that may affect the mother’s ability to parent effectively. An alternative way in which fatherless families can be formed is when lesbian women become mothers, either singly or with a female partner. These families are similar to single heterosexual mother families in that there is no father present, but differ with respect to the sexual orientation of the mother, and possibly in the presence of a female coparent. Research

44

Social Diversity and Single-Parent Families

on lesbian mothers began in the 1970s and initially focused on women who had become mothers within a heterosexual relationship, and had then separated from the father and come out as lesbian (for reviews, see Golombok, 1999; Patterson, 2002). At that time, women who identified as lesbian often lost custody of their children because of concerns about the impact of being raised by a lesbian mother on the psychological development of children. It was argued that due to the stigma attached to homosexuality, children would be teased and ostracized by their peers, leading to social isolation and psychological dysfunction. There was also concern that the lack of a father figure and the presence of one or two mothers who were not following conventional sex-typed behavior could result in children showing atypical gender development, i.e., boys being less masculine and girls less feminine. In addition, it was suggested that the children of lesbian mothers were themselves more likely to grow up to be lesbian or gay, which was considered an undesirable outcome by courts of law. The findings from the early investigations of lesbian mother families were strikingly consistent. With regard to the children themselves, there was no evidence of raised levels of emotional or behavioral problems, or of difficulties relating to peer adjustment, self-esteem, or gender development (Golombok, Spencer, & Rutter, 1983; Green, Mandel, Hotvedt, Gray, & Smith, 1986; Hoeffer, 1981; Huggins, 1989; Kirkpatrick, Smith, & Roy, 1981), and this remained the case in a recent community study of lesbian mother families (Golombok et al., 2003). When a group of children of lesbian mothers were followed up into early adulthood, they were found no more likely to identify as lesbian or homosexual than their counterparts from heterosexual homes (Golombok & Tasker, 1996). Moreover, these young men and women continued to function well and to maintain good relations with both their mothers and their mothers’ partners. In terms of quality of parenting, lesbian mothers were found to be just as warm and responsive (Golombok et al., 1983), just as nurturing (Mucklow & Phelan, 1979), and just as child-oriented (Kirkpatrick, 1987) as comparison groups of single heterosexual mothers. In the early lesbian mother research, as with the first studies of single mothers, the children had lived with their father during the early years of their life, and had experienced the breakdown of their parents’ relationship. More recently, a growing number of lesbian women have turned to assisted conception, in particular donor insemination, to have a child without the involvement of a man from the outset. In line with the previous investigations, comparisons between these families and two-parent heterosexual families have found lesbian mothers to show a high quality of parenting and positive relationships with their children (Brewaeys, Ponjaert, Van Hall, & Golombok, 1997; Chan, Raboy, & Patterson, 1998; Flaks, Ficher, Masterpasqua, & Joseph, 1995). In addition, no differences

Children Raised in Fatherless Families from Infancy

45

were identified between children from lesbian families and those from heterosexual families in terms of psychological well-being or gender development. A study from Belgium also found that children from lesbian families were no more likely to experience teasing by peers, although they were more prone to family-related teasing incidents (Vanfraussen, Ponjaert-Kristoffersen, & Brewaeys, 2002). One of the initial studies of children raised in families without the presence of a father since their first year of life was carried out in the United Kingdom (Golombok, Tasker, & Murray, 1997). Although the sample comprised volunteers, the families were recruited according to strict criteria, and are of interest because the children were raised from the outset in a father-absent family. Thirty lesbian mother families and 42 families headed by a single heterosexual mother were compared with 41 twoparent heterosexual families when the children were six years old. It was found that mothers in fatherless families showed greater warmth toward their children and interacted more with them than did mothers in fatherpresent families. In addition, disputes between mothers and children were found to be more serious, but no more frequent, in fatherless families. Although the children in fatherless families did not show raised levels of emotional and behavioral problems, they perceived themselves as less cognitively and physically competent. There were no differences identified between families headed by lesbian mothers and those headed by single heterosexual mothers, suggesting that the sexual orientation of the mother did not have any effect. This article reports on a longitudinal study of the fatherless families investigated by Golombok et al. (1997) as the children reached age 12, focusing on parent-child relationships and the children’s socioemotional development. It is the first study to have followed up to adolescence children raised in fatherless families throughout the preschool years. In the first phase of the study the children may have been too young for any negative impact of being raised without a father to be seen, since it is conceivable that the effect of father-absence in early infancy may not become apparent until the adolescent years. Adolescence is a time when issues of identity formation assume great importance and parent-child conflict becomes more frequent (Coleman & Hendry, 1999). Thus the absence of a father may become more salient at this stage for both mothers and children. Aspects of parenting considered particularly important for the psychological adjustment of the adolescent child include parental warmth in combination with appropriate levels of control, and the fostering of autonomy (Baumrind, 1991; Collins, 1990; Lamborn, Mounts, Steinberg, & Dornbusch, 1991; Steinberg, 1990, 2000). Studies of the impact of fatherabsence on parenting have shown that, on average, children in singleparent homes experience a poorer quality of parenting than children who live with both their mother and their father. In the study by Dunn et al.

46

Social Diversity and Single-Parent Families

(1998), greater maternal negativity toward the child was shown by single mothers than by mothers in two-parent heterosexual families, and found to be associated with a higher rate of behavioral problems in children. Similarly, McLanahan & Sandefur (1994) reported that single mothers exert less control over their children in terms of supervision and establishing rules than do mothers in two-parent heterosexual families. The poorer quality of parenting shown by single mothers may be explained, in part at least, by the higher rates of psychological problems, particularly depression, found among single mothers. Depression is thought to interfere with parents’ emotional availability and sensitivity to their children and also with their control and discipline of them (Cummings & Davies, 1994). A number of studies have shown that depressed parents tend to be either very lenient with their children or very authoritarian, often switching between the two (Kochanska, Kuczynski, Radke-Yarrow, & Welsh, 1987). Insofar as children in fatherless families experience good quality parenting from their mothers, they would not necessarily be expected to show negative consequences arising from their family structure. Thus, it was predicted that problems would arise for these adolescents to the extent that the absence of a father interfered with the quality of the mothers’ relationship with their child. With respect to gender development, different theoretical perspectives lead to different expectations about the consequences of father absence for children’s sex-typed behavior. Although theories that stress the importance of prenatal (Collaer & Hines, 1995), cognitive (Martin, 1991; Martin & Halverson, 1981) or peer group (Maccoby, 1998) processes in the acquisition of gender role behavior would not predict differences between children in father-absent and father-present homes, theories that view parents as influential (Bandura, 1986; Bussey & Bandura, 1999; Mischel, 1970) would predict less feminine behavior in daughters and less masculine behavior in sons. METHODS Participants At the time of the initial study, families were asked for permission to be contacted again for follow-up, and all agreed to this (for details of recruitment of participants to the first study, see Golombok et al., 1997). Families were then approached either by telephone or letter as close as possible to the child’s 12th birthday. Twenty-five lesbian mother families, 38 single heterosexual mother families, and 38 two-parent heterosexual families agreed to participate in the follow-up study, giving response rates of 83 percent, 90 percent, and 93 percent respectively. Two lesbian mother families, two single mothers, and one two-parent heterosexual family

Children Raised in Fatherless Families from Infancy

47

refused (without giving their reasons), and seven families had moved house and could not be traced. Therefore, excluding those families who were untraceable, the responses rates were 93 percent, 95 percent, and 97 percent, respectively. Examining the data from the first phase, the nonparticipating families did not differ from the participating families at that stage on the measures of quality of parenting and parent-child interaction. Therefore, there is no evidence that those families not taking part are experiencing more problems in family functioning. In the original study, 11 of the lesbian mothers were single parents, while 14 lived with partners who acted as coparents to the child. By the time of the follow-up study, seven of these couples were still cohabiting, six had separated but were still co-parenting, and in one case the biological mother had died leaving the child to be raised by the comother. However, due to the small sample sizes that would arise if dividing the group into two-parent lesbian mother families and single lesbian mother families, all of the lesbian mothers were analyzed as one group. Of the 38 single heterosexual mothers, 29 were still living alone, 8 were cohabiting with a new male partner, and one mother was living with the child’s father. Thirty-three of the heterosexual two-parent families were still married or cohabiting, and five couples had separated or divorced. In those families that had separated, the child still had regular contact with the nonresident parent, parenting decisions were still made jointly by the couple, and both parents were making financial contributions to the child’s upbringing, so these were considered two-parent families for the purposes of the analysis. There were similar proportions of boys and girls in each family type. The age of the target child differed significantly between groups, F(2, 98)  3.37, p  .05. The children of lesbian mothers were the oldest with a mean age of 12 years 1 month, whereas the single mother children and two-parent heterosexual family children had mean ages of 11 years 10 months and 11 years 9 months, respectively. Similarly, a significant group difference was found for the age of the mothers, F(2, 98)  4.74, p  .05. The single mothers were the youngest (mean age  43 years), whereas the lesbian mothers and the mothers from two-parent heterosexual families were slightly older (mean age  46 years). The groups were also found to differ significantly with respect to family size, F(2, 98)  22.72, p  .005, with more children in the two-parent heterosexual families than in the other two types. There was no difference between family types for social class as measured by mother’s occupation, using a modified version of the Registrar General’s classification (OPCS and Employment Department Group, 1991) ranging from 1 (professional/managerial) to 4 (partly skilled or unskilled). Neither was there a group difference for geographical location, i.e., whether the families lived in urban or rural areas. As a significant group difference was found for the child’s age,

48

Social Diversity and Single-Parent Families

mother’s age, and family size, these demographic variables were entered into all further analyses as covariates.

PROCEDURE The families were visited at home by a researcher who was trained in the study techniques. Data were collected from the mother and the target child by tape-recorded interview and questionnaire. Information obtained by interview was rated according to a standardized coding scheme, and regular meetings were held to minimize rater discrepancy. In most cases, two visits were made, the first to interview the mother and the second to interview the child. Ninety-nine percent of mothers and 94 percent of children were interviewed (one single mother and one lesbian mother family had moved abroad, so they could not be interviewed but did complete questionnaires). Questionnaire data were obtained from 94 percent of mothers and children. The groups did not differ with respect to the proportion of children taking part. Measures Mothers’ psychological state. Mothers were administered the State-Trait Anxiety Inventory (STAI; Spielberger, 1983) and the Beck Depression Inventory (BDI; Beck & Steer, 1987; Steer, Beck, & Garrison, 1986) to assess anxiety and depression, respectively. Both instruments have been shown to have good reliability and to discriminate well between clinical and nonclinical groups. Mother-child relationships. Interviews with mothers: The mothers were interviewed using an adaptation of a standardized interview designed to assess parenting (Quinton & Rutter, 1988). The interview lasted one to two hours and was tape-recorded. This procedure has been validated against observational ratings of mother-child relationships in the home, and has demonstrated a high level of agreement between global ratings of the quality of parenting by interviewers and observers (concurrent validity; r  .63). Detailed accounts were made of the child’s behavior and the mother’s response to it, with reference to the child’s progress at school, peer adjustment, and relationships within the family unit. Particular attention was paid to mother-interactions relating to issues of maternal warmth and control, and to the child’s social and emotional development. In a previous study using this interview by the same researchers (Golombok, MacCallum, Goodman, & Rutter, 2002), 57 randomly selected interviews were coded by a second interviewer who was “blind” to family type. Pearson’s product-moment correlation coefficients between raters for individual variables are given in the relevant sections below.

Children Raised in Fatherless Families from Infancy

49

Overall ratings of the quality of parenting were made, taking into account information obtained from the entire interview: (1) expressed warmth was rated on a 6-point scale from 0 (none) to 5 (high). Aspects of warmth considered for this rating included tone of voice, facial expression, and gestures when speaking about the child, spontaneous expressions of warmth, sympathy, and concern about the child’s difficulties (if any) and interest in the child as a person; (2) sensitive responding was rated on a 5-point scale from 0 (none) to 4 (very sensitive responding) and represents the mother’s ability to recognize and respond appropriately to her child’s fears and anxieties; (3) emotional involvement was rated on a 3-point scale from 0 (no over-involvement) to 2 (enmeshed). This rating takes account of the extent to which the parent is over-concerned or overprotective toward the child, and the extent to which the parent has interests apart from those relating to the child; and (4) disciplinary aggression was rated on a 6-point scale, ranging from 0 (none) to 5 (abusive). This rating measures irritability, loss of temper, and physical aggression shown by the mother toward the child during disciplinary interactions. Pearson’s product-moment inter-rater reliability coefficients for expressed warmth, sensitive responding, emotional involvement and disciplinary aggression were .52, .63, .51, and .60, respectively. In addition to these overall ratings, the following individual variables were rated from the interview material: (1) mother to child warmth and (2) child to mother warmth were both rated on a 4-point scale from 0 (little or none) to 3 (marked) and represent the level of demonstrably affectionate behavior between the mother and the child; (3) confiding of child to mother was rated on a 4-point scale from 0 (none) to 3 (some intimate disclosure) and assesses the level of sharing of intimate or personal information; (4) enjoyment in motherhood was rated on a 4-point scale from 0 (none) to 3 (a great deal) and represents the degree of positive feelings expressed about being a parent; (5) activities where parents not informed was rated on a 5-point scale from 0 (none) to 4 (major problem) and is concerned with the extent to which the child engages in activities or is absent from the house without the parent’s knowledge of his or her whereabouts; and (6) severity of disputes was rated on a 4-point scale from 0 (no confrontations) to 3 (major battles) and assesses the intensity of disputes during conflict with the child. Inter-rater reliability coefficients for the above variables were found to be .70, .84, .79, .46, .67, and .81, respectively. Interviews with children: Children were interviewed using the Child and Adolescent Functioning and Environment Schedule (CAFE; John & Quinton, 1991), a semi-structured interview designed to obtain information on the child’s functioning at school, relationships with peers, and relationships with parents. The interview lasted 1 to 11⁄2 hours and was tape-recorded. The following ratings relating to children’s perceptions of their relationships with their mothers were made from the interview. All variables were

50

Social Diversity and Single-Parent Families

measured on a 4-point scale with a higher score representing a higher level of the behavior: (1) warmth from mother is a measure of the mother’s overt affectionate and caring behavior; (2) confiding in mother assesses how often the child confides difficulties and anxieties to the parent; (3) shared interests/activities with mother is a measure of the amount of time the child spends directly involved in interests or activities with the mother; (4) availability of mother assesses the child’s perception of the mother’s availability to the child when the child wants or needs to make contact; (5) dependability of mother represents the mother’s reliability and trustworthiness; and (6) quality of maternal discipline measures the mother’s level of disciplinary control. Inter-rater reliabilities for warmth from mother, confiding in mother, shared interests with mother, availability of mother, maternal dependability, and maternal discipline, respectively, were found to be .50, .77, .68, .84, .74, and .50. Questionnaire measures: The Expression of Affection Inventory (EAI; Hetherington & Clingempeel, 1992) was completed by both the mothers (regarding the child) and the children (regarding the mother). Internal consistency for the total affection score for the present sample was calculated to be .76 for the maternal questionnaire and .68 for the child questionnaire. A higher score represented greater affection. The Conflict Tactics Scale (CTS; Straus, 1979) was also administered to mothers and children, to assess how each one acts during conflict. The CTS yields three subscales; reasoning (e.g., discussing issues), symbolic aggression (e.g., sulking), and physical aggression (e.g., hitting). Internal consistencies for this sample for the reasoning, symbolic aggression, and physical aggression scores, respectively, were .67, .65, and .77 for the maternal questionnaire and .57, .60, and .83 for the child questionnaire. A higher score represented a higher level of the behavior during disputes between mother and child. Children’s socioemotional development. Interviews with mothers: The child’s psychiatric state was assessed using a standardized procedure, the reliability and validity of which are well established (Graham & Rutter, 1968). Detailed descriptions were obtained of any behavioral or emotional problems shown by the child. These descriptions of actual behavior, which included information about where the behavior was shown, severity of the behavior, frequency, precipitants, and course of the behavior over the past year, were transcribed and rated “blind” to the knowledge of family type by an experienced child psychiatrist. Psychiatric disorder, when identified, was rated according to severity (on a 4-point scale from 0  no abnormality to 3  definite and marked abnormality) and type (1  emotional disorder, 2  conduct disorder, and 3  mixed emotional and conduct disorder). The following ratings regarding the child’s adjustment were also made from the interview with the mother: (1) interest in school work was rated

Children Raised in Fatherless Families from Infancy

51

on a 5-point scale from 0 (no interest/effort) to 4 (keen on most subjects) and concerns the child’s interest in academic subjects; (2) worries about relationships at school was measured on a 4-point scale from 0 (none) to 3 (major) and is a rating of worries expressed to the mother by the child concerning relationships with other children at school, e.g., not having any friends or being bullied; and (3) peer problems was rated on a 4-point scale from 0 (no problems) to 3 (very many problems) and assesses the mother’s perception of the extent to which the child appears to have difficulties in making and keeping friends. Interviews with children: The child’s views of his or her own school and peer adjustment were assessed using the following ratings from the CAFE: (1) interest/effort in school work was rated on a 5-point scale from 0 (no interest/effort) to 4 (above average interest in most areas) and assesses the child’s interest and effort in both academic and nonacademic subjects; (2) confidence in school performance is a rating on a 4-point scale from 0 (none in any subject) to 3 (very confident in most) and measures the child’s confidence in his or her own abilities at school; and (3) bullying was rated on a 4-point scale from 0 (never) to 3 (chronic and more serious) and assesses whether the child has ever been bullied and the severity of the incidents. The Social Adjustment Inventory for Children and Adolescents (SAICA; John, Gammon, Prusoff, & Warner, 1987) was administered to each child by the interviewer. Items are presented in the some “childrenother children” format developed by Harter (1985), with higher scores representing greater problem behavior. In the present study, the following four SAICA scales were used: functioning in school, peer relationships, physical self-esteem, and global self-esteem. The SAICA has been found to discriminate well between adolescents with and without psychiatric disorder. The school functioning and peer relationship scales of the SAICA have also been validated against maternal reports from the Child Behavior Checklist (CBCL; Achenbach, 1980) subscales relating to school performance and peer relations with correlations of .69 and .59, respectively. Internal consistencies for this sample were .60, .70, .93, and .91 for school functioning, peer relationships, physical self-esteem, and global self-esteem, in that order. Questionnaire measures: The presence of behavioral or emotional problems in the children was also assessed using the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1994, 1997) administered to mothers and to the children’s teachers. Eighty percent of mothers gave permission for their child’s teacher to be contacted, and 80 percent of the teachers who were contacted returned completed questionnaires. The SDQ produces an overall score of the child’s adjustment (total deviance score). The questionnaire has been found to have good inter-rater reliability with correlations between parent and teacher total deviance scores reported to be .62.

52

Social Diversity and Single-Parent Families

Evidence for validity comes from the high correlations between the total deviance score of the SDQ and the total score of the Rutter Parent Questionnaire, r  .88 (Rutter, Tizard, & Whitmore, 1970) and the Rutter Teacher Questionnaire, r  .92 (Rutter, 1967), which were designed to assess child psychiatric disorder. In addition, the SDQ discriminates well between psychiatric and nonpsychiatric samples. In addition, the Children’s Sex Role Inventory (CSRI; Boldizar, 1991) was administered to the children. This questionnaire assesses the gender role orientation of children and yields two separate independent subscale scores, one of masculinity and one of femininity. Significant gender differences have been found for each subscale, and the validity of the scales has also been demonstrated by correlations with other measures of children’s self-perceptions. Internal consistency scores for the masculinity and femininity scales for this sample were .80 and .81, respectively.

DISCUSSION In this follow-up of children raised in fatherless families from infancy, adolescents’ relationships with their mothers were found to differ in some ways depending on the presence or absence of a father in the home. In fatherless families, children perceived their mothers as interacting more with them and as being more available and dependable. The mothers themselves did not show any differences in warmth toward their children, but mothers in father-absent families reported more serious disputes and more irritability and loss of temper during disciplinary interactions. There were no differences between father-absent and fatherpresent households in the mother’s psychological state. Neither were there differences with respect to the presence of a father on the children’s level of emotional and behavioral problems, school adjustment, peer relationships, or self-esteem. However, boys being raised without a father showed more feminine characteristics, although no less masculine ones, in terms of gender role orientation. Overall it seems that there are not serious negative consequences for children raised in fatherless families from infancy as they reach adolescence, with regard to the quality of parenting that they experience or their own social and emotional development. The finding from the initial study (Golombok et al., 1997) that mothers in fatherless families interacted more with their children is still apparent at age 12. This is perhaps not surprising in light of the fact that in the majority of these families, the mother is the only parent present, and therefore has sole responsibility for interaction with the child. However, the greater maternal warmth found in fatherless families in the first phase of the study has not continued. Although children in two-parent heterosexual families saw their mothers

Children Raised in Fatherless Families from Infancy

53

as less available and dependable than those in fatherless families, they felt equally warm toward them and able to confide in them. With respect to control issues, the increased severity of disputes in fatherless families is also consistent with findings from the initial study. It is possible that in heterosexual two-parent families, it is still traditional for the father to take on more of the disciplining role. However, mothers in female-headed households were not showing dysfunctional levels of disciplinary aggression and the children themselves did not report their mothers to be more punitive. A good level of control combined with high parental warmth creates an authoritative parenting style, proposed by Baumrind (1989) as optimal for children’s development. In line with this, the children in fatherless families showed no evidence of raised levels of problems in emotions, behavior, or relationships. At age six, the children in father-absent families perceived themselves to be less cognitively and physically competent than the children from father-present homes but this difference seems to have dropped out by age 12, with school adjustment and self-esteem being similar across family types. One explanation for the original discrepancy was that since perceptions of competence and self-worth are strongly related to approval by others (Harter, 1993), children may have been reacting to the perception that female-headed families are less valued in society. The growing number of father-absent families and the changing social climate toward greater acceptance of nontraditional family forms may explain why these children no longer feel themselves to be inferior in these aspects. The finding that boys from fatherless families are significantly more feminine in their gender role orientation, although no less masculine, than their counterparts from father-present homes seems to contradict previous research on preschool children, which found that gender role behavior did not differ for boys in father-absent families (Stevens, Golombok, Beveridge, & ALSPAC Study Team, 2002). In a review of previous studies of children of lesbian mothers, Stacey and Biblarz (2001) criticized researchers for downplaying the differences that have been identified in gender development between children of lesbian parents and children from heterosexual homes. However, Stacey and Biblarz made no distinction between core aspects of children’s gender development, such as gender role behavior, on the one hand, and children’s attitudes, such as occupational preferences, on the other. The present study used a measure that examined attitudes and selfperceptions, rather than actual activities and behaviors, as were assessed with the younger children by Stevens et al. Children’s attitudes toward gender-related issues are more open to influence by parents than is their gender role behavior or identity (Egan & Perry, 2001; Jodl, Micheal, Malanchuk, Eccles, & Sameroff, 2001), so it may be that the single and lesbian mothers are explicitly encouraging their sons to have more sensitive

54

Social Diversity and Single-Parent Families

and caring attitudes than the stereotypical male. During the interview, several of the mothers in father-absent families spoke about how they were trying to teach their children, both sons and daughters, to be considerate and to appreciate the feelings of others, suggesting that social learning and social cognitive processes may be involved (Bandura, 1986; Bussey & Bandura, 1999; Mischel, 1970). It is noteworthy that this difference in gender development applied to children from all fatherless families and not just to those with lesbian mothers. In fact, there were very few differences found overall between single heterosexual mother families and lesbian mother families, apart from the higher levels of disciplinary aggression from single heterosexual mothers, which is possibly due to a greater proportion of lesbian mothers having a coparent to share the disciplinarian role. It seems, therefore, that the sexual orientation of the mother does not in itself influence the quality of parenting or the psychological well-being of the child. In addition, there was no evidence to support the concern that children of lesbian mothers would experience more teasing or bullying and more difficulties in their relationships with their peers. The majority of the single heterosexual and lesbian mothers in this study were in occupations classed as “professional/managerial,” as were most of the mothers from two-parent heterosexual mothers. A measure of mother’s educational level (which correlated highly with occupational level) found that a large proportion from all three groups had some form of further education past the age of 18, and there was no difference between the groups in level of maternal education. Thus, the sample is highly educated and relatively affluent, unlike groups of fatherless families in some previous studies (McLanahan & Sandefur, 1994). Therefore, the study allows an evaluation of the impact on family life of being raised without a father, without the confounding influence of a lower socioeconomic status. It must be remembered that the samples of lesbian mothers and single heterosexual mothers in the present study are relatively small. However, the findings are consistent with Ferri’s (1976) data from a general population sample of never-married heterosexual mothers. They are also in line with other investigations of younger children raised from infancy without a father (Brewaeys et al., 1997; Flaks et al., 1995). In addition, one must be aware of the potential for social desirability bias since these mothers may try to present themselves and their children in the best possible light, due to the prejudice still present regarding their family structures. The use of multiple measures (standardized interviews and questionnaires) and multiple respondents (mothers, children, and teachers) goes some way toward combating this problem. Of particular value is the finding that teachers did not report more problems for the children from father-absent families.

Children Raised in Fatherless Families from Infancy

55

A further difficulty comes from the heterogeneity of the samples. Some of the single-mother families now have a resident stepfather, which presents the child with a new array of psychological issues. Of the lesbian mothers, some are in two-parent families and some are single mothers; indeed, one lesbian mother said that the main challenge in parenting for her was being single, not being lesbian. Therefore, it is difficult to determine whether the differences found have arisen due to the presence of only one parent or due to the specific absence of a father. The small cell sizes when the lesbian families were divided according to the number of parents, and the inclusion of several lesbian couples who were coparenting but not cohabiting, meant that statistical comparisons between one-parent and two-parent lesbian families could not be carried out to address this question. Although family circumstances had changed for seven of the lesbian mothers and eight of the single heterosexual mothers (and for five of the heterosexual couples), during the primary school-age years the children had all been in father-absent families (or father-present families for the control group). Therefore the study provides data on the effects of early father absence on adolescent development, and is the first follow-up to adolescence of a group of children raised in fatherless families from birth or early infancy. There will always be difficulties associated with single motherhood, such as poverty and low social status, but it is important to understand that single mothers can vary in their circumstances to the same extent as two-parent families. Though factors associated with fatherless families such as financial hardship (McLanahan & Sandefur, 1994), parental conflict (Amato, 1993), and maternal psychiatric disorder (Dunn et al, 1998; Hetherington & Stanley-Hagan, 2002) do appear to place children at risk, the findings of this follow-up study of children in fatherless families from the outset who were not exposed to financial hardship, parental conflict, or maternal psychiatric disorder in their early years suggest that the absence of a father per se does not necessarily result in psychological disadvantages for children.

REFERENCES Achenbach, T. M. (1980). The child behavior checklist and child behavior profile. Burlington, VT: University of Vermont. Amato, P. (1993). Children’s adjustment to divorce: Theories, hypotheses, and empirical support. Journal of Marriage and the Family, 55, 23–38. Bandura, A. (1986). Social foundations of thought and action: Asocial cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Baumrind, D. (1989). Rearing competent children. In W. Damon (Ed.), Child development today and tomorrow (pp. 349–378). San Francisco: Jossey-Bass.

56

Social Diversity and Single-Parent Families

Baumrind, D. (1991). Effective parenting during the early adolescent transition. In P. A. Cowen & M. Hetherington (Eds.), Family transitions. Hillsdale, NJ: Erlbaum. Beck, A., & Steer, R. (1987). The Beck Depression Inventory manual. San Diego, CA: Psychological Corporation. Biller, H. B. (1974). Parental deprivation. Lexington, MA: Heath. Boldizar, J. P. (1991). Assessing sex typing and androgyny in children: The Children’s sex role inventory. Developmental Psychology, 27, 505–515. Brewaeys, A., Ponjaert, I., Van Hall, E. V., & Golombok, S. (1997). Donor insemination: Child development and family functioning in lesbian mother families. Human Reproduction, 12, 1349–1359. Broman, S. H., Nichols, P. L., & Kennedy, W. A. (1975). Pre-school IQ: Parental and early development correlates. Hillsdale, NJ: Erlbaum. Bussey, K., & Bandura, A. (1999). Social cognitive theory of gender development and differentiation. Psychological Review, 106, 676–713. Chan, R. W., Raboy, B., & Patterson, C. J. (1998). Psychosocial adjustment among children conceived via donor insemination by lesbian and heterosexual mothers. Child Development, 69, 443–457. Coleman, J. C., & Hendry, L. (1999). The nature of adolescence (3rd ed.). London, UK: Routledge. Collaer, M. L., & Hines, M. (1995). Human behavioral sex differences: A role for gonadal hormones during early development? Psychological Bulletin, 118, 55–107. Collins, W. A. (1990). Parent-child relationships in the transition to adolescence: Continuity and change in interaction, affect, and cognition. In R. Montemayor, G. R. Adams, & T. P. Gullato (Eds.), From childhood to adolescence: A transition period? Advances in adolescent development (vol. 2, pp. 85–106). Newbury Park, CA: Sage. Crockett, L. J., Eggebeen, D. J., & Hawkins, A. J. (1993). Fathers’ presence and young children’s behavioral and cognitive adjustment. Journal of Family Issues, 14, 355–377. Cummings, E. M., & Davies, P. T. (1994). Maternal depression and child development. Journal of Child Psychology and Psychiatry, 35, 73–112. Dunn, J., Deater-Deckard, K., Pickering, K., O’Connor, T. G., Golding, J., & The ALSPAC Study Team. (1998). Children’s adjustment and prosocial behaviour in step-, single-parent, and non-stepfamily settings: Findings from a community study. Journal of Child Psychology and Psychiatry, 39, 1083–1095. Egan, S. K., & Perry, D. G. (2001). Gender identity: A multidimensional analysis with implications for psychosocial adjustment. Developmental Psychology, 37, 451–463. Ferri, E. (1976). Growing up in a one parent family. Slough, UK: NFER. Flaks, D. K., Ficher, I., Masterpasqua, F., & Joseph, G. (1995). Lesbians choosing motherhood: A comparative study of lesbian and heterosexual parents and their children. Developmental Psychology, 31, 105–114. Golombok, S. (1999). Lesbian mother families. In A. Bainham, S. D. Sclater, & M. Richards (Eds.), What is a parent? A socio-legal analysis (pp. 161–180). Oxford: Hart. Golombok, S., MacCallum, F., Goodman, E., & Rutter, M. (2002). Families with children conceived by donor insemination: A follow-up at age 12. Child Development, 73, 952–968.

Children Raised in Fatherless Families from Infancy

57

Golombok, S., Perry, B., Burston, A., Murray, C., Mooney-Somers, J., Stevens, M., et al. (2003). Children with lesbian parents: A community study. Developmental Psychology, 39, 20–33. Golombok, S., Spencer, A., & Rutter, M. (1983). Children in lesbian and singleparent households: Psychosexual and psychiatric appraisal. Journal of Child Psychology and Psychiatry, 24, 551–572. Golombok, S., & Tasker, F. (1996). Do parents influence the sexual orientation of their children? Findings from a longitudinal study of lesbian families. Developmental Psychology, 32, 3–11. Golombok, S., Tasker, F., & Murray, C. (1997). Children raised in fatherless families from infancy: Family relationships and the socioemotional development of children of lesbian and single heterosexual mothers. Journal of Child Psychology and Psychiatry, 38, 783–791. Goodman, R. (1994). A modified version of the Rutter Parent Questionnaire including extra items on children’s strengths: A research note. Journal of Child Psychology and Psychiatry, 35, 1483–1494. ———. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581–586. Graham, P., & Rutter, M. (1968). The reliability and validity of the psychiatric assessment of the child: II. Interview with the parent. British Journal of Psychiatry, 114, 581–592. Green, R., Mandel, J. B., Hotvedt, M. E., Gray, J., & Smith, L. (1986). Lesbian mothers and their children: A comparison with solo parent heterosexual mothers and their children. Archives of Sexual Behavior, 15, 167–184. Harter, S. (1985). The self-perception profile for children: Revision of the perceived competence scale for children. Unpublished manual, University of Denver. ———. (1993). Causes and consequences of low self-esteem in adolescents and adults. In R. F. Baumeister (Ed.), Self-esteem The puzzle of low self-regard. New York: Plenum Press. Herzog, E., & Sudia, C. E. (1973). Children in fatherless families. In B. Campbell, & H. N. Riccuiti (Eds.), Review of child development research. Chicago: University of Chicago Press. Hetherington, E. M., & Clingempeel, W. G. (1992). Coping with marital transitions: A family systems perspective. Monographs of the Society for Research in Child Development, 57, 242. Hetherington, E. M., & Stanley-Hagan, M. M. (2002). Parenting in divorced and remarried families. In M. H. Bornstein (Ed.), Handbook of parenting (vol. 3, pp. 287–315). Hillsdale, NJ: Lawrence Erlbaum Associates. Hoeffer, B. (1981). Children’s acquisition of sex-role behavior in lesbian mother families. American Journal of Orthopsychiatry, 51, 536–544. Huggins, S. L. (1989). A comparative study of self-esteem of adolescent children of divorced lesbian mothers and divorced heterosexual mothers. In F. Bozett (Ed.), Homosexuality and the family (pp. 123–135). New York: Harrington Park Press. Jodl, K. M., Micheal, A., Malanchuk, O., Eccles, J. S., & Sameroff, A. (2001). Parents’ roles in shaping early adolescents’ occupational aspirations. Child Development, 72, 1247–1265. John, K., Gammon, G. D., Prusoff, B. A., & Warner, V. (1987). The Social Adjustment Inventory for Children and Adolescents (SAICA): Testing of a new

58

Social Diversity and Single-Parent Families

semistructured interview. Journal of American Academy of Child and Adolescent Psychiatry, 26, 898–911. John, K., & Quinton, D. (1991). Child and adolescent functioning and environment schedule (Rev.). London: MRC Child Psychiatry Unit. Kirkpatrick, M. (1987). Clinical implications of lesbian mother studies. Journal of Homosexuality, 13, 201–211. Kirkpatrick, M., Smith, C., & Roy, R. (1981). Lesbian mothers and their children: A comparative survey. American Journal of Orthopsychiatry, 51, 545–551. Kochanska, G., Kuczynski, L., Radke-Yarrow, M., & Welsh, J. D. (1987). Resolution of control episodes between well and affectively ill mothers and their young child. Journal of Abnormal Child Psychology, 15, 441–456. Lamborn, S. D., Mounts, N. S., Steinberg, L., & Dornbusch, S. M. (1991). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful homes. Child Development, 62, 1049–1065. Maccoby, E. E. (1998). The two sexes: Growing up apart, coming together. Cambridge, MA: Harvard University Press. Martin, C.L. (1991). The role of cognition in understanding gender effects. In H. Reese (Ed.), Advances in child development and behavior (vol. 23, pp. 113–164). New York: Academic Press. Martin, C. L., & Halverson, C. (1981). A schematic processing model of sex typing and stereotyping in children. Child Development, 52, 1119–1134. McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press. Mischel, W. (1970). Sex-typing and socialization. In P. H. Mussen (Ed.), Carmichael’s manual of child psychology (vol. 2). New York: Wiley. Mucklow, B. M., & Phelan, G. K. (1979). Lesbian and traditional mothers’ responses to child behavior and self-concept. Psychological Reports, 44, 880–882. Office of National Statistics. (2002). Social Trends, 32. UK: HMSO. Office of Population and Census Statistics (OPCS) and Employment Department Group. (1991). Standard classification of occupations. London: Her Majesty’s Stationery Office. 882. Patterson, C. J. (2002). Lesbian and gay parenthood. In M. C. Bornstein (Ed.), Handbook of parenting (vol. 3, pp. 317–338). Hillsdale, NJ: Lawrence Erlbaum and associates. Quinton, D., & Rutter, M. (1988). Parenting breakdown: The making and breaking of intergenerational links. Aldershot, UK: Avebury Gower Publishing. Rutter, M. (1967). A children’s behaviour questionnaire for completion by teachers: Preliminary findings. Journal of Child Psychology and Psychiatry, 8, 1–11. ———. (1971). Parent-child separation: Psychological effects on the children. Journal of Child Psychology and Psychiatry, 12, 233–260. Rutter, M., Tizard, J., & Whitmore, K. (1970). Education, health and behaviour. London: Longman. Spielberger, C. (1983). The Handbook of the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. Stacey, J., & Biblarz, T. (2001). (How) does the sexual orientation of parents matter? American Sociological Review, 66, 159–183.

Children Raised in Fatherless Families from Infancy

59

Steer, R., Beck, A., & Garrison, B. (1986). Applications of the Beck Depression Inventory. In N. Sartorius & T. Ban (Eds.), Assessment of depression (pp. 123–142). Berlin: Springer-Verlag. Steinberg, L. (1990). Autonomy, conflict, and harmony in the family relationships. In S. S. Feldman & G. R. Elliot (Eds.), At the threshold: The developing adolescent (pp. 225–276). Cambridge, MA: Harvard University Press. ———. (2000). We know some things: Parent-adolescent relations in retrospect and prospect. Journal of Research on Adolescence, 11, 1–19. Stevens, M., Golombok, S., Beveridge, M., & ALSPAC Study Team. (2002). Does father absence influence children’s gender development? Findings from a general population study of pre-school children. Parenting: Science and Practice, 2, 49–62. Straus, M. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and the Family, 41, 75–88. Vanfraussen, K., Ponjaert-Kristoffersen, I., & Brewaeys, A. (2002). What does it mean for youngsters to grow up in a lesbian family created by means of donor insemination? Journal of Reproductive and Infant Psychology, 20, 237–254. Weinraub, M., Horvath, D. L., & Gringlas, M. B. (2002). Single parenthood. In M. H. Bornstein (Ed.), Handbook of parenting (vol. 3, pp. 109–139). Hillsdale, NJ: Lawrence Erlbaum Associates.

Chapter 5

Family Structure and Children’s Success: A Comparison of Widowed and Divorced Single-Mother Families Timothy J. Biblarz and Greg Gottainer*

Does single motherhood produced by divorce have more negative consequences for children than single motherhood produced by the death of the father, and if so, why? Are widows with dependent children different kinds of mothers—in tastes, values, or lifestyles—than divorced mothers? Do widows occupy advantaged social and economic positions relative to divorced mothers? Our premise is that not all single-mother families are alike. By examining heterogeneity within the type, factors that allow children from single-mother families to do more or less well can be identified. This article examines long-term differences in children’s attainment and well-being across types of single-mother families using a nationally representative sample (the 1972–1996 General Social Surveys [GSS]). Compared with children raised in single-mother families produced by the death of the father, children raised in single-mother families produced by divorce have significantly greater odds of not completing high school, lower odds of entering and graduating from college, a lower average occupational status, and a lower average level of happiness in adulthood. In an attempt to uncover why the one type of single-mother family has

* T. J. Biblarz & G. Gottainer (2000). Family Structure and Children’s Success: A Comparison of Widowed and Divorced Single-Mother Families. Journal of Marriage and Family, 62(2), 533–548. Copyright 2000, Wiley. Reproduced with permission of Blackwell Publishing Ltd.

Family Structure and Children’s Success

61

such different implications for children than the other, widowed and divorced single mothers with dependent children are then compared on a variety of attitudes and behaviors. No significant differences in health (physical or psychological), education, religion, family values, or other dimensions of lifestyle and social behavior are observed. In contrast, there are significant differences in socioeconomic position and financial stress in which, relative to widows, divorced mothers are at a substantial disadvantage.

BACKGROUND Is the single-parent family structure by its very nature less capable of providing fully for children than the two-biological-parent form, or do the observed negative effects on children actually reflect children’s greater likelihood of exposure to certain kinds of problematic family processes that may be correlated with single-parent families but are not a necessary part of them? One way to investigate this family structurefamily process question is to compare the attainments of children from similar family structures produced by different family processes, as in the case of single-mother families created by the death of the father and those created by the divorce of the parents. (Children from single-parent families produced by nonmarital childbearing are given less attention, because data limitations precluded their inclusion in the statistical analyses later in the paper). Studies that examine variation in achievement among children from single-mother families show mixed results. Some studies show that children from single-parent families who experienced the death of a parent have higher attainments than those who experienced their parents’ divorce (Acock & Keicolt, 1989; Amato & Keith, 1991a, 1991b). Other evidence suggests that children from both types of single-parent families have greater odds of delinquency and lower academic achievement (Evans et al., 1995; Rankin, 1983). Findings of difference in child outcomes (or not) are generally not accompanied by systematic theorizing about the nature of diversity in experience of different types of single-mother families. One possibility— a selection or fitness argument—is that single parents who divorce (and their ex-spouses) may be fundamentally different kinds of people and parents than single-parent widows and their deceased spouses. Another is that single-parent families are accorded greater or lesser social and economic support depending on antecedent factors. The main theoretical perspectives on this issue are outlined below. While all the theories predict that the two-parent home will provide certain advantages for children, they make contrasting predictions about

62

Social Diversity and Single-Parent Families

the consequences for children being raised in different kinds of singlemother families. Family Structure Model A basic family structure model emphasizes the fundamental importance of family structure for children’s attainment. Having been raised by two biological parents, or not, is the crucial determinant of child outcomes. With two parents, children will learn about the structure of authority relations and how to successfully interact with authority figures (Nock, 1988). When that structure is removed, such as in the case of the single-parent family, parent-child relations can become more peerlike and children will not learn how to deal with power holders, and so the children will miss important, subtle lessons about how to achieve in market activity. The absence of one parent will also risk subjecting children to higher levels of parental authoritarianism or neglect and to lower levels of parental involvement and supervision (Astone & McLanahan, 1991; Bronfenbrenner, 1979; Thomson, Hanson, & McLanahan, 1994; Thomson, McLanahan, & Curtin, 1992). The family structure model would therefore predict no differences in outcomes between children from widowed single-mother families and those from divorced single-mother families because the two share the same basic family structure. Support for this prediction is found in studies that show that children from both types of families have higher rates of delinquency (running away or truancy) and emotional problems (depression or low self-esteem), and lowered school performance (Evans et al., 1995; Rankin, 1983; Wadsworth & Maclean, 1986). Household Economics Model The economic model views households as acting as singular units to maximize collective utility. Utility comes partly from children whose human capital is developed by parental investments in market activity and household services. The two-parent family is a particularly efficient system for maximizing utility (including that of children), because two parents provide time and money in a complementary way (Becker, 1964; Becker & Tomes, 1986). The economic resources (and equivalent services) that parents provide determine how well the children will fare. Children from any type of single-mother household will lack the economic resources that fathers provide, and so they will not do as well as children from twoparent homes (Biblarz, Raftery, & Bucur, 1997; Bogess, 1998; McLanahan, 1985). But among single-parent families, differences in child outcomes between single divorced-mother households and single

Family Structure and Children’s Success

63

widowed-mother households would be observed only if one type of household had substantially higher levels of economic resources, on average, than the other type. Evolutionary Perspectives The evolutionary view (e.g., Emlen, 1997; Trivets, 1972) proposes that parents seek to maximize reproductive fitness through investment in children. Because the number of children that women could potentially have is lower than that of men, more of the potential reproductive investment of the mother than of the father is tied up in the child or children at hand (Trivets, 1972). Accordingly, mothers will generally invest more of their resources in existing children than will fathers. Single-divorced and single-widowed mothers have the same level of fitness represented in their children, and so both types of mothers would have the same level of incentive to invest highly in their children. Like the economic model, the evolutionary perspective predicts that child outcomes would vary between widowed and divorced single-mother families only if the absence of the husband meant a greater loss of material resources for one type of family than the other. Even though both divorced and widowed families experience a large decline in standard of living and family income, the decline may be particularly steep for divorced mothers (Holden & Smock, 1991). At the government level, the benefits available to widowed and divorced mothers differ enormously. Widows with dependent children are entitled to social security survivor’s benefits, a program that provides each child a monthly benefit until age 18. The mother can choose to stay at home with her children and also collect a monthly payment until her children are 16–18 years old. If the mother chooses to work in the paid labor force, her monthly benefit continues until her own earnings surpass a particular threshold (the children’s benefits will continue to be paid regardless). The family does not need to pass an assets test to collect benefits, and the benefits are paid regardless of income from any combination from savings, pension plans, or life insurance (Sugarman, 1993, 1995). In contrast, the main program available to divorced mothers with dependent children has been Aid to Families with Dependent Children (AFDC). To be eligible, single-divorced mothers must deplete any liquid assets remaining from their divorce and must demonstrate a standard of living below the poverty level. The possibility of using AFDC benefits as a supplement to labor market earnings is constrained by the requirement that after four months of employment, AFDC payments are reduced by one dollar for every dollar earned on the job. The amount of AFDC payments alone is not sufficient to raise the standard of living of singleparent families above the poverty level (Sugarman, 1993, 1995).

64

Social Diversity and Single-Parent Families

In 1996, AFDC was replaced with the Temporary Assistance to Needy Families (TANF) program as part of the Personal Responsibility and Work Opportunity Reconciliation Act. Grounded in a “no work-no pay” approach, TANF limits the amount of consecutive and lifetime months that divorced single mothers (and others who are eligible) can collect benefits, typically to between 24 and 60 months, respectively (U.S. Administration for Children and Families, 1998). Children from divorced single-mother families are entitled to a relatively low level of government support, in part because of the belief that nonresidential fathers (along with residential mothers) are responsible for providing for the children. However, only half of divorced mothers receive the full amount of child support awarded, and a quarter receive no child support at all from the ex-spouse (another 15 percent are not awarded child support) (Maccoby & Mnookin, 1992). Thus, widowed mothers and their children may have greater access than divorced mothers to certain kinds of socioeconomic supports and consequently may enjoy advantaged socioeconomic positions (McLanahan, Garfinkel, & Ooms, 1987). The Parental Fitness Model Fitness models propose that the negative effect of divorce on children is a selection effect. People who divorce may have preexisting qualities— alternative value systems or a lack of competency at family life—that make both divorce and problems for children more likely (Booth, 1999). Support for this kind of perspective comes from longitudinal data showing that behavioral problems in children typically associated with divorce were actually present in the children prior to the parental divorce (Cherlin et al., 1991; Kiernan, 1997; but see Cherlin, Chase-Lansdale, and McRae, 1998). An early study of juvenile delinquency attributed the greater likelihood of delinquent behavior among children who experienced parental divorce, separation, or desertion (relative to those that had experienced the death of a parent) as a reflection of their greater likelihood of exposure to “disturbed” behavior in parents (see Birtchnell, 1969). Children from divorced single-mother families have also been described by teachers as having significantly lower levels of educational stimulation from parents and greater levels of parental rejection than have those who experienced the death of a parent (Felner, Farber, Ginter, Boike, & Cohen, 1980). The parental fitness perspective proposes that divorce proxies for or reflects family competency in a way that widowhood does not. Competency involves parental behavior, but it may also involve value orientations (Popenoe, 1996). Widowed mothers will have more

Family Structure and Children’s Success

65

traditional family values and lifestyles than divorced mothers. Unlike single mothers who divorce, widowed single mothers did not choose an alternative family structure for themselves and their children. The kinds of values and lifestyles held by widows will be more beneficial to children’s success in the larger society than will the alternative values of divorcees. For all of these reasons, the parental fitness perspective predicts that children from widowed single-mother families will do substantially better than will children from divorced single-mother homes. The Marital Conflict Model Another kind of selection hypothesis is that children who experience their parents’ divorce are likely to have had prolonged exposure to conflict between their parents. Open conflict between parents is frightening and distressing for children and can have both short- and long-term negative effects on self-esteem and educational attainment (Amato, 1986, 1993; Amato & Booth, 1997; Emery, 1982). Children from widowed singlemother homes would be expected to do substantially better than would those from divorced single-mother homes because they will have been less likely, on average, to have had prolonged exposure to parental conflict. In addition, the experiences surrounding divorce and the father’s moving out of the residential household often lead to the development of hostile feelings toward the father on the part of the children (Parish & Kappes, 1980; Rozondal, 1983; Wallerstein & Kelly, 1980). In contrast, when fathers die, children tend to develop warm and positive inner constructions of their fathers (Silverman, Nickman, & Worden, 1992). Children build a set of memories of feelings and actions with the deceased parent that helps them gain a sense of meaning about the loss, facilitates the process of mourning and acceptance, and helps them incorporate the deceased parent into their life in a positive and ongoing way. These theories make competing predictions about the effects on children of single-mother families produced by the death of a parent rather than by divorce. Some (the parental fitness and marital conflict perspectives) argue that there are differences in competencies, childcenteredness, and values between mothers whose husbands died and parents who got divorced and that, as a result, there are differences in the success of their children. Others (the economic and evolutionary perspectives) propose that differences in child outcomes across types of single-mother families can be accounted for more by the contrasting structural positions of families than by cultural positions or psychological functioning. The baseline family structure perspective predicts that the single-parent family structures will have roughly the same effect on children, irrespective of the underlying processes that gave rise to them.

66

Social Diversity and Single-Parent Families

DATA Data are the 1972–1996 pooled General Social Surveys (GSS, N = 35,284). Each survey is a nationally representative sample of adults ages 18 and over living in the continental United States (Davis & Smith, 1996; also http://www.icpsr. umich.edu/GSS99). The GSS are one of few national-level surveys available that include not only primary family structure experienced during childhood but also the cause structure (death, divorce, etc.) underlying alternative family structures and that, when pooled, have an adequate number of respondents from each of the family types to do meaningful statistical analyses. One drawback is the necessity of pooling information over 25 years in order to gain the adequate sample size (although period can be taken into account in statistical analyses). However, recent evidence suggests that family structure effects on children’s attainment have been relatively constant over this period (Biblarz & Raftery, 1999). In the first stage of the analysis, the GSS are used to describe basic aggregate differences in three adult outcomes of respondents who at age 16 were living in one of five family types: (a) a two-biological-parent family; (b) a single-mother family caused by the death of the father; (c) a single-mother family caused by the divorce of the parents; (d) a motherand-stepfather family preceded by the death of the biological father; and (e) a mother-and-stepfather family preceded by the divorce of the biological mother and father. Adult children from single-mother homes whose mothers were never married cannot be distinguished in the GSS—they are pooled with other kinds of respondents in an “Other” reason category and are not included in the analysis. The three adult outcomes are (a) educational attainment (logistic regression models), (b) respondent’s current occupational status (least squares regression models), and (c) respondent’s general psychological well-being: Taken all together, how would you say things are these days? Would you say that you are very happy, pretty happy, or not too happy? The question is scored in such a way that very happy = 1; otherwise, the score is 0 (logistic regression models). Following Mare (1980, 1981, 1995), educational attainment is treated as a series of transitions, beginning with the completion (or not) of ninth grade and ending with graduation from college (four-year degree or higher). Analysis of ninth-grade completion includes the full sample, analysis of high school completion includes only those who had completed ninth grade, analysis of entry into college includes only those who had completed high school, and analysis of college completion includes only those who had begun college. From 1972–1987, the GSS coded respondent’s occupation on the basis of 1970 Census occupational titles. From 1988 to the present, occupation is

Family Structure and Children’s Success

67

coded on the basis of 1980 Census titles. There is no singular socioeconomic index (SEI) of occupations developed uniformly for both the 1970 and (changed) 1980 Census occupational titles. Hence in the analysis of occupational attainment for 1970-basis GSS years, we use Hauler and Featherman’s (1977) socioeconomic status scores for 1970 Census titles; for 1980-basis years, we use Hauser and Warren’s (1997) socioeconomic status scores for 1980 Census titles. The main focus of interest is whether there are significant differences in these outcomes between respondents raised in divorced and widowed single-mother families. All models include dummy variables for each year and dummy variables for respondent’s age (five-year age intervals).

CONCLUSIONS The attainments of children from widowed single-mother families are approximately the same as those from two-biological-parent families, and they are substantially higher than those of children from divorced singlemother families. The family structure model discussed above—that the same family structure should lead to the same outcome—is rejected by the evidence. Why is one type of single-parent family more successful than the other? The parental fitness model proposes that widows will be more competent parents than divorced mothers. Although direct measures of parental behavior were not available, no evidence to support this view was detected from the variables at hand. Widowed and divorced single mothers had approximately the same values for children and level of religiosity, and they reported similar levels of physical health and healthrelated behavior, psychological well-being, and social behavior. A variant of the fitness model emphasizes family values. Here, too, we found weak evidence of differences. On items reflecting approval of nontraditional family behavior and gender role egalitarianism, the responses of widowed, divorced, and never-married single mothers were generally similar. The marital conflict model proposed that children who experience their parents’ divorce will have had more exposure to family conflict than children whose fathers died, and so they will have lower attainments. The GSS has no measure of exposure to family conflict, and this hypothesis was not tested. As a roundabout way of touching on the issue, we speculated that in the same way that marital conflict has negative effects on children’s psychological well-being, it may also have negative effects on the psychological well-being of the single mothers who were most likely to experience it—those who had divorced. However, no significant

68

Social Diversity and Single-Parent Families

psychological differences were observed. Divorced single-mothers were no more or less jaded than widows, were no more or less distrustful toward the world, were no more or less pessimistic about the present or the future, and reported finding life equally exciting and happy. When we turned to the structural and socioeconomic variables highlighted by the economic and evolutionary perspectives, significant differences were found. The patterns involving employment, occupational status, and financial situation suggest that widows occupy an advantaged position in the social structure. There is some evidence that the advantage may be growing over time, but this requires further testing. The differences in employment, occupation, and financial stress are not likely to be explainable on the basis of individual-level variation in human capital, because there were no differences in the average educational attainment of the three groups of mothers. They may be due in part to collective decisions about policy that discriminate between kinds of single-parent families, decisions that are then realized in substantially more favorable public support for widows (Sugarman, 1995). On the basis of our evidence, the family structure model is rejected, the parental fitness model is unsupported, the marital conflict model could not be fully tested, and the economic and parental investment models are supported. The results suggest that the family’s position in the social structure may be an important starting point for understanding variation in attainments among children from different kinds of alternative families. Although significant differences in socioeconomic position were found between widowed and divorced single mothers, we were not able to directly test whether these could explain the variation in children’s longterm outcomes that were observed. The contrasting child outcomes may be due to the socioeconomic advantages and disadvantages of the different kinds of single mothers, but a direct assessment of this hypothesis requires bringing together in the same model the variables from each side of the analysis. One limitation we faced was that women’s (and mothers’) position in the social structure has not traditionally been given substantial attention in national surveys. For example, mother’s occupation was not asked in the GSS until 1994 (whereas father’s occupation has been asked consistently from 1972 to the present). Thus, adult children’s retrospective reports of their mothers’ occupational and employment positions could not be included in the analysis of children’s attainment and well-being. A second limitation is in the availability of longitudinal data on a large number of widowed and divorced single mothers with dependent children. The NSFH, for example, is longitudinal, but the sample of focal children from widowed single-mother families interviewed in the second wave is very small. One exception is the National Education Longitudinal

Family Structure and Children’s Success

69

Survey (NELS), which includes baseline interviews with 24,599 5th graders and 22,651 of their parents in 1988, and reinterviews every two years. Our next step in this research is to use the NELS to further explore links between the social contexts and economic positions of different kinds of single-mother families, as well as to explore the attainments and well-being of the children. Though we observed few differences (other than socioeconomic ones) between types of single mothers, we were not able to make comparisons involving fathers, or children’s perceptions of fathers. The role of fathers could be considered by comparing children’s relationship with, and feelings toward, the living father (in the case of divorce) with children’s psychic relationship with, and feelings toward, the deceased father (in the case of the death of the parent). Similarities and differences in the consequences for children of these feelings and psychic vis-à-vis actual or behavioral relationships with absent fathers could then be investigated. This could be coupled with an investigation of how, depending upon the type of loss (death or divorce) experienced, the custodial family and its members are responded to by actors outside the family such as extended family members, friends, coworkers, the community, schools, and the government. The causes of the contrasts in the relative social and socioeconomic standing of different kinds of single-mother families are an important focus for further research.

REFERENCES Acock, A. C., & Demo, D. H. (1994). Family diversity and well-being. Thousand Oaks, CA: Sage. Acock, A. C., & Keicolt, J. K. (1989). Is it family structure or socioeconomic status? Family structure during adolescence and adult adjustment. Social Forces, 68, 553–571. Alwin, D. F. (1984). Trends in parental socialization values: Detroit, 1958–1983. American Journal of Sociology, 90, 359–382 ———. (1986). Religion and parental child-rearing orientations: Evidence of a Catholic-Protestant convergence. American Journal of Sociology, 92, 412–440. ———. (1988). From obedience to autonomy: Changes in traits desired in children, 1924–1978. Public Opinion Quarterly, 52, 33–52. ———. (1990). Cohort replacement and changes in parental socialization values. Journal of Marriage and the Family, 52, 347–360. Amato, P. R. (1986). Marital conflict, the parent-child relationship, and child self esteem. Family Relations, 35, 403–410. ———. (1993). Children’s adjustment to divorce: Theories, hypotheses, and empirical support. Journal of Marriage and the Family, 55, 23–38. Amato, P. R., & Booth, A. (1997). A generation at risk: Growing up in an era of family upheaval. Cambridge, MA: Harvard University Press.

70

Social Diversity and Single-Parent Families

Amato, P. R., & Keith, B. (1991a). Parental divorce and adult well-being: A metaanalysis. Journal of Marriage and the Family, 53, 43–58. ———. (1991b). Separation from a parent during childhood and adult socioeconomic attainment. Social Forces, 70, 187–206. Astone, N. M., & McLanahan, S. S. (1991). Family structure, parental practices, and high school completion. American Sociological Review, 56, 309–320. Becker, G. S. (1964). Human capital. New York: National Bureau of Economic Research. ———. (1981). A treatise on the family. Cambridge, MA: Harvard University Press. Becker, G. S., & Tomes, N. (1986). Human capital and the rise and fall of families. Journal of Labor Economics, 4, S1-S39. Biblarz, T. J., & Raftery, A. E. (1999). Family structure, educational attainment, and socioeconomic success: Rethinking the “pathology of matriarchy.” American Journal of Sociology, 1999. Biblarz, T. J., Raftery, A. E., & Bucur, A. (1997). Family structure and social mobility. Social Forces, 75, 1319–1339. Birtchnell, J. (1969). The possible consequences of early parental death. British Journal of Medical Psychology, 42, 1–12. Bogess, S. (1998). Family structure, economic status, and educational attainment. Journal of Population Economics, 11, 203–222. Booth, A. (1999). Causes and consequences of divorce: Reflections on recent research. In R. A. Thompson & P. R. Amato (Eds.), The post-divorce family (p. 2948). Thousand Oaks, CA: Sage. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press. Bumpass, L. L., & Sweet, J. A. (1989). Children’s experience in single-parent families: Implications of cohabitation and marital transitions. Family Planning Perspectives, 21, 256–260. Cherlin, A. J. (1978). Remarriage as an incomplete institution. American Journal of Sociology, 84, 634–650. Cherlin, A. J., Chase-Lansdale, P. L., & McRae, C. (1998). Effects of parental divorce on mental health throughout the life course. American Sociological Review, 63, 239–249. Cherlin, A. J., & Furstenberg, F. F. (1994). Stepfamilies in the United States: A reconsideration. Annual Review of Sociology, 20, 359–381. Cherlin, A. J., Furstenberg, E. E., Chase-Lansdale, P. L., Kiernan, K., Morrison, D. R., & Teitler, J. (1991). Longitudinal studies of effects of divorce on children in Great Britain and the United States. Science, 252, 1386–1389. DaVanzo, J., & Rahman, M. O. (1993). American families: Trends and correlates. Population Index, 59, 350–386. Davis, J. A., & Smith, T. W. (1996). General social surveys, 1972–96: Cumulative codebook. Principal investigator, J. A. Davis; Director and co-principal investigator, T. W. Smith. Chicago: National Opinion Research Center. Duncan, B. (1967). Education and social background. American Journal of Sociology, 72, 363–372. Duncan, B., & Duncan, O. D. (1969). Family stability and occupational success. Social Problems, 16, 273–285. Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 92, 310–330.

Family Structure and Children’s Success

71

Emlen, S. T. (1997). The evolutionary study of human family systems. Social Science Information Sur Les Sciences Sociales, 36, 563–589. Evans, M. D. R., Kelley, J., Borgers, M., Dronkers, J., & Grullenberg, L. (1995, July 17). Parent divorce and children’s education: Australian evidence. Worldwide Attitudes [Online serial], 1–8. Felner, R. D., Farber, S. S., Ginter, M. A., Boike, M. F, & Cohen, E. L. (1980). Family stress and organization following parental divorce or death. Journal of Divorce, 4, 67–76. Furstenberg, F. F., & Cherlin, A. J. (1991). Divided families. Cambridge, MA: Harvard University Press. Gals-Sternas, K. (1995). Single parent widows: Stressors, appraisal, coping resources, grieving responses and health. Marriage and Family Review, 20, 411–444. Goldscheider, F. K., & Goldscheider, C. (1989). Family structure and conflict: Nest leaving expectations of young adults and their parents. Journal of Marriage and the Family, 51, 87–97. ———. (1991). The intergenerational flow of income: Family structure and the status of Black Americans. Journal of Marriage and the Family, 53, 499–508. Hauler, R. M., & Featherman, D. L. (1977). The process of stratification: Trends and analysis. New York: Academic Press. Hauser, R. M., & Warren, J. R. (1997). Socioeconomic indexes for occupations: A review, update, and critique. Sociological Methodology, 27, 177–298. Holden, K. C., & Smock, P. J. (1991). The economic costs of marital dissolution: Why do women bear a disproportionate cost? Annual Review of Sociology, 17, 51–78. Keith, P., & Schafer, R. B. (1982). A comparison of depression among employed single parent and married women. Journal of Psychology, 110, 239–247. Kiernan, K. (1997). The legacy of parental divorce: Social, economic and demographic experiences in adulthood. Centre for Analysis of Social Exclusion Paper 1. London School of Economics. Kurdek, L. A., & Fine, M. A. (1991). Cognitive correlates of satisfaction for mothers and stepfathers in stepfather families. Journal of Marriage and the Family, 53, 565–572. Lenski, G. (1962). The religious factor. New York: Doubleday. Maccoby, E. E., & Mnookin, R. H. (1992). Dividing the child: Social and legal dilemmas of custody. Cambridge, MA: Harvard University Press. Mare, R. D. (1980). Social background and school continuation decisions. Journal of the American Statistical Association, 75, 295–305. Mare, R. D. (1981). Change and stability in educational stratification. American Sociological Review, 46, 72–87. Mare, R. D. (1995). Changes in educational attainment and school enrollment. In R. Farley (Ed.), State of the Union: America in the 1990s (pp. 155–214). New York: Russell Sage Foundation. McLanahan, S. S. (1985). Family structure and the reproduction of poverty. American Journal of Sociology, 90, 873–901. McLanahan, S. S., Garfinkel, L., & Ooms, T. (1987). Female-headed families and economic policy: Expanding the clinician’s focus. In J. C. Hansen & M. Lindblad-Goldberg (Eds.), Clinical issues in single parent households (pp. I-18) Rockville, MD: Aspen Publishers.

72

Social Diversity and Single-Parent Families

McLanahan, S. S., & Sandefur, G. (1994). Growing up with a single parent. Cambridge, MA: Harvard University Press. Nakau, K., & Treas, J. (1994). Updating occupational prestige and socioeconomic scores: How the new measures measure up. In P. V. Marsden (Ed.), Sociological Methodology 1994 (pp. 1–72). Washington, DC: American Sociological Association. Nock, S. L. (1988). The family and hierarchy. Journal of Marriage and the Family, 50, 957–966. Parish, T. S., & Kappes, B. M. (1980). Impact of father loss on the family. Social Behavior and Personality, 8, 107–112. Popenoe, D. (1993). American family decline, 1960–1990: A review and appraisal. Journal of Marriage and the Family, SS, 527–541. Popenoe, D. (1996). Life without father. New York: Free Press. Rankin, J. (1983). The family context of delinquency. Social Problems, 30, 466–479. Rozondal, E. G. (1983). Halos versus stigmas: Long term effects of parent’s death or divorce on college students’ concepts of the family. Adolescence, 18, 947–955. Silverman, P. R., Nickman, S., & Worden, J. W. (1992). Detachment revisited: The child’s reconstruction of a dead parent. American Journal of Orthopsychiatry, 62, 494–503. Simons, R. L., Johnson, C., & Lorenz, F. O. (1996). Family structure differences in stress and behavioral predispositions. In R. L. Simons and associates, Understanding differences between divorced and intact families (pp. 45–64). Thousand Oaks, CA: Sage. Sugarman, S. (1993). Reforming welfare through social security. University of Michigan Journal of Law Reform, 26, 849–851. ———. (1995). Financial support of children and the end of welfare as we know it. Virginia Law Review, 81, 2523–2573. Thomson, E., Hanson, T. L., & McLanahan, S. S. (1994). Family structure and child well being: Economic resources vs. parental behaviors. Social Forces, 73, 221–242. Thomson, E., McLanahan, S. S., & Curtin, R. B. (1992). Family structure, gender, and parental socialization. Journal of Marriage and the Family, 54, 368–378. Trivets, R. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man. Chicago: Aldine de Gruyter. U.S. Administration for Children and Families. (1998). Temporary Assistance for Needy Families (TANF) program: First annual report to Congress. Washington, DC: U.S. Department of Health and Human Services. Wadsworth, M. E. J., & Maclean, M. (1986). Parents’ divorce and children’s life chances. Children and Youth Services Review, 8, 145–161. Wallerstein, J. S., & Kelly, J. B. (1980). Surviving the breakup: How children and parents cope with divorce. New York: Basic Books. White, L. K., & Booth, A. (1985). The quality and stability of remarriages: The role of stepchildren. American Sociological Review, 50, 689–698.

Chapter 6

Influences on Delinquency: The Role of the Single-Parent Family Amy L. Anderson*

Both individual- and aggregate-level theories suggest that family structure is an important factor related to delinquency. Although both levels have been examined in isolation, there is no research explicitly modeling both effects simultaneously. Furthermore, each level of explanation implies different things, increasing the importance of properly separating the two levels of effects. This study advanced the criminological literature concerning the link between single-parent families and delinquency by exploring the theoretically important relationships in a methodologically appropriate way, using a multilevel design in order to separate the two levels of effects. Furthermore, this research argues that schools are an appropriate unit for assessing contextual effects on delinquency. This argument is based on (1) research suggesting that delinquency is often a group event, and (2) the nature of school in terms of both pulling together same-aged children and creating a context favorable toward friendship formation and maintenance.

INDIVIDUAL-LEVEL RESEARCH There are many possible reasons why the absence of a parent in the home is associated with an adolescent’s risk for delinquency, such as lower income (McLanahan, 1985) or higher residential mobility (Astone & McLanahan, 1994). The prevailing criminological notion follows research

* A. L. Anderson. Individual and Contextual Influences on Delinquency: The Role of the Single-Parent Family. Journal of Criminal Justice, 2002, 30(6), 575–587. Reprinted with permission from Elsevier.

74

Social Diversity and Single-Parent Families

indicating that two parents are better able to care for, supervise, and socialize children than one parent (Amato & Keith, 1991; Hirschi, 1969; McLanahan & Sandefur, 1994). In general, both parents are important and the absence of one weakens family functioning. These general concepts are reflected in the notion of social control, particularly informal social control. The family is an important socializing and supervision agent, and the notion is that a child is exposed to lower levels of both types of social control when one of the two parents is missing (Gottfredson & Hirschi, 1990; Mulligan, 1960; Nye, 1957). Lower levels of supervision and effective socialization are associated with delinquency whether the child has one parent or two. Children with one parent are at higher risk of delinquency, then, because there is one less person capable of supervision. Thus, parental absence, in a very broad sense, is likely to reduce the level of social control to which the child is exposed. Recent criminological research concerning the effect of single-parent families on delinquency is rather sparse. Primarily due to backlash over the Moynihan Report in 1965, criminologists stopped looking directly at the relationship between broken homes and delinquency (Wilkinson, 1974). Instead, researchers focused on which family processes increased an adolescent’s risk for delinquency when weakened. This shift in focus created a base of research with mixed results because studies did not examine the direct effect of single-parent families. This variation is partly attributable to the difference between older research that examined the total effect of living in a single-parent family and current research that usually reports the family structure effect after controlling statistically for other variables. Two meta-analyses illustrate this variability. First, Loeber and Stouthamer-Loeber (1986) found 33 of 40 analyses with a strong, statistically significant association between broken homes or parental absence and a child’s delinquency or aggression for cross-sectional studies. In contrast, Lipsey and Derzon (1998) reported broken homes as one of the weakest predictors of violent or serious delinquency for ages six to fourteen. Neither of these reviews discussed whether the effect sizes for family structure on delinquency are direct, indirect, or total effects. Both reviews suggested that living in a singleparent family increased an adolescent’s risk for delinquency when compared with living with two parents. In sum, most individual-level studies of delinquency include a measure of family structure, but only as a control variable. Nevertheless, there is evidence that living in a single-parent family does increase an adolescent’s risk for delinquency as compared with living with two parents. Unfortunately, current research has not adequately explored this relationship.

Influences on Delinquency: The Role of the Single-Parent Family

75

AGGREGATE-LEVEL RESEARCH There is also an aggregate-level relationship posited between family structure and the delinquency rates of an area. The aggregate-level relationship is different from the individual-level relationship discussed above. Specifically, whereas individual-level research explores the effects of living with one parent on the delinquent behavior of a particular adolescent, aggregate-level research explores how the proportion of singleparent families within some social unit potentially place a social unit, and by extension all children within that unit, at risk for higher rates of delinquency. Currently, most aggregate-level research stems from social disorganization theory and uses the neighborhood as the unit of analysis. The essence of social disorganization theory is that, at the aggregate level, informal social control mechanisms are not working effectively; the community is not able to manage the behavior of its residents (Bursik, 1988; Bursik & Grasmick, 1993, p. 39; Sampson & Groves, 1989, p. 777; Sampson & Lauritsen, 1994, p. 44). The ability of a community to overcome common problems is hindered when formal and informal ties are not developed. Thus, the community is not able to rally against criminogenic forces that result from the weak levels of control because there is a deficit in the collective monitoring of individuals. For example, one important dimension of Shaw and McKay’s (1969) framework of social disorganization was the ability to control peer groups (see Sampson & Groves, 1989, p. 778; Sampson & Lauritsen, 1994, p. 58). Peer groups are of considerable importance because delinquency often is a group event (Shaw & McKay, 1942; Thrasher, 1927). The ability of a community to control these peer groups is thought to be directly and positively correlated with the community’s delinquency rates. Shaw and McKay (1942, p. 20) suggested that residential mobility, poverty, and ethnic/racial heterogeneity undermined formal and informal community ties by decreasing communication and increasing anonymity among residents, thereby decreasing the chances someone will intervene to control the behavior of children (Bursik & Grasmick, 1993, p. 7). In 1987, Sampson argued that family disruption affects crime in a similar fashion. One reason was that a high level of single-parent families in a community weakens informal social controls (Sampson, 1987, p. 351; Sampson & Lauritsen, 1994, p. 56). Informal controls are those controls most likely to affect unsupervised peer groups, one of the leading predictors of higher neighborhood delinquency rates (Sampson & Groves, 1989, p. 778). Essentially, due to the presence of many households with absent adults, there are fewer adults available for the day-to-day monitoring of their own children and other children in the area (Bursik & Grasmick, 1993; Sampson, 1987, p. 353). Informal social controls are weakened when there are high numbers of missing parents. This reasoning

76

Social Diversity and Single-Parent Families

suggests an emergent property of single-parent families. That is, the adolescent in the setting with many single-parent families is at higher risk for delinquency than the adolescent in the setting with few single-parent families, regardless of any specific adolescent’s family structure. Interestingly, some aggregate-level research points to family disruption as a leading predictor of juvenile delinquency, above and beyond the “usual” social disorganization variables, such as unsupervised peer groups and other formal and informal social controls. For example, Sampson and Groves (1989, p. 789) tested social disorganization theory using neighborhood characteristics and found that family disruption was significantly related to rates of various types of crime after controlling for mediating measures of social disorganization (local friendship networks, unsupervised peer groups, and organizational participation). A reexamination by Veysey and Messner (1999) found that after including other structural and mediating variables generally thought to influence levels of social organization, the structural effect of living in a singleparent family is still large. Those neighborhoods with high proportions of single-parent families also experience higher rates of delinquency. Moreover, this effect also holds for rural areas. Osgood and Chambers (2000) tested social disorganization theory in rural areas and found that a high proportion of female-headed households is associated with higher levels of rural violent delinquency, excluding homicide. Thus, there is aggregatelevel support for a significant relationship between the proportion of single-parent families in a setting and the delinquency rates of that setting. In sum, social disorganization theory indicates that the supervision of children by families is an important buffer against high rates of delinquency. At the individual level, a child in a single-parent home may be at higher risk for delinquency because fewer controls are placed over the child due to the absence of an adult in the home. Within the social disorganization framework, this notion is extended to a collective level: if there is a high proportion of single-parent homes in a social setting (e.g., neighborhood), the youths within the setting are at higher risk for delinquency regardless of their particular family arrangement. In such settings, delinquency rates are higher due in part to the ineffectual level of control over all the residents including, and especially, adolescents. Alternatively, there is some benefit for all children when intact families surround them.

MULTILEVEL RESEARCH As discussed above, there is evidence that children in single-parent families are at higher risk for delinquency and that areas with high proportions of single-parent families have higher delinquency rates. However, research conducted at either the individual or the aggregate level is

Influences on Delinquency: The Role of the Single-Parent Family

77

susceptible to cross-level misspecification. Cross-level misspecification refers to assuming the unit of analysis determines the unit of causation (Sampson & Lauritsen, 1994, p. 80). In statistical terms, the coefficients that are obtained with research conducted at either level are pooled estimates and may contain any proportion of influence from either level. The impossibility of differentiating the two levels is also the source of the ecological fallacy, which is improperly inferring individual effects from aggregate data (Robinson, 1950). Aggregate studies of delinquency explain variation in delinquency rates by the collective characteristics of individuals in the social unit, but say nothing about which people are involved or affected (see Sampson & Lauritsen, 1994, p. 3). The current research used individual and aggregate or contextual level data in order to separate the two different effects and to explore the relationship between single-parent families and delinquency. Multilevel research has not examined the total relationship between both levels of family structure and delinquency. One reason is that many researchers take a factor analytic approach, which combines a number of aggregate-level variables into one ”factor” that represents the variables’ dominant theme (Elliott et al., 1996; Sampson, Morenoff, & Earls, 1999; Sampson, Raudenbush, & Earls, 1997; Taylor & Covington, 1988). The factor analytic approach is useful particularly in ecological research because many aspects of neighborhoods are related (e.g., minority population and poverty rates). Specifically, collapsing several correlated variables into a single factor adjusts for the multicolinearity, yielding stable regression coefficients and greater statistical power. However, the result of this approach is that the independent effects of particular components of a factor cannot be determined. Such is the case for family disruption; multilevel studies include some aggregate measure of family disruption, but it is combined with other variables in a larger factor rendering interpretation impossible. Thus, there is a lack of multilevel research concerning the relationship between single-parent families and delinquency. As mentioned earlier, most contextual research is geographic in nature. The assumption is that individuals in close proximity to each other are likely to be similarly affected by the structural conditions of the spatial unit. Large spatial areas may be too geographically dispersed, however, to provide meaningful insight into contextual effects. Alternatively, macrolevel (or emergent) effects are evident in many contexts that are not necessarily rooted in residence. Based on the group nature of delinquency and the structure of schools for creating and maintaining friendship networks, the current research argues that schools are an appropriate context for examining delinquency. The family is a primary context for child development but it does not exist in isolation; rather, contexts of development are nested within other contexts (Bronfenbrenner, 1986). Literature on adolescent development

78

Social Diversity and Single-Parent Families

suggests that three of the most important contexts in which adolescents are embedded are the family, peers, and school (Steinberg & Darling, 1994; Vazsonyi & Flannery, 1997). Due to the natural interaction between the family and school, these two contexts are often used to predict adolescent developmental outcomes such as school achievement (Astone & McLanahan, 1991; Thompson, Alexander, & Entwisle, 1988). For example, Pong (1997, 1998) examined the individual and school contextual effect of single-parent families on the reading and math achievement of each student. She found that schools with high proportions of single-parent families created a significant negative contextual effect on the math and reading achievement of eighth and tenth graders, over and above the effect of an adolescent’s membership in a single-parent family. Criminologists seldom examine schools in multilevel research although they may be important for understanding delinquency. Schools bring together large groups of same-aged children to an environment with a highly skewed adult-child ratio, leaving children more room to operate away from watchful adults (Corsaro & Eder, 1990, 1995). Importantly, schools offer an opportunity for adolescents to form peer groups or to maintain peer groups already formed before they enter into school. Indeed, Ennett and Bauman (1993) reported that 95 percent of friendship ties were between students attending the same school. There is extensive criminological literature concerning the group nature of delinquency (Sampson & Groves, 1989, p. 778; Shaw & McKay, 1942, p. 166; Warr, 1996; Zimring, 1981). Moreover, studies of social disorganization suggest that the presence of unsupervised peer groups is one of the strongest predictors of juvenile delinquency (Sampson & Groves, 1989, p. 798; Veysey & Messner, 1999, p. 169). Given that unsupervised peer groups contribute to high neighborhood delinquency rates, delinquency is often a group event, and school is the dominant setting for friendship formation and/or maintenance, it seems reasonable to expect the school to play a part in an adolescent’s risk for delinquency. The purpose in the current study was not to determine the impact of the school environment; rather, schools were used to identify aggregates of adolescents who were available to one another as potential companions (see also Osgood & Anderson, 2002). To date, two multilevel studies in criminological research used the school as a context for delinquency. The first examined the effects of academic values and the subculture of violence on interpersonal violence, theft/vandalism, and rule breaking (Felson, Liska, & South, 1994). A measure of family stability was included at both the individual and school level as control variables, and both relationships were small and insignificant across all three dependent variables. The second employed an HLM analysis in order to determine individual, school, and community effects on school disorder (Welsh, Greene, & Jenkins, 1999). A factor analytic approach was taken with the concentration of single-parent families

Influences on Delinquency: The Role of the Single-Parent Family

79

included in the community poverty factor, rendering interpretation impossible. It was evident that multilevel studies had not addressed the question of family structure effects on delinquency. The current research fills an empirical gap by describing the total effects of single-parent families on delinquency from both an individual and a structural perspective. This research addresses two questions: First, are children from single-parent families more at risk for delinquency compared with children who live with two parents? Second, are adolescents who attend a school with a high proportion of single-parent families at higher risk for delinquency than those who attend school with mostly two-parent families? This question implies that all children are at risk for delinquency in schools with a high proportion of single-parent families, regardless of family structure. DATA The data used to examine the individual and contextual effects of singleparent families on delinquency was a sample from the Gang Resistance Education and Training (GREAT) evaluation (for details, see Esbensen & Osgood, 1999, p. 201). The original sample consisted of 5,935 eighth-grade students, aged 13 to 15 from 42 schools at 11 sites. Attendance was necessary for survey completion. The attendance rates across schools on the day of the survey varied from 75 to 93 percent; 98 to 100 percent of the students at these sites participated. These schools were selected because they offered the program being studied in the evaluation and so could not be considered random. However, the sample was diverse at both the individual and aggregate levels. The data were cross-sectional and based on self-reported measures. Schools with fewer than 40 respondents were excluded as the low number of students called into question the reliability of the estimate (eliminating 176 respondents). One site did not include the school codes and was excluded, eliminating 408 respondents. All adolescents with missing data on any of the independent variables (except one, see below) were excluded from this analysis (n  680). Due to these restrictions, the sample for this research was reduced to 4,671 respondents from 35 schools (10 sites). The average number of children in each school was 135, ranging from 40 to 389 students. RESULTS While controlling for gender, minority group status, and mother’s education, findings suggest that adolescents living in a single-parent family were at significantly higher risk for status, property, and person delinquency than adolescents who lived with two parents.

80

Social Diversity and Single-Parent Families

The school-level, single-parent family variable was significant for person offenses (P < .05), and was marginally significant for status and property offenses (P < .10). This suggests that students attending school where a high proportion of students live with only one parent are at significantly higher risk for person delinquency than those attending school where a low proportion of students live with only one parent, and they were somewhat more at risk for committing status offenses. The results also indicate that adolescents attending schools with a high proportion of single-parent families were less likely to commit property offenses. For the control variables, being male was significantly related to all three types of delinquency. Minority group members were significantly more at risk for status and person offenses but were not significantly related to property offending. Mother’s education has a borderline significant relationship to status and property offending, with adolescents whose mothers had more education being at lower risk. Status and person offenses follow the same general pattern, with adolescents living with two parents and attending school where only 9 percent of students who live with one parent are least at risk. Adolescents who live with two parents but attend school where 74 percent of the students live with only one parent commit slightly more status and person offenses than adolescents who live with only one parent and attend school where 91 percent of the students live with two parents. This suggests that attending school where a higher proportion of the students have two-parent families can reduce some of the risk involved with an adolescent living in a single-parent family for these two offenses. Additionally, the average number of property offenses is lowest for those living with two parents but attending a school where 74 percent of the adolescents live with only one parent, followed by the adolescents who live with only one parent but attend a school where 74 percent of the adolescents live with only one parent. The highest risk offenders were those who attended school where 91 percent of the kids lived with two parents, with the highest risk offender living with only one parent. Structural Disadvantage The results indicate that an adolescent who attended a school with a high proportion of single-parent families was at significantly higher risk for person delinquency than an adolescent in a school with a low proportion, even after controlling for school-level disadvantage. For status offenses, the magnitude of the single-parent family effect is slightly smaller than in the full model discussed earlier. The coefficient for property offenses remained negative and increased, meaning the effect became smaller and further reduced the risk of committing a property offense for those attending school with a high proportion of single-parent families.

Influences on Delinquency: The Role of the Single-Parent Family

81

Limitations There are several threats to validity that warrant mentioning. First, due to the cross-sectional nature of the data, there is no way of determining whether single-parent families “cause” delinquency, or whether delinquency “causes” single-parent families, creating ambiguity about the direction of causal influence (Cook & Campbell, 1979). Although it is possible that delinquency affects family structure, chances are low that the effect is strong enough to render this analysis futile. Second, there is no way to discern whether the self-reported delinquent acts were committed while attending the present school or a different school. The overall movement of all students among schools, however, may work to cancel each other out, not greatly affecting the results. At worst, the current analysis suffers from a reduction in accuracy. There may be some selection issues concerning the sample and students. Specifically, students filled out the interview in school, and those students who were absent the day it was administered were not included. These absent students might also be more delinquent than the children who were attending that day. Finally, these schools were chosen as part of an evaluation effort. Thus, the sample students may not be representative of eighth grade students and the generalizability of the results is suspect. Nevertheless, the schools vary on many community level dimensions, such as size of school and minority population, potentially minimizing bias (see Esbensen & Osgood, 1999, p. 201).

DISCUSSION AND CONCLUSION The current research sought to determine whether there was an individual- and aggregate-level relationship between family structure and delinquency. This research is grounded in theoretical and empirical evidence that adolescents placed at higher risk for delinquency both if they live in a single-parent family and if they are in a context with a high proportion of adolescents who live with only one parent. The current research used multilevel analyses to test these notions about family structure and found evidence of both types of family structure effects. Three measures of self-reported delinquency were used in this analysis: status, property, and person offenses. The results suggest independent effects for both living in a single-parent family and attending school where a higher proportion of adolescents live with only one parent. An important thing to notice about the results is that it matters how many single-parent families a student is exposed to, regardless of whether the student has one or two parents in the home. It appears that there is an effect of being surrounded by intact families, for status and person

82

Social Diversity and Single-Parent Families

offending. The effect for property offending went in the opposite direction. That is, kids in one-parent families were likely to commit more property offenses than those in two-parent families, but adolescents who attended schools where a high proportion of students live with only one parent were less likely to commit property offenses than adolescents who attended schools where a high proportion of students live with two parents. It could be that there were fewer things worth stealing as the proportion of single-parent families increased, or that the middle-class children were the ones committing property crime. The current study used a different context than most contextual- or aggregate-level research of this type. In particular, the use of schools as the context is an interesting yet meaningful way of grouping adolescent children and is a context that is particularly salient to youth. There is much developmental and educational research suggesting the importance of the meaning of school in children’s lives as they grow, including the declining amount of time spent with the family relative to friends (e.g., Larson & Richards, 1991), the increase in the relative importance of friends in school and the function of schools themselves in pulling together same-aged children. With these notions in mind, the school context itself is not the focus of this article. Rather, schools serve as an important marker for grouping children as well as providing a pool of potential companions for adolescents. REFERENCES Agresti, A. (1996). An introduction to categorical data analysis. New York: Wiley. Amato, P. R., & Keith, B. (1991). Parental divorce and the well-being of children: A meta-analysis. Psychological Bulletin, 110, 26–46. Astone, N. M., & McLanahan, S. (1991). Family structure, parental practices and high school completion. American Sociological Review, 56, 309–320. Astone, N. M., & McLanahan, S. (1994). Family structure, residential mobility, and school dropout: A research note. Demography, 31, 575–584. Blalock, H. M. (1984). Contextual-effects models: Theoretical and methodological issues. Annual Review of Sociology, 10, 353–372. Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: research perspectives. Developmental Psychology, 22, 723–742. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods (Vol. 1). Newbury Park, CA: Sage Publications. Bryk, A. S., Raudenbush, S. W., & Congdon, R. (1996). Hierarchical linear and nonlinear modeling with HLM/ 2L and HLM/3L programs. Chicago: Scientific Software International. Bursik, R., & Grasmick, H. G. (1993). Neighborhoods and crime: The dimensions of effective community control. New York: Lexington Books. Bursik, R. J., Jr. (1988). Social disorganization and theories of crime and delinquency: Problems and prospects. Criminology, 26, 519–551.

Influences on Delinquency: The Role of the Single-Parent Family

83

Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Boston: Houghton Mifflin. Corsaro, W. A., & Eder, D. (1990). Children’s peer cultures. Annual Review of Sociology, 16, 197–220. Corsaro, W. A., & Eder, D. (1995). Development and socialization of children and adolescents. In K. S. Cook, G. A. Fine, & J. S. House (Eds.), Sociological perspectives on social psychology (pp. 421–451). Boston: Allyn and Bacon. Elliott, D. S., Wilson, W. J., Huizinga, D., Sampson, R. J., Elliott, A., & Rankin, B. (1996). The effects of neighborhood disadvantage on adolescent development. Journal of Research in Crime and Delinquency, 33, 389–426. Ennett, S. T., & Bauman, K. E. (1993). Peer group structure and adolescent cigarette smoking: A social network analysis. Journal of Health and Social Behavior, 34, 226–236. Esbensen, F. A., & Osgood, D. W. (1999). Gang Resistance Education and Training (GREAT): Results from the national evaluation. Journal of Research in Crime and Delinquency, 36, 194–225. Felson, R. B., Liska, A. E., & South, S. J. (1994). The subculture of violence and delinquency: Individual vs. school context effects. Social Forces, 73, 155–173. Gardner, W., Mulvey, E. P., & Shaw, E. C. (1995). Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models. Psychological Bulletin, 118, 392–404. Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press. Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press. Iversen, G. R. (1991). Contextual analysis. Newbury Park, CA: Sage Publications. Kreft, I. G., & De Leeuw, J. (1998). Introducing multilevel modeling. Thousand Oaks, CA: Sage Publications. Larson, R., & Richards, M. H. (1991). Daily companionship in late childhood and early adolescence: Changing developmental contexts. Child Development, 62, 284–300. Liao, T. F. (1994). Interpreting probability models: Logit, probit, and other generalized linear models. Thousand Oaks, CA: Sage Publications. Lipsey, M. W., & Derzon, J. H. (1998). Predictors of violent and serious delinquency in adolescence and early adulthood. In R. Loeber & D. P. Farrington (Eds.), Serious and violent juvenile offenders: risk factors and successful interventions (pp. 86–105). London: Sage Publications. Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors as correlates and predictors of juvenile conduct problems and delinquency. In M. Tonry & N. Morris (Eds.), Crime and justice: an annual review of the research (vol. 7, pp. 29–150). Chicago: University of Chicago Press. McLanahan, S. (1985). Family structure and the reproduction of poverty. American Journal of Sociology, 90, 873–901. McLanahan, S., & Sandefur, G. D. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press. Mulligan, R. (1960). Theory and juvenile delinquency. Journal of Educational Sociology, 33, 365–372. Nye, I. F. (1957). Child adjustment in broken and in unhappy homes. Marriage and Family Living, 19, 356–361.

84

Social Diversity and Single-Parent Families

Osgood, D. W. (2000). Poisson based regression analysis of aggregate crime rates. Journal of Quantitative Criminology, 16, 21–43. Osgood, D. W., & Anderson, A. L. (2002). Routine activities and differences among schools in rates of delinquency. Unpublished manuscript. Osgood, D. W., & Chambers, J. M. (2000). Social disorganization outside the metropolis: An analysis of rural youth violence. Criminology, 38, 81–115. Pong, S. L. (1997). Family structure, school context and eighth-grade math and reading achievement. Journal of Marriage and the Family, 59, 734–746. ———. (1998). The school compositional effect of single parenthood on 10th-grade achievement. Sociology of Education, 71, 23–42. Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15, 351–357. Sampson, R. J. (1987). Urban Black violence: The effect of male joblessness and family disruption. American Journal of Sociology, 93, 348–382. Sampson, R. J., & Groves, B. (1989). Community structure and crime: Testing social disorganization theory. American Journal of Sociology, 94, 774–802. Sampson, R. J., & Lauritsen, J. L. (1994). Violent victimization and offending: Individual level, situational, and community-level risk factors. In A. J. Reiss (Ed.), Understanding and preventing violence (pp. 1–114). Washington, DC: National Academy Press. Sampson, R. J., Morenoff, J. D., & Earls, F. (1999). Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review, 64, 633–660. Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924. Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas: A study of rates of delinquents in relation to differential characteristics of local communities in American cities. Chicago: University of Chicago Press. ———. (1969). Juvenile delinquency and urban areas: A study of rates of delinquents in relation to differential characteristics of local communities in American cities (Rev. ed.). Chicago: University of Chicago Press. Steinberg, L., & Darling, N. (1994). The broader context of social influence in adolescence. In R. K. Silbereisen & E. Todt (Eds.), Adolescence in context: The interplay of family, school, peers, and work in adjustment (pp. 25–49). New York: Springer-Verlag. Taylor, R. B., & Covington, J. (1988). Neighborhood changes in ecology and violence. Criminology, 26, 553–589. Thompson, M. S., Alexander, K. L., & Entwisle, D. R. (1988). Household composition, parental expectations, and school achievement. Social Forces, 67, 424–451. Thrasher, F. M. (1927). The gang: A study of 1,313 gangs in Chicago. Chicago: University of Chicago Press. Vazsonyi, A. T., & Flannery, D. J. (1997). Early adolescent delinquent behaviors: Associations with family and school domains. Journal of Early Adolescence, 17, 271–293. Veysey, B. M., & Messner, S. F. (1999). Further testing of social disorganization theory: An elaboration of Sampson and Groves’s “community structure and crime.” Journal of Research in Crime and Delinquency, 36, 156–174.

Influences on Delinquency: The Role of the Single-Parent Family

85

Warr, M. (1996). Organization and instigation in delinquent groups. Criminology, 34, 11– 37. Welsh, W. N., Greene, J. R., & Jenkins, P. H. (1999). School disorder: The influence of individual, institutional, and community factors. Criminology, 37, 73–115. Wilkinson, K. (1974). The broken family and juvenile delinquency: Scientific explanation or ideology? Social Problems, 21, 726–739. Zimring, F. (1981). Kids, groups, and crime: Some implications of a well-known secret. Journal of Criminal Law and Criminology, 72, 867–885.

Chapter 7

Parental Incarceration: Implications of Child Welfare Elizabeth I. Johnson and Jane Waldfogel*

Although advocates, child welfare workers, and, more recently, academics have become concerned about the challenges posed for children, families, and the child welfare system by parental incarceration, relatively little is known about the magnitude and growth of parental incarceration. Estimates of the proportion of inmates who are parents vary widely (Johnston, 1995). Part of this disparity is definitional; some studies count parents with children of any age, while others include only parents with minor children (Schafer & Dellinger, 1999). Perhaps a larger share of the problem, however, is methodological. Studies rely on various data sources, including national surveys, samples from different regions of the country, and personal interviews with small samples of inmates (usually women). It has been difficult to use these disparate data to identify national trends over time.

BACKGROUND A considerable portion of the research on incarcerated parents focuses on their children, especially children whose mothers are incarcerated. These studies suggest that parental incarceration can negatively affect emotional, behavioral, and psychological development (Baunach, 1985; Bloom & Steinhart, 1993; Stanton, 1980). For instance, problems such as aggressive behavior and withdrawal (Baunach, 1985); criminal

* E. I. Johnson and J. Waldfogel. Parental Incarceration: Recent Trends and Implications for Child Welfare. Social Service Review, 2002, 76(3), 460–479. Reprinted with permission from University of Chicago Press.

Parental Incarceration: Implications of Child Welfare

87

involvement (Johnston, 1991, 1992); and depression, difficulty sleeping, and concentration problems (Kampfner, 1995) have been reported among children whose parents are incarcerated. There has also been some attention paid to where children are placed when a parent goes to prison. There is evidence that child placements depend on the gender of the parent, with children of incarcerated mothers more likely to be displaced from the home than children of incarcerated fathers (Johnston, 1991; Mumola, 2000). Though living with a grandparent is the most common arrangement for children of incarcerated mothers (Mumola, 2000), a larger share of these children end up in the child welfare system than children of incarcerated fathers (Bloom & Steinhart, 1993; Johnston, 1991; Mumola, 2000). Estimates of the number of children with incarcerated parents who end up in the foster care system vary. For children with mothers in prison, these estimates range from 10 percent (Johnston, 1993) to 14 percent (McGowan & Blumenthal, 1978). But research also indicates that children of incarcerated parents may be at risk long before their parents are incarcerated. The U.S. Department of Justice reports that nearly 60 percent of women in state prisons used drugs in the month prior to their offense and that approximately 50 percent described themselves as regular substance users (Greenfeld & Snell, 1999). Moreover, 65 percent of women and 77 percent of men in state prisons have a history of prior convictions (Greenfeld & Snell, 1999). Histories of sexual and physical abuse, mental illness, and parental incarceration (U.S. Department of Justice, Bureau of Justice Statistics, 1993) and low socioeconomic status (Baunach, 1985; Bloom & Steinhart, 1993; Kampfner, 1995) are also common among inmates. Overall, research indicates the presence of multiple risk factors in the lives of incarcerated parents and their children. Concern about the long-term consequences of incarceration for children and families has been heightened by the fact that child welfare legislation, such as the Adoption Assistance and Child Welfare Act of 1980 (P.L. 96–272) and the Adoption and Safe Families Act of 1997 (ASFA; P.L. 105–89), has created tensions between what parents are expected to do and what incarcerated parents are capable of doing from the confines of prison (Beckerman, 1989, 1991, 1998; Genty, 1995, 1998). Though barriers to fulfilling legal obligations for incarcerated parents are not new, ASFA heightened these tensions by imposing a requirement that states must file a petition to terminate parental rights when a child has been in state care for 15 of the last 22 months. Given the increase in sentence lengths and amount of time served, this requirement may be particularly consequential for incarcerated parents (Genty, 1998). The ASFA permits states to opt out of this requirement if the child is living with a relative, though data from the National Conference of State Legislatures suggests that not all

88

Social Diversity and Single-Parent Families

states are exercising this right (National Conference of State Legislatures, 2001). There is also some evidence that agency-level policies and practices heighten the tensions between the demands of permanency planning and the constraints imposed by parental incarceration. In one of the first studies of the role of child welfare services in correctional facilities, Dorothy Zietz (1963) discovered a general lack of information and confusion about mothers’ rights and responsibilities (among mothers and caseworkers). Brenda McGowan and Karen Blumenthal (1978) were also early contributors to this literature, providing estimates on the number of children with parents in prison, describing children’s living arrangements during incarceration, and delineating the unique service needs of incarcerated parents. This work also revealed limited communication between mothers and caseworkers, particularly regarding legal issues. More recently, Adela Beckerman (1994) observes that many incarcerated mothers with children in foster care received no correspondence (49 percent) from caseworkers and were often uninformed of child custody hearings (28 percent). Twothirds (66 percent) of mothers reported not receiving a copy of their child’s case plan. Moreover, there are often no specific child welfare policies for dealing with children of incarcerated parents. A relatively recent survey of 500 child welfare, law enforcement, and correctional officials (Smith & Elstein, 1994) revealed that 80 percent of protective service agencies surveyed had no specific policies or guidelines for placing children whose mothers had been arrested. Although more than 50 percent of child welfare administrators acknowledged an increase in the number of children of incarcerated parents requiring placement, most reported a lack of formal procedures for working with this population. A 1997 survey by the Child Welfare League of America (1998) revealed low levels of state-sponsored services and policies for incarcerated parents and their families. For example, of the 38 states that responded, only 6 percent reported specific policies pertaining to children with incarcerated parents. Only 25 states provide transportation for children to visit incarcerated parents (Child Welfare League of America, 1998), even though the majority of incarcerated parents (62 percent of parents in state prison and 84 percent of parents in federal prison) are incarcerated more than 100 miles away from their homes (Mumola, 2000). Before considering an adequate policy and service response, it is first necessary to determine the number of parents in prison and the number of their children. Information on how parents’ demographic characteristics and children’s living arrangements have changed over time will also contribute to understanding the problem. This study builds on existing work by examining these issues on a national scale from 1986 to 1997.

Parental Incarceration: Implications of Child Welfare

89

DATA AND METHOD We analyze trends in parental incarceration using the 1986 and 1991 Survey of Inmates in State Correctional Facilities and the 1997 Survey of Inmates in State and Federal Correctional Facilities (U.S. Department of Justice, Bureau of Justice Statistics 1993b, 1994, 2000). The 1986 and 1991 surveys contain data on state inmates only (separate surveys collected data on federal inmates in these years); the 1997 survey contains data on both state and federal inmates. Except for estimates on the growth of parents in prison that utilize data on state and federal inmates, our substantive analyses are limited to state inmates. Inmate surveys have been conducted by the U.S. Census Bureau every five years since 1974. Examining the three most recent waves should sufficiently capture the dramatic increase in rates of incarceration of parents (especially mothers), which is largely thought to be a function of the shift in national drug policy that occurred during the mid- to late-1980s (i.e., the Anti Drug Abuse Acts of 1986 and 1988 and the Sentencing Reform Act of 1984 that changed sentencing laws) concurrent with the rapid increase in use of crack cocaine. The inmate surveys contain detailed information on inmates’ criminal history, drug and alcohol use, prison activities, conditions of confinement, family background, demographic characteristics, and a number of other variables. For 1986, 1991, and 1997, personal interviews were conducted in about 270–275 correctional facilities, yielding large samples of between 13,000 and 17,000. A stratified, two-stage selection process was used, whereby prisons were selected first and inmates in sampled prisons were subsequently sampled. Second-stage response rates were high—approximately 93 percent for all three years. Each year contains a roughly comparable set of variables and, when weighted, is nationally representative of inmates in state prisons. Weights provided by the Bureau of Justice Statistics were used in all analyses. Parents were selected for analysis if they had at least one child under the age of 18. For estimates of the number of children with parents in state prison, we aggregated the number of children per inmate per year.1 The proportion of children in each living arrangement was calculated by weighting arrangements (which are reported at the level of the family rather than the child) by the parent’s number of children. As noted above, most of the analyses in this article refer to children of state prisoners only. However, we do use data on federal inmates to estimate the overall magnitude of the problem. For these analyses, figures provided by the Bureau of Justice Statistics’ Correction Population in the United States, 1986 and 1991 (U.S. Department of Justice, 1989, 1993a) and the Incarcerated Parents and Their Children report (Mumola, 2000) were used to estimate the number of parents in federal prison for 1986 and 1991; data on federal

90

Social Diversity and Single-Parent Families

inmates for 1997 were obtained from the 1997 Survey of Inmates in State and Federal Correctional Facilities (U.S. Department of Justice, Bureau of Justice Statistics, 2000). RESULTS Growth in the Number of Parents in Prison The number of parents in prison more than doubled during this 11-year period, from 273,045 in 1986 to 637,309 in 1997. The number of children with a parent in prison also rose dramatically, from a little over 600,000 in 1986 to more than 1.3 million in 1997. Put differently, about 10 in every 1,000 U.S. children had a parent in state or federal prison in 1986, while nearly 20 in every 1,000 children had a parent in prison during 1997. The number of women in prison more than tripled during this period. We turn next to analyses of state prisoners, who make up the bulk of inmates and for whom we have comparable data for 1986, 1991, and 1997. In the following sections, we use these data to analyze the characteristics of the population of parents in state prison and how those characteristics have changed over time. Demographic Changes On average, parents incarcerated in state prisons during 1997 had fewer children, were less likely to be married or previously married, and were better educated than parents incarcerated in state prisons during 1986. Larger shares of both mothers and fathers report Hispanic ethnicity in 1991 and 1997 than in 1986. A slightly larger percentage of both mothers and fathers identified as African American in 1997 than in 1986. Across all three points in time, more mothers than fathers report having custody of children prior to incarceration. However, fewer mothers and fathers reported living with their children in 1997 than in 1986 or 1991: 78 percent of 1986 mothers, 71.7 percent of 1991 mothers, and only 64.3 percent of 1997 mothers lived with their children prior to incarceration. The percentage of fathers who lived with their children prior to incarceration also decreased, from 50.5 percent in 1986 to 43.8 percent in 1997. Family Background Characteristics The majority of parents incarcerated in state prison grew up in a twoparent home with a mother and father. The second most common childhood living arrangement was with a lone mother. The proportion of parents (both mothers and fathers) who lived primarily in foster care or

Parental Incarceration: Implications of Child Welfare

91

in the care of agency declined from 1986 to 1997: 3.3 percent of mothers and 2.4 percent of fathers in 1986 declared foster or agency care as the primary living arrangement in 1986, while only 2.2 percent and 1.2 percent did for 1997. Similarly, the proportion of inmates who ever lived in foster care declined slightly from 1991 (16.3 percent for mothers, 14.6 percent for fathers) to 1997 (15.2 percent for mothers, 12.8 percent for fathers).2 Parents incarcerated in state prisons during 1997 identified risk factors such as a history of physical and sexual abuse and incarceration of their own parents with greater frequency than did parents incarcerated in state prisons during 1986 and 1991. For instance, 36.1 percent of 1986 mothers and 47.6 percent of 1997 mothers report having ever been physically abused. This increase was less pronounced for fathers: 10.4 percent of 1986 fathers and 12.8 percent of 1997 fathers report a history of physical abuse. The reported prevalence of sexual abuse also increased for both mothers and fathers. Overall, reported physical or sexual abuse histories are much more prevalent among mothers, though part of this discrepancy may reflect underreporting by fathers. Nonetheless, it is noteworthy that reported rates of abuse increased for both men and women between 1986 and 1997. Considerably more parents in state prison reported a history of their parents having been incarcerated in 1997 than in 1986. Only 3.1 percent of mothers incarcerated in state prisons during 1986 reported having a mother incarcerated, compared with 8.6 percent of mothers in state prisons in 1997. The same pattern exists for fathers: 1.3 percent of fathers incarcerated during 1986 and 4.7 percent of those incarcerated during 1997 reported having a mother who was incarcerated. The figures for paternal incarceration are also substantially higher for 1997 than for 1991 and 1986. When we consider the share of parents who had either parent incarcerated, we find an especially sharp increase from 1986 to 1997: 8.5 percent of mothers and 7.1 percent of fathers in 1986 had either a mother or father incarcerated, compared with 22.6 percent of mothers and 18.8 percent of fathers in 1997. Drug Involvement and Current Offense In 1997, about 35 percent of mothers and 23 percent of fathers in state prison were incarcerated for drug-related crimes, compared with a mere 12.3 percent of mothers and 9.7 percent of fathers in 1986. For both mothers and fathers, this shift occurred primarily between 1986 and 1991, when 33.9 percent of mothers and 23.6 percent of fathers in state prison were incarcerated for drug-related offenses. Across years, the most common reason for fathers’ incarceration was a violent offense. For mothers, however, the most common offense involved violence in 1986 and drugs in both 1991 and 1997. The proportion of parents who reported at least

92

Social Diversity and Single-Parent Families

one period of incarceration prior to this incarceration rose steadily between 1986 and 1997: 39.3 percent of 1986 mothers and 57.6 percent of 1986 fathers reported being previously incarcerated, compared with 53.1 percent of 1997 mothers and 67.1 percent of 1997 fathers. Regular crack-cocaine use or other cocaine use also increased substantially between 1986 and 1997 among state prison inmates. In 1986, 26 percent of mothers and 24 percent of fathers reported ever having used crack regularly, compared with 54 percent of mothers and 36 percent of fathers in 1997. The proportion of inmates who reported ever having used heroin or other opiates decreased between 1986 and 1997. Histories of experimental and regular use of marijuana or hash increased slightly across these years. Children’s Living Arrangements and Correspondence with Parents The most common living arrangement for children with mothers in prison is with a grandparent, though the most common arrangement for children with a father in prison is with a mother or stepmother. For instance, nearly two-thirds (63.4 percent) of children with mothers in state prison were in the care of a grandparent or other relative in 1997, compared with only 12 percent of children with fathers in state prison during this year. This pattern was consistent across years. The proportion of incarcerated fathers’ children who lived with a parent was consistent from 1986 to 1991 (85 percent and 84.5 percent, respectively), though slightly lower for 1997 (82.2 percent). Correspondence variables are available only in the 1991 and 1997 surveys, so we are unable to determine how correspondence changed over the entire 11-year period. Data suggest, however, that the frequency of all forms of correspondence between parents in state prison and their children declined from 1991 to 1997. For instance, the proportion of mothers who reported monthly visits with children fell from 17.9 percent in 1991 to 14.3 percent in 1997. Similarly, the share of fathers who reported monthly personal visits declined from 14.8 percent in 1991 to 13.2 percent in 1997. More than half of parents reported no personal visits with children, and more than one-third reported no phone correspondence with children. Children of State and Federal Inmates as a Growing Share of Children Living in Foster Care and with Grandparents Here we look at the overall population of children of both state and federal inmates and analyze what share these children make up of children living in foster care and of children living with grandparents. We find that children with incarcerated parents became an increasingly large

Parental Incarceration: Implications of Child Welfare

93

share of the total U.S. nonrelative foster care population between 1986 and 1997 (which includes children living in foster care, an agency, or an institution). About 229,600 U.S. children were in out-of-home care with nonrelatives in 1986, compared with 337,000 in 1997.3 The share of U.S. children in nonrelative foster care with an incarcerated parent increased from 5.7 percent in 1986 to 7.1 percent in 1997, an increase of 25 percent over the level in 1986.4 This figure peaked in 1991, with children of incarcerated parents constituting 8.5 percent of the U.S. foster care population for that year. Thus, children with incarcerated parents became an increasingly large share of the foster care population between 1986 and 1997. Although the overall number of children of incarcerated parents living in foster care increased from 1986 to 1997, the proportion of mothers’ children living in out-of-home care (i.e., foster care, agency, or an institution) declined from 10.7 percent in 1986 to 7.9 percent in 1997. Over the same time period, proportionately more parents reported children living with a grandparent. This suggests that grandparents are caring for some of the children who previously would have been in foster care, which makes sense given the increase over this period in kinship care. Survey response categories make it impossible to discern whether these kinship arrangements are formal (i.e., grandparent subject to state regulations in exchange for foster care maintenance payments) or more informal arrangements. Nonetheless, it is clear from our data that children with incarcerated parents are becoming an increasingly large share of U.S. children living with a grandparent. Based on figures on the total number of U.S. children living with grandparents (U.S. Census Bureau, 2001), we estimate that children with an incarcerated parent made up 5.8 percent of U.S. children who lived with a grandparent caregiver without a parent present in 1986, 10.7 percent of this population in 1991, and 11.8 percent of this population in 1997.

DISCUSSION Although it is well known that the U.S. prison population has grown dramatically over the last two decades, it is less clear how the population of parents in prison has grown and changed. Our work examines this growth over an 11-year period using three large, nationally representative surveys of inmates in state correctional facilities (we also conduct some analyses of inmates in both state and federal facilities using data from the 1997 survey, which included both populations). We find that state and federal inmates were parents to over 1.3 million children in 1997—a near tripling of the 1986 figure. In our more detailed analyses of state prisoners, one of the most surprising demographic changes we observe is that fewer parents (both

94

Social Diversity and Single-Parent Families

mothers and fathers) had custody of fewer children prior to their incarceration in 1997 than in 1986. We also find, by several measures, that parents incarcerated in state prisons during 1997 were a much more troubled population than parents incarcerated during 1986. More of the 1997 parents reported histories of physical or sexual abuse, prior sentences to incarceration, incarceration of their own parents, and regular drug use. One interesting exception is that a larger share of 1997 parents completed high school or earned a GED prior to entering prison than did 1986 parents. Rates of high school and GED completion were highest in 1991, and the prevalence of a history of abuse was lowest for both mothers and fathers during 1991. These findings perhaps indicate that changes in drug policy initially caught a broader spectrum of people. Paralleling the latter part of the crack-cocaine epidemic and the shift in national drug control policy, the proportion of parents (especially mothers) who were incarcerated in state prisons for drug-related offenses increased dramatically between 1986 and 1991. A notable growth in the number of parents who regularly used crack cocaine and the number of parents who reported prior sentences to incarceration also occurred during this period. Taken together, these findings may help to explain why fewer parents had custody of fewer children in 1997 than in 1986. Parents may have already lost custody of their children for drug- or incarcerationrelated reasons prior to this incarceration. When we consider the overall population of children of parents incarcerated in state or federal prison, results indicate that children with incarcerated parents became an increasingly large share of the foster care population between 1986 and 1997. The proportion of U.S. children in nonrelative foster care who had an incarcerated parent increased by about 25 percent during this period. The largest increase occurred between 1986 and 1991, with 5.7 percent of foster care children having a parent in prison in 1986 and 8.5 percent in 1991. Although this figure declined slightly to 7.1 percent in 1997, it is nevertheless concerning that an increasing number of children of incarcerated parents are in the foster care system. Data also indicate that inmates’ own parents are assuming a considerable amount of care for this population of children, especially for children whose mothers are in prison. Children with a parent incarcerated in state or federal prison made up about 12 percent of the population of U.S. children living with a grandparent in 1997, compared with only 5.8 percent in 1986. Data from the National Survey of Households and Families indicates that more than half of grandparents raising their grandchildren do so for three or more years and that about one-quarter live below the poverty level (Fuller-Thomson, Minkler, & Driver, 1997). The fact that many grandparents are raising young grandchildren for several years on a limited budget, coupled with the more general stress of familial incarceration, suggests that policy makers and service providers

Parental Incarceration: Implications of Child Welfare

95

should give more thought to financial and supportive services for this population. Our analysis also suggests that some children who formerly would have been placed in foster care are now living with grandparents instead. It is important to determine what types of families these children come from and whether they face the same risk factors as children in foster care (Johnson & Waldfogel, 2002). Though concerns about visitation have been widely articulated, there remains considerable variability in how child welfare agencies and prison facilities deal with incarcerated parents and their families. The Bureau of Prisons “encourages visiting by family, friends, and community groups,” yet visitation regulations require only that facilities provide at least four hours of visiting per month (U.S. Department of Justice, Federal Bureau of Prisons, 1999). Child welfare agencies also have discretion in the extent to which they provide services to incarcerated parents. California and New York are examples of two states that have set special standards for the nature of “reasonable efforts” in the case of incarcerated parents (Genty, 1998), such as arranging visits and providing transportation. Though this is encouraging, we know little about whether states’ efforts have actually improved correspondence between inmates and their children. In fact, our data suggest that correspondence between state inmates and their children actually declined from 1991 to 1997. Given the potential importance of parent-child correspondence for child well-being and criminal justice outcomes (Boudouris, 1996; Hairston, 1991; Koban, 1983; Wooldredge, 1999), future research should examine factors that promote or impede parent-child contact during imprisonment. It would be helpful for practitioners and policy makers if researchers could tease apart the individual (e.g., demographic and social characteristics of parents, caregivers, and children) and institutional constraints (e.g., prison facilities, child welfare services) on correspondence. Given that most children of incarcerated parents are not under state care, it is important that more systematic reporting occur at the level of prison facilities and child welfare agencies. The Child Welfare League of America publicized the seriousness of the problem in its 1998 survey of state agencies and in the special issue of Child Welfare on parental incarceration. The survey was conducted in 1997; it would be useful if funding were provided to update this survey. Have efforts of child welfare agencies to address the needs of children of incarcerated parents been improved? Improved reporting of programs and policies regarding incarcerated parents would help us to determine whether states’ efforts to improve visitation have been successful. Social science research can also make important contributions by conducting additional studies on the consequences of parental incarceration for children. Such studies should look at the incarceration of fathers as

96

Social Diversity and Single-Parent Families

well as mothers. As one anonymous reviewer pointed out, much of the literature focuses on mothers and thus may underestimate the impact of parental incarceration on families and communities. Denise Johnston (2001) notes that much of the existing research on this topic has been hampered by conceptual and methodological issues such as a focus on the isolated offender, difficulty identifying representative samples, data accuracy, and assumptions about inmates and their families. By more carefully identifying the consequences of the problem for children and families, we can begin to explore child risk and protective factors. Although our analysis represents an important attempt to present a picture of the problem over time using large, nationally representative survey data, several limitations are noteworthy. First, it should be noted that our analyses are limited to children of parents in prison and do not include parents who are involved in the criminal justice system in other ways (e.g., in jail or on probation) or who have been incarcerated or involved with criminal justice in the past. Thus, we do not consider the full extent of involvement of children’s parents with the criminal justice system. Second, our analyses do not take into account the length of time that parents are in prison, which has increased over the period we studied (Ditton & Wilson, 1999; Hughes, Wilson, & Beck, 2001; Sabol & McGready, 1999), meaning that children are separated from their parents for longer periods of time. For all incarcerated parents, this may influence their ability to care for their children. For incarcerated parents whose children are not in the care of a relative, this may have important consequences for maintaining parental rights. A third concern is that we may have slightly overcounted the number of children with incarcerated parents if some children have two parents in prison. Because we know of no national estimates on the number of children with two incarcerated parents, we were unable to adjust our figures downward. A related issue is that we used rough estimates of the number of children in nonrelative foster care. Fourth, even though we use estimates on federal prisoners to get at the magnitude of the problem, our substantive analyses were limited to state prisoners. As an anonymous reviewer pointed out, state and federal inmates may differ in several important ways. Thus, our results should not be generalized to all inmates. Fifth, response categories used in the survey may have constrained our estimates, a problem often encountered with secondary analyses. For example, response categories do not distinguish between children who are living with relatives and those who are in formal kinship care arrangements. In assessing the impact of parental incarceration on the child welfare system (as well as on grandparents), this is an important distinction. A related concern is that survey data do not distinguish between children who were already in foster care prior to their parent’s incarceration and those who were placed in the child welfare system as a consequence of the parent’s incarceration. Nonetheless, our figures suggest that a share of U.S. children in the foster care population may be affected by incarceration at a point in time and thus may have special casework needs.

Parental Incarceration: Implications of Child Welfare

97

Despite these limitations, this article contributes to our understanding of how the problem of parental incarceration has grown and changed over time and confirms earlier work that suggests that children of incarcerated parents may be an especially vulnerable population. The implications of parental incarceration are far reaching. Not only does incarceration itself cost an enormous amount of money each year, but also has important effects on children, families, and the child welfare system. Parental incarceration may also have longer-run detrimental effects for children and families that threaten to incur additional costs to society (e.g., recidivism, intergenerational crime, and incarceration). The seriousness of the problem, especially for children, is recognized in the 2001 Amendments of the Safe and Stable Families Act (H.R. 28732), which allocates $67 million to mentoring programs for children with incarcerated parents. In summary, children of incarcerated parents are a group that is increasing in size and faces special risks. In the absence of a coordinated policy and service response on the part of criminal justice and child welfare agencies, these children may be at risk for a number of emotional, behavioral, and academic difficulties, and their parents may be at risk for termination of parental rights and recidivism. With children of incarcerated parents now making up at least 7 percent of foster children and 12 percent of children living with grandparents, it is time to develop such a response. NOTES 1. It is possible for a child to have two incarcerated parents, which would result in double counting of children. However, because reliable estimates on the number of children who have two incarcerated parents are not available, we are unable to adjust the figures. 2. Data not available for 1986. 3. We calculate the number of children in out-of-home care with nonrelatives by taking the total number of children in out-of-home care for each year and subtracting the number estimated to be in kinship foster care with a relative (using data from U.S. House of Representatives’ Overview of Entitlement Programs, or Green Book [U.S. House of Representatives 2000]). See appendix for details. 4. Our estimates of the proportion of U.S. children living in foster care, an agency, or an institution include children with parents in state and federal prison. Because the inmates in federal prisons were included in the survey in 1997 only, we use the data on the living arrangements of their children in 1997 to impute the living arrangements of children of federal inmates in the two earlier years.

REFERENCES Baunach, P. J. (1985). Mothers in prison. New Brunswick, NJ: Transaction Books. Beckerman, A. (1989). Incarcerated mothers and their children in foster care: The dilemma of visitation. Children and Youth Services Review, 11, 175–183.

98

Social Diversity and Single-Parent Families

———. (1991). Women in prison: The conflict between confinement and parental rights. Social Justice, 18(3), 171–183. ———. (1994). Mothers in prison: Meeting the prerequisite conditions for permanency planning. Social Work, 39(1), 9–14. ———. (1998). Charting a course: Meeting the challenge of permanency planning for children of incarcerated mothers. Child Welfare, 77(5), 513–529. Bloom, B., & Steinhart, D. (1993). Why punish the children? A reappraisal of the children of incarcerated mothers in America. San Francisco: National Council on Crime and Delinquency. Boudouris, J. (1996). Parents in prison: Addressing the needs of families. Lanham, MD: American Correctional Association. Child Welfare League of America. (1998). State Agency Survey on Children with Incarcerated Parents. Washington, DC: Child Welfare League of America. Ditton, P. M., & Wilson D. J. (1999). Truth in sentencing in state prisons. NCJ 170032. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Fuller-Thomson, E., Minkler, M., & Driver, D. (1997). A profile of grandparents raising grandchildren in the United States. Gerontologist, 37, 406–411. Genty, P. M. (1995). Termination of parental rights among prisoners: A national perspective. In K. Gabel & D. Johnston (Eds.), Children of incarcerated parents. (pp. 167–182). New York: Lexington Books. ———. (1998). Permanency planning in the context of parental incarceration: Legal issues and recommendations. Child Welfare, 77, 543–559. Greenfeld, L. A., & Snell, T. L. (1999). Women offenders. NCJ 175688. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Hairston, C. F. (1991). Family ties during imprisonment: Important to whom and for what? Journal of Sociology and Social Welfare, 18, 87–104. Hughes, T. A., Wilson, D. J., &. Beck, A. J. (2001). Trends in state parole, 1990–2000. NCJ 184735. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Johnson, E. I., & Waldfogel, J. (2002). Children of incarcerated parents: Where are they living and what risks do they face? Manuscript. Ann Arbor: University of Michigan, and New York: Columbia University. Johnston, D. (1991). Jailed mothers. Pasadena, CA: Pacific Oaks Center for Children of Incarcerated Parents. ———. (1992). Children of offenders. Pasadena, CA: Pacific Oaks Center for Children of Incarcerated Parents. ———. (1993). Caregivers of prisoners’ children. Pasadena, Calif.: Center for Children of Incarcerated Parents. ———. (1995). Effects of parental incarceration. In K. Gabel & D. Johnston (Eds.), Children of incarcerated parents (pp. 59–88). New York: Lexington Books. ———. (2001). Incarceration of women and effects on parenting. Paper presented at the Conferences on the Effects of Incarceration for Children and Families, Northwestern University, Institute for Policy Research, Chicago, May. Kampfner, C. J. 1995. Post-traumatic stress reactions in children of imprisoned Mothers. In K. Gabel & D. Johnston (Eds.), Children of incarcerated parents (pp. 89–100). New York: Lexington Books. Koban, L. A. 1983. Parents in prison: A comparative analysis of the effects of incarceration on the families of men and women. Research in law, deviance, and social control, 5, 171–183.

Parental Incarceration: Implications of Child Welfare

99

McGowan, B., & Blumenthal, K. (1978). Why punish the children? Hackensack, NJ: National Council on Crime and Delinquency. Mumola, C. J. (2000). Incarcerated parents and their children. NCJ 182335. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. National Conference of State Legislatures. (2001). Summaries of State Legislation Enacted in Response to the Adoption and Safe Families Act. Searchable online database available at http://www.ncsl.org/statefed/cf/asfasearch.htm (accessed on April 5, 2002). Sabol, W. J., & McGready, J. (1999). Time served in prison by federal offenders, 1986–1997. NCJ 171682. Washingon, DC: U. S. Department of Justice, Bureau of Justice Statistics. Schafer, N. E., & Dellinger, A. B. (1999). Jailed parents: An assessment. Women and criminal justice, 10(4), 73–91. Smith, B. E., & Elstein, S. G. (1994). Children on hold: Improving the response to children whose parents are arrested and incarcerated. Washington, DC: American Bar Association, Center for Children and the Law. Stanton, A. (1980). When mothers go to jail. Lexington, MA: Lexington Books. U.S. Census Bureau. (2001). Statistical abstract of the United States, 2000. Washington, DC: U.S. Census Bureau. U.S. Department of Justice. (1989). Correctional population in the United States, 1986. Washington, DC: Bureau of Justice Statistics. ———. (1993a). Correctional population in the United States, 1991. Washington, DC: Bureau of Justice Statistics. ———. (1993b). Survey of state prison inmates, 1991. NCJ 136949. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. U.S. Department of Justice, Bureau of Justice Statistics. (1993). Survey of inmates in state correctional facilities, 1991. Computer file. Conducted by U.S. Department of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research (producer and distributor). ———. (1994). Survey of inmates in state correctional facilities, 1986. Computer file. Conducted by the U.S. Department of Commerce, Bureau of the Census. 2d ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research (producer and distributor). ———. (2000). Survey of inmates in state and federal correctional facilities, 1997. Computer file. Compiled by the U.S. Department of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research (producer and distributor). U.S. Department of Justice, Federal Bureau of Prisons. (1999). Visiting regulations. P.S. 5267.06. Available online at http://www.bop.gov (accessed May 1, 2002). U.S. House of Representatives. (2000). Overview of Entitlement Programs: 2000 Green Book. Background Material and Data on Programs within the Jurisdiction of the Committee on Ways and Means. Washington, DC: U.S. Government Printing Office. Waters Boots, S., & Geen, R. (1999). Family care or foster care? How state policies affect kinship caregivers. Washington, DC: Urban Institute. Wooldredge, J. D. (1999). Inmate experiences and psychological well-being. Criminal Justice and Behavior, 26(2), 235–250. Zietz, D. (1963). Child welfare services in a women’s correctional facility. Child Welfare, 42, 185–190.

Chapter 8

Grandmothers Raising Grandchildren Catherine Goodman and Merril Silverstein*

During the past 30 years, the number of American children raised in grandparent-headed households has more than doubled, growing from 2.2 million in 1970 to 4.5 million in 2000—currently 6.3 percent of all children. In the 1980s, increases were primarily in three-generation households with a parent also present in the family. Since 1990, the increases have been greatest in households without a parent present (Bryson, 2001; Casper & Bryson, 1998; Saluter, 1996). In this study, we examined the social etiologies, filial stressors, and psychological well-being experienced by two types of grandmothers raising their grandchildren: those for whom the parent is absent (custodial grandmothers) and those for whom the parent is present (coparenting grandmothers). We used a unique sample that was designed to provide near-equal representations of African American, Latino, and white grandmothers across custodial and coparenting household types. Custodial and coparenting families are compared for stressors resulting from the parent’s circumstances (e.g., reasons the grandmothers assumed care and intergenerational relationship quality) and grandmother well-being within each ethnic group. The intent of this analysis is to examine the influence of household structure on the antecedents and consequences of raising grandchildren as a grandparent. Over the past 10 years, a great deal of attention has focused on custodial grandparent-headed families owing to the important societal gap these grandparents fill (Cox, 2000b; Hayslip & Goldberg-Glen, 2000; Minkler &

* C. Goodman and M. Silverstein. Grandmothers Raising Grandchildren: Family Structure and Well-Being in Culturally Diverse Families. The Gerontologist, 2002, 42, 676–689. Reprinted by permission of Oxford University Press.

Grandmothers Raising Grandchildren

101

Roe, 1993). Grandparents typically assume full-time care for grandchildren under disruptive circumstances, usually the consequence of serious problems experienced by the parents (Jendrek, 1994; Minkler & Roe, 1993). Jendrek found that mothers’ emotional or mental problems and drug or alcohol abuse were common reasons for grandmothers to assume custodial care of their grandchildren. These parental stresses erode family relationships, and low family cohesion has been linked to anxiety among primary caregiving grandparents (Sands & Goldberg-Glen, 2000). Particularly when substance abuse is an issue, the absent parent’s level of involvement is often ambiguous and of uncertain duration, producing depression in the custodial grandparent (Hirshorn, Van Meter, & Brown, 2000). Minkler and Roe have documented the difficult lives of African American grandparents raising grandchildren abandoned by parents who were victims of the crack cocaine epidemic. As a result of the strained family conditions under which care is typically assumed, custodial grandparenthood is often unanticipated, involuntary, and indefinite and is therefore a risk factor for psychological distress (Pearlin, 1993). Several studies comparing grandparent caregivers to noncaregivers have identified greater depression and worse health (Fuller-Thomson & Minkler, 2000b; Solomon & Marx, 2000) in national samples, and caregiving grandparents were more depressed than noncaregivers in an Alameda County study (Strawbridge, Wallhagen, Shema, & Kaplan, 1997). Grandparents who house both grandchildren and their parent(s) are typically partners in child rearing, in contrast to parent-headed families that may provide for an older, often ill grandparent (Bryson & Casper, 1999). Studies of coparenting families have focused primarily on grandparents assisting young mothers. However, the advantages attributed to multigenerational coresidence—such as family closeness and increased resources (Stevens, 1984)—may be counteracted by disengaged parents, conflict over child rearing, low family cohesion, and loss of privacy. For example, the grandmother’s interference in child rearing (e.g., unwanted advice and assistance) and emotional friction (e.g., criticism and antagonism) have been found to cause strain among young parents living with their mothers (Richardson, Barbour, & Bubenzer, 1991). Low family cohesion has been related to depression (Kalil, Spencer, Spieker, & Gilchrist, 1998), and intergenerational conflict has been found to increase the risk of depression among both mothers and grandmothers in coparenting families (Caldwell, Antonucci, & Jackson, 1998). Only a few studies have compared the well-being of coparenting and custodial grandparents in grandparent-headed families (Jendrek, 1993; Musil, 2000; Pruchno & McKenney, 2000). Most have found lower wellbeing among custodial grandparents in terms of restrictions in privacy and on time use (Jendrek, 1993), levels of parenting stress and available supports (Musil, 2000), and caregiver burden and reduced life satisfaction

102

Social Diversity and Single-Parent Families

(Bowers & Myers, 1999). One study showed custodial grandparents to be more burdened, although they did not differ in caregiver satisfaction and mood (Pruchno & McKenney, 2000). Thus, the weight of evidence supports the view that custodial grandparents are at risk of social isolation and elevated emotional distress compared with coparenting grandparents. Ethnicity and Grandparent Caregivers Although large cultural groupings mask many subgroups with diverse experiences, national backgrounds, and socioeconomic strata, we have pursued census-accepted and inclusive categories of African American, Latino, and white for this study. The incidence of grandparents who are raising their grandchildren varies by these ethnic groups, suggesting culturally relevant pathways to caregiving roles, shaped by different family composition and values and unique role expectations regarding grandparenting. African American children are more apt to live with grandparents than other groups: Thirteen percent of African American children lived in grandparent-headed households compared with 5.7 percent of Latino and 3.9 percent of white children in 1994 (Saluter, 1996), and a higher proportion of African American grandparent-headed families than other groups have no parent at home, reflecting a cultural tradition of surrogate parenting (Burton & Dilworth-Anderson, 1991). Although fewer Latino children are raised in grandparent-headed families compared with African American children, proportionally more Latino grandparentheaded families coparent (Saluter, 1996). Three-generation living is common in parent-headed Latino families, particularly among immigrant families (Bryson & Casper, 1999) and to provide for older adults (Lubben & Becerra, 1987). White families have the lowest proportion of children raised in grandparent-headed families, and more of them are in coparenting families (Saluter, 1996). Research on ethnicity and grandparenting has shown important differences in family composition and role expectations. African American grandparents have historically served as kinkeepers (Burton & DilworthAnderson, 1991) and have often raised their grandchildren as a result of African tradition, family survival during slavery, and the parents’ search for economic opportunity in the North (Hunter & Taylor, 1998). In spite of the tradition, qualitative studies of African American grandmothers in urban communities have described neighborhood dangers, multiple caregiving roles, financial drain, and stress resulting from raising grandchildren (Burton, 1992). Coparenting in African American families has traditionally been a response to the needs of single and teen parents, and studies with African American samples have found that grandmothers played highly utilitarian and active roles in terms of transmitting knowledge about infant develop-

Grandmothers Raising Grandchildren

103

ment to their daughters (Stevens, 1984) and serving as a parental replacement, supplement, and support (Apfel & Seitz, 1991). On the other hand, coparenting in the same household has been related to lower levels of maternal competency than when mothers lived independently (Wakschlag, Chase-Lansdale, & Brooks-Gunn, 1996). In contrast to African American grandparents, little has been written about Latino grandparents raising grandchildren until recently (Burnette, 1999; Cox, 2000a). A training program for Latino and African American caregiving grandparents showed Latino grandmothers were more likely to be involved with parents and to be providing day care, in contrast to custodial African American grandmothers (Cox, 2000a). Even custodial Latino grandmothers often had a nonparent adult child living at home to help (Burnette, 1999). Thus, Latino grandparents were more likely to play support roles to the parent, to coparent in intergenerational households, and to rely on adult children even in custodial situations, consistent with the values of familism. The preference for familism involves frequent and close family contacts and the expectations of reciprocal mutual aid, including filial responsibility in old age (Vega, 1995). These values ultimately have a bearing on the quality of family relationships and emotional well-being in custodial and coparenting Latino families. The surrogate parent role for white grandmothers emerged in the literature in an early typology developed by Neugarten and Weinstein (1964), and the tradition for a small proportion of children to be raised by custodial grandparents has been documented back to 1940. Coparenting is also a consistent, somewhat more prevalent tradition in response to single parenthood or economic need (Uhlenberg & Kirby, 1998). Nevertheless, the style of grandparenting that predominates is companionate, and white grandmothers are less apt to discipline and correct their grandchildren than African American grandparents (Cherlin & Furstenberg, 1992). Pruchno (1999) found white grandparents to be less familiar and more burdened with the custodial role than African American grandmothers. Although longstanding, the tradition has not been widespread, and the lack of well-known, shared norms for primary grandparent caregiving may limit the support available to white grandmothers as they assume the role.

HYPOTHESES The literature suggests that grandparents who assume care for grandchildren tend to have many difficult challenges that produce elevated levels of stress. At the same time, cultural and ethnic perspectives and traditions shape expectations and values about grandparent roles, and therefore may shape grandparent well-being in this increasingly

104

Social Diversity and Single-Parent Families

prevalent grandparent-headed type of family. Given the different pathways by which grandparents achieve caregiving with respect to parental coresidence, we hypothesized that custodial grandmothers would have more serious reasons for assuming care than coparenting grandparents and would experience lower intergenerational cohesion as a result. Owing to stressors related to the etiology of role adoption and lower family cohesion, we expected that custodial grandmothers would have lower psychological well-being than coparenting grandparents. As a result of differences in ethnic traditions and expectations about grandparent caregiving roles, we further hypothesized that departure from ethnic traditions in household structure would result in lower well-being among caregiving grandmothers. Therefore, we predicted that Latino and white grandmothers would have lower wellbeing when they were in custodial circumstances than in coparenting circumstances and that the coparenting advantage would be less pronounced for African American grandmothers. METHODS Sample We addressed our hypotheses using a sample of custodial or coparenting grandmothers living in Los Angeles County in 1998–2001. Criteria for selection were a grandparent head of the household; school-aged grandchild living in the household; residence in Los Angeles County; and African American, Latino, or white grandmother. The sample was recruited in order to obtain near equal representation of African American, Latino, and white grandmothers. A total of 1,058 grandmothers were interviewed for the study: 360 African American (247 custodial and 113 coparenting), 354 Latino (158 custodial and 196 coparenting), and 344 white (176 custodial and 168 coparenting). Ethnicity of grandmothers was defined by self-report. Custodial grandmothers were differentiated from coparenting grandmothers on the basis of whether the parent was a current resident in the household. Grandmothers were recruited into the study through the schools and the media. Almost two-thirds (63 percent) of the sample was recruited through grandchildren attending public schools in the Los Angeles Unified School District. Notices that announced the study and incentive ($15 payment for the grandmother and a $5 McDonald’s gift certificate for the grandchild) were sent home to all students in targeted schools. Notices were distributed in 223 of 792 schools in the district and reached 32 percent of all students. To supplement this sampling technique, grandmothers were also recruited, using the same incentives, through media ads (24 percent) and community contacts (13 percent). One-hour

Grandmothers Raising Grandchildren

105

interviews were completed in the grandmothers’ homes by ethnically and linguistically matched interviewers by the Survey Research Center, Institute for Social Science Research, University of California, Los Angeles. Because the sample was recruited purposively, it did not provide the advantages of a sample randomly selected from a known population. Sample selection decisions in the study of relatively rare populations must be viewed in light of the unique nature of the population being studied. The most often used random sample of a couple hundred caregiving grandparents (see Fuller-Thomson & Minkler, 2000b; Minkler, FullerThomson, Miller, & Driver, 1997) presents low power and makes subgroup analyses difficult. Pruchno (Pruchno, 1999; Pruchno & McKenney, 2000), in her national study of African American and white caregiving grandmothers, recruited a large sample of over 700 through media ads. Our use of several recruitment strategies optimized sample size while representing a good deal of the ethnic, economic, and lifestyle diversity in the study population. Nevertheless, owing to the unique characteristics of our quota sample, findings on the total sample cannot be generalized to the population of caregiving grandmothers. The full sample of custodial and coparenting grandmothers consisted of 43 percent married grandmothers, with 43 percent working full or part time. The mean age for the total sample was 56.7 years. Overall educational level was 11.5 years, which is almost at the level of a high school graduate. Average yearly family income level was between $30,000 and $35,000. The distribution of income was broad, with 46 percent having a yearly family income of $30,000 or less, 38 percent having a yearly family income of $30,001 to $60,000, and 15 percent having a yearly family income of $60,001 or more. There was a 23 percent poverty rate. Measures Survey questions included those concerning grandmothers’ and grandchildren’s demographic and social characteristics, reasons for caregiving, family relationships, and psychological well-being. Survey questions that were not already available in Spanish were translated and reviewed by bilingual persons of three different nationalities (Mexican, Spanish, & Salvadoran) to account for regional variation. Demographic characteristics of grandmothers included their marital status, age, health, ethnicity, education, work status, yearly family income, and per capita income. Acculturation was assessed for Latino grandmothers on the basis of responses to a four-item measure (Marín, Sabogal, Marin, Otero-Sabogal, & Perez-Stable, 1987) that tapped preferred language use in different social contexts. Responses were rated on a five-point scale ranging from only Spanish to only English. Coefficient alpha for this sample was .95.

106

Social Diversity and Single-Parent Families

One target grandchild was selected—the child in the targeted school or with the most recent birthday—and a series of detailed questions was asked, including those concerning child welfare involvement, and the child’s gender, the child’s age, and the number of years lived with grandmother. The child’s behavior problems were listed using a 10-item Behavior Rating Index for Children, rated on a five-point scale ranging from rarely or never to most or all of the time (Stiffman, Orme, Evans, Feldman, & Keeney, 1984). The coefficient alpha for this sample (.78) was slightly lower than at the time of development, but was acceptable. Stress resulting from the parent’s circumstance was described using a problem list developed by Jendrek (1994) through in-depth interviews, which elucidated the reason the grandmother assumed care. The final list of 36 reasons was adapted for the current study. The resident parent—mother, father, or couple—was also described for coparenting families. Family stress was also measured as the quality of intergenerational relations, which were assessed using the affectual solidarity and conflict scales developed by Bengtson (1991). The grandmother described her relationship with the target grandchild and with the father and mother of the child and provided an estimate of the grandchild’s relationship with each parent. A five-item version of the affectual solidarity scale was adopted, including assessments of understanding, getting along, emotional closeness, communication, and affection. In addition, a single conflict item was used that reflects conflict, tension, or disagreement in the relationships (Bengtson, 1991). Coefficient alphas for affectual solidarity ranged from .83 for grandmother and grandchild to .93 to .95 for other relationships. Psychological well-being of grandmothers was measured using four different scales discussed below, all available in Spanish as well as in English. Each scale taps a unique aspect of well-being related to depression, general affect, global mental health, and satisfaction with aspects of daily life. The Center for Epidemiological Studies–Depression scale (Radloff, 1977) is a 20-item scale designed to measure depression in the general population and has also proved useful for clinical and psychiatric research. Items are rated on a four-point frequency scale that assesses frequency over the past week, ranging from rarely or none of the time to most or all of the time. The measure has been and continues to be used with Spanish-speaking samples and with older adults. Coefficient alpha of the 20 summed items was .89. The Short Form-36 health survey (Ware, 1993) was used to assess the physical and mental health of the grandmothers. This 36-item measure has scales measuring physical functioning, role disability due to physical health problems, bodily pain, general health perceptions, vitality,

Grandmothers Raising Grandchildren

107

social functioning, role disability due to emotional health problems, and general mental health. The measure meets generally recommended reliability standards for group comparisons. Subscale internal consistency for this sample ranged from .74 for the social functioning component to .93 for the physical functioning component. Algorithms provided were used to construct general physical health and mental health component scores from the subscales. General physical health was used as a control variable, and general mental health was used as a measure of well-being. The Positive and Negative Affect Scale (PANAS) was used to measure positive and negative mood (Watson, Clark, & Tellegen, 1988). This is a 20-item measure of mood that uses descriptive positive adjectives, such as interested, strong, enthusiastic, or negative adjectives, such as distressed, scared, or hostile. Respondents described how they felt during the past few weeks, by means of a five-point scale ranging from very slightly or not at all to extremely. Given the literature showing the independence of positive affect and negative affect, we formed two scales, each consisting of pure markers of positive or negative affect. Variables were selected, based on factor loadings from a principal-components analysis, if they had substantial (4) loading on one factor and a near-zero loading on the other. Coefficient alpha for this sample was .87 for positive affect and .88 for negative affect. Life satisfaction, a cognitive judgment of subjective well-being, was measured using the Satisfaction With Life Scale (Diener, Emmons, Larsen, & Griffin, 1985). The measure consists of five items scored on a sevenpoint Likert scale ranging from strongly agree to strongly disagree. Life satisfaction also met criteria for discriminant validity from affective components as measured by the PANAS (Lucas, Diener, & Suh, 1996). The coefficient alpha for this sample was .84.

RESULTS The first three sections of our presentation of results compare custodial and coparenting grandmothers within each of the three ethnic groups with respect to their demographic and social characteristics, the reasons they adopted the caregiving role, and the quality of their intergenerational relationships. In these sections, chi-square tests and Fisher exact tests were used to test household type differences on categorical variables, and one-way analysis of variance and F-tests were used to examine the mean differences between household type for interval-level variables. In the fourth section, multivariate analyses were used to examine differences in well-being between custodial and coparenting grandmothers with and without various risk factors of distress controlled.

108

Social Diversity and Single-Parent Families

Demographic Differences Between Household Types The total sample showed custodial families to have a smaller proportion of married grandmothers, higher educational level, lower yearly family income, and higher per capita yearly income compared with coparenting grandmothers. There were no differences in age, working status, health, or proportion in poverty, using the 1999 poverty guidelines (Annual Update, 1999). Educational differences resulted from less acculturated and less educated Latino grandmothers in coparenting families. Higher yearly income among coparenting grandmothers was a result of larger household size, and per capita income was higher in custodial families. Thus, grandmothers in custodial and coparenting families were similar on many important factors. Where differences existed, they were not consistent across all ethnic groups. In terms of the target grandchild’s characteristics, there were no gender differences in custodial versus coparenting families. As might be expected, there were more custodial grandchildren supervised by child welfare agencies. They were also older, had more behavioral problems, and had stayed longer with their grandmothers. Child welfare oversight was consistent across ethnic groups, but other differences were not. The resident parent was typically the mother, although Latino families had a high proportion of couples living in the family.

Reasons to Assume Care in Custodial Versus Coparenting Families Reasons for assuming care were dramatically different for custodial and coparenting grandmothers. Roughly two-thirds of the reasons listed showed significant family structure differences, and the most frequently selected reasons were consistent for all ethnic groups. Custodial grandmothers assumed care most often because of the mother’s drug use, mental or emotional problem, or child neglect or the father’s drug use. Grandmothers also often assumed care to avoid foster care for their grandchild. Less frequently selected but significant differentiating reasons were mother’s alcohol problem, trouble with the law, physical illness, and physical abuse to the child and father’s alcohol problem, trouble with the law, emotional or mental problem, and child neglect or sexual abuse. The child’s medical problem was also a reason more frequently selected in custodial families. Thus, the most serious parental problems were most often found in custodial families. There were only a few variations within ethnic groups that departed from the overall pattern of custodial versus coparenting families, although proportions of serious problems varied considerably across ethnicity. In contrast, the most common reason coparenting grandmothers assumed care was to help financially, although working mother or father and parents divorcing were also significantly more prevalent among

Grandmothers Raising Grandchildren

109

coparenting families. Teenage mother/father and mother/father in school were equally prevalent in both groups. Unmarried parents, better school district, avoiding day care, or “gave the grandmother something to do” were reasons that were equally endorsed by custodial and coparenting grandmothers. Only a few differences emerged for one ethnic group when the total sample showed no differences between custodial and coparenting families, although the proportions selecting a particular reason varied by ethnicity. Intergenerational Relationships in Custodial Versus Coparenting Families There were no differences between custodial and coparenting grandmothers in terms of closeness with the target grandchild. However, all intergenerational relationships involving parents showed differences in closeness by household type, with closer relationships characteristic of coparenting families. These results were consistent within ethnic groups, with one exception. The relationships involving African American fathers were equally close in custodial and coparenting families. Conflict was higher between grandmother and grandchild in custodial families than in coparenting families across all ethnic groups. Thereafter, conflict was greater in custodial families in relationships with the mother, except for African American families, which showed no differences in conflict in relationships with the parent. Well-Being of Custodial Versus Coparenting Grandmothers We used multivariate linear modeling to examine the differences in well-being of grandmothers across the household types, using all five indicators of well-being simultaneously. Analyses were conducted separately for each ethnic group so as not to obscure the distinct patterns by ethnicity. We built our models hierarchically, such that the impact of household type could be examined after sequentially controlling for many of the stresses that differentiate grandparents in the two family structures. In the first model, no control variables were applied, and results reflected these grandmothers as they appeared in the community. The second model added demographic factors as controls and thus isolated the impact of household type on well-being after variances due to socioeconomic correlates of well-being were removed (age, education, per capita income, working status, marital status, physical health, recruitment method, and acculturation for Latino grandmothers). The third model followed with the addition of reasons for assuming care and the quality of intergenerational relationship, therefore statistically removing stress resulting from the parent generation.

110

Social Diversity and Single-Parent Families

To reduce the number of variables in Model 3, the most frequently described reasons relating to the parents were selected, and related reasons were summed for each parent. Substance abuse was the sum of drug and alcohol abuse; child abuse was the sum of child neglect and physical and sexual abuse. The reasons used in this analysis were mother’s substance abuse, mother’s child abuse, mother’s emotional or mental problem, father’s substance abuse, and financial assistance for the parents, all incorporating variables endorsed by 40 percent or more of custodial or coparenting grandmothers. Model 3 also added intergenerational relationships to the existing controls. Because of the high bivariate correlations, grandmother/mother closeness and mother/child closeness were summed for a mother closeness index. Similarly, father closeness, mother conflict, and father conflict indices were constructed to reflect their closeness and conflict within the family. Preliminary analyses using a broader range of parental control variables showed essentially the same results for custodial versus coparenting family structure, although significance for control variables shifted somewhat. The reduced multivariate model appears here for ease of presentation. Control variables did not apply uniformly to well-being across all ethnic groups; only health status was important for all ethnic groups. For African American grandmothers, there were no significant differences in household type among the five well-being indicators for Model 1 (no controls) or Model 2 (demographic factors controlled). When the reasons for assuming care and intergenerational relationships were controlled in Model 3, household type differences emerged. African American grandmothers in custodial families showed less negative affect, more positive affect, and better mental health. Statistically removing the problems of the parent generation revealed a greater well-being in response to custodial over coparenting circumstances for African American grandmothers. In marked contrast, Latino grandmothers had significantly lower negative affect and greater life satisfaction in coparenting households in Model 1, which had no controls. Demographic controls in Model 2 extended the result, and greater well-being in coparenting grandmothers was also evident in higher positive affect and better mental health, four of five indicators. In Model 3, controlling for parental stresses, all well-being differences between custodial and coparenting grandmothers disappeared. In short, there was little difference between the household types for Latino grandmothers beyond the stresses from the parent generation. White custodial grandmothers showed higher positive affect in Model 1 when there were no controls. In Model 2, they showed higher negative and positive affect, suggesting the custodial situation is emotionally charged, with greater affect overall once demographic factors were controlled. In Model 3, when the stresses from the parent’s circumstances

Grandmothers Raising Grandchildren

111

were controlled, they continued to show higher positive affect in the custodial situation, although the multivariate F did not reach significance. Thus, without the stress from the parent, custodial and structures were roughly equal or somewhat more positive for white grandmothers.

DISCUSSION In these analyses, we investigated the impact of household structure on the stress and distress experienced by grandmothers who are raising their grandchildren. Using a unique sample of African American, Latino, and white caregiving grandmothers, we found evidence that household structure is associated with differential exposure to stressful experiences and that the impact of those experiences on well-being is conditioned by the cultural context in which expectations of grandparents are formed. African American grandmothers showed no differences as a result of household structure until the stressful parental conditions and precursors associated with custodial grandparenting were statistically controlled. With these controls, African American custodial grandmothers had greater well-being than their coparenting counterparts—signifying acceptance of a surrogate parenting role that has a long historical precedent in the African American community. By contrast, the coparenting arrangement was associated with greater well-being among Latino grandmothers, the product of a different cultural role ideal—that of close intergenerational relationships and reciprocal contact. White grandmothers showed no difference on most indices of well-being but found custodial grandparenthood emotionally evocative, demonstrating higher positive and negative affect. For Latino and white grandmothers, wellbeing was the same in custodial and coparenting families once parental stress factors were statistically removed—demonstrating the importance of parental factors in the grandmother’s response to these family structures. Although the sample provided great strengths in terms of its cultural diversity, as a quota sample it overrepresented some groups and cannot be compared to the U.S. population of grandmothers who are raising their grandchildren. Only grandmothers raising school-aged children were included, and the sampling frame of urban Los Angeles restricts generalization beyond the urban context, particularly among Latino grandmothers, who were more likely to be immigrants than in other regions of the nation. In comparison with national data on coresident grandparents, grandmothers in our sample had lower household income level (approximately $34,000 compared with $43,783 nationally in 1997) but virtually the same poverty rate (23.4 percent compared with 23 percent nationally; Casper & Bryson, 1998). Compared with those of

112

Social Diversity and Single-Parent Families

other studies, the sample was younger (Fuller-Thomson & Minkler, 2000a; Pruchno, 1999), fewer were married (Fuller-Thomson & Minkler, 2000a; Pruchno, 1999), and fewer were employed (Casper and Bryson, 1998). These differences from national studies are the result of quota sampling; we adjusted for these apparent biases by conducting analyses within ethnic groups. Another limitation of the analysis is related to the study’s cross-sectional nature. We were unable to determine to what extent differences in well-being in custodial or coparenting families were a result of the household structure or the social forces predisposing grandmothers to adopt their particular living arrangement. We sought to remedy this limitation by controlling for the most salient precursors and consequences of living in a custodial or coparenting household that are also risk factors for distress among caregiving grandparents. Stress and Family Disruption in Custodial Versus Coparenting Families Although not surprising, our hypothesis that greater disruption from the parent’s problem would characterize custodial families was confirmed. We affirmed, as have other studies (Jendrek, 1994; Pruchno & McKenney, 2000), that parental drug use was a reason to assume care in around half of custodial families, whereas it was a reason for one in five or fewer grandmothers in coparenting families. Similarly, financial reasons motivated most grandmothers in coresident situations but not those in custodial situations. The grandmother-grandchild relationship was equally close in both family types, although this relationship was more conflicted in custodial families, suggesting that some conflict accompanies parental responsibility for discipline of the grandchildren. Because all grandmothers in our sample participated to some extent in the care of their grandchildren, our finding is consistent with Fuller-Thomson and Minkler (2000b), who found that self-defined, primary caregiving grandmothers had greater closeness to their grandchildren than noncaregiving grandmothers. As expected, we found that closeness was lower in custodial families in all relationships with the parent. The only exception was the relationship with the father in African American families. Thus, the relationship with the mother more consistently reflected the stresses of custodial grandparenting. Custodial African American families showed less relationship disruption in terms of conflict with the parent generation than custodial Latino or white families. The African American tradition of grandparent surrogate care is longstanding and did not originate with the parent’s inability to parent (Burton & Dilworth-Anderson, 1991), a history that may reduce stigma and conflict with the parent, in spite of contemporary parental problems

Grandmothers Raising Grandchildren

113

and reduced closeness with the mother. Latino custodial grandmothers may experience conflict with the mother, as their expectations of familism and close intergenerational relationship are not fulfilled. The intricate interdependence in Latino families implies future reciprocity: Adult children are expected to care for their parents in old age, as well as engage in ongoing social and supportive family involvement (Vega, 1995). White families have a tradition of custodial care with very low prevalence (Uhlenberg & Kirby, 1998), and the problematic circumstances of custodial care may generate conflict with the parents in light of norms for independence and noninterference. Thus, the stresses experienced as a result of the parent are both common and unique, with a subtle effect on the quality and dynamics of intergenerational relationships resulting from ethnic traditions and expectations. Grandmother Well-Being in Custodial Versus Coparenting Families We also hypothesized that disruptions and stress resulting from the parent’s problem and from problematic intergenerational relationships would result in lower well-being among custodial families than among coparenting families. This hypothesis was confirmed unequivocally for Latino grandmothers. In contrast, African American grandmothers showed no significant differences in well-being as a result of custodial or coparenting circumstances. For white grandmothers, three of five indicators showed no differences, but the custodial family was related to higher affect (positive and negative), a somewhat ambiguous result. Most studies of grandparent well-being have focused on the comparison between caregiving and noncaregiving grandparents, showing greater depression and lower physical well-being for the former (Fuller-Thomson & Minkler, 2000b; Solomon & Marx, 2000). One study compared custodial and coparenting grandparents with noncaregiving grandparents (Szinovacz, DeViney, & Atkinson, 1999). Szinovacz and colleagues found greater depression when a grandchild entered a grandparent-headed family for both custodial and coparenting grandmothers, suggesting that both circumstances exert considerable stress on grandmothers. Although household structure is typically a response to the parent’s problems, we can address the impact of the structure itself, separated from the parent’s problems by statistical control. Among African Americans, without the stresses of the parent’s serious problems, grandmothers would appear to adapt with greater well-being to custodial situations. Coparenting also presents struggles and hazards, including the tendency of some young mothers to abdicate parenting responsibility (Apfel & Seitz, 1991) or exhibit reduced parenting competence in threegeneration families (Wakschlag et al., 1996). Latino and white grandmothers in custodial and coparenting households fared equally well after

114

Social Diversity and Single-Parent Families

the effects of parental and relationship problems were statistically removed. Latino family traditions include the expectation that grandmothers will be involved in the lives and upbringing of their grandchildren (Williams & Torrez, 1998). Once the disappointments over the parent’s circumstances are removed, the preference for coparenting disappears. For white families, this statistical simulation resulted in similar well-being in both types of families, although univariate positive affect remained higher in custodial families. Both custodial and coparenting roles may be contrary to cultural ideals of independence and noninterference, and the preferred role is companionate (Cherlin & Furstenberg, 1992). Implications for Practice and Research Because of the failure to identify a coparenting advantage for African American and white caregiving grandmothers, the findings of this study point to more in-depth analysis of coparenting stress. Many stress factors associated with multigenerational living were not captured in this research (e.g., extent of shared parenting, parental competence and individuation, and conflict over child-rearing decisions; Apfel & Seitz, 1991; Wakschlag et al., 1996). The lack of a stronger coparenting advantage alerts the professional to the needs of coparenting families as well as custodial families. For custodial families, the problems of the parent are a major source of stress. Controlling for parental stresses shifted results for all ethnic groups toward improvement of well-being for custodial grandmothers. Important practice goals are dealing with the grief and disappointment over the parent’s situation and helping to salvage whatever is possible of the relationship with the parent. Although not the focus of the study, demographics of the sample showed that poverty was a concern for a large proportion of the grandmothers (a third of Latino grandmothers and 25 percent of custodial African American grandmothers). Latino families, in addition, have lower education, and many are recent immigrants. They may have an acculturation and language gap in accessing resources for their families, and many families do not have legal immigrant status, which would qualify them for most types of assistance. Welfare reform under the Personal Responsibility and Work Opportunity Act of 1996 (P.L. 104–193) has mandated work requirements and time limits for welfare (Temporary Assistance to Needy Families, or TANF), and the status of caregiving grandparents in relation to these regulations is still being clarified by the states. States are allowed to exempt 20 percent of their caseload from time limits (Minkler, Berrick, & Needell, 1999). In California, time limits do not apply if caregivers are older (60 or older); however, the average age of grandparent caregivers in this study was 57 years of age.

Grandmothers Raising Grandchildren

115

Regulations excluding persons with drug-related felonies and requiring teen mothers to live with their parents will certainly affect the parent generation, who may be rendered ineligible or must meet behavioral standards to qualify (Minkler et al., 1999). In addition to welfare (TANF), some grandparents assume care for grandchildren through the child welfare system as a result of parental abuse or neglect. If children are wards of the court, grandparents may be paid as foster parents if they qualify for state licensing, although state regulations vary widely and continue to shift (Gleeson, 1999). As these policies are further defined over the next decade, the interests of caregiving grandparents should be represented and protected. Grandparents often make enormous personal sacrifices to assume roles they are not legally obligated to fulfill and that provide a contribution to society and to the well-being of the family. Finally, the study makes it clear that cultural ideals, norms, and traditions make a difference in assuming custodial and coparenting roles. These ideas and traditions may provide a protection against the daily stresses inherent in both these family types. Where there are no cultural templates that normalize a family situation, subcultural enclaves can be created to provide support and a positive ideology. Support groups are one example that are effective for some caregivers. Other means that create and modify institutions also need to be developed in order to provide greater support for the courageous and dedicated efforts grandparents are making to provide for the next generation.

REFERENCES Apfel, N. H., & Seitz, V. (1991). Four models of adolescent mother-grandmother relationships in black inner-city families. Family Relations, 40, 421–429. Bengtson, V. L. (1991). The longitudinal study of three-generation families: 1991. Unpublished survey instrument, Andrus Gerontology Center, University of Southern California. Bowers, B. F., & Myers, B. J. (1999). Grandmothers providing care for grandchildren: Consequences of various levels of caregiving. Family Relations, 48, 303–311. Bryson, K. (2001, November). New Census Bureau data on grandparents raising grandchildren. Paper presented at the 54th Annual Scientific meeting of the Gerontological Society of America, Chicago, IL. Bryson, K., & Casper, L. M. (1999). Current populations reports: Special studies (P23–198). Washington, DC: U.S. Bureau of the Census. Retrieved from http://www.census.gov/prod/99pubs/p23–198.pdf. Burnette, D. (1999). Social relationships of Latino grandparent caregivers: A role theory perspective. The Gerontologist, 39, 49–58. Burton, L. M. (1992). Black grandparents rearing children of drug-addicted parents: Stressors, outcomes, and social service needs. The Gerontologist, 32, 744–751.

116

Social Diversity and Single-Parent Families

Burton, L. M., & Dilworth-Anderson, P. (1991). The intergenerational family roles of aged Black Americans. Marriage and Family Review, 16, 311–330. Caldwell, C. H., Antonucci, T. C., & Jackson, J. S. (1998). Supportive/conflictual family relations and depressive symptomatology: Teenage mother and grandmother perspectives. Family Relations, 47, 395–402. Casper, L. M., & Bryson, K. R. (1998). Co-resident grandparents and their grandchildren: Grandparent maintained families (Population Division Working Paper No. 26). Washington, DC: U.S. Bureau of the Census. Retrieved from http://www.census.gov/population/www/documentation/twps0026/ twps0026.html. Cherlin, A. J., & Furstenberg, F. F., Jr. (1992). The new American family: A place in the family, a life apart. London: Harvard University Press. Cox, C. (2000a). Empowerment practice: Implications for interventions with African American and Latina custodial grandmothers. Mental Health and Aging, 6(4), 385–397. ———. (2000b). To grandmother’s house we go and stay: Perspectives on custodial grandparents. New York: Springer. Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71–75. Diener, E., & Lucas, R. E. (1999). Personality and subjective well-being. In D. Kahneman, E. Diener, & N. Schwarz (Eds.), Well-being: The foundations of hedonic psychology (pp. 213–229). New York: Russell Sage. Fuller-Thomson, E., & Minkler, M. (2000a). America’s grandparent caregivers: Who are they? In B. Hayslip, Jr. & R. Goldberg-Glen (Eds.), Grandparents raising grandchildren: Theoretical, empirical, and clinical perspectives (pp. 3–22). New York: Springer. ———. (2000b). The mental and physical health of grandmothers who are raising their grandchildren. Journal of Mental Health and Aging, 6, 311–323. Gleeson, J. P. (1999). Kinship care as a child welfare service. In R. L. Hegar & M. Scannapieco (Eds.), Kinship foster care: Policy, practice, and research (pp. 28–53). New York: Oxford University Press. Hayslip, B., Jr., & Goldberg-Glen, R. (2000). Grandparents raising grandchildren: Theoretical, empirical, and clinical perspectives. New York: Springer. Hirshorn, B. A., Van Meter, J. V., & Brown, D. R. (2000). When grandparents raise grandchildren due to substance abuse: Responding to a uniquely destabilizing factor. In B. Hayslip, Jr., & R. Goldberg-Glen (Eds.), Grandparents raising grandchildren: Theoretical, empirical, and clinical perspectives (pp. 269–288). New York: Springer. Hunter, A. G., & Taylor, R. J. (1998). Grandparenthood in African American families. In M. E. Szinovacz (Ed.), Handbook on grandparenthood (pp. 70–86). Westport, CT: Greenwood. Jendrek, M. P. (1993). Grandparents who parent their grandchildren: Effects on lifestyle. Journal of Marriage and the Family, 55, 609–621. ———. (1994). Grandparents who parent their grandchildren: Circumstances and decisions. The Gerontologist, 34, 206–216. Kalil, A., Spencer, M. S., Spieker, S. J., & Gilchrist, L. D. (1998). Effects of grandmother coresidence and quality of family relationships on depressive symptoms in adolescent mothers. Family Relations, 47, 433–441.

Grandmothers Raising Grandchildren

117

Lubben, J. E., & Becerra, R. M. (1987). Social support among Black, Mexican, and Chinese elderly. In D. E. Gelfand & C. H. Barresi (Eds.), Ethnic dimensions of aging (pp. 130–144). New York: Springer. Lucas, R. E., Diener, E., & Suh, E. (1996). Discriminant validity of well-being measures. Journal of Personality and Social Psychology, 71, 616–628. Marín, G., Sabogal, F., Marín, B. V., Otero-Sabogal, R., & Pérez-Stable, E. J. (1987). Development of a short acculturation scale for Latinos. Latino Journal of Behavioral Sciences, 9, 183–205. Minkler, M., Berrick, J. D., & Needell, B. (1999). Impacts of welfare reform on California grandparents raising grandchildren: Reflections from the field. Journal of Aging and Social Policy, 10, 45–63. Minkler, M., Fuller-Thomson, E., Miller, D., & Driver, D. (1997). Depression in grandparents raising grandchildren. Archives of Family Medicine, 6, 445–462. Minkler, M., & Roe, K. M. (1993). Grandmothers as caregivers: Raising children of the crack cocaine epidemic. Newbury Park, CA: Sage. Musil, C. M. (2000). Health of grandmothers as caregivers: A ten month follow-up. Journal of Women and Aging, 12(1/2), 129–145. Neugarten, B. L., & Weinstein, K. K. (1964). The changing American grandparent. Journal of Marriage and the Family, 26, 199–204. Office of the Assistant Secretary for Planning and Evaluation (1999). Annual update of the HHS poverty guidelines (64 Fed. Reg. 13428–13430). Retrieved from http://www.aspe.hhs.gov/poverty/99fedreg.htm. Pearlin, L. I. (1993). The social contexts of stress. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects, 2nd ed. (pp. 303–315). New York: Free Press. Personal Responsibility and Work Opportunity Reconciliation Act of 1996. Pub. L. No. 104–193, 110 stat. 2105 et seq. Pruchno, R. (1999). Raising grandchildren: The experiences of Black and White grandmothers. The Gerontologist, 39, 209–221. Pruchno, R., & McKenney, D. (2000). The effects of custodial and co-resident households on the mental health of grandmothers. Journal of Mental Health and Aging, 6, 291–310. Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. Richardson, R. A., Barbour, N. B., & Bubenzer, E. G. (1991). Bittersweet connections: Informal social networks as sources of support and interference for adolescent mothers. Family Relations, 40, 430–434. Saluter, A. F. (1996). Marital status and living arrangements: March 1994. In Current Population Reports (Series P20–484, pp. A–7). Washington, DC: U.S. Government Printing Office. Sands, R. G., & Goldberg-Glen, R. S. (2000). Factors associated with stress among grandparents raising their grandchildren. Family Relations, 49, 97–105. Solomon, J. D., & Marx, J. (2000). The physical, mental, and social health of custodial grandparents. In B. Hayslip, Jr., & R. Goldberg-Glen (Eds.), Grandparents raising grandchildren: Theoretical, empirical, and clinical perspectives (pp. 183–206). New York: Springer. Stevens, J. H. (1984). Black grandmothers’ and black adolescent mothers’ knowledge about parenting. Developmental Psychology, 20, 1017–1025.

118

Social Diversity and Single-Parent Families

Stiffman, A. R., Orme, J. G., Evans, D. A., Feldman, R. A., & Keeney, P. A. (1984). A brief measure of children’s behavior problems: The Behavior Rating Index for Children. Measurement and Evaluation in Counseling and Development, 17, 83–90. Strawbridge, W. J., Wallhagen, M. I., Shema, S. J., & Kaplan, G. A. (1997). New burdens or more of the same? Comparing grandparent, spouse, and adult-child caregivers. The Gerontologist, 37, 505–510. Szinovacz, M. E., DeViney, S., & Atkinson, M. P. (1999). Effects of surrogate parenting on grandparents’ well-being. Journal of Gerontology: Social Sciences, 54B, S376–S388. Uhlenberg, P., & Kirby, J. B. (1998). Grandparenthood over time: Historical and demographic trends. In M. E. Szinovacz (Ed.), Handbook on grandparenthood (pp. 23–39). Westport, CT: Greenwood Press. Vega, W. A. (1995). The study of Latino families: A point of departure. In R. E. Zambrana (Ed.), Understanding Latino families: Scholarship, policy, and practice (pp. 3–17). Thousand Oaks, CA: Sage. Wakschlag, L. S., Chase-Lansdale, P. L., & Brooks-Gunn, J. (1996). Not just “ghosts in the nursery”: Contemporaneous intergenerational relationships and parenting in young African-American families. Child Development, 67, 2131–2147. Ware, J. E. (1993). SF-36 health survey: Manual and interpretation guide. Boston: The Health Institute, New England Medical Center. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. Williams, N., & Torrez, D. J. (1998). Grandparenthood among Hispanics. In M. E. Szinovacz (Ed.), Handbook on grandparenthood (pp. 87–96). Westport, CT: Greenwood Press.

Part Three

Single-Parent Families: The Lived Experience

As we witnessed in Part Two, single-parent families are extremely diverse and are becoming an increasingly common fixture in the social landscape of the United States. We wanted to continue the look at diversity of this family type and add what has come to be called “the lived experience,” meaning that variations in economic conditions, religion, race, family size, and many other factors contribute to the experiences of living in a single-parent family much to the same degree that others experience life in a two-parent family. Although much research has been conducted on income and well-being, especially among families and children, we begin Part Three with the economics of single parenthood. Mark Lino identifies first how people end up as single parents and the economic consequences thereof, the sources of income for these families, including employment and child support issues, and finally how these families spend their money. As we have seen, divorce and separation, births to never-married women, and widowhood are the primary causes of single-parent family formation. Of course, the economic ramifications of these paths to single parenthood are of great significance. All of the paths mentioned have negative economic impacts, but births to unwed, and especially teenage, women result in the most negative impacts. With the newer trend of more affluent women and men wanting families at a later age, the expectation is that these families will be less economically stressed than those who have children earlier. A very interesting profile emerges regarding the sources of income for single-parent families, and these income sources have been the center of significant policy debate in both welfare and child support policy reform

120

Single-Parent Families: The Lived Experience

efforts. As might be expected, widowed single parents generally had the highest household incomes, followed by those affected by divorce or separation, and women who were never married with children. Single-parent families maintained by a mother only often face problems such as the low earning potential of these mothers due to such things as low educational and skills acquisition, lack of job experience, discrimination, and child care issues. All create hindrances to income growth and maintenance of these families. Lack of employment or underemployment is often exacerbated by the lack of consistent child support from noncustodial parents. Several factors, such as the ex-spouse’s marital status, distance from the child, and other arrangements made child support payments more likely, while some women do not want support because of friction between exspouses and other relatives. Single mothers who were not well educated, never married, and black were much less likely to receive support than others. In terms of economic expenditures, two-parent families have approximately double the income to spend than do single-parent households. One fact has remained consistent regarding custodial parents: mothers are more often awarded either sole custody or a greater percentage of custodial rights than fathers. Researchers themselves have often ignored that many antecedent factors may account for variations in the child’s wellbeing in father-only and mother-only households. The role of the parent’s sex on the well-being of the child and the parent-child interactional quality is the focus of “Sex of Parent and Children’s Well-Being in SingleParent Families.” Is the comparison between two-parent families and single-mother families really a fair and indeed an accurate one? In short, no. Individualistic and structural comparisons provide extremely different views of this subject. No matter the family structure type—mother-only, father-only, or an array of other single-parent type families—the law and the legal system itself are involved at some level, often spanning many years. “Single Parents and the Law” represents an excellent overview of the history of law regarding single parents and the evolving legal matrix that the increasing diverse and common nature of this family type presents. One significant feature, economic stability, infuses any inquiry into family, whether it is the impact of divorce, custody, child support, or the many other aspects of family disruption. As we have seen, single mothers, especially out-of-wedlock mothers, fair far worse, as do their children, than any of the other types of single-parent families. One of the constantly significant issues regarding income and the quality of child-parent interaction is the educational level of the parent, especially post-secondary education level. If the single mother or father or both have a four-year degree, economic impact is more positive than if they are less well educated, as one would anticipate.

Single-Parent Families: The Lived Experience

121

Most laws that affect single-parent families have to do with the continuing parent-child relationship after marital disruption, especially divorce, and obviously many of these laws are being applied to never-married parents and families. Divorce law, child custody law, child support laws, and tax and welfare reform law are all of central concern here, and the authors find discrimination in legal environment a common theme. The consequences of discrimination are important since the environment in which law is practiced is not as uniform as the law itself. Certainly, there are potent myths, both cultural and societal, that have pervaded the stereotype of divorced women or women who have children out of wedlock. As example of how this might be intrinsic in the application of the law, the authors point out that the equal protection clause of the Fourteenth Amendment was not made applicable to women until 1971, and it was 1979 before the Supreme Court interpreted the amendment to apply to discrimination in family relations. Although the gender issues have narrowed and seem more aligned with modern reality, men are still more likely to suffer during child custody decision because of the historic notion that women are more natural parents and possess nurturing skills that men lack, a belief often referred to as maternal preference. As one might correctly assume, homosexual parents have an even more difficult time not only in the legal environment but also with the constantly changing and ambiguous interpretation of law. In fact, promiscuous custodial parents, cohabiting custodial parents, and even religious practices can legally affect court decisions of custody and support. The law as it relates and reacts to the changing dynamics of the singleparent family must deal with a multitude of changes in societal attitudes and realities. Fathers are perhaps seen as nurturing, grandparents seek more control legally of their grandchildren, especially if marital disruption has occurred, and even homosexuals who have custody or who adopt vie for more and more of the court’s attention. Perhaps nowhere else has the dramatic changes in the family become so evident as in the legal arena. “Idle hands are the devil’s tools.” Obviously, criminologists long ago determined to a large degree, and common sense would suggest, that children and adolescents involved in conventional activities such as sports are much less likely to become involved in delinquent and even criminal activities. Does a two-parent family influence the development of social capital in children at the same or higher level than in single-parent families as it relates to children’s time-use patterns? The time-use patterns of adolescents in single-mother families as compared to two-parent families tend to be less family-oriented than what the single mother brings to her role as a human capital source. Marital status has much less impact than other factors such as the mother’s education, employment status (income), and age. Of course as adolescents age, they increasingly

122

Single-Parent Families: The Lived Experience

shift their time use patterns to outside paid work and away from school and household-related tasks in single-mother households. The authors created three distinct hypotheses to test the effects of mother-adolescent relationships by sex, the extent to which a neighborhood had concentrations of single-parent families, and race composition on premarital sexual experiences of adolescents, especially the number of sexual partners they might have. Consistency was found in regard to twoway interactions between neighborhood level of single-parent families, and a positive correlation was found in regard to healthy motheradolescent relationships. The increased involvement in premarital sexual relations can have extremely negative consequences for these teens and families, especially intergenerational early motherhood and the consequences of low income and opportunities. Single motherhood is stressful, especially for adolescent mothers. Sorenson, Grindstaff, and Turner investigate the role that religion plays in relieving stress and offering social support to adolescent mothers. Does religiosity, the positive influence of religion in a person’s life, have an effect on well-being? Does the involvement of religion in adolescent mothers’ lifes lead to better coping skills and positive well-being for these women? Of course, teen pregnancy is often at odds with religious beliefs and values, which have differing effects than support networks may offer. Indeed, religious values may exacerbate rather than moderate the stress associated with teen pregnancy, which may cause a more negative emotional state. In most cases religion does lead to better coping skills, and the mental health and outlook of the person is positively affected. In this research, the married teen mothers did not feel as depressed or were less negatively affected by pregnancy than those who were unmarried, with religiosity being a positive factor. Nonmarital pregnancy produces the most stress for those who tend to have higher levels of religiosity due to the assertion that this runs counter to their religious values and beliefs.

Chapter 9

Expenditures on Children and Financial Difficulties Mark Lino*

One of the most dramatic changes in American family life during the past 20 years has been the growth in the number of single parents and their children. In 1970, 13 percent of all family groups with children under age 18 (3.8 million) were single-parent situations. By 1991, 29 percent of all family groups with children under age 18 (10.1 million) were singleparent situations (U.S. Bureau of the Census, 1992a). Single parents with children, as a proportion of all family groups, are expected to continue increasing in the future (Martin & Bumpass,1989). Although single parenthood may be viewed as transitional, given that most people marry or remarry (Saluter, 1989), it is estimated that more than half of all children will live, at some time in their lives, with only one parent. For children, average time spent in a single-parent situation is 6 years (Bumpass, 1984).

ECONOMICS OF BECOMING A SINGLE PARENT Single parenthood basically occurs as a result of divorce or separation, births to never-married women, or death of a spouse. In 1991, 60 percent of single-parent situations resulted from divorce or separation, 34 percent from births to never-married women, and 6 percent from death of a spouse (U.S. Bureau of the Census, 1992a). Of these various paths to single parenthood, the economic ramifications of divorce or separation have been most extensively examined.

* M. Lino. The economics of single parenthood: Past research and future directions. Marriage and Family Review, 1996, 20(1), 99–114. Reprinted with permission of Taylor & Francis Group.

124

Single-Parent Families: The Lived Experience

One of the most widely cited studies on the economic consequences of divorce or separation was conducted by Weitzman (1981). This study was based on interview data of 228 divorced men and women in Los Angeles County who had been divorced for approximately 1 year in 1978. One year after divorce, women who had been married less than 10 years had incomes between 29 and 71 percent of their predivorce household income. Comparable incomes of men 1 year after divorce were between 74 and 78 percent of their predivorce household income. Women in the highest household income group before divorce experienced the largest drop in household income after marital dissolution. Their postdivorce income, however, remained the highest of all women in the sample. Since total household income does not adjust for different family needs, Weitzman derived an income-to-needs standard (from the Bureau of Labor Statistics’ lower standard budget for an urban family of four) to measure economic well-being. On average, women experienced a 73 percent decline and men experienced a 42 percent rise in their economic well-being, as measured by this standard, 1 year after marital dissolution. Although the Weitzman study is often cited, it has also been criticized for its geographically specific sample. Weiss (1984), Duncan and Hoffman (1985), and Stirling (1989) examined the economic consequences of divorce using nationally representative data from the Panel Study of Income Dynamics. Weiss (1984) studied mothers who were divorced or separated between 1969 and 1974. For women in upper-income families prior to divorce or separation, household income in real terms 1 year after marital dissolution was approximately one-half that reported in the last year of marriage. For women in middle income families, this income was about two-thirds of what it had been, and for women in lower income families, income was about three-fourths of that in the last year of marriage. Family income in real terms for these newly created families maintained by females remained at this lower level over a 5-year observation period. Mother’s earnings became a more important source of household income after marital dissolution for all income groups. Duncan and Hoffman (1985), using an income-to-needs standard derived from U.S. poverty thresholds, found the economic consequences of divorce or separation for women with children to be less severe than found in earlier studies. Economic consequences, however, were still negative. For women who did not remarry, the income-to-needs standard declined to 87 percent of its predivorce or separation level 1 year after marital dissolution and rose to 94 percent of its predivorce or separation level by the fifth year after marital dissolution. In addition, families that experienced a marital dissolution were found to have a lower predivorce or separation income than families that did not. This further compounded the adverse economic impact of single parenthood. As the last year of marriage may be associated with a downward trend in economic status, Stirling (1989) argued that measuring postdivorce

Expenditures on Children and Financial Difficulties

125

income in relation to family income during this last year may yield biased results. Therefore, Stirling developed an income-to-needs ratio standard using average income during the last 3 years of marriage. This standard declined 30 percent for women the first year after divorce and fluctuated 2–4 percent each year in the 4 succeeding years. McLindon (1987), Wishik (1986), and Rowe and Morrow (1988) examined the economic consequences of divorce or separation using regional samples. Their results were similar to those of previous studies; divorce or separation had negative economic consequences for women who retained primary physical care of children. Rather than examine the economic consequences of divorce or separation on women, Bianchi and McArthur (1991) examined the short-term economic effects of marital dissolution on children, using the 1984 Panel of the Survey of Income and Program Participation. Children in households where the father departed experienced a 23 percent drop in average family income in real terms over a 32-month observation period. Those children who remained with two parents had an 8 percent increase in average household income in real terms, and those in households where the mother remarried or reconciled with the father had a 115 percent increase. The increase in real income of the latter group was due to the addition of another earner to the family. An income-to-needs standard derived from U.S. poverty thresholds measured a 13 percent decline in economic well-being over the 32-month period for children in households where the father departed, compared with a 7 percent increase for children in households where the parents stayed together and a 90 percent increase for those where a “father” entered the household. This decline in real income and in the income-to-needs standard for children in households where the father departed was not the result of a decrease in mothers’ labor force participation. In fact, there was an increase in labor force participation by women after marital dissolution. The economics of situations caused by widowhood or births to never married women have received less attention than situations caused by divorce or separation. Duncan and Morgan (1976) examined changes in economic status among wives who became widowed between 1967 and 1973 using the Panel Study of Income Dynamics. Of those who were nonpoor in 1967 (before the death of their spouse), 24 percent were poor in 1973 (after the death of their spouse). These figures include women both with and without dependent children.

INCOME AND INCOME SOURCES FOR SINGLE PARENTS Single-parent families maintained by mothers typically have the lowest income of all family groups. The 1991 Current Population Survey showed that average before-tax family income for single mothers with

126

Single-Parent Families: The Lived Experience

children under age 18 was $17,747; for single fathers, $30,445; and for married couple families with children, $48,737. Adjusting for family size, per capita family income for single mothers with children under age 18 was $5,506; for single fathers, $10,040; and for married-couple families with children, $11,668 (U.S. Bureau of the Census, 1992b). A substantial percentage of single-parent families fall below the U.S. poverty threshold for their family size. Poverty thresholds for 1991 were $8,865 for a two-person family, $10,860 for a three-person family, and $13,924 for a four-person family (U.S. Department of Agriculture, 1992). In 1991, 47 percent of all families composed of single mothers with children under age 18 were in poverty, compared with 20 percent of their male counterparts and 8 percent of married-couple families with children under age 18 (U.S. Bureau of the Census, 1992c). The poverty rate of single mothers and their children over the 1980s was relatively constant but always well above the rates of families maintained by single fathers and married couples. Money income does not include the value of noncash benefits such as food stamps, Medicaid, public housing, and employer-provided fringe benefits. When taxes and noncash benefits were taken into consideration, the median effective family income of single mothers in 1990 increased from 34 to 45 percent that of married couples with children (U.S. Bureau of the Census, 1991). Generally, transfer programs raise the effective income of families maintained by mothers, and taxes reduce that of married couples. Accounting for various transfer programs also reduced the percentage of families maintained by mothers in poverty, although a significant percentage, at least 32 percent, still were poor. Garfinkel and McLanahan (1986) undertook one of the few studies that examined the income and income sources of single-parent families by marital status. They looked separately at white and black families with children maintained by mothers using data from the 1983 Current Population Survey. For white families, those maintained by widows had the highest income, and those maintained by never-married women, the lowest income—less than half that of widows ($17,799 vs. $7,812). Earnings of white widows accounted for 29 percent of total household income—a lower percentage than for divorced women (69 percent), separated women (60 percent), and never-married women (58 percent). Public assistance accounted for the second largest share of income for white families maintained by never-married mothers, with 53 percent receiving such assistance. For black families maintained by mothers, there was less dispersion in income by marital status. Families maintained by divorced women had the highest income, and those maintained by separated women the lowest ($11,187 vs. $8,221). Earnings of divorced, separated, and never-married black mothers composed the bulk of household income. Social Security

Expenditures on Children and Financial Difficulties

127

was the major source of household income for widowed single parents. As with white families, public assistance accounted for the second largest share of income for black families maintained by never-married mothers, with 59 percent receiving such assistance. Garfinkel and McLanahan concluded that the low income level of single-parent families maintained by mothers reflects the low earnings of women in general. They hypothesized that women receive lower earnings because they have less education and job experience and because of market discrimination. In addition, they speculate that female single parents are hindered in working outside the home because of child care problems. The overall income of single-parent families has also declined in real terms over the years. Using Current Population Survey data, Danziger and Gottschalk (1986) examined the real income of single-parent families maintained by mothers. Over the 1973–1984 period, this income fell 8 percent for white, 9 percent for black, and 13 percent for Hispanic singleparent families. A decline in cash transfer payments accounted for some of this decrease. Employment Status A study on the employment status of single parents was done by Rawlings (1989) using the Current Population Survey. For all children under age 18 living with a single parent in 1988, 46 percent of parents were employed full time at the time of the survey, 9 percent part time, 8 percent were unemployed, and 37 percent were not in the labor force (unemployed persons are those actively looking for work, whereas those not actively looking are considered not in the labor force). Because such a large percentage of single parents were not in the labor force, the average income of their families is reduced. Using 1979 Current Population Survey data, Beller and Graham (1985) examined factors affecting the labor force participation of divorced and separated women with children. The longer the time these women had been divorced or separated and the greater the amount of public assistance and child support payments received, the less likely these women were to be in the labor force, all else equal; presence of young children also reduced this probability. The greater the educational level and age of these women, the more likely they were to be in the labor force. Divorced and separated black women and those of Hispanic origin were also less likely than white women to be in the labor force. To measure the work constraint of child care, a special 1982 Census Bureau survey asked nonemployed mothers with preschool children if they would look for work if child care were available to them at a reasonable cost (O’Connell & Bloom, 1987). Not surprisingly, many mothers with young children said they would. This was especially true for unmarried

128

Single-Parent Families: The Lived Experience

and never-married mothers. For unmarried mothers in 1982, 59 percent were in the labor force, and 17 percent stated they would look for work if reasonably priced child care were available, resulting in a potential labor force participation rate of 76 percent for this group. For never-married mothers, 50 percent were in the labor force, and 24 percent stated they would look for work if reasonably priced child care were available, resulting in a potential labor force participation rate of 74 percent. Although these potential labor force participation rates may not be realized even if reasonably priced child care were available, it does suggest that child care is a labor force constraint for many single mothers. Child Support Child support is the income source of single-parent families that has received the most research attention. Periodic supplements to the Current Population Survey have tracked child support over time as a source of income for families maintained by mothers. In Lester (1991), among women with children from an absent father, 50 percent were awarded and due child support in 1989, 42 percent had not been awarded child support, and 8 percent were awarded but not due child support that year (this support was due in a following year). Of the women not awarded child support, 64 percent desired it, but the father was deemed unable to pay, could not be located, or the mother did not pursue it even though it was wanted; 22 percent stated they did not want any child support; and 14 percent either had a final agreement pending or had made other financial arrangements. Of the women due child support, 51 percent reported receiving the full amount awarded, 24 percent received less than the full amount, and 25 percent did not receive any support at all. For single mothers being paid child support, the mean amount received in 1989 was $2,995, accounting for 19 percent of their before-tax household income. The mean amount of child support received by nevermarried mothers being paid such support was $1,888, which accounted for 20 percent of their before-tax income. Never-married mothers, however, were less likely to receive child support than divorced or separated women. Difficulties in establishing paternity and lower incomes of fathers in these situations probably are contributing factors. In real terms, child support payments received did not change much from 1981 to 1989, averaging $2,898 (in 1989 dollars) in 1981 and $2,995 in 1989. The percentage of women due child support payments, who received them, slightly increased from 72 percent in 1981 to 75 percent in 1989. Studies have examined factors affecting receipt of child support income. Teachman (1991) used data from the National Longitudinal Study of the High School Class of 1972 to do this. The higher the father’s income and amount of child support due, his being remarried, having a voluntary

Expenditures on Children and Financial Difficulties

129

support agreement, and his visiting children at least as often as stipulated in the divorce agreement, the more likely child support payments were made, all else equal. The longer the time since divorce and the farther away a father lived from his children, the less likely child support payments were made. Peterson and Nord (1990) found similar results using the 1984 Survey of Income and Program Participation. Also, single mothers who were black, had less than a high school education, or were never married were less likely to receive child support of any amount, all else equal.

EXPENDITURES OF SINGLE PARENTS Research on expenditures of single-parent families typically has used the Consumer Expenditure Survey. This survey collects data on expenditures and major socioeconomic and demographic characteristics of a sample of approximately 20,000 households weighted to the population. Using 1990 Consumer Expenditure Survey data, total expenses of singleparent families averaged $19,230 (U.S. Department of Labor, 1991). Of this total, 36 percent went for housing, 18 percent for food, 14 percent for transportation, 8 percent for clothing, 6 percent for personal insurance and pensions (including Social Security taxes), 5 percent for entertainment, 3 percent for health care, and 10 percent for other expenses (education, personal care items, and miscellaneous goods and services). Although housing consumed slightly more than one-third of the total outlay of single-parent families, only 35 percent were homeowners. Total expenditures of single-parent families were approximately half those of married couples with children; the latter group had total expenses averaging $38,999 in 1990 (U.S. Department of Labor, 1991). Married couples with children allocated a smaller proportion of their expenses to housing than single-parent families, although 76 percent were homeowners. Boyle (1989) found per capita expenditures of single-parent families also to be less than those of married couples with children. The average family size of married couples with children was four and that of single-parent families, three. Lino (1989a) examined the total expenditures of single-parent families by sex, age, and race of the parent as well as family size using the 1984–1985 Consumer Expenditure Survey. Families maintained by male single parents had higher total expenses than those maintained by female single parents ($24,626 vs. $15,022) with housing and transportation composing the major shares for both groups. Families with a single parent over age 40 had higher total expenses than those with a single parent under age 30 ($20,086 vs. $9,565); those with a single parent age 30–40 had total expenses between these two amounts. Housing accounted

130

Single-Parent Families: The Lived Experience

for a larger share of expenses in families with a parent under age 30. However, only 12 percent of these families were homeowners, compared with 69 percent of families with a single parent over age 40. White singleparent families had higher total expenses than nonwhite families ($18,739 vs. $9,991), with nonwhite families allocating nearly 40 percent of their expenses to housing. Transportation accounted for only 11 percent of total expenses in nonwhite families as 56 percent did not own an automobile. For many groups of single-parent families, total expenditures exceeded after-tax income, which probably indicates these families went into debt to cover their expenses. In another study, Lino (1989b) analyzed the expenditures of singleparent families by marital status. Total expenditures of single-parent families maintained by a widowed parent amounted to $22,071; those by a divorced or separated parent, $16,426; and those by a never-married parent, $7,741. Families maintained by a widowed and divorced or separated parent each allocated 35 percent of total expenses to housing, with 75 percent of the widowed group and 51 percent of the divorced/ separated group being homeowners. Families maintained by a nevermarried parent allocated 40 percent of their total expenses to housing, with 10 percent being homeowners. Examining expenditures of singleparent families over the 1980s, Lino (1991) found the total expenses of these families increased 3 percent in real terms from 1980 to 1989, with housing slightly increasing and food decreasing as a budgetary share. Of the individual expenses of single-parent families, housing has received the most attention. Grail (1992) used the 1989 American Housing Survey to examine the housing status of single parents and their children. Thirty-five percent of single-parent households owned a home, compared with 74 percent of married couples with children. Single parents resided in smaller living quarters than married couples with children whether they owned a home or not. For single-parent households, the median size of owned living quarters was 1,646 square feet and that of rented units, 1,202 square feet; for married couples with children, the median size of owned living quarters was 1,914 square feet and that of rented units, 1,342 square feet. Using data from the 1988 Survey of Income and Program Participation, Fronczek and Savage (1991) studied the affordability of home ownership for single mothers with children. Overall, 87 percent of these families (including current homeowners) could not afford to purchase a medianpriced home in the region where they lived. Among single mothers with children who rented a home, 97 percent could not afford to purchase a median-priced home in the region where they lived. Horton and Hafstrom (1985) examined factors affecting selected expenditures of single-parent families maintained by mothers using data from the 1972–1973 Consumer Expenditure Survey. Family income had a

Expenditures on Children and Financial Difficulties

131

positive effect on all expenditures, all else equal. Nonblack (white and other races) families spent more on shelter, less on food at home, and less on clothing than black families. The older the single mother, the more was spent on food. Lino (1990) expanded on the Horton and Hafstrom study by including single-parent families maintained by fathers in the analysis and using the 1984–1985 Consumer Expenditure Survey. Single-parent families maintained by fathers did not have different expenditure patterns than families maintained by mothers for housing, transportation, and food, all else equal. Families maintained by fathers spent less on clothing than those maintained by mothers. In addition to looking at changes in income after marital dissolution, Weiss (1984) examined changes in housing and food expenditures for women divorced or separated between 1969 and 1974 using the Panel Study of Income Dynamics. Women with children in lower and middle income families prior to marital dissolution experienced no reduction in housing expenses in real terms within 5 years. For single mothers in higher income families prior to marital breakup, real housing expenses declined sharply in the year following divorce or separation and continued at this reduced level over the next 5 years. Regardless of income, housing expenses consumed a larger percentage of income after divorce or separation than before. For all three income groups, food expenditures in real terms declined the year following marital dissolution and continued at this reduced level. For employed single mothers child care can be a large expense. Using data from the 1988 Survey of Income and Program Participation, O’Connell and Bachu (1992) examined child care arrangements and weekly child care costs. Among employed single mothers, 39 percent made payments for some type of child care. The large percentage of employed mothers without child care expenses had care provided by relatives and others at no charge. For employed single mothers with the expense, the mean weekly child care cost was $47, which accounted for 10 percent of their monthly family income. These women worked an average of 39 hours per week in the labor force.

ASSETS AND LIABILITIES OF SINGLE PARENTS Given the low income levels of single-parent families, little research has focused on their assets. Kennickell and Shack-Marquez (1992) examined the financial and nonfinancial assets and liabilities of households, including those maintained by unmarried persons under age 55 with children, using data from the 1989 Survey of Consumer Finances. This survey contains information from approximately 3,100 households weighted to the population.

132

Single-Parent Families: The Lived Experience

Sixty-eight percent of these single-parent families owned some type of financial asset (checking, savings, money market, and retirement accounts; certificates of deposit; stocks; bonds; trusts; and other). The most common financial asset owned was a checking account, held by 55 percent of these families. The median value of their financial assets was approximately $3,000. By comparison, among married couples under age 55 with children, the median value of financial assets was approximately $11,000. Regarding nonfinancial assets (principal residence, vehicles, business, investment real estate, and other), 72 percent of single-parent families reported owning one or more such assets. A vehicle was the most common nonfinancial asset held (owned by 65 percent of families). The median value of nonfinancial assets held by single-parent families was approximately $23,000; for their married counterparts the median value of nonfinancial assets was $86,000. Seventy percent of single-parent families were in debt (home mortgage, investment real estate, home equity lines of credit, credit cards, car loans, and other). Median debt was approximately $7,000 for these households. For their married counterparts, the median amount of debt carried was $31,000. Credit card debt was the most common individual liability of single-parent families, carried by 33 percent. From 1983 to 1989, the real value (in 1989 dollars) of financial assets held by families maintained by unmarried persons under age 55 with children increased by approximately $1,000, their nonfinancial assets decreased by $16,000, and their liabilities remained nearly constant. Another study by Canner and Luckett (1991) used data from the 1989 Survey of Consumer Finances to examine payment of household debt. Among all marital status groups, families maintained by divorced or separated persons had the highest rate of payment difficulties. Of indebted families maintained by divorced or separated persons, 27 percent experienced payment difficulties (such difficulties being defined as having missed or been late in debt payments in the preceding year).

CONCLUSION This review of the literature on the economics of single parenthood shows the negative economic impact associated with such a status. Family income of single-parent families maintained by mothers, who account for most single parents, is usually the lowest of all household groups. In addition, single mothers often face child care constraints that hinder their labor force participation, and they do not always receive the full amount of child support due. For most single-parent families, total expenditures consumed a large part of their income. For some, total expenditures

Expenditures on Children and Financial Difficulties

133

exceeded income with debt a likely result. Housing accounted for the largest share of total expenses for these families, although most were renters. The low home ownership rate among single-parent families contributed to their lower level of assets.

REFERENCES Beller, A. H., & Graham, J. W. (1985). Variations in economic well-being of divorced women and their children: The role of child support income. In M. David & T. Smeeding (Eds.), Horizontal equity, uncertainty, and measures of well-being, NBER studies in income and wealth (No. 50, pp. 471–509). Chicago, IL: University of Chicago Press. Bianchi, S., & McArthur, E. (1991). Family disruption and economic hardship: The short-run picture for children. (Current Population Reports Series P-70, 23). U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. Boyle, M. (1989). Spending patterns and income of single and married parents. Monthly Labor Review, 112(3), 37–41. Bumpass, L. L. (1984). Children and marital disruption: A replication and update. Demography, 21(1), 71–82. Canner, G. B., & Luckett, C. A. (1991). Payment of household debts. Federal Reserve Bulletin, 77, 218–229. Danziger, S., & Gottschalk, P. (1986). Families with children have fared worst. Challenge, 29(2), 40–47. Duncan, G. J., & Hoffman, S. D. (1985). A reconsideration of the economic consequences of marital dissolution. Demography, 22, 485–497. Duncan, G. J., & Morgan, J. N. (1976). Introduction, overview, summary, and conclusions. In G. J. Duncan & J. N. Morgan (Eds.), Five thousand American families—Patterns of economic progress (vol. IV, pp. 1–22). Ann Arbor, MI: The University of Michigan, Survey Research Center, Institute for Social Research. Fronczek, P. J., & Savage, H. A. (1991). Who can afford to buy a house? (Current Housing Reports H121/91–1). U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. Garfinkel, L., & McLanahan, S. S. (1986). Single mothers and their children. Washington, DC: The Urban Institute. Grail, T. S. (1992). Housing characteristics of one-parent households, 1989. (Current Housing Reports H121/92–2). U.S. Department of Housing and Urban Development and U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. Horton, S. E., & Hafstrom, J. L. (1985). Income elasticities for selected consumption categories: Comparison of single female-headed and two-parent families. Home Economics Research Journal, 13, 292–303. Kennickell, A., & Shack-Marquez, J. (1992). Changes in family finances from 1983 to 1989: Evidence from the Survey of Consumer Finances. Federal Reserve Bulletin, 78(January), 1–18.

134

Single-Parent Families: The Lived Experience

Lester, G. H. (1991). Child support and alimony: 1989. (Current Population Reports Series P-60, 173). U.S. Bureau of the Census. Washington DC: U.S. Government Printing Office. Lino, M. (1989a). Financial status of single-parent households. Family Economics Review, 2(1), 2–7. ———. (1989b). Financial status of single-parent households headed by a never married, divorced/separated or widowed parent. In R. Walker (Ed.), Families in transition: Structural changes and effects on family life (pp. 151–160). Proceedings of the 1989 pre-conference workshop of the Family Economics–Home Management Section of the American Home Economics Association. ———. (1990). Factors affecting expenditures of single-parent households. Home Economics Research Journal, 18, 191–201. ———. (1991). Changes in income and expenditures for families with children in the 1980s. In J. Bauer (Ed.), Family economic well-being in the next century: Challenges, changes, continuity (pp. 159–166). Proceedings of the 1991 pre conference workshop of the Family Economics–Home Management Section of the American Home Economics Association. Martin, T. C., & Bumpass, L. L. (1989). Recent trends in marital disruption. Demography, 26, 37–51. McLindon, J. B. (1987). Separate but unequal: The economic disaster of divorce for women and children. Family Law Quarterly, 21, 351–405. O’Connell, M., & Bachu, A. (1992). Who’s minding the kids? Child care arrangements: Fall 1988. (Current Population Reports Series P70–30). U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. O’Connell, M., & Bloom, D. E. (1987). Juggling jobs and babies: America’s child care challenge. Issue Number 12, Population Trends and Public Policy. Population Reference Bureau, Inc. Peterson, J. L., & Nord, C. W. (1990). The regular receipt of child support: A multistep process. Journal of Marriage and the Family, 52, 539–551. Rawlings, S. W. (1989). Single parents and their children. (Current Population Reports Series P-23,162). U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. Rowe, B. R., & Morrow, A. M. (1988). The economic consequences of divorce in Oregon after ten or more years of marriage. Willamette Law Review, 24, 463–484. Saluter. A. F. (1989). Changes in American family life. (Current Population Reports Series P-23. 163). U.S. Bureau of the Census. Washington, DC: U.S. Government Printing Office. Stirling, K. J. (1989). Women who remain divorced: The long-term economic consequences. Social Science Quarterly, 70, 549–561. Teachman, J. D. (1991). Who pays? Receipt of child support in the United States. Journal of Marriage and the Family, 53, 759–772. U.S. Bureau of the Census. (1991). Measuring the effects of benefits and taxes on income and poverty: 1990. (Current Population Reports Series P-60, 176-RD). Washington, DC: U.S. Government Printing Office. ———. (1992a). Household and family characteristics: March 1991. (Current Population Reports Series P-20, 458). Washington, DC: U.S. Government Printing Office.

Expenditures on Children and Financial Difficulties

135

———. (1992b). Money income of households, families, and persons in the United States: 1991. (Current Population Reports Series P-60, 180). Washington, DC: U.S. Government Printing Office. ———. (1992c). Poverty in the United States: 1991. (Current Population Reports Series P-60. 181). Washington, DC: U.S. Government Printing Office. U.S. Department of Agriculture, Family Economics Research Group. (1992). Poverty thresholds. Family Economics Review, 5(3), 36. U.S. Department of Labor, Bureau of Labor Statistics. (1991). Consumer expenditures in 1990. News, 91, 607. Weiss, R. S. (1984). The impact of marital dissolution on income and consumption in single-parent households. Journal of Marriage and the Family, 46, 115–127. Weitzman, L J. (1981). The economics of divorce: Social and economic consquences of property, alimony, and child support awards. UCLA Law Review, 28, 1181–1268. Wishik, H. R. (1986). Economics of divorce: An exploratory study. Family Law Quarterly, 20(1), 79–103.

Chapter 10

The Psychological Well-Being of Single Parents Anna-Marie Cunningham and Chris Knoester*

In the last three decades, the number of single-parent families has increased dramatically. Single-parent families are the fastest-growing family type in the United States and now constitute 31 percent of all families. Single-mother households represent approximately 85 percent of single-parent families, or 10 million households. The number of singlefather households has more than tripled in the last three decades. In 1970 single-father families comprised only 1 percent of all families, with less than 400,000 households. By 2000, the number of single-father households had reached 2 million (Fields & Casper, 2000). Because of these changes in family structure, it is especially important to understand the relationships between family structure, gender, and parental well-being. There is consistent evidence that raising children is negatively associated with parents’ psychological well-being, and the relationship is exacerbated for single parents (Bird 1989; Evenson & Simon, 1997; Hughes, 2005). Also, it appears that mothers are more likely to be affected by the burdens of child rearing than are fathers (Nomaguchi, Milkie, & Bianchi, 2005; Mattingly & Sayer, 2006). However, the mechanisms through which parenting affects well-being are not well understood. In addition, the relative scarcity of single fathers has prevented them from being a central focus in the literature. The research that examines the psychological well-being of parents also fails to consider that men and women may manifest their well-being differently.

* A. Cunningham, C. Knoester (2007). Marital Status, Gender, and Parents’ Psychological Well-Being. Sociological Inquiry, 77(2), 264–287. Copyright © 2007, Wiley. Reproduced with permission of Blackwell Publishing Ltd.

The Psychological Well-Being of Single Parents

137

The purpose of this study is to analyze the extent to which being a man or a woman and being married or single are associated with parents’ feelings of well-being. The sample for this study includes parents who have biological or adopted children under the age of 18 residing in the home. Measures of psychological well-being include reports of depressive symptoms and alcohol abuse. The economic, work, and emotional burdens associated with having children are assessed as mediators and moderators of the relationships between gender, marital status, and parents’ psychological well-being. We focus on the extent to which the heightened burdens of single parenthood and of the larger share of the division of household labor that mothers have traditionally fulfilled, relative to fathers, which can explain differences in the depressive symptoms of parents of different genders and marital statuses.

CONCEPTUAL FRAMEWORK Our conceptual framework is based on expectations of how gender and marital status may be related to parents’ psychological well-being. First, we introduce theory about the extent to which the burdens of parenting can explain gender and marital status differences in parents’ psychological well-being. Then, we discuss evidence for why parenting burdens may mediate gender and marital status differences in parents’ psychological well-being and why the impact of parenting burdens on psychological well-being may depend on parents’ gender and marital status. Finally, we suggest that traditional measures of depression may be biased toward women and that there are reasons to expand our conception of the symptoms of depression.

Explanations for Gender and Marital Status Differences in Parents’ Psychological Well-Being

Individualist and Structuralist Expectations The literature that attempts to explain the differences in how men and women experience the roles of parenthood often relies on the individualist paradigm. That is, gender roles are internalized as enduring personality characteristics due to a powerful socialization process. In recent years, many have argued that this perspective offers a biased explanation and does not examine the role that the social environment plays in contributing to different parenting experiences among men and women (Risman, 1987). The effects of parenthood, according to a structuralist perspective, are a function of the social conditions that parents experience. We expand on both arguments

138

Single-Parent Families: The Lived Experience

and how each perspective offers different predictions about gender differences in the consequences of parenting. The individualist perspective argues that gender roles are internalized as personality traits. By adulthood, men and women have developed distinct personality traits that are gendered behaviors. Men become more competitive and work-oriented, while women become more nurturing and child-centered (Risman, 1987). Individualists attribute these differences to the combination of childhood socialization and biological differences between men and women, but especially emphasize the socialization process (Downey, Ainsworth-Darnell, & Durfur, 1998; Risman, 1986). Gendered personality traits are expected to enable women to adjust better, relative to men, to parenting demands. Therefore, women are also expected to engage in more parenting work. Structuralism offers a different explanation for gender differences in parenting experiences. A structuralist paradigm claims that parenting experiences result from different opportunities for men and women, shaped by factors such as economic forces, access to social networks, and gendered divisions of labor. Behaviors are not established by early childhood socialization but emerge as people adapt to daily interactions (Risman, 1987). Structuralists contend, for example, that single fathers are no less prepared for parenting than are single mothers. In fact, all single parents confront undefined aspects of the parental role and cross gender boundaries when raising children (Hilton, Desrocher, & Devall, 2001). Thus, if women and men encounter similar opportunities then similar levels of well-being should result. Individualist and structuralist perspectives have important implications for understanding how parenting may affect psychological wellbeing. The individualist argument for women’s socialization into the parental role suggests that women manage parenthood better than do men even in situations in which their responsibilities may be similar— such as when the burdens of parenting are equal. Thus, single mothers might be expected to exhibit higher levels of psychological well-being than single fathers. In contrast, a structuralist perspective predicts that fathers and mothers respond to parenthood similarly under the same circumstances. If the structural pressures and expectations of single fatherhood and single motherhood are the same, both single mothers and single fathers might be expected to experience similar levels of psychological well-being. However, to the extent that there are substantial gender differences in the amount of parenting burdens that mothers and fathers take on, their levels of psychological well-being may vary.

Marital Status, the Division of Labor, and the Marriage Premium Single and married parents may also differ in their psychological well-being because of different divisions of labor. In addition, marriage may offer a premium

The Psychological Well-Being of Single Parents

139

to one’s psychological well-being beyond its potential to reduce one’s share of parenting responsibilities. First, marital status may predict parents’ psychological well-being because the presence of a spouse may alleviate the burdens of parenting. A spouse may be expected to offer help in performing parenting tasks, contribute economic resources, and provide support to reduce the negative impact of parenting burdens on a parent’s psychological well-being. Single parents are less likely to have their parenting burdens alleviated by a committed coparent (McLanahan & Sandefur, 1994; Mirowsky & Ross, 2003; Nomaguchi & Milkie, 2003). Furthermore, differences may exist in the psychological well-being of married versus single parents that are not a function of the different burdens to which single and married parents are exposed. Married individuals report less depression and less anxiety and feel happier than nonmarried individuals (Gove, Hughes, & Style, 1983; Ross, Mirowsky, & Goldsteen, 1990). House, Umberson, and Landis (1988) posit that close interpersonal ties and social relationships such as marriage provide a sense of well-being. Waite and Gallagher (1995, 2000) argue that not only does marriage protect individuals against life’s uncertainties but it also provides individuals with a sense of obligation and feelings of greater meaning in life. The psychological benefits of marriage may also stem from the social expectations centered on the institution of marriage. Married individuals may obtain some peace of mind after they fulfill the social norm and common personal goal of marrying (Simon, 1997). In sum, an individualist perspective of gender and a marriage premium perspective of marital status suggest that mothers, compared to fathers, and married parents, compared to single parents, are less likely to experience lower levels of psychological well-being at equal levels of parenting burdens. Nevertheless, consistent with a structuralist perspective of gender and a division of labor focus on marriage, there is evidence consistent with the notion that gender and marital status differences in parenting burdens may explain gender and marital status differences in parents’ psychological well-being. Yet, the influence of parenting burdens on psychological well-being may also depend on parents’ gender and marital status. We discuss this evidence in the following section. Gender and Marital Status Differences in Parenting Burdens and Parents’ Psychological Well-Being We propose that the primary mechanisms through which parenting influences psychological well-being are economic strains, work demands, and emotional burdens. These mechanisms may help to explain differences in psychological well-being not only between parents and nonparents but also between parents of different genders and marital statuses (Bianchi et al., 1993; Mirowsky & Ross 2000; Richards &

140

Single-Parent Families: The Lived Experience

Schmiege, 2003). Social support and religious involvement may allow parents to better cope with the burdens of parenting and partially explain differences between mothers’ and fathers’ feeling of well-being, as well as the differences between married and single parents’ well-being. Parenting burdens may also moderate gender and marital status differences in parents’ psychological well-being. One mechanism through which parenthood affects psychological wellbeing is through the economic strains that caring for children produce. The presence of children in the home increases economic strains on families and may increase the risk of depression among parents (Mirowsky & Ross, 2003; Ross & Huber, 1985). There are substantial gender differences in economic strain. Fathers usually earn higher incomes than mothers. Single-mother families have a poverty rate that is more than twice as high as that in single-father families (Hilton, Desrocher, & Devall, 2001; Saluter, 1993). Yet, there is also reason to believe that economic strain may affect fathers’ feelings of well-being more than it does mothers’ because men are expected to be better economic providers than are women (Mills et al., 1992). Indeed, Ross and Huber (1985) find that economic hardship has a greater impact on husbands’ well-being than it does on wives’ well-being. Economic strains are pronounced for single parents because they often assume the entire responsibility for their child’s financial needs (Richards & Schmiege, 1993). Indeed, the poverty rate among single-parent families is two to five times greater than among two-parent families (Hilton, Desrocher, & Devall, 2001; Saluter, 1993), and many of the difficulties experienced by single parents are attributable to economic strain (McLanahan & Adams, 1987; Mirowsky & Ross, 2003). A second mechanism is the increase in work-related burdens that accompany parenthood. Time spent in housework is associated with decreases in psychological well-being for both men and women (Glass & Fujimoto, 1994). Indeed, Nomaguchi and Milkie (2003) find that married mothers and both single mothers and fathers spend more hours of housework than do their childless counterparts. The demands of caring for children’s needs can also lead to lower levels of psychological well-being (McLanahan & Adams, 1987; Nomaguchi & Milkie, 2003). More tasks and responsibilities are required to maintain the household among families with children than in households with no dependent children. Previous work notes the gender gap in the division of household labor. Women consistently perform more cooking, cleaning, and child care tasks than do men (Coltrane, 1996; Hochschild, 2003; Nomaguchi & Milkie, 2003; South & Spitze, 1994). South and Spitze (1994) find that a gender gap exists among all marital statuses (never married, cohabiting, married, divorced, and widowed), but the largest gap in household labor is found between husbands and wives. Although divorced and widowed men

The Psychological Well-Being of Single Parents

141

perform substantially more housework than do any other group of men, women in the same status category still perform substantially more housework and child care than do men. Yet, because household labor and child care is largely viewed as women’s work, equal levels of caretaking responsibilities may have a greater negative impact on fathers’ psychological well-being than on mothers’ psychological well-being. Furthermore, in contrast to married parents, single parents generally undertake the entire responsibility for housework and child care while often working full time. Richards and Schmiege (1993) find that aside from financial concerns, task overload is the second highest problem identified by single parents. The ability to balance work, child care, and household responsibilities is a common difficulty for both single mothers and fathers. Single parents rarely live with other adults who can offer supplemental support when competing demands are encountered (Wright, 1989). A third mechanism through which parenthood affects psychological well-being involves the emotional demands of parenting. Even beyond the financial and work burdens of parents, the responsibility for a child’s physical and emotional well-being can wear on parents. Child-related concerns can adversely affect parents’ psychological well-being (Belsky, Robins, & Gamble, 1984; Greenberger & O’Neill, 1990). However, social support can alleviate the emotional burdens of parenting (Gerster, Kohler, & Rosenfield, 1985; Ross, Mirowsky, & Goldsteen, 1990). There are also differences in the receipt of social support among men and women (Mirowsky & Ross, 2003). Turner and Marino (1994) report that women experience substantially higher levels of social support than do men and men may not seek out help even when they need it. Because of the buffering effects of social support on psychological well-being, this finding suggests that women may be better able to cope with a burdensome situation than men. Furthermore, due to expectations that men are not to express emotions or a need for help (Balswick & Pete, 1974), the receipt of social support may be more beneficial to a mother’s psychological well-being than to a father’s psychological well-being—although the receipt of social support can have adverse effects on families as well (Cohen, 1996; Robertson et al., 1991; Walen & Lachman, 2000). Similarly, Weiss (1979) argues that a fundamental problem in singleparent families is the lack of immediately available support. Stress within two-parent homes is limited because there are, by definition, two individuals to absorb the strain. When one parent experiences difficulty with child management or maintaining the home, another parent is available to help. The consensus is that, overall, nonmarried individuals have fewer psychological resources and smaller social networks than do married individuals (McLanahan, 1983; Turner & Marino, 1994).

142

Single-Parent Families: The Lived Experience

Religious organizations also provide support to individuals through the social networks that they offer and the psychological benefits of religious beliefs and practices. Because parents are more likely to attend church than do individuals without children, they may be especially likely to use these resources (Stolzenberg, Blair-Loy, & Waite, 1995). Women tend to have higher levels of participation in religious organizations than do men (Cornwall, 1989; Stolzenberg, Blair-Loy, & Waite, 1995). To the extent that women are more religious than men, social support proffered through religious involvement may be more likely to help mothers cope with the burdens of parenting when compared to fathers. In addition, married parents may be more likely to benefit from the social and psychological support offered by religious organizations because religious institutions generally encourage marriage and view it as sacred (Iannaccone, 1990; Stolzenberg, Blair-Loy, & Waite, 1995). Thus, single parents may not feel as comfortable attending religious services as married parents. Indeed, the probability of belonging to a religious organization is four to eight times higher for married individuals than for those not married (Stolzenberg, Blair-Loy, & Waite, 1995). In sum, the combination of economic, work-related, and emotional burdens of raising children appears to lead to lower levels of psychological well-being among parents. Social support and religious participation may ease the weight of these burdens. There are substantial gender and marital status differences in the experiences and expectations of parenting that may explain gender and marital status differences in psychological well-being. However, the impact of parenting burdens on parents’ psychological well-being may also depend on parents’ gender and marital status. Gendered Measures of Psychological Well-Being The third part of our conceptual framework emphasizes that traditional measures of psychological well-being may be inadequate. These measures often consist of questions designed to assess depressive symptoms; however, the questions focus on internalized symptoms and generally fail to consider that men may be more likely than women to exhibit depressive symptoms through externalizing behaviors. Thus, measures of drug abuse, alcohol abuse, and the frequency of alcohol consumption are increasingly being used to assess men’s levels of depression, in particular (Horwitz, White, & Howell-White, 1996; Pearlin & Radabaugh, 1976). For example, Simon (2002) argues that men are more likely to manifest their depression in an external form (e.g., alcohol abuse), but women are more likely to report internal symptoms (e.g., feeling sad). She finds that men and women respond to marital transitions differently; women report higher levels of depression, but men report more alcohol abuse

The Psychological Well-Being of Single Parents

143

than do women. In fact, she contends that alcohol use and self-reports of depression are functionally equivalent measures of emotional problems. Similarly, in their study of the effects of marriage among young adults, Horwitz and White (1991) find that men have significantly higher rates of alcohol abuse than do women. However, women are significantly more likely to report being depressed. These findings again suggest that there are significant differences in male and female depressive symptoms. Previous research has not adequately explored the extent to which parenting burdens may be associated with externalized depressive symptoms. Pearlin and Radabaugh (1976) find that alcohol consumption is often used as a coping mechanism in the presence of strain. In fact, nearly one-fourth of the people who drink do so to assuage stressful situations (Mulford & Miller, 1963; Polich & Orvis, 1979). Therefore, in this study, we use reports of traditional symptoms of depression as well as alcohol abuse to measure psychological well-being.

HYPOTHESES The conceptual framework of this study and previous research suggest the following hypotheses. First, single parents are expected to be more likely to report traditional symptoms of depression and rates of alcohol abuse than are married parents. Second, mothers are expected to report more traditional symptoms of depression, and fathers are expected to report more alcohol abuse under similar structural circumstances. Finally, economic strain, responsibilities for child care and housework, social support, and religious involvement are expected to at least partially mediate the relationships between gender, family structure, and depressive symptoms. Previous research also suggests potential moderating effects such that equal levels of parenting burdens may be more positively associated with fathers’ depressive symptoms, compared to mothers’, because of societal expectations that men be better breadwinners than women (Coltrane, 1996; Hochschild, 2003), that they perform less of the proportion of household labor and child care (Hochschild, 2003; West & Zimmerman, 1987), and that they be less likely to ask for and receive social support (Balswick & Peek, 1971; Turner & Marino, 1994). Single parents’ depressive symptoms, compared to married parents’ depressive symptoms, may also be more positively associated with parenting burdens because single parents are more likely to lack contributions from a coparent (McLanahan & Sandefur, 1994; Nomaguchi & Milkie, 2003) and do not experience the buffering effects of an additional marriage premium on their risk of experiencing depression (Waite, 2000).

144

Single-Parent Families: The Lived Experience

Our theoretical perspectives predict different outcomes for the mediating and moderating influences of parenting burdens. Evidence that parenting burdens completely mediate gender differences in depressive symptoms is consistent with a structural perspective of gender because it suggests that mothers and fathers respond similarly to the same structural opportunities. An absence of mediating effects is consistent with an individualist perspective of gender if the findings suggest that mothers are less depressed than are fathers under equal structural circumstances. Evidence that parenting burdens completely mediate marital status differences in depressive symptoms is consistent with the perspective that a division of labor of parenting burdens can explain marital status differences in parents’ depressive symptoms. An absence of mediating effects is consistent with the view that an additional marriage premium negatively affects parents’ depressive symptoms. Evidence that parenting burdens interact with gender is most consistent with an individualist perspective if the interaction effect suggests that the burden is more positively associated with fathers’ depressive symptoms, compared to mothers’ symptoms, all else being equal. Yet, this finding is not incompatible with a structuralist perspective if it represents gendered parenting expectations (i.e., mothers may have a greater pressure to adequately fulfill parenting roles than do fathers) rather than the relative success of mothers and fathers in coping with equal parenting burdens. Finally, evidence that parenting burdens interact with marital status in predicting depressive symptoms is consistent with a marriage premium perspective on the benefits of marriage for parents’ psychological well-being if the finding suggests that single parents’ depressive symptoms are more positively associated with parenting burdens than are married parents’ depressive symptoms.

ANALYTIC STRATEGY Our analysis of the associations between marital status, gender, and parents’ psychological well-being occurs in three main stages. Using Ordinary Least Squares (OLS) regression, we first assess the associations between gender and marital status with reports of traditional symptoms of depression and alcohol abuse. Then, we test for interactions between gender and marital status in predicting depressive symptoms, prior to accounting for the additive effects of our hypothesized mechanisms. We run separate models for two different indicators of depressive symptoms: reports of traditional depression symptoms and alcohol abuse. Second, we assess the relative roles of financial security, work demands at home, social support, and religious involvement as mediating factors between our independent and dependent variables. To do so, we add measures of these indicators of parenting burdens to our basic additive models.

The Psychological Well-Being of Single Parents

145

Finally, after accounting for the additive effects of our measures of the hypothesized mechanisms, we explore potential interaction effects between gender and marital status, and between these independent variables and our measures of burdens of parenting, in predicting psychological well-being. We test for these interactions by adding a multiplicative interaction term to our full additive models, one at a time. If an interaction is significant, we graph it to better understand its effects. In graphing interactions, we calculate expected values for the dependent variable after using various combinations of values for the variables involved in the interaction effect and setting the values of all other variables in the equation to their means. To efficiently display interaction effects, we present a full interaction model for each dependent variable that contains all of the significant interaction terms (at p < .05) that were significant when they were analyzed separately. In all analyses, we use listwise deletion of missing data.

METHODS Data This study uses data from the second wave of the National Survey of Families and Households (NSFH), which is a national probability sample of households in the United States (Sweet & Bumpass, 1996). The NSFH was originally administered in between 1987 and 1988 to 13,017 individuals. The second wave of interviews was administered in 1992 to 10,008 respondents. Due to oversampling in Wave 1, higher proportions of single-parent families and racial minorities are included in the NSFH. We use the second wave of data for our main outcome variables in the analysis because it contains comprehensive questions on alcohol abuse that were not included in Wave 1. Specifically, Wave 1 contains a question that only asks whether one has a problem of alcohol abuse, while Wave 2 contains more detailed questions about how often and how much one drinks. Also, we are able to control for reports of depressive symptoms and alcohol abuse at Time 1 in our analysis in an attempt to partially account for preexisting propensities toward reporting traditional depressive symptoms and alcohol abuse. To assess whether the attrition of respondents between waves may affect our results, we performed supplementary analyses that used the Heckman (1979) method for adjusting for attrition. This procedure did not change our findings, and we present our original models for the sake of parsimony. Because this study focuses on the depressive symptoms of parents, our sample is limited to main respondents coresiding at least 50 percent of the time with their biological or adopted child under 18 years of age (N = 3975).

146

Single-Parent Families: The Lived Experience

Dependent Variables Two measures of parental well-being are examined: traditional depressive symptoms and alcohol abuse. We measure traditional depressive symptoms with 12 items (e.g., how many days in a week have you felt sad? felt lonely? etc.), from the Center for Epidemiological Studies Depression Scale (CES-D). This is a commonly used measure of depressed mood that has high construct validity and internal consistency (Radloff, 1977). The responses range from 0 to 7 days per week and were summed to form a scale that ranges from 0 to 84. We measure alcohol abuse by the question “On about how many days did you have five or more drinks on the same occasion during the past 30 days?” (see Simon, 2002). Scores range from 0 to 30. Independent Variables The primary independent variables are gender (1 = Father) and marital status (1 = Single). Financial and work demands are mechanisms through which gender and marital status may influence parents’ risk of depression. We use the question “How often do you worry that your total family income will not be enough to meet your family’s expenses and bills,” to measure economic strain. Responses range from 1 to 5, higher responses indicate greater economic strain, (1 = Never; 5 = All most all of the time). Time spent on household tasks includes respondents’ estimates of the number of hours per week they usually spent in preparing meals, washing dishes, cleaning house, outdoor tasks, shopping, washing and ironing clothes, auto maintenance, paying bills, and driving. Responses range from 0 to 112. In order to correct for overestimates, we recode any reports that are two standard deviations above the mean, to the mean number of hours by gender, plus two standard deviations. The amount of time respondents spend with children involves reports of time spent (1) in leisure activities away from home; (2) at home working on a project or playing together; (3) having private talks; and (4) helping with reading or homework. Responses range from 1 to 6 (1 = Never; 6 = Almost every day) and are summed to form a scale that ranges from 4 to 24. Social support is a protective mechanism that may offset the financial, work, and emotional demands of parenting. Instrumental support includes four indicators that denote if respondents received help (0 = No, 1 = Yes) from persons not living in the household (parents, siblings, friends and coworkers, and other relatives) with (1) babysitting or child care; (2) transportation; (3) repairs to home or car; and (4) other kinds of work around the house. Responses are summed to form a scale that ranges from 1 to 12. Emotional support is assessed by asking if respondents received advice, encouragement, or moral support during the last

The Psychological Well-Being of Single Parents

147

month from the same group of individuals (1 = Yes). Responses were summed and range from 0 to 4. Participation in religious services consists of the number of times one attends per year (0  Never, 1 = 111 times, 2  1224 times, 3  2452 times, 4  52  times). Control Variables We control for variables that are associated with both marital status and psychological well-being. Marriage is positively associated with education and being employed (Marini, 1989). African Americans and Hispanics are less likely to be married than are whites (Waite, 1995). Married individuals also tend to be older and have more children than nonmarried individuals (Spanier, 1983). Further, 15–37 percent of singleparent families have another adult, often times a parent, present in the household who may help to alleviate burdens associated with maintaining the household (Bianchi, 1995). Education, employment status, race, age, marital status, and number of children are also associated with psychological well-being (Liem & Liem, 1978; McLanahan, 1983; Mirowsky & Ross, 1992). Therefore, the background characteristics of years of education, usual hours worked per week, Black, Hispanic, age, recent marital breakup, presence of a coresident adult or romantic partner, and the number of children are controlled for in our models. We also control for the Wave 1 measures of our dependent variables in our models. Traditional depressive symptoms at Time 1 are derived from the same questions that are used to create our (Time 2) dependent variable. Alcohol abuse at Time 1 is drawn from a question that asks whether one has a problem with alcohol abuse. RESULTS Overall, the results of the analyses indicate that single parents are significantly more likely to report traditional symptoms of depression and alcohol abuse than are their married counterparts. In addition, mothers are more likely to report traditional symptoms of depression, and fathers are more likely to abuse alcohol. Our measures of the burdens of parenthood partially mediate marital status and gender differences in reports of depression and alcohol abuse. Finally, we find some evidence of interaction effects among our independent variables, and between our independent variables and the burdens of parenthood, in predicting psychological well-being. Below, we detail the results for each analysis. In the first set of analyses, we examine gender and marital status differences in parents’ traditional symptoms of depression. Results indicate that, as expected, fathers report significantly lower levels of traditional depressive symptoms than do mothers (b  1.11, p  .05),

148

Single-Parent Families: The Lived Experience

even after controlling for background characteristics and Time 1 depression. Parents that are single report more depressive symptoms than do their married counterparts (b  4.24, p  .001). Because marital status and gender are our primary independent variables, we also test for a potential interaction between them in predicting depression prior to controlling for the indicators of parenting burdens. As expected, there is a significant interaction between gender and marital status that signifies that, when we do not control for the influence of the burdens of parenting, being a single parent instead of a married parent is more positively associated with traditional depressive symptoms for mothers than it is for fathers. In Model 2, it is evident that accounting for the burdens of parenting mediates the relationship between gender and parents’ depression, with gender no longer being significant (b  1.11, p  .05 to b  .48, p  .10). In addition, parenting burdens partially mediate the association between marital status and psychological well-being (b  4.24, p  .001, to b  2.28, p  .001). These results are consistent with a structural perspective of gender and the benefits of a division of labor among married parents. In supplementary analyses (results not shown), each hypothesized mechanism was entered into the model by itself, in order to gauge the extent to which it explains the decrease in the association between our independent and dependent variables. We find that emotional support accounts for almost the entire difference between mothers’ and fathers’ reports of depression. Economic strain appears to account for nearly the entire difference between single and married parents’ reports of depression. As expected, each indicator of parenting burdens is significantly associated with parents’ reports of traditional depressive symptoms. As expected, economic strain (b  3.85, p  .001) and spending time in household labor (b  .03, p  .05) are positively associated with traditional depressive symptoms. Religious involvement is negatively related to traditional depressive symptoms (b  .48, p  .001). However, unexpectedly, we find that spending time with children is negatively associated with reports of depression (b  .13, p  .05). One interpretation of this finding is that our measures capture the amount of free time that parents spend with their children, rather than parents’ performance of monotonous child care tasks. Time with children seems to allow parents an escape from the burdens of parenting if they take pleasure in their children’s company and development. Another unexpected finding is that receiving instrumental support and emotional support from individuals outside of the home is positively associated with depression (b  .41, p  .05, b  .64, p  .01). Seeking help or emotional advice from outside of the home is not without its costs, in part because obtaining outside help may be seen by parents as a

The Psychological Well-Being of Single Parents

149

symbol of inadequate parenting (Robertson et al., 1991; Walen & Lachman, 2000). An offer of supplementary help or encouragement may also be an indication of existing problems not related to the burdens of parenting per se. The full interaction model for predicting depression contains the interaction terms between our independent variables and indicators of parenting burdens that were significant at p  .05 when entered into Model 2 alone. After accounting for the effects of parenting burdens, we find there is no longer a significant interaction between marital status and gender; therefore, a marital status–gender interaction term is not included in our full interaction model (i.e., Model 3). This result suggests that the interaction between marital status and gender in predicting depression is a function of differences in the burdens of parenting. To better explore this process, we entered measures of each one of the hypothesized mechanisms into the model by itself to see if it accounted for the change in significance of the interaction variable. We found that controlling for economic strain resulted in the interaction term between marital status and gender becoming nonsignificant. This finding is consistent with previous work that suggests that economic strain is a particularly oppressive parenting burden for single mothers (McLanahan & Sandefur, 1994). In Model 3, we present our full interaction model for predicting traditional symptoms of depression. First, we find evidence that economic strain, time with children, instrumental support, and emotional support interact with gender in predicting parents’ depression. Consistent with an individualistic perspective of gender, equal levels of spending time with one’s children and receiving emotional support are more positively associated with traditional symptoms of depression among fathers than among mothers. Unexpectedly, equal levels of economic strain and instrumental support are more positively associated with mothers’ traditional depressive symptoms compared to fathers’ symptoms. Thus, we do not find support for the individualist hypothesis that economic strain has a greater impact on fathers’ depressive symptoms, but we do find support for the expectation that spending time with children is more positively associated with fathers’ depressive symptoms. Furthermore, we find mixed support for our expectation that the receipt of social support is less likely to be associated with lower levels of depressive symptoms for fathers than for mothers. As expected, emotional support does not appear to buffer the burdens of parenting among fathers as effectively as it might among mothers; however, the receipt of instrumental support is more positively associated with depressive symptoms among mothers than it is among fathers. Economic strain, time with children, and emotional support also interact with marital status in predicting parents’ depression. Consistent with perspectives on the benefits of marriage, economic strain is more

150

Single-Parent Families: The Lived Experience

positively associated with single parents’ reports of depression, and emotional support appears to play a larger role in buffering the risk of depression among single parents compared to their married counterparts. Spending time with children is also more negatively associated with traditional depressive symptoms among single parents than among married parents. We now turn to predicting alcohol abuse. As expected, the coefficients from Model 1 indicate that fathers abuse alcohol at significantly higher levels than mothers (b  .93, p  .001). Parents who are single also report significantly higher rates of alcohol abuse than do married parents (b  .32, p  .001). We test for potential interaction effects between our independent variables prior to adding indicators of parenting burdens to our model (results not shown). As expected, there is a significant interaction between marital status and gender that indicates that, when we do not control for the influence of the burdens of parenting, being a single parent instead of a married parent is more positively associated with reports of alcohol abuse for fathers than it is for mothers. As expected, the results from Model 2 indicate that the burdens of parenthood partially mediate the relationship between gender and alcohol abuse (b  .93, p  .001 to b  .88, p  .001) and marital status (b  .32, p  .001 to b  .21, p  .05). In supplementary analyses (results not shown), we entered each indicator of our hypothesized mechanisms for the burdens of parenting into the model, one at a time, and found that religious involvement accounts for most of the modest mediation of gender and marital status differences in alcohol abuse. Indeed, in Model 2, religious involvement is negatively associated with the alcohol abuse (b  .17, p  .001). In contrast to our predictions about the burdens of parenting, spending time with children is negatively associated with the abuse of alcohol (b  .02, p  .01). This may once again reflect that parents relish spending time with their children in many activities and do not view this time together as burdensome. In addition, parents who spend a lot of time with their children may have relatively few opportunities or desires to consume alcohol. The receipt of instrumental support (b  .08, p  .01) is positively associated with alcohol abuse. Again, this may reflect the costs of receiving support and that one is more likely to seek out or be given help only after it is absolutely necessary. We present our full interaction model for predicting alcohol abuse in Model 3. There are four significant interactions. First, the gender–marital status interaction indicates a greater disparity in the abuse of alcohol between married and single fathers than between married and single mothers. Second, religious involvement is more negatively associated with fathers’ abuse of alcohol than it is with mothers’ abuse. Third, there is evidence that economic strain has a stronger positive relationship with alcohol abuse among fathers than among mothers. Finally, our results indicate that

The Psychological Well-Being of Single Parents

151

receiving instrumental support is more positively associated with fathers’ alcohol abuse than with mothers’ alcohol abuse. Overall, these interaction effects are largely consistent with expectations that fathers are more likely than mothers to exhibit externalizing depressive symptoms. In addition, the interactions provide some support for the expectations that fathers may be more likely than mothers to exhibit depressive symptoms when experiencing equal levels of economic strains and receipt of social support.

CONCLUSION Overall, this study advances our understanding of the relationship between marital status, gender, and parental psychological well-being. First, our findings provide some evidence that marriage protects parents from the economic, work, and emotional demands that accompany the raising of children. Nevertheless, even after accounting for the impact of these burdens, married parents report fewer traditional symptoms of depression and lower rates of alcohol abuse than do single parents. These findings are consistent with our theoretical perspectives of the benefits of marriage for parents’ psychological well-being and recent research on the relationship between marital status and parents’ depression (Evenson & Simon, 2005; Nomaguchi & Milkie, 2003). Second, our findings support previous arguments that men and women may exhibit different depressive symptoms. We find that even after controlling for Time 1 levels of depressive symptoms, mothers are more likely to report traditional depressive symptoms and fathers are more likely to abuse alcohol. In addition, interaction effects involving gender and marital status indicate that being single, relative to being married, is more positively associated with mothers’ reports of traditional depressive symptoms and to fathers’ alcohol abuse. These results suggest that scholars should continue to consider gender differences in depressive symptoms as they attempt to understand relationships between gender, marital status, and parents’ psychological well-being. This is especially important because some recent research suggests that marriage and parenting experiences are more negatively associated with women’s psychological well-being (Mattingly & Sayer, 2006; Nomaguchi, Milkie, & Bianchi, 2005), while other work reports finding no gender differences in the relationship between parenthood and psychological well-being (Evenson & Simon, 2005). Third, our results suggest that parenting burdens partially mediate gender and marital status differences in parents’ depressive symptoms, as expected. Consistent with a structuralist perspective, we find that differences in parenting burdens appear to completely explain gender

152

Single-Parent Families: The Lived Experience

differences in traditional depressive symptoms in our full additive model. Consistent with a perspective that emphasizes a division of labor within marriage, parenting burdens largely mediate marital status differences in traditional depressive symptoms as well. We find evidence of very modest mediating effects of gender and marital status differences in alcohol abuse. Finally, this study demonstrates that parenting burdens moderate the relationships between gender, marital status, and parents’ depressive symptoms. There is some support for the individualist perspective that equal levels of parenting burdens may be less likely to lead to depressive symptoms among mothers than among fathers. For example, time with children is more positively associated with fathers’ traditional depressive symptoms and economic strain is more positively associated with fathers’ alcohol abuse. Yet, these findings are also consistent with a structuralist perspective to the extent that they reflect violated expectations of mothers performing the bulk of childrearing responsibilities and fathers being successful breadwinners. From a structuralist perspective, the interaction effects involving gender suggest that mothers and fathers do not experience parenthood in the same way—there are gendered expectations and pressures that differentially affect mothers’ and fathers’ depressive symptoms. Consistent with a marriage premium perspective, equal levels of economic strain are more positively associated with traditional depressive symptoms among single parents than among married parents. It is important to recognize that, like all studies, this study has limitations. One limitation is that there is relatively little change in the marital and parenthood statuses of the respondents in the sample between waves of data collection. Thus, we are unable to adequately analyze associations between marital transitions, parenthood transitions, and changes in parents’ depressive symptoms for men and women. Instead, we focus on trying to understand marital and gender differences in parents’ depressive symptoms by focusing on the presence of parenting burdens and controlling for Time 1 depressive symptoms. Also, there may be some contamination of the association between economic strain and traditional depressive symptoms, because of the subjective nature of our economic strain measure. However, the use of objective indicators of economic status did not affect our substantive findings in supplementary analyses (results not shown). Similarly, social support may be endogenous to psychological well-being. That is, it is unclear whether social support is primarily affecting parents’ depressive symptoms or whether the presence of depressive symptoms is encouraging social support. Another limitation is that the types of men who become single fathers may be considerably different from the types of women who become single mothers. Although custodial mothers are much more common than

The Psychological Well-Being of Single Parents

153

custodial fathers, the number of fathers that have sole or shared custody is beginning to rise (Cancian & Meyer, 1998). Yet, certain factors are associated with receiving child custody following a divorce. Fox and Kelly (1995) and Christensen, Dahl, and Rettig (1990) find that having older and male children increase the likelihood of father custody. Father custody may also be more likely to occur among men with higher incomes (Christensen, Dahl, & Rettig, 1990), although Fox and Kelly (1995) find the opposite relationship. Mendes (1976) argues that the most crucial dynamic in a single father’s parenting experience is whether he sought custody. Thus, marital status differences among fathers are more likely to entail unmeasured personal characteristics than marital status differences among mothers. Future work should consider the reality and implications of this possibility. Another limitation of this study is that we do not focus on differentiating individuals who are single from those who are single and in a cohabiting relationship. It could be argued that cohabitation is similar to marriage in that there is a built-in support system from one’s partner. However, cohabitors appear to differ from married couples in many ways (Rindfuss & VandenHeuvel, 1990). Therefore, we do not examine cohabitation as a separate marital category in this study, which may provide a conservative estimate of the predictive power of marital status for parents’ psychological well-being. Instead, we take into account the aid and support that any additional adult in the household can provide by using a control variable in our models that indicates the presence of an adult who was not the respondent’s spouse. Despite these limitations, this study sheds light on the associations between gender, marital status, and parents’ psychological well-being. This study is particularly unique in its consideration of how parenting experiences may affect both single mothers’ and fathers’ depressive symptoms. We find that being single is positively associated with both traditional reports of depressive symptoms and alcohol abuse for both mothers and fathers, compared to being married. Also, whereas previous studies that have examined mothers’ and fathers’ psychological wellbeing have focused on internalized symptoms, our consideration of both traditional depressive symptoms as well as alcohol abuse as indicators of psychological well-being reveal gender differences in the expression of depressive symptoms. Finally, our analysis of mechanisms that may link marital status and gender to parents’ psychological well-being provides a clearer understanding of the extent to which the burdens of parenting are associated with depressive symptoms. We find that parenting burdens mediate gender and marital status differences in parents’ depressive symptoms—especially for traditional reports of depressive symptoms. Yet, there is also evidence that parenting burdens moderate gender and marital status differences in depressive symptoms,

154

Single-Parent Families: The Lived Experience

as well. Future works need to continue to explore the mechanisms through which parenthood influences mothers’ and fathers’ psychological well-being. REFERENCES Balswick, J. O., & Peek, C. W. (1971). The inexpressive male: A tragedy of American society. The Family Coordinator, 20, 363–368. Belsky, J., Robins, E., & Gamble, W. (1984). The contextual determinants of parental competence: Towards a contextual theory. In M. Lewis (Ed.), Genesis of Behavior beyond the Dyad (pp. 251–279). New York: Plenum. Bianchi, S. M. (1995). The changing demographic and socioeconomic characteristics of single parent families. Marriage and Family Review, 20, 71–98. Bianchi, S. M., Milkie, M., Sayer, L. C., & Robinson, J. P. (2000). Is anyone doing the housework? Trends in the gender division of household labor. Social Forces, 79(1), 191–228. Bird, C. E. (1997). Gender differences in the social and economic burdens of parenting and psychological distress. Journal of Marriage and Family, 59, 809–823. Cancian, M., & Meyer, D. R. (1998). Who gets custody? Demography, 35, 147–157. Christensen, D. H., Dahl, C. M., & Rettig, K. D. (1990). Noncustodial mothers and child support: Examining the larger context. Family Relations, 39, 388–394. Cohen, O. (1996). The personal well-being of single-parent family heads rearing their children by themselves: A comparative study. Contemporary Family Therapy, 18, 129–146. Coltrane, S. (1996). Family man: Fatherhood, housework, and gender equity. New York: Oxford University Press. Cornwall, M. (1989). The determinants of religious behavior: A theoretical model and empirical test. Social Forces, 68, 572–592. Downey, D. B., Ainsworth-Darnell, J. W., & Durfur, M. J. (1998). Sex of parent and children’s well-being in single-parent households. Journal of Marriage of Family, 60, 878–893. Evenson, R. J., & Simon, R. W. (2005). Clarifying the relationship between parenthood and depression. Journal of Health and Social Behavior, 46, 341–358. Fields, J. M., & Casper, L. C. (2000). America’s families and living arrangements. March 2000 Current Population Reports P20–537. Washington, DC: U.S. Census Bureau. Fox, G. L., & Kelly, R. (1995). Determinants of child custody arrangements at divorce. Journal of Marriage and the Family, 57, 693–708. Gerster, N., Kohler, C. C., & Rosenfield, S. (1985). Explaining the symptomatology of separated and divorced women and men: The role of material conditions and social networks. Social Forces, 64, 84–101. Glass, J., & Fujimoto, T. (1994). Housework, paid work, and depression among husbands and wives. Journal of Health and Social Behavior, 35, 179–191. Gove, W. R., Hughes, M., & Style, C. B. (1983). Does marriage have positive effects on the psychological well-being of the individual? Journal of Health and Social Behavior, 24, 122–131.

The Psychological Well-Being of Single Parents

155

Greenberger, E., & O’Neil, R. (1990). Parents’ concerns about their child’s development: Implications or father’s and mothers’ well-being and attitudes toward work. Journal of Marriage and the Family, 52, 621–635. Heckman, J. J. (1979). Sample selection bias as specification error. Econometrics, 47, 153–67. Hilton, J. M., Desrocher, S., & Devall, E. (2001). Comparison of role demands, relationships and child functioning in single-mother, single-father, and intact families. Journal of Divorce and Remarriage, 35, 29–57. Hochschild, A. (2003). The second shift: Working parents and the revolution. New York: Avon Books. Horwitz, A., & White, H. R. (1991). Becoming married, depression, and alcohol problems among young adults. Journal of Health and Social Behavior, 32, 221–237. Horwitz, A., White, H. R., & Howell-White, S. (1996). The use of multiple outcomes in stress research: A case study of gender differences in response to marital dissolution. Journal of Health and Social Behavior, 37, 278–291. House, J. S., Umberson, D., & Landis, K. (1988). Structures and processes of social support. Annual Review of Sociology, 14, 293–318. Hughes, M. (1989). Parenthood and psychological well being among the formerly married. Journal of Family Issues, 10, 463–481. Iannaccone, L. R. (1990). Religious practice: A human capital approach. Journal for the Scientific Study of Religion, 52, 325–345. Liem, R., & Liem, J. (1978). Social class and mental illness reconsidered: The role of economic stress and social support. Journal of Health and Social Behavior, 19, 139–156. Marini, M. M. (1989). Sex differences in earnings in the United States. Annual Review of Sociology, 15, 343–380. Mattingly, M. J., & Sayer, L. (2006). Under pressure: Gender differences in the relationship between free time and feeling rushed. Journal of Marriage and Family, 68(1), 205–221. McLanahan, S. (1983). Family structure and stress: A longitudinal comparison of two parent and female-headed families. Journal of Marriage and the Family, 45, 347–357. McLanahan, S., & Adams, J. (1987). Parenthood and psychological well being. Annual Review of Sociology, 13, 237–257. McLanahan, S., & Sanderfur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press. Mendes, H.. (1976). Single fathers. The Family Coordinator, 25, 438–444. Mills, R. J., Grasmick, H. G., Morgan, C. S., & Wenk, D. (1992). The effect of gender, family satisfaction, and economic strain on psychological well being. Family Relations, 41, 440–445. Mirowsky, J., & Ross, C. E. (1992). Age and depression. Journal of Health and Social Behavior, 33, 181–205. ———. (2003). Social causes of psychological distress. New York: Aldine de Gruyter. Mulford, H., & Miller, D. E. (1963). Drinking in Iowa: III. A scale of definitions alcohol related to drinking behavior. Quarterly Journal of Studies on Alcohol, 21, 267–278.

156

Single-Parent Families: The Lived Experience

Nomaguchi, K. M., & Milkie, M. M. (2003). Cost and rewards of children: The effects of becoming a parent on adults’ lives. Journal of Marriage and the Family, 65, 356–374. Nomaguchi, K., Milkie, M. M., & Bianchi, S. M. (2005). Do dual-earner mothers and fathers differ? Journal of Family Issues, 26, 756–792. Pearlin, L. I., & Radabaugh, C. W. (1976). Economic strains and the coping function of alcohol. The American Journal of Sociology, 82, 652–663. Polich, J. M., & Orvis, B. R. (1979). Alcohol problems: Patterns and prevalance in the U.S. Airforce. Santa Monica, CA: The RAND Corporation. Radloff, L. S. (1977). The CES-D scale: A self-reported depression scale for research in the general population. Applied Psychological Measurements, 1, 385–401. Richards, L., & Schmiege, C. J. (1993). Problems and strengths of single parent families. Family Relations, 42, 277–285. Rindfuss, R., & VandenHeuvel, A. (1990). Cohabitation: A precursor to marriage or an alternative to being single? Population and Development Review, 16, 703–726. Risman, B. (1986). Can men mother? Life as a single father. Gender and Society, 1, 632. ———. (1987). Intimate relations from a microstructural perspective: Men who mother. Gender and Society, 1, 6–32. Robertson, E. B., Elder, G. H., Skinner, M. L., & Conger, R. D. (1991). The costs and benefits of social support in families. Journal of Marriage and the Family, 53, 403–416. Ross, C. E.., & Huber, J. (1985). Economic hardship and depression. Journal of Health and Social Behavior, 26, 312–337. Ross, C. E., Mirowsky, J., & Goldsteen, K. (1990). The impact of the family on health: The decade in review. Journal of Marriage and the Family, 52, 1059–1078. Saluter, A. F. (1993). Marital status and living arrangements: March 1992. Current Population Reports Series P-20, No. 0468. Washington, DC: U.S. Bureau of the Census. Simon, R. (1997). The meanings individuals attach to the role identities and their implications for mental health. Journal of Health and Social Behavior, 38, 256–274. ———. (2002). Revisiting the relationships among gender, marital status, and mental health. American Journal of Sociology, 107, 1065–1096. South, S. J., & Spitze, G. (1994). Housework in marital and non-marital households. American Sociological Review, 59, 327–347. Spanier, G. B. (1983). Married and unmarried cohabitation in the United States: 1980. Journal of Marriage and the Family, 45, 277–288. Stolzenberg, R. M., Blair-Loy, M., & Waite, L. (1995). Religious participation in early adulthood: Age and family life cycle effects on church membership. American Sociological Review, 60, 84–103. Sweet, J. A., & Bumpass, L. L. (1996). The National Survey of Families and Households Waves 1 and 2: Data description and documentation. Madison, WI: Center for Demography and Ecology, University of Wisconsin. Turner, J. R., & Marino, F. (1994). Social support and social structure: A descriptive epidemiology. Journal of Health and Social Behavior, 35, 193–212. Waite, L. J. (1995). Does marriage matter? Demography, 32, 483–507. Waite, L. J., & Gallagher, M. (2000). The case for marriage: Why married people are happier, healthier and better off financially. New York: Doubleday.

The Psychological Well-Being of Single Parents

157

Walen, H. R., & Lachman, M. E. (2000). Social support and strain from partner, family, and friends: Costs and benefits for men and women in adulthood. Journal of Social and Personal Relationships, 17(1), 5–30. Weiss, R. S. (1979). Going it alone: The family life and social situation of the single parent. New York: Basic Books. West, C., & Zimmerman, D. H. (1987). Doing gender. Gender and Society, 1, 125–151. Wright, D. W. (1989). Single parents in the workplace: Conserving and increasing human capital. In G. L. Bowen & D. K. Orthner, The organization family: Work and family linkages in the U.S. military (pp. 79–96). New York: Praeger.

Chapter 11

Single Parenthood and the Law Lynda H. Walters and Carla Abshire*

With few exceptions, laws that affect single parents are focused on the continuing relationship of parent and child following the dissolution of marriage; both custodial and noncustodial parents are affected. Although never-married parents come under these laws, most laws were developed with reference to formerly married single parents. However, with the increase in the number of never-married parents, application of laws will increasingly involve never-married, single-parent families. Most single parents are female, who have, historically, faced discrimination in the law (Kanowitz, 1969; “Sex Discrimination,” 1950) and in the legal environment (Johnston & Knapp, 1971). Important to the examination of law and single parenthood are the consequences of discrimination. It is true that with the revisions of domestic relations laws in most states, laws have become more gender neutral; however, the legal environment in which laws are applied is not gender neutral. According to Schafran (1988, p. 39), when a woman goes to court she has “three strikes against her: she is a woman, she lacks the resources to retain adequate counsel, and her issues are those with which the legal system would prefer not to deal.” Task forces appointed by the supreme courts of both New Jersey and New York (see Schafran, 1987, for overview of these task force reports) have confirmed that although law reform efforts have resulted in a gender-neutral statutory framework for divorce, in practice women continue to be disadvantaged by courts. Indeed, according to Lonsdorf (1989), many women are disadvantaged in negotiations before they ever reach court. Although single fathers also encounter discrimination in

* L. H. Walters and C. R. Abshire (1996). Single Parenthood and the Law. Marriage & Family Review, 20(1), 166–188. Reprinted with permission of Taylor & Francis Group.

Single Parenthood and the Law

159

courts (McCant, 1987; Weitzman & Dixon, 1979), they do so far less frequently and to less extent than women. In order to gain perspective on the context in which laws affecting single parents have been developed and are applied, some of the myths about women and biases against women in courts are reviewed. Then, following a brief comment on the historical development of domestic relations law, child custody, child support, tax, and welfare law are considered. Child custody is reviewed with attention to factors that affect custody decisions; these are highly complex issues involving married, formerly married, and never-married parents. Child support issues are reviewed with attention to reasons for modification and termination and to efforts to enforce support obligations. Because of importance to single parents, developments in the area of tax law are briefly reviewed. Finally, noting that many single-mother families fall below the poverty threshold and considering that welfare statutes and regulations may have the most direct effect of any laws on poor, never-married, singlemother families, the role of courts in shaping welfare policy is reviewed.

LEGAL CONTEXT OF DECISIONS AFFECTING SINGLE PARENTS More people who live in female-headed households live below the poverty level than any other group (Harrington, 1984; Lugaila, 1992). In female-headed families, median family income was below the poverty threshold, and it was less than 58 percent of the median income in husband-wife families in 1990 (Lugaila, 1992). Approximately 69 percent of all mother-only families with children under 18 receive means-tested assistance (U.S. Bureau of the Census, 1992). Considering these data and the social problems associated with poverty, it is readily understandable that single-parent families are often thought to represent some form of social pathology. This is not a presumption that has evolved in recent years, but dates back prior to the 17th century, when even widows with children, if they received public help, were treated punitively and with concern for the kind of moral influence they might have on a community—and there was more sympathy for widows than for divorced, abandoned, or never-married mothers (Kamerman & Kahn, 1988)! In a system in which poverty has historically been considered a moral issue and women have been considered legally incompetent, derogatory myths about single mothers should not be unexpected. Reviews of some of the widely held myths that, in part, have shaped the legal environment can be found in Mulroy (1988), Weisberg (1982), and the New York Task Force on Women in the Courts (“Report of the New York Task Force,” 1987).

160

Single-Parent Families: The Lived Experience

Examples of myths include the belief that women who are alone have made themselves undesirable to men—especially if they are divorced. If they have never married but have children, they have what they deserve. When they have sought child support that had been awarded but not paid, they have been told by judges that they were vindictive, moneygrubbing, and that they had made their beds and must now lie in them (“Report of the New York Task Force,” 1987). It is generally believed that if a woman works hard, she will not be in poverty; if she turns to welfare, she is lazy and therefore unworthy. If families headed by women live in substandard housing in central city neighborhoods, it is because it is their choice—perhaps motivated by fear of the unknown or perhaps by choosing to be with people like themselves. Finally, one of the most powerful myths is that women can improve their situation if they want to and work hard enough. Biases in Courts Along with cultural and societal myths, bias toward women in courts is also caused by the fact that women have not always been considered legally competent. Women have been thought to deserve the protection of the law, but not as a constitutional right; furthermore, protection has been defined from a male perspective. When women have lived up to the male image of womanhood, they have been more likely to be protected (Weisberg, 1982). Indeed, the Equal Protection Clause of the Fourteenth Amendment was not thought to apply to women (Ginsberg, 1978; Johnston & Knapp, 1971); however, in 1971 (Reed v. Reed) the Supreme Court interpreted the amendment to apply to both men and women. It was not until 1979 (Orr v. Orr) that it was interpreted to apply to gender discrimination in family relationships. Still, many judges have difficulty overcoming the myths they have long accepted. For example, in September 1992, the Atlanta Journal Constitution carried a full-page story about a judge in North Carolina who had put a woman in jail because she could only pay $55 of the $60 court costs when she filed a petition to require her former husband to pay child support. This judge was described by friends as a pillar of the community, “a nice fellow, one of the few who would have stopped to help the woman at the well.” It was not until one woman was murdered by a man the judge had found not guilty of beating her and a young reporter took the story of abuse and outrageous treatment in the court to the newspaper that the judge reexamined his views. To his credit he apologized to two of the women who had been treated poorly in his court and set up a separate court to hear domestic violence cases. This judge appears to differ from many other judges primarily in his willingness to admit his mistake; the local bar was “astonished” that he admitted he was wrong.

Single Parenthood and the Law

161

In another forum, a blue-ribbon panel reported to the New York state chief judge that “Women are often denied equal justice, equal treatment and equal opportunity” (“Report of the New York Task Force,” 1987, p. 17). For example, it is widely known that child support payments are not made at an acceptable rate. Not only are child support awards not paid (a problem of law enforcement rather than law making or interpretation), the amount of awards is low (see for example, “The First Year Report,” 1986). Fathers rarely pay more than one-third of their incomes, and with little scrutiny, what fathers say they can pay is often accepted by courts (Weitzman, 1985; Yee, 1979). Indeed, family courts have been accused of making women feel that attempts to gain support for children from fathers is “vindictive, unimportant or even a joke” (“Report of the New York Task Force,” 1987, p. 86). Men, too, face some biases in courts, especially in terms of child custody. Even with the move away from the maternal preference for custody, many courts still view the best interest of children to be with a mother who is “naturally” better able to nurture them. However, considered case by case, stereotyped gender-based expectations for men appear to be balanced by an inclination for male judges to understand and sympathize with the situation of another male—something that is most unlikely to happen in the response of a male judge to a woman (see Johnston & Knapp, 1971). It would be unwise for a single mother seeking custody to count on the maternal preference. For example, if a woman does not work and stays at home where she can maximize her nurturing function, she and her child usually must live in poverty because it is rare for a court to award enough child and spousal support to relieve her of the need to work. On the other hand, if she seeks employment, the bias of the court shifts to favor the male, especially if he has remarried and the child could live with the father and a mother figure. In other words, the maternal presumption is easily overcome or inapplicable if the mother is employed outside the home (Raymond, 1988). For women seeking custody, the maternal preference is balanced by traditional stereotypes and myths about women and poverty.

PERSPECTIVE ON DOMESTIC RELATIONS LAW The law of domestic relations as it relates to child custody and child support has changed significantly throughout this century. It was developed from state statutory and common law. In creating statutes and deciding cases, legislatures and courts have reviewed other state laws and have considered decisions in other state courts as precedent for their own actions. These efforts have culminated into what is now viewed as a national tradition of the law of domestic relations (Schuele, 1988–89).

162

Single-Parent Families: The Lived Experience

There are still differences between and among jurisdictions, but there are also national trends that cut across jurisdictions. Although the “best interests of the child” has been adopted in the United States as the standard by which custody determinations are made, remnants of the Roman and English common law view of children as chattel were influential on the development of the law, especially as it relates to child support. For example, the notion that children were property of the father gave the father certain rights and responsibilities, including the right to services and custody of his children, which led to the responsibility or duty to support his children. Child support has not always, however, been considered a legally enforceable duty of the parent or right of the child. Whereas child custody has been considered in terms of the best interests of the child, child support has been considered in terms of contract law or in terms of a community’s interest in keeping citizens from becoming a public charge (Schuele, 1988–89). Even the child support enforcement program that was initiated in 1950 was motivated primarily out of a desire to keep families from becoming dependents of the state (Anders, 1990; Fleece, 1981–82). As would be expected, laws of custody and laws of support although related, affect single parents differently.

CHILD CUSTODY Most state statutes give courts considerable discretion in making custody determinations. Many would-be custody battles are settled by agreement of the parties before the case reaches the courtroom (Ellman et al., 1991; Mnookin, 1978), but the court, not the parties, decides whether the agreement is in the child’s best interests. Even if the court accepts an agreement made by divorcing parents, it is free to reopen the case if a change in circumstances warrants reconsideration. This rule may be viewed as one which greatly limits parental power (Mnookin, 1978). However, the propensity of courts to rubberstamp these agreements may exacerbate a situation in which one parent has little bargaining power. For example, the parent most emotionally involved with the child (usually the mother) may be willing to accept less child support and other monetary or property awards as a result of threats from the other parent (usually the father) to sue for sole custody (Lonsdorf, 1989; and see, for example, the opinion of the court in Garska, 1981). With extensive judicial discretion, it is difficult to avoid interjecting personal values and opinions in custody and visitation decisions (Neeley, 1984; Uviller, 1978). These decisions require judgments regarding the characteristics of all persons involved in the case as well as consideration of the facts. As a rule, any circumstance relevant to the best interests of the child is admissible—even past behaviors of parents. Adding to the

Single Parenthood and the Law

163

difficulty of courts to be fair, state statutes that enumerate circumstances relevant to the child’s best interests generally do not include the weight to be given to each factor in making the custody determination (Wadlington, Whitebread, & Davis, 1983). Because of the continuing inclination to view mother-custody as best for children, fathers can be at a disadvantage in obtaining custody and visitation. Weitzman & Dixon (1979) found, however, that by 1977, twothirds of fathers who requested it were awarded custody in California courts. In contrast, according to Albert & Brodek (1989), courts are insensitive to noncustodial fathers who wish to continue a relationship with their children and to those fathers who are frustrated in their efforts to relate to their children. They report that when one man requested maximum visitation with his children, the judge became exasperated and belittled fathers who seek more time with their children. Additional influences on custody decisions, including putative fathers, mobility, sexual behavior, religious practices, and race, are reviewed below. Both custodial and noncustodial single parents are considered. Putative Fathers A series of Supreme Court cases in the 1970s and 1980s defined the unwed, or putative, father’s relationship with his child as the basis of a liberty interest. (A liberty interest is not exactly a right but it is similar in that it has been granted constitutional protection.) This liberty interest is thought to arise from belief in the importance of the family unit; it is primarily a function of relationship rather than biological kinship. The importance of a putative father’s relationship with his child as compared to the importance of biological kinship is addressed in what is known as the Stanley line of cases: Stanley (1972), Quillion (1978), Caban (1979), and Lehr (1983). In this line of cases, the United States Supreme Court held that (1) biological kinship provides an opportunity for the putative father to develop a relationship with his child; (2) the nature of the relationship between the putative father and his child may give rise to a liberty interest in the relationship; and (3) the liberty interest gives rise to the constitutionally protected right to assert paternity. Although decisions in the Stanley line of cases emphasized the importance of the psychological relationship, a more recent case (Michael H., 1989) demonstrates that the liberty interest will not always prevail. In this case, a child was born to a woman who was married to a man other than the biological father, and the plurality opinion was that the notion of a protectable liberty interest growing out of the father-child relationship is irrelevant. Instead, the Court relied on the specific tradition of the marital presumption and held that the rights of a putative father are subordinate to the state’s interest in protecting the integrity of the marital unit. Only

164

Single-Parent Families: The Lived Experience

in one concurring opinion and two dissenting opinions was this father’s relationship with his child considered in light of the Stanley line of cases. Thus, in Michael H., the last case decided by the Supreme Court, the Stanley line of cases was essentially ignored, and the more conservative position of the marital presumption (which had not been addressed in the previous cases) held. Although the unwed father may be denied the right to gain custody, to visit, or to stop an adoption proceeding, especially if the mother is married to another man, an illegitimate child is entitled to support from the biological father (Gomez, 1973). Thus, the state or the mother (whether or not married to another man; see Smith v. Cole, 1989) may bring an action to obtain support for a child. For excellent analyses of this issue, see Anders (1990), Batty (1990), Kisthardt (1991), Solari (1989), Sylvain (1990), and Winjen (1990). Mobility of the Single Custodial Parent The standards applied to cases involving the relocation of a parent outside the jurisdiction in which a custody decree exists varies by jurisdiction. Requirements in different states range from allowing the custodial parent to move for any reason so long as the noncustodial parent does not prove that the move is adverse to the child’s best interest (Minnesota) to requiring the custodial parent to show exceptional circumstances in order to obtain permission to move (New York). In many states, only a challenge by the noncustodial parent will require the custodial parent to provide a reason for moving. In a recent decision (Seessel, 1988), the Tennessee Supreme Court ruled that the custodial parent could not move without showing that the move was in the best interest of the child; in this case, problems associated with remarriage, career, and the welfare of the family were ignored (Gooch, 1989). Indeed, many times the custodial parent’s desire/need to move to another state is pitted against the desire to retain custody. This may be particularly true for women because courts tend to be more demanding of sacrifice for the sake of children from mothers and question a woman’s pursuit of a career (Ellman et al., 1991; Weitzman & Dixon, 1979). An illustration of the restrictiveness of courts toward women is found in Lozinak (1990), in which the appellate court forced a mother to choose between retaining custody of her daughter or moving to another state with her new husband. In doing so, the court stated that if the mother were really concerned with the child’s best interests, as opposed to her own interests, then she would remain in the initial state. Because women are expected to move with a husband, if a woman remarries, restriction of mobility can create a conflict between two deeply held values: the best interest of her child and commitment to her husband.

Single Parenthood and the Law

165

In the past, if dissatisfied with a custody decision, a parent could move with a child to another state and bring an action for custody. Although not condoned, “forum shopping,” as this practice has been called, was possible because jurisdiction rules have not been applied in custody cases as they have in other cases (Bodenheiner & Neeley-Kvarme, 1979). However, parents are finding this practice less viable today. To address problems of jurisdiction in custody cases, the Uniform Child Custody Jurisdiction Act (UCCJA) was designed as a model to shift the focus from jurisdiction over the parent to the needs of the child. The purposes were to reduce chaos created by interstate jurisdictional competition and to promote cooperation between jurisdictions, as well as to promote the stable environment that most consider to be in a child’s best interests. Through the concept of continuing jurisdiction, the act forbids the modification of a child custody decree in one state when the original decree was issued in another state that still has jurisdiction over the case (Schorsch, 1982). The model act is said to have partially laid aside the confusion created by Halvey (1947), in which the Supreme Court held that in order to protect the best interests of the child, a court in one state could modify a custody decree from another state. Although most states have adopted the principles in the UCCJA, the issues are complicated, and there are still many problems. For an excellent review of the application of and problems with UCCJA, see Coombs (1982). Sexual Behavior of Custodial Parents As has been seen, courts tend to be biased in favor of traditional assumptions about family life and the roles of men and women. Such biases may be best illustrated in disputes involving the sexual behavior of parents. It should be noted at the outset that jeopardy to custody because of a nonmarital sexual relationship is a greater problem for women than for men (see Sack, 1992, note 65; and Woods, Been, & Schulman, 1983). Whether the parent is involved in a nonmarital heterosexual or homosexual relationship, the court may grant custody on the condition that the parent not live with certain persons or even that the parent terminate certain relationships. The court may require that a juvenile authority supervise the parent’s behavior, may require a change of the parent’s conduct, or may prohibit the parent from participating in overnight stays with certain persons—all of which is done in order to insulate the child from behavior considered inappropriate by the court (Allen, 1985). Although courts at all levels may let their views of appropriate behaviors influence custody decisions, these biases appear to be most prevalent at the trial court level. Sack (1992) has provided a review of West Virginia cases involving sexual behaviors of parents in which lower courts denied custody to mothers (primary caretakers) and

166

Single-Parent Families: The Lived Experience

appellate courts reversed the decision because the mother’s sexual relationship was unrelated to her relationship with the child and her fitness as a parent. Homosexual parents have a more difficult time than heterosexual parents. Where actual harm to the child is presumed, courts will generally prohibit the custodial parent from overnight visitation with lovers (Ellman et al., 1991). Among other things, this indicates that courts are reluctant to expand the definition of family in visitation and custody disputes to reflect trends in the makeup of the American family. For example, in Allison D. (1991), a New York court refused visitation rights to a lesbian ex-partner with the biological mother of the child because the statute in question only applied to biological parents. The biological mother had been artificially inseminated, and the lesbian ex-partner had acted as the child’s parent during the relationship; after the relationship was terminated, she had provided financial support for the child. The New York Court of Appeals affirmed the lower court, stating that the ex-partner had no standing to bring a petition for visitation (“Family Law,” 1992). (When the issues do not involve custody, courts have not been quite so reluctant to consider homosexual unions a family; see for example, Braschi, 1989.) It is common for courts to worry that a homosexual parent will influence a child’s sexual orientation even though the evidence clearly indicates this is not a problem. Courts may be concerned that others will harass a child whose parent is homosexual or may conclude that a homosexual parent is unacceptable because the behavior is in violation of sodomy statutes. The degree of limitation on personal behavior/freedom of parents varies according to the court’s beliefs about and knowledge of sexuality (“Sexual Orientation,” 1990). Cohabitation is another situation in which presumed sexual behavior can cause difficulty with custody decisions. A disgruntled spouse may use the traditional assumptions of the court to manipulate and control the behavior of the other spouse. For example, in Parrillo (1989), in response to the mother’s motion for modification of custody, the father also filed a petition to modify custody. He did not seek custody of his children, but sought to have the mother cease relations with her boyfriend (Carter, 1991). The custodial parent (mother) had sought modification of a custody decree in order to establish specific times at which the noncustodial parent (father) could visit their children. Apparently, the father had not only refused to pay the amount of child support stipulated in the divorce decree, he had also crashed his car into the mother’s boyfriend’s car and had threatened the family. The trial court, without specific findings of harm to the children, stated that overnight visitations by the mother’s boyfriend were not conducive to the children’s well-being, but that if the mother were to marry the boyfriend there would be no issue. The

Single Parenthood and the Law

167

Supreme Court of Rhode Island upheld the trial court’s decision to modify the custody provision, which, in effect, restrained the mother from having overnight visits with her boyfriend. Not only did this court extend the definition of cohabitation to include sexual relations (Knauerhase, 1989), but also, in effect, it allowed an exspouse to control the sexual behavior of the custodial parent (Carter, 1991). This decision forced the custodial parent to choose between two constitutional rights, freedom of association and personal privacy from unwarranted governmental intrusion. Although the United States Supreme Court has expressly provided that the government cannot force an individual to choose between two constitutionally guaranteed rights (Aptheker, 1964; Knauerhase, 1989), in custody cases, it is the child’s best interests, not the parents interests, that are the issue. Religious Practices When religious practices of parents are thought to jeopardize the mental health or physical safety of a child, they can be a determining factor in a custody decision. In Hadeen (1980), the mother’s religious practices included fasting and spanking until the will of a child is broken. The trial court awarded custody to the father, but, on appeal, the court specifically considered whether the religious practices of the mother constituted reasonable likelihood that the child would be impaired. The appellate court reversed and remanded the case for further consideration by the trial court in order to ensure that the trial court had not put too much weight on the religious factor (Ellman et al., 1991). Justice Dore, dissenting, disagreed with the reversal, stating that the facts of the case indicated that the children were in jeopardy. One compelling incident reported in testimony was of Mrs. Hadeen spanking one child for two hours while her other children held the child down. In addition to noting that religion is a thorny issue in custody disputes (see for example, Beschle, 1989; Paul, 1989; and “The Establishment Clause,” 1984), it is important to recognize that the religious interests of parents are more often seen as commensurate with the best interests of children—even when children are directly affected and hurt (Hadeen, 1980)—than are the relationship interests of parents, even if children are neither directly affected nor hurt (recall Parrillo, 1989). In Hadeen, the court appeared to be more concerned with protecting Mrs. Hadeen’s right to freedom of religious practices than with protecting the best interests of the children. In neither of these cases (Hadeen and Parrillo) was attention clearly focused on the children. Taken together, the cases seem to say that it is not really the effect on children but the values of courts that ultimately determine the outcome. Contrasting outcomes in Hadeen and Parrillo, it appears to be more acceptable to discipline a child in a manner

168

Single-Parent Families: The Lived Experience

that constitutes child abuse than to have relationships with members of the same or opposite sex, regardless of whether the nature of the relationship is known by the children or whether the relationship can be shown to be harmful to the children. Interracial Relationships Another issue that has been specifically addressed by courts in custody determinations is interracial relationships. Because of the civil rights movement, issues of race are more protected by equal rights legislation and less vulnerable to biases of the court than issues of moral unfitness and sexual behavior of a parent (Johnston & Knapp, 1971). Race was the issue in a case before the United States Supreme Court (Palmore, 1984) in which the custodial parent was cohabiting with a member of another race. The Court held that, although actual harm sustained by the child due to prejudices of the community may be an appropriate factor to consider in a custody decision, the potential harm of a mixed-race relationship to the child is not a permissible factor to be considered (Ellman et al., 1991). There is both greater clarity in the law and greater sensitivity on the part of judges for handling issues of race than of sexuality; consider this decision in contrast to those in which the focus is on sexuality (e.g., Allison D., 1991, and Parrillo, 1989). Although laws governing the parent-child relationship of single parents are not specific to race or ethnicity, minority parents can be expected to have more difficulty in courts than do white parents—especially if they are poor. Minority, low-income women face the greatest difficulty in our legal system. Their problems may be similar to those of other single parents, but it is more difficult for them to obtain qualified legal representation, and they are not received well in courts (Schafran, 1988) even though race per se will not be an issue. Joint Custody Joint custody arrangements have become increasingly popular. Joint legal custody, as opposed to joint physical custody, generally connotes a situation in which the parents share equal responsibilities in child rearing; it is generally not tied to the amount of time that a child is in the presence of one or the other parent (Mnookin, 1978; WIshnew, 1988). The joint custody arrangement has its roots in concern for the best interests of the child (Wishnew, 1988); however, there are still many arguments for and against joint custody (see Folberg, 1991, and Sack, 1992, for reviews of issues and arguments). Joint custody is necessarily a complex arrangement, and partly because of the individual circumstances in each case, there is little agreement on the best policies to be followed (Kass, 1989).

Single Parenthood and the Law

169

It is interesting that despite the fact that the success of joint custody arrangements is predicated on the willingness of parents to work together, some jurisdictions allow joint custody over the objection of a parent. Some courts feel so strongly about the importance of both parents being involved with the child(ren) that spousal attitudes toward each other can influence custody decisions. Insistence that both parents have access to a child may be warranted in most cases, but where there is a history of family violence or harassment, a parent’s objection to an award of joint custody should not prejudice the court (Keenan,1985). Although there are cases in which a court has ruled that a parent’s good faith objection to joint custody should not prejudice a request for sole custody (see, e.g., In re Marriage of Weidner, 1983), one can imagine a situation in which a battered spouse has no specific proof to present to the court in support of a request for sole custody (see Sack, 1992, pp. 311–316; and Ashe, 1992, for an interesting discussion of the “bad mother”). Additionally, joint custody arrangements may result in reduction in the allowance or in no allowance for child support. In her essays about child custody, Judge Kass has noted that it is not only unfair but also illogical to place equal financial responsibility on parents who are not equivalent in terms of financial resources (Kass,1989).

CHILD SUPPORT Regardless of financial resources, parents have a legal duty during and after marriage to support children below the age of majority (i.e., legal age), including those born outside of wedlock. Postdivorce support of children is intended to maintain the marital standard of living. Amount of support required is based on current needs; therefore, it is modifiable if there has been a change in circumstances that warrants reconsideration. Almost all states have laws that hold mothers and fathers equally responsible for child support (Ellman et al., 1991; Mnookin, 1978; see also 67A C.J.S., 1978). Likewise, most recent court decisions have interpreted the Fourteenth Amendment or state equal rights amendments to require gender-neutral support statutes (“Constitutionality of Gender-Based Classifications,” 1982). However, gender neutral statutes that essentially eliminate the father’s primary duty to support have received some criticism. Not only does level of living improve for men and get worse for women after divorce (Hoffman & Holmes, 1975; Phillip Morris Family Survey, 1987, p. 11; Weitzman, 1985; see especially Sorensen, 1992), Bregande (1989) has noted that, because the custodial parent bears the primary responsibility for the care of children, earning potential often must be sacrificed, whereas the noncustodial parent gains earning

170

Single-Parent Families: The Lived Experience

potential as a result of not having responsibility for the day-to-day care of children. Because the majority of custodial parents are women, and women generally earn less than men, it is the woman who will, proportionately, contribute most to the financial support of children. It is ironic that, in the face of an established parental duty to support, the early courts looked to the custodial parent’s financial stability rather than considering the noncustodial parent’s duty to support (Schuele, 1988–89). Today some courts are unwilling to increase child support if the increase would improve the custodial parent’s standard of living along with the child’s. At the same time, even though a court may be reluctant to award enough support for a mother to stay home or obtain excellent care for children, the same court may question whether a working mother should get and/or retain custody (Raymond, 1988). Modification of Support There are a number of circumstances under which modifications are made. Examples include a reduction in the income of the supporting parent (however, this is rarely effective even if the supporting parent has been incarcerated; see Rohloff, 1987; Knights, 1988; Koch, 1990) or either a formal or informal agreement to a reduction (see also 67A C.J.S., 1978 51–58; Krause,1977; Lieberman, 1986). Ordinarily only modifications of future obligations are allowed, but when a parent has requested that the obligation be reduced and has not been able to pay at the current level, retroactive modifications have been made. The problem is that some noncustodial parents have made reduced payments expecting that if the custodial parent seeks enforcement of the original support obligation, it would be possible to get a retroactive reduction (see, e.g., Towne, 1988). Although it is clearly possible in some cases, retroactive reduction of arrearages in child support are prohibited by federal law if the custodial parent receives funds from Aid to Families with Dependent Children (42 U.S.C.A. 666(a)(9)(c), 1990). Termination of Support There are several reasons why support may be terminated before a child reaches the age of majority. In some states if the supporting parent dies, support is terminated, but in some, death does not result in automatic termination; inheritance and life insurance are used to continue support (see In re North Carolina Inheritance Taxes, 1987). Another reason for terminating support is the emancipation of the minor child; that is, the child is accorded legal status before reaching the age of majority. Examples include marriage, entry into military service, leaving the parent’s home and refusing to follow parent’s wishes, or becoming

Single Parenthood and the Law

171

economically independent through employment (see Gottesfeld, 1981, for a discussion of emancipation issues). Adoption by a parent who replaces the supporting parent is also a possible reason for termination of a support obligation. It is also possible for support obligations to be terminated if visitation by the supporting, noncustodial parent is denied. It is commonly thought that the reason why men refuse to pay child support is that they are denied opportunity to visit their children. Although Weitzman (1985) found no correlation between child support compliance and complaints about visitation, some courts have interpreted visitation and support decrees as mutually dependent contracts. Indeed, according to spring 1990 data, when visitation privileges or joint custody were awarded to fathers, almost 80 percent of mothers were awarded child support, but when neither visitation privileges nor joint custody were awarded to fathers, only about 30 percent of mothers were awarded child support (U.S. Bureau of the Census, 1992a). In cases in which a child has refused to see the custodial parent, courts have terminated support (Tyrrell, 1978; Cohen, 1983). Also, in extreme cases, such as concealment of the child by the mother, some courts have refused to make fathers pay unpaid support, maintaining that concealment is a waiver of support (Washington, 1987) even though the child or spouse has been abused by the father (Clark, 1990; Ellman et al., 1991). Czapanski (1989) maintained that rules that link child support and visitation and those that do not are inherently biased. When support and visitation are linked, the child and custodial parent’s need for financial support may be subordinated to the noncustodial parent’s need for a relationship with the child. On the other hand, when the rules are not connected, the noncustodial parent’s financial support may be considered more important than the nurturance received from either the custodial or the noncustodial parent. This is a problem because the law is more apt to recognize a tangible contribution such as money than it is to recognize an intangible contribution such as nurturance. Under some partially connecting rules, a noncustodial parent may withhold support payments if the custodial parent interferes with his or her visitation rights, but the custodial parent must have court permission to interfere with the visitation rights of the noncustodial parent even if child support has not been paid (Czapanski, 1989). In other words, this partially connecting rule seems to place the noncustodial parent’s right to visit above the child’s right to support. Child Support Enforcement The problems most common to child support are those associated with enforcement (Krause,1977; Lieberman,1986; see also 59 American

172

Single-Parent Families: The Lived Experience

Jurisprudence, 1987; and 67A C.J.S., 1978 59–61, 63). Courts have used various civil and criminal remedies to enforce the payment of child support obligations, including contempt of court, income withholding in cases of arrearages, criminal prosecution, income tax refund offsets, guarantees for nonpayment such as bonds and liens, and termination of parental rights (59 American Jurisprudence, 1987 69; 67A C.J.S., 1978 59–61, 63; Ellman et al., 1991). Since the mid-1970s, the federal government has been actively involved in enforcement efforts (Schorsch, 1982). There are two sides to federal involvement in child support enforcement. Both have grown out of efforts to obtain child support from fathers of children who receive AFDC benefits. The Child Support Enforcement Act (Pub. Law No. 93–647) was passed in 1975 and included establishment of state child support agencies, state and federal parent locator services, and the federal Office of Child Support Enforcement operated by the department of Health and Human Services. This law and its amendments put in place a number of strategies for finding fathers and forcing them to pay child support. On the negative side of federal involvement, mothers of children who receive AFDC must identify the father of the child, a requirement that only affects poor, unmarried mothers. Drawing on logic of the right of privacy that has developed over the past 50 years, Anders (1990) concluded that this legislation gives the government an unconstitutional right to intrude in the lives of poor, unmarried mothers. The United States Supreme Court agreed in Shapiro v. Thompson (1969), indicating that the interest of the state in preserving its fiscal resources did not justify an invasion of the constitutional rights of poor women. Later, when the issue was unannounced home visits by social service personnel to insure that there was no father in the home, the Court took a different position. The visits had been objected to as a violation of Fourth Amendment protection from unlawful search (a part of the logic of the right of privacy), but were upheld by the majority opinion of the Court (Wyman v. James, 1971) with dissents from Justices Douglas, Marshall, and Brennan, who pointed out the jeopardy to constitutional rights brought on by acceptance of welfare benefits. To date, this violation of the rights of poor, unmarried mothers to privacy and familial autonomy is a problem that has not been resolved either by statute or Court interpretation. On the positive side, for a mother who wants to find a father and enforce his payment of child support, resources have been improved. This is especially true in cases in which parents no longer live in the same state. Although all states now have laws to deal with problems of enforcement of support obligations when one parent moves out of the state, getting these laws in place has been a long and frustrating problem for both parents and courts. Initially, when a parent moved out of the

Single Parenthood and the Law

173

jurisdiction of a court, the laws that were used to exercise jurisdiction were those that had been developed to apply to commercial activities. Some courts have been unwilling to extend a long-arm statute that applied to commercial activities to a child support case because, obviously, they address different issues (Hughes, 1979–80). In Shaffer (1977) the Court would not exercise jurisdiction (i.e., exercise authority to hear the case) over the nonresident parent who had support obligations because the nature of the ties between that parent, the state, and the precedent for litigation (commercial activities) were not related. In Kulko (1978), the Court agreed that the use of commercial long-arm statutes was not appropriate in child support cases. In this case, the mother moved from New York (the marital domicile of the couple) to California, where she later obtained custody of her two children. She subsequently filed a motion to modify child support. The only contacts the child’s father had with the state of California were (1) a three-day stay years before at the time the couple was married, (2) the act of buying one child’s airline ticket to move in with her mother, and (3) the fact that his children were located in that state. The Supreme Court, citing Shaffer (1977), ruled that this was insufficient to warrant the exercise of jurisdiction over the father by the state of California. The Court not only noted that from these acts alone the defendant would not have foreseen being summoned to a California court in order to defend himself, but also noted that his actions neither amounted to personal benefits derived from activities within the state nor caused an effect in California that amounted to criminal activities. Thus, the Court concluded that it would not be fair to require the father to defend himself in California. The Court further stated that the mother could have sought relief through the Uniform Reciprocal Enforcement of Support Act (URESA), which would not require the father to leave the state of New York and would satisfy her needs. The URESA and later the Revised URESA (RURESA) were designed as models to provide for interstate enforcement of a child support obligation (Schorsch, 1982). The drafters of RURESA sought to reduce the possibility of unnecessary or inappropriate modifications of support decrees by requiring that, among other things, motions to modify be reviewed by the court that delivered the initial support decision. In the Federal Child Support Enforcement Amendments of 1984, Congress required all states to have some form of the RURESA. Although state reciprocal enforcement acts have resulted in improvements over past attempts to use long-arm statutes designed for commercial activities, there are still some problems. In this as well as other parenting issues, courts may find themselves juggling a concern for the best interest of the child with often unacknowledged biases that arise from assumptions about relationships between husbands and wives. .

174

Single-Parent Families: The Lived Experience

TAX IMPLICATIONS FOR SINGLE PARENTS Single custodial parents who file federal income tax as single head of household are allowed to take the income tax deduction for a dependent child. However, according to the Internal Revenue Code 152(e) (Title 26 U.S.C.A.), in the past it was possible for a noncustodial parent to claim the dependency tax exemption if more than $1,200 was paid in child support in one year and if the custodial parent did not provide more support in the same calendar year. The Tax Reform Act of 1984 (Public Law No. 98–369, 423(a), 98 Stat. 494, 799) clarified that the parent who has custody for the greater part of the year will be considered to provide more than half of a child’s support. The major exception to this rule is the case in which the custodial parent releases the dependency exemption. The release is a specific IRS form that must be signed by the custodial parent and filed with taxes by the noncustodial parent. The question then arose as to whether a trial court could require a custodial parent to sign the release. The most common situation in which the requirement might be imposed is when the custodial parent earns less income than the noncustodial parent. In such a case, the value of the exemption is presumed to be greater for the noncustodial parent—a fact that would certainly be true if the custodial parent earned so little income that no income tax payments were required. Also, it has been suggested that if the noncustodial parent has the exemption and saves money on taxes, an equivalent amount can be added to child support payments, thus providing greater benefit to the child; however, if support is not increased, there will be no advantage to the child or mother, and the discrepancy between the resources available to the father and the custodial mother will be increased. Some state courts have focused their attention on the interest of the Internal Revenue Service in avoiding determination of which parent should be allowed the exemption. In these states, trial courts have been allowed to award the exemption to the noncustodial parent and to require the custodial parent to sign the release. Additional tax issues that are of interest to a single parent include, for example, tax consequences of property settlements including retirement plans. These issues are complex and highly specific and are not discussed here. Excellent reviews can be found in Hjorth (1990) and Langbein & Wolk (1990).

SOCIAL POLICY AND THE COURTS In this country, social policy has most often been designed to provide a support system that enables individuals to work and help themselves and/or provides income for those who cannot work. In the eighteenth

Single Parenthood and the Law

175

and nineteenth centuries, aid was most generously provided for those who were considered “deserving” or for whom there was hope that they would eventually be able to help themselves. It was not until the late 1960s that the notion that disadvantaged persons had a right to government assistance had gained a significant foothold. Policy and policy rhetoric have vacillated between feelings of responsibility for the less fortunate and expectations that those who receive help should work for it. Nearly all policy decisions have been influenced by feelings about deservedness and potential for independence from assistance—it has just been more obvious at some times than at others. Since the late 1960s, the courts have been participants in the development of policy, primarily through the reform of policy. (See Kamerman & Kahn, 1988, Chapter 2, and Sard, 1988, for summaries of the history of the development of AFDC and court involvement.) In the 1960s, government funding to provide legal services for poor people was begun. Until that time they had no way to challenge laws or practices that disadvantaged them or ignored their basic rights (Sard, 1988). It was through early welfare litigation that the right of those in need to receive aid was legitimated. Welfare had been considered discretionary by the courts as well as by those who designed and administered the laws. Those who dispensed aid decided who would receive it; there was neither guarantee that aid could be obtained nor that the amount of aid received would be similar to others in the same circumstances. In Goldberg v. Kelly (1970, p. 262), the Supreme Court, drawing on the position articulated by Reich in 1964 and 1965, said that benefits “are a matter of statutory entitlement for persons qualified to receive them.” At about the same time, the Court applied the Equal Protection Clause of the Fourteenth Amendment to rule that states could not require time-limited residency that resulted in denial of aid to families with dependent children (Shapiro v. Thompson, 1969), but refused to use the Equal Protection Clause to remove the ceiling from welfare grants so that children in large families would not be disadvantaged. Unsuccessful attempts have been made to use equal protection to correct racial imbalances in amount of aid paid to recipients (Jefferson v. Hakney, 1971). Women and illegitimates have been considered more vulnerable to discrimination and lack of protection, a fact that resulted in a 1979 decision against limiting interpretation of the AFDC-Unemployed Parent program. The Court reiterated that no one group of families could be excluded from needed benefits (Wescott v. Califano, 1970). Although AFDC is a part of the Social Security Act, it is a state-run program. States submit the plan for their program, and if it conforms to the terms of the Social Security Act (SSA), it must be approved. States receive 50 percent or greater federal reimbursement for AFDC expenditures; however, states set the standard of need and the level of benefits (in

176

Single-Parent Families: The Lived Experience

all states, benefits are set below the poverty line). In other words, most of the control of AFDC is in the states. Because states set the standard of need, they determine eligibility conditions. It was not until 1968, more than 30 years after the passage of the SSA, that the Supreme Court ruled that states would have to use eligibility conditions contained in the SSA (King v. Smith, 1968); the standard of need in SSA was less restrictive than that used in many states. The limitations of the Court are clearly seen in the history of AFDC decisions. The Court has been interpreting a federal statute that can be changed by Congress; thus, an interpretation that has developed over years can become irrelevant because of a congressional change. Indeed, in the 1980s, at the initiation of the Executive Branch, Congress successfully placed greater restrictions on eligibility for AFDC. The Court can hold Congress to the basic premises of the act in its current form (see Heckler v. Turner, 1985), but it cannot keep Congress from changing the act. Courts have had the most effect on policy via AFDC through protection of procedures for assuring fairness and promptness. Once welfare benefits were considered an entitlement, or property (see Goldberg, 1970), the due process clause of the Fourteenth Amendment could be used for protection of “fair hearing” and “reasonable promptness” guarantees of the statute. Prior to that time, there was no point in questioning authority because there was no legal protection for fair hearing rights. Although few recipients of AFDC have taken advantage of the right to a fair hearing, this right is an important ingredient in a social system that rejects arbitrary authority. The Court has limited itself to procedural protections, refusing to become involved in attempts to use litigation as a strategy to increase benefits (see Rosado v. Wyman, 1970). However, when state laws contain language about levels of AFDC benefits, state courts have greater leeway than federal courts to make decisions and issue orders that affect state appropriations. If a state is not funding at the level its own law requires, the state court can uphold the language of the law and require compliance (see for example “Massachusetts Coalition,” 1987). The important point here is that when the issue is access to AFDC benefits, single-parent, mother-headed families who live in poverty are among those who potentially benefit most from Court involvement in welfare reform. To date, reform has been most successful in requiring the application of due process and equal protection. These protections eliminate the possibility for an administrator of welfare to decide who may or may not receive government support on the basis of personal moral position or bias, and they create minimal protection from the myths about the poor that have had their greatest impact on women and children. (Recall, however, the discussion of child support enforcement in which it was

Single Parenthood and the Law

177

pointed out that court involvement has left in place a statutory requirement to identify the father that jeopardizes the rights of poor, unmarried mothers.)

CONCLUSIONS There are many implications for both practice and policy that can be drawn from this discussion; however, one is particularly important. In this time of evolving family forms and merging sex roles, it is easy to be distracted from serious, ongoing problems of single-parent families. We have discovered that fathers can be nurturant and some want custody of children following divorce. Grandparents are asserting their rights to visit and even to have custody of grandchildren. Gay and lesbian persons are seeking ways to fulfill their desire to parent. Foster and adoptive parents push for their rights as psychological parents of children they love and have cared for. All these are important issues and deserve careful thought and study. The fact remains, however, that the vast majority of single parents are women. Their lives are shaped by myths and stereotypes that even they often do not recognize—they hold many of the stereotypes themselves. They are poor, often educationally unprepared to work for an adequate wage, and believe that they should be at home to care for their children. Decisions are made about their lives by people who feel that they should be good mothers and stay home with their children while, at the same time, believing that they should be hard workers and support themselves. The double standard that is held for men and women following divorce is well documented at every step of the legal process. Although “best interest of the child” has emerged as the dominant guiding principle for law-making and court decisions that affect singleparent families, it is our observation that before the interests of children can be well served, it will be necessary to resolve conflicting feelings and beliefs and recognize that we must also promote the best interest of the parent. To maximize the effectiveness of either practice or policy, it will be necessary to set aside myths and stereotypes and address the issue of equality. When we understand what we mean by equality, we will be able—without regard to gender—to educate children to assume the variety of roles that are required to function well as a single parent or a dual parent. When we start from the assumption that every individual should be prepared to function well in both instrumental and expressive roles, policy and practice will not be so heavily colored by myths and stereotypes. Until we achieve this goal, we can make progress by questioning our assumptions with every decision that is made regarding a single-parent family. Not only will such a process

178

Single-Parent Families: The Lived Experience

improve the conditions of single-mother families, it will enable us to think more constructively and clearly about emerging forms of singleparent families.

REFERENCES Albert, J. A., & Brodek, G. A. (1989). Habeas corpus—A better remedy in visitation denial cases. Maine Law Review, 41, 239–271. Allen, D. M. (1985). Propriety of provision of custody or visitation order designed to insulate child from parent’s extramarital sexual relationships. American Law Review (4th), 40, 812. Allison D. v. Virginia M., 572 N.E.2d 27 (N.Y. 1991). 59 American Jurisprudence (1987). Parent and child, 14, 23–36, 41–49, 57–60, 69 Anders, C. (1990). State intervention into the lives of single mothers and their children: Toward a resolution of maternal autonomy and children’s needs. Law and Inequality, 4, 566–610. Aptheker v. Secretary of State, 378 U.S. 500 (1964). Ashe, M. (1992). The “bad mother” in law and literature: A problem of representation. Hastings Law Journal, 43, 1017–1037. Batty, D. L. (1990). Michael H. v. Gerald D.: The constitutional rights of putative rights of fathers and a proposal for reform. Boston College Law Review, 31, 1173–1207. Beschle, D. L. ( 1989). God bless the child? The use of religion as a factor in child custody and adoption proceedings. Fordham Law Review, 58, 383–426. Bodenheimer, B. M., & Neeley-Kvarme, L. (1979). Jurisdiction over child custody and adoption after Shaffer and Kulko. University of California at Davis, 12, 229–253. Braschi v. Stahl Assocs., 543 N.E.2d 49 (N.Y 1989). Bregande, M. (1989). No longer the father’s primary duty to support: Legal equality or economic disparity among the sexes? Saint Louis University Law Journal, 34, 133–147. Caban v. Mohammed, 441 U.S. 380 (1979). Carter, E. L. (1991). Parrillo v. Parrillo: Less must be seen of the paramour but in whose best interest? New England Law Review, 25, 1223–1250. Clark, L. M. G. (1990). Wife battery and determinations of custody and access: A comparison of U.S. and Canadian findings. Ottawa Law Review, 22, 691–724. Cohen v. Schnepf, 463 N.Y.S.2d 29 (N.Y. App. 1983). Constitutionality of gender-based classifications in criminal laws proscribing nonsupport of spouse or child. (1982). American Law Review (4th), 14, 717. Coombs, R M. (1982). Interstate child custody: Jurisdiction, recognition, and enforcement. Minnesota Law Review, 66, 711–864. 67A Corpus Juris Secundum (CJ.S.) (1978). Parent and child, 49–61, 63. Czapanski, K. (1989). Child support and visitation: Rethinking the connections. Rutgers Law Journal, 20, 619–665. Ellman, L. M., Kurtz, P. Me, & Bartlett, K. T. (1991). Family law: Cases, text, problems (2nd ed). Charlottesville, VA: Michie.

Single Parenthood and the Law

179

The Establishment Clause and religion in child custody disputes: Factoring religion into the best interests equation. (1984). Michigan Law Review, 82, 1702–1738. Family law-visitation rights—New York Court of Appeals refused to adopt a functional analysis in defining family relationships Allison D. v. Virginia M., 572 N.E.2d 27 (N.Y. 1991). (1992). Harvard Law Review, 105, 941–946. First Year Report of the New Jersey Supreme Court Task Force on Women in the Courts. (1986). Women’s Rights Law Reporter, 9, 129–177. Fleece, S. M. (1981–82). A review of the child support enforcement program. Journal of Family Law, 20, 489–521. Folberg, J. (Ed.). (1991). Joint custody and shared parenting (2nd ed.). New York: Guilford. Garska v. McCoy, 278 S.E.2d 357 (W.Va 1981). Ginsberg, R B. (1978). Women, men and the constitution: Key Supreme Court rulings. In W. L. Hepperle & L. Crites (Eds.), Women in the courts (pp. 21–46). Williamsburg, VA: National Center for State Courts (Publication #R0037). Goldberg v. Kelly, 397 U.S. 254 (1970). Gomez v. Perez, 409 U.S. 535 (1973). Gooch, K. S. (1989). Family law—Seessel v. Seessel: The burden of proof when a custodial parent wishes to remove a child from a court’s jurisdiction. Memphis State University Law Review, 19, 265–274. Gottesfeld, H. J. (1981). The uncertain status of the emancipated minor: Why we need a Uniform Statutory Emancipation of Minors Act (USEMA). University of San Francisco Law Review, 15, 473–507. Halvey 330 U.S. 610 (1947). Harrington, M. (1984). The new American poverty. New York: Penguin. Heckler v. Turner, 470 U.S. 184 (1985). Hjorth, R. (1990). The effect of federal tax consequences on amount of property allocated to spouses in state court dissolution proceedings. Family Law Quarterly, 24, 247–278. Hoffman, S. P., & Holmes, J. (1976). Husbands, wives and divorce. In G. Duncan & J. N. Morgan (Eds.), Five thousand American families—Patterns of economic progress (vol. 4). Ann Arbor: Institute for Social Research, University of Michigan. Hughes, L. (1979–80). Interstate enforcement of support obligations through long arm statutes and URESA. Journal of Family Law, 18, 537–563. In re Marriage of Hadeen, 619 P.2d 374 (Wash App. 1980). In re Marriage of Weidner, 338 N.W. 2d 351 (Iowa, 1983). In re North Carolina Inheritance Taxes, 277 S.E.2d 403 (N.C. 1987). Jefferson v. Hackney, 406 U.S. 535 (1971). Johnston, J. D., & Knapp, C. L. (1971). Sex discrimination by law: A study in judicial perspective. New York University Law Review, 46, 675–747. Kamerman, S. B., & Kahn, A. J. (1988). Mothers alone: Strategies for a time of change. Dover, MA: Auburn House. Kanowitz, L. (1969). Women and the law: The unfinished revolution. Albuquerque: University of New Mexico Press. Kass, A. (1989). A view from the bench: Children and custody. American Journal of Family Law, 3, 301–313.

180

Single-Parent Families: The Lived Experience

Keenan, L. R. (1985). Domestic violence and custody litigation: The need for statutory reform. Hofstra Law Review, 13, 407–441. King v. DcManneville, 102 Eng. Rep. 1054 (KB. 1084). King v. Smith, 392 U.S. 316 (1968). Kisthardt, M. K. (1991). Of fatherhood, families and fantasy: The legacy of Michael H. v. Gerald D. Tulane Law Review, 65, 585–661. Knauerhase, E. C. (1989). The sexually active custodial parent: A contradiction in terms? Parrillo v. Parrillo. Cooley Law Review, 6, 545. Knights v. Knights, 71 N.Y.2d 865, 522 N.E.2d 1045, 527 N.Y.S.2d 748 (1988). Koch v. Williams, 456 N.W.2d 299 (N.D. 1990). Kotkin, R. (1985). Grandparents versus the state: A constitutional right to custody. Hofstra Law Review, 13, 375–406. Krause, M. D. (1977). Family law in a nutshell. St. Paul, MN: West. Kulko v. Superior Court, 436 U.S. 84 (1978). Langbein, J., & Wolk, B. (1990). Pension and employment benefit law. Westbury, NY: Foundation Press. Lehr v. Robinson, 463 U.S. 248 (1983). Lieban, J. I. (1986). Child support in America: Practical advice for negotiating and collecting a fair settlement. New Haven, CT: Yale University Press. Lonsdorf, B. J. (1989). Coercion: A factor affecting women’s inferior outcome in divorce. American Journal of Family Law, 3, 281–300. Lozinak v. Lozinak, 569 A.2d 353 (Pa. Super 1990). Lugaila, T. (1992). Households, families and children: A 30-year perspective. Current Population Reports (Series P-23, No. 181). U. S. Bureau of the Census. Washington, DC: U. S. Government Printing Office. Massachusetts Coalition for the Homeless v. Secretary of Human Services, 400 Mass. 806 (1987). McCant, J. W. (1987). The cultural contradiction of fathers as nonparents. Family Law Quarterly, 21, 127–143. Michael H. v. Gerald D., 491 U.S. 110 (1989). Mnookin, R. H. (1978). Child, family and state: Problems and materials on children and the law. Boston: Little, Brown. Mulroy, E. A. (1988). Introduction. In E. A. Mulroy (Ed.), Women as single parents: Confronting institutional barriers in the courts, the workplace, and the housing market (pp. 3–12). Dover, MA: Auburn House. Neeley, R. (1984). The primary caretaker parent rule: Child custody and the dynamics of greed. Yale Law and Policy Review, 3, 168–186. Orr v. Orr, 440 U.S. 268 (1979). Palmore v. Sidoti, 466 U.S. 429 (1984). Parrillo v. Parrillo, 554 A.2d 1043 (R.L 1989), cert. denied, 110 S. CL 364 (1989). Paul, J.C. (1989). “You get the house. I get the car. You get the kids. I get their souls”: The impact of spiritual custody awards on the free exercise rights of custodial parents. University of Pennsylvania Law Review, 138, 583–613. Philip Morris Family Survey. (1987). New York: Louis Harris Quillion v. Walcott, 434 U.S. 246 (1978). Raymond, E. L. (1988). Mother’s status as “working mother” as factor in awarding child custody. American Law Review (4th), 62, 259–300. Reed v. Reed, 404 U.S. 71 (1971).

Single Parenthood and the Law

181

Reich, C. A. (1964). The new property. Yale Law Journal, 73, 733–787. ———. (1965). Individual rights and social welfare: The emerging legal issues. Yale Law Journal, 74, 1245–1257. Report of the New York Task Force on Women in the Courts (1987). Fordham Urban Law Journal, 15, 11–177. Rohloff v. Rohloff, 411 N.W.2d 484 (Mich. App. 1987). Rosado v. Wyman, 397 U.S. 397 (1970). Sack, L. (1992). Women and children first: A feminist analysis of the primary caretaker standard in child custody cases. Yale Journal of Law and Feminism, 4, 291–328. Sard, B. (1988). The role of the courts in welfare reform. In E. A. Mulroy (Ed.), Women as single parents: Confronting institutional barriers in the courts, the workplace and the housing market (pp. 167–202). Dover, MA: Auburn House. Schafran, L. H. (1987). Documenting gender bias in the courts: The task force approach. Judicature, 70, 280–290. ———. (1988). Gender bias in the courts. In E. A. Mulroy (Ed.), Women as single parents: Confronting institutional barriers in the courts, the workplace, and the housing market (pp. 39–72). Dover, MA: Auburn House. Schorsch, D. (Ed.). (1982). Federal regulation of family law. Charlottesville, VA: Michie. Schuele, D. (1988–89). Origins and development of the law of parental child support. Journal of Family Law, 27, 807–841. Seessel v. Seessel, 748 S.W.2d 422 (Tenn.1988). Sex discrimination and the constitution. (1950). Law Review, 2, 691. Sexual orientation and the law/Harvard Law Review. (1990). Cambridge, MA: Harvard University Press. Shaffer v. Heitner, 433 U.S.186 (1977). Shapiro v. Thompson, 394 U.S. 633 (1969). Smith v. Cole, 553 S.2d 847 (LA,1989). Solari, F. P. (1989). Custody of the illegitimate child. North Carolina Central Law Journal, 18, 18–41. Sorensen, A. (1992). Estimating the economic consequences of separation and divorce: A cautionary tale from the United States. In L. Weitzman & M. Maclean (Eds.), Economic consequences of divorce: The international perspective (pp. 263–282). Oxford: Clarendon. Stanley v. Illinois, 405 U.S. 645 (1972). Sylvain, J. C. (1990). Michael v. Gerald D.: The presumption of paternity. Catholic University Law Review, 39, 831–858. Towne v Towne, 552 A.2d 404 (VL 1988). Tuter v. Tuter, 120 S.W.2d 203 (Mo. Ct App. 1938). Tyrrell v. Tyrrell, 359 So.2d 62 (Fla. App. 1978). 42 U.S.C.A. 666(a)(9)(c) (1990). U.S. Bureau of the Census. (1992a). Child support and alimony: 1989. Current Population Reports (Series P.60, No. 173). Washington, DC: United States Government Printing Office. ———. (1992b). Poverty in the United States: 1991. Current Population Reports (Series P-60, No. 181). Washington, DC: United States Government Printing Office. Uviller, R. K. (1978). Father’s rights and feminism: The material presumption revisited. Harvard Women’s Law Journal, 1, 107–130.

182

Single-Parent Families: The Lived Experience

Wadlington, W., Whitebread, C. H., & Davis, S. M. (1983). Cases and materials on children in the legal system. Mineola, NY: Foundation Press. Washington ex rel Burton v. Leiser, 196 Cal. App.3d 451 (1987). Weisberg, D. K. (Ed.). (1989). Women and the law: A social historical perspective, Vol. 2 Property, family and the legal profession. Cambridge, MA: Schenkman. Weitzman, L. (1985). The divorce revolution: The unexpected social and economic consequences for women and children in America. New York: The Free Press. Weitzman, L. J., & Dixon, R. B. (1979). Child custody awards: Legal standards and empirical patterns for child custody, support and visitation after divorce. University of California at Davis Law Review, 12, 473–521. Wescott v. Califano, 443 U.S. 89 (1970). Winjen, G. (1990). Make room for daddy: A putative father’s rights to his children. New England Law Review, 24, 1059–1093. Wishnew, J. A. (1988). Joint custody agreements. Trial, 29–30. Woods, L., Been, V., & Schulman, J. (1983). Sex and economic discrimination in child custody awards. Clearinghouse Review, 16, 1130–1134. Wyman v. James, 400 U.S. 309 (1971). Yee, L. M. (1979). What really happens in child support cases: An empirical study of establishment and enforcement of child support orders to Denver District Court. Denver Law Journal, 57, 21–68.

Chapter 12

The Impact of Parents’ Marital Status on the Time Adolescents Spend in Productive Activities Cathleen Zick and Corrine Allen*

The amount of debate regarding how family structure affects children’s welfare has increased markedly as the number of children living in divorced and never-married families has risen. Much of the scholarly research in this area (reviewed in Demo & Acock, 1991) has focused on the negative effects of divorce on children’s social, psychological, and academic outcomes. For example, in their recent longitudinal study, Haveman & Wolfe (1994) found an inverse relationship between the number of years a minor child lives in a single-parent household and the child’s educational attainment. Explanations for the observed associations between family structure and various childhood outcomes sometimes allude to differences in children’s time use patterns as an intermediate factor. The conventional wisdom is that children in single-parent households have less parental supervision than their counterparts in two-parent households. Indeed, several studies (reviewed in McLanahan & Booth, 1991) have found that single parents are less involved in monitoring their children’s social and school activities than their two-parent counterparts. Less parental supervision may lead these children to spend less time in school work. At the same time, children in single-parent families may feel greater pressure to do housework to help the parent better balance work and family responsibilities. These children also may want to do more paid work so that they can acquire the consumer goods that their more affluent peers in twoparent households have. Wallerstein & Blakeslee (1990) go so far as to

* C. Zick and C. Allen (1996). The Impact of Parents’ Marital Status on the Time Adolescents Spend in Productive Activities. Family Relations, 45(1), 65–71. Copyright © 1996, Wiley. Reproduced with permission of Blackwell Publishing Ltd.

184

Single-Parent Families: The Lived Experience

suggest that a shift away from leisure activities for children whose parents have divorced, coupled with the psychological stress of the divorce, may mean the end of a care-free childhood for many of these children. The theory of social capital developed by Coleman (1990) can be used to formalize this conventional wisdom and generate testable hypotheses about the impact that marital status has on adolescents’ time use. Coleman argues that individuals provide resources to other individuals in various forms that include obligations, expectations, information, norms, sanctions, authority, and organizational structures. These social relationships that can be used by an individual to facilitate his/her self-interest are termed social capital. Within the context of the family, social capital is hypothesized to influence adolescents’ human capital acquisition and, consequently, the time they allocate to various productive activities. Specifically, when a mother and father are both present in the home, there is the opportunity for more parental supervision and communication with a teenager regarding expected time use (e.g., reinforcement of setting aside time to do homework) compared to single-parent households. If parents in two-parent families take advantage of these additional opportunities to bolster positive activities and discourage less fruitful uses of time, then adolescents’ time allocation is likely to vary across these two family types. The limited research that has made use of the social capital framework (Asmussen & Larson, 1991; Haveman & Wolfe, 1994) has found that the absence of a parent alters the mix of children’s time spent with family members and their early adulthood outcomes. Both of these findings are consistent with our hypotheses. Although assessing the influence of family structure is the primary focus of the current analysis, other available indicators of the family’s social capital will also be examined. Specifically, the mother has historically been viewed as the primary socializing agent within the family. The norms, sanctions, and authority that a mother exercises as a parent are hypothesized to be a function of her age, her educational attainment, her employment status, and the demands that a younger child (i.e., a preschooler) puts on her time. In addition, household income is included in the model to test for the possibility that adolescents in higher income households spend less time in paid work than adolescents in lower income households. Such a relationship might exist because adolescents from higher income households may more readily obtain money from their parents to cover their current consumer wants and/or their savings desires. The influence of social capital will be assessed in the current investigation in the context of several structural factors that have been demonstrated to affect time use (Timmer, Eccles, & O’Brien, 1985). These structural factors include the adolescent’s age, the season of the year, and

The Impact of Parents’ Marital Status

185

the day of the week. It should be noted that season may not be a particularly powerful predictor in the current study because the data used here come from a state where many (but not all) public schools operate on a year-round schedule. This type of school scheduling may reduce the seasonal variation in productive activities among teenagers, particularly in the case of school work. Although there is a sizeable literature on adolescent time use (Cogle, Tasker, & Morton, 1982; Duckett, Raffaelli, & Richards, 1989; Duncan & Sanik, 1989; Leone & Richards, 1989; Sanik & Stafford, 1985, 1986; Stafford & Evertson, 1983; Timmer, Eccles, & O’Brien, 1985), surprisingly few studies have investigated the extent to which children’s time use varies by parents’ marital status, and only one of these investigations (Asmussen & Larson, 1991) is guided by the social capital framework. The focus of most of these studies has been on how marital status affects children’s (typically those age 6 and older) time spent in housework (e.g., meal preparation, house cleaning, caring for siblings; Clark, 1980; Demo & Acock, 1993; Hilton & Haldeman, 1991; Key & Sanik, 1985; Lovett & Abdel-Ghany, 1988; Lyerly, 1969; Peters & Haldeman, 1987). Only two of these studies (Demo & Acock, 1993; Lyerly, 1969) have found statistically significant differences in children’s time spent in housework by the parents’ marital status. Work by Douthitt (1991) examined how marital status affects children’s time use in four domains: household work, leisure, school work, and paid work. Her findings showed that marital status is an important predictor of children’s leisure time, but that it does not appear to influence children’s household work, school work, or paid work. It may be that the impact of marital status on children’s time use has been rarely studied because of the general absence of time-diary surveys that contain data on sufficient numbers of single-parent families. With the exception of two unpublished master’s theses (Clark, 1980; Lyerly, 1969), all of the time-diary studies have made use of data that were gathered in California during the late 1970s on children age 6 and older who were living in 81 single- and 210 two-parent households (Douthitt, 1991; Hilton & Haldeman, 1991; Key & Sanik, 1985; Lovett & Abdel-Ghany, 1988; Peters & Haldeman, 1987). Each family in the California sample had two children. Alternative sources of time-use information are available from (1) experience sampling method (ESM) studies in which participants wear a pager and are asked to record what they are doing each time the pager sends out a signal and (2) survey reports of usual or typical time spent in various activities. The study by Asmussen & Larson (1991) used the ESM approach to compare the overall time use of fifth through ninth grade students living in two-parent (N  354) and single-parent (N  42) families. They found that time-use patterns and subjective experience did not vary by the children’s living arrangements. In contrast, Demo & Acock (1993)

186

Single-Parent Families: The Lived Experience

used survey reports from the 1987–88 National Survey of Families and Households (NSFH) to examine how parents’ and children’s housework time varies by family type. Their analysis of children utilized all boys and girls (grouped together) under age 19; theirs is one of the few studies to find that children in divorced families spend significantly more time in housework than children in first-married families (a mean of 5.9 hours per week versus 2.9 hours per week, respectively). Some of the differences in study conclusions may be a function of differences in data collection, age restrictions, and/or the theoretically informed control variables entered in the multivariate analyses. In particular, methodological studies suggest that time diaries provide the most reliable and valid measures of household time use (Andorka, 1987; Robinson, 1985). Thus, when possible, time-diary data should be used to test hypotheses regarding time use. To date, no research that used time diary data has focused specifically on how adolescents’ productive time use varies by family type. Yet, adolescents are likely to have the least parental supervision of—and hence the most control over—their time when compared to younger children. Thus, the question is whether differences in social capital between singlemother and two-parent families result in different time-use patterns for the adolescents who live in these homes. The purpose of the current study is to shed light on this issue as it relates to three key productive domains: housework, school work, and market work.

DATA The data for the present study come from a 1987–88 time-diary survey of 214 two-parent, two-child families and 99 single-mother, two-child families living in Utah. Although the generalizability of our study findings may be limited by the geographic and demographic restrictions imposed on the sample, this data set is appealing because of the detailed time-use information contained in the diaries. The adult with the major responsibility for operating the household was asked to complete the time diaries for all household members aged 6 and over. This self-identified person was the mother in all of the participating households. Time diaries were kept for two 24-hour periods. On the first day, the mother was asked to recall family members’ time use for the previous day. For the second day, she was asked to complete the time diaries as events occurred. The mother was urged to consult with other family members to verify the accuracy of the time she recorded for them on both days. Despite the consultations, the use of the mothers’ reports of adolescent time use is clearly a limitation that may affect the validity of these data.

The Impact of Parents’ Marital Status

187

In the analyses that follow, the two diary days have been averaged to construct a representative day for each adolescent. The averaging was performed in order to negate possible reporting biases. Specifically, although recall day information does not contain any behavioral bias, it may contain reporting errors due to memory loss. On the record day, memory error is less likely to be a problem, but it is possible that these observations suffer from social desirability biases and conditioning effects. That is, on the record day, family members may have changed their time use so as to make their behavior appear more socially desirable (e.g., they spend more time doing housework and less time watching television). Interviews were scheduled so that diary information was collected fairly evenly over all 7 days of the week and all 12 months of the year. In addition to completing the diaries, all respondents were asked to answer a series of questions designed to gather information about the family’s socioeconomic status and the physical characteristics of their homes. For the purposes of the current research, three categories of productive time were identified. Housework was defined to include time spent by the adolescent in food preparation; dishwashing; housecleaning; maintenance of home, yard, car, and pets; care of clothing; construction of clothing; physical care of other family members; nonphysical care of other family members; and household management. (Shopping time has been purposely omitted from the list of household work activities because, although adults may view shopping as housework, many teens identify shopping as a leisure activity.) Paid work includes all time spent by the adolescent working for pay. School work is defined to be all of the time spent by the adolescent in classes that relate to future employment and time spent preparing for those classes (e.g., doing homework).

ANALYSES The statistical analysis consisted of three parts: First, z scores were utilized to ascertain whether the proportion of adolescents participating in the three productive activities varied by the parents’ marital status at the bivariate level. Second, a multivariate analysis was used to identify whether the observed relationships continue to hold once other indicators of social capital and structural factors were controlled. The social capital variables included in the multivariate analysis were marital status of the mother, age of the mother, the mother’s education level, and presence of a preschooler, the mother’s employment status, and annual family income. Structural factors were measured by season of the year, weekend versus week day, and age of the adolescent. A sizable percentage of both boys and girls reported spending no time in one or more of the productive activities. To correct for this sample

188

Single-Parent Families: The Lived Experience

censoring, tobit, rather than ordinary least squares, was used in the multivariate analysis. Tobit is a maximum-likelihood estimation routine that is appropriately used when the distribution of the dependent variable is continuous over a range but then censored at some value, typically zero (Kmenta, 1986). Finally, the results of the tobit analysis were used to generate predicted values of the total amount of time these adolescents would spend in productive activities between the ages of 12 and 17. These figures are reported by marital status where significant differences exist. The average adolescent in the sample was 15 years old, had a mother who was 41 years old with 2 years of college education, and lived in a family with an annual income of $33,626 in 1987–88. Rarely (i.e., 7 percent of the time) was the adolescent’s sibling under age 6. Fifty-four percent of the adolescents’ diaries contained a weekend day observation. Forty-four percent of the diaries were recorded in the summer months of May through August. On average, adolescents reported spending 58 minutes in housework, 246 minutes in school work, and 64 minutes in paid work each day. These time-use figures differed by the gender of the child. Girls reported spending a mean of 64 minutes in housework, 264 minutes in school work, and 53 minutes in paid work each day. In contrast, the corresponding means for the boys were 53 minutes, 233 minutes, and 72 minutes per day. An examination of family characteristics by marital status reveal several differences—none of which are too surprising. Specifically, total household incomes averaged $40,368 in the two-parent households, whereas those households headed by single mothers averaged $23,667. Married mothers were significantly more likely than single-mothers to have a preschooler in the home (10 percent vs. 1 percent), and they were significantly less likely to be employed outside of the home (66 percent vs. 95 percent). On other dimensions, the background variables were quite similar.

CONCLUSION Do adolescents in single-mother families allocate their time to productive activities differently than adolescents in two-parent families? The current analysis shows that the impact of a parent’s marital status on adolescent time use pales in comparison to the impact of other social capital and structural measures. The other domains of social capital examined in this analysis are the resources that the mother brings to her parenting role as measured by her education, her age, and her employment status. It would appear that these resources exert considerable influence on adolescents’ productive time use, particularly as they relate to adolescent girls’ housework and paid employment.

The Impact of Parents’ Marital Status

189

The differences in social capital that are reflected in single-mother and two-parent households appear to have little impact on the time use of adolescents in these analyses. Consistent with the majority of prior studies that looked at the impact of family type on children’s housework (Clark, 1980; Douthitt, 1991; Hilton & Haldeman, 1991; Key & Sanik, 1985; Lovett & Abdel-Ghany, 1988; Peters & Haldeman, 1987), the multivariate analyses revealed no statistically significant effect of mother’s marital status on either adolescent girls’ or adolescent boys’ housework time. By broadening the categories of productive time to include school work and paid employment, we did find some evidence that girls in single-mother families may be spending more time in paid employment and boys in single-mother families may be spending less time in school work than are their counterparts in two-parent families. In addition, the answer to the above question must be viewed in light of both the weaknesses and the strengths of the analysis approach used here. Specifically, the generalizability of these findings may be limited by the fact that the data were from two-child families in Utah. The two-child sampling restriction should not be a serious concern, however, because this is the most common number of children in both single-parent and two-parent families in the United States (U.S. Bureau of the Census, 1993, Table 70). Furthermore, comparisons of earlier time-diary data collected in Utah and ten other states revealed no significant state variations in family time use (Family Time Use, 1981). Another possible concern is that the adolescents’ time diaries were recorded by their mothers, which could lead to over-reporting of the more socially desirable activities such as housework and school work. Yet, limited checks against a national time-diary study conducted seven years earlier, in which the adolescents kept their own diaries, revealed little cause for concern about the validity of the mothers’ reports. The strengths of the current analysis are threefold. First, the use of recently collected, time-diary information increases the applicability of the estimates to today’s teenagers. Second, by focusing the analysis on adolescents’ time, we can start to understand how social capital and structural factors influence the time-allocation choices of teenagers who are beginning to exercise more control over various domains in their lives. Finally, by looking at three potentially competing uses of productive time, we can begin to develop a picture of the trade-offs that teenagers are making. Within both the popular press and the academic literature, there has been considerable debate recently regarding the impact of mothers’ employment on children’s welfare. Asmussen & Larson (1991) put forth, but do not test, the proposition that the mother’s employment is not related to adolescents’ time use. The current analyses suggest that within some domains (i.e., girls’ paid employment and boys’ school work), this may not be true. These differences may occur because mothers who work

190

Single-Parent Families: The Lived Experience

outside of the home impart a different mix of social capital than mothers who are not employed by influencing their children’s aspirations and/or their norms about work and school. Structural factors also appear to play critical roles in adolescents’ time allocation. Specifically, older adolescent boys and girls increasingly shift their time out of housework and school work and into paid employment. Looking at the changes in the age-specific means for school work and paid employment, we found that the average 12-year-old girl spent 337 minutes in school work and only 20 minutes in paid employment. In contrast, the average 17-year-old girl spent 229 minutes in school work and 117 minutes in paid employment. Comparing the average 12-year-old boy to the average 17-year-old boy, we found that school work time declined from 293 minutes to 195 minutes per day and market work time increased from 4 minutes to 251 minutes per day. The analyses conducted in this article reveal only selective evidence of differences in adolescents’ productive time use by parents’ marital status. The estimated 3.6 hours more per week that adolescent girls in singlemother families spend in paid employment appear to come at the expense of leisure time rather than other productive activities. Likewise, the estimated 3.5 hours per week less that boys spend in school work when they live in single-mother families appear to add to their leisure time rather than to other productive activities. To the extent that adolescents in single-mother households miss out on some social capital because their fathers are absent, it would appear that this situation does not translate into large shifts in adolescent time use. This result suggests that family life educators and family counselors should not necessarily attribute adverse schooling and/or socio-emotional outcomes experienced by adolescents in single-mother households to the premature adoption of adult work roles. If differences in social capital between single-mother and two-parent families affect children’s socio-emotional development and their human capital acquisition, then these differences must be manifest in more subtle ways than adolescents’ time use. Perhaps they affect qualitative dimensions of their time use that were not measured here (e.g., whether they elect to take college preparatory classes rather than vocational classes at school) or perhaps they affect the range of adult role models that teenagers consider when thinking about the future. The absence of significant social capital effects—as captured by marital status in the current study—suggests that the relationship between family structure and childhood outcomes is more complex than what has been alluded to in the literature. More important than social capital in the current investigation appears to be the pervasive impact of age. As adolescents grow older, they spend less time investing in family (via housework) and their own future (via

The Impact of Parents’ Marital Status

191

school work), and they spend more time in paid work, the proceeds of which may be used for immediate gratification and/or savings. The view of paid employment as a way to enhance current consumption is reinforced by Mortimer et al. (1990), who found that 69 percent of the employed adolescents in their study said they were working “to buy things.” The dramatic shift in the composition of productive activities over the adolescent years noted here and elsewhere (Mortimer et al., 1990) has sparked interest in identifying to what extent adolescents’ paid employment time influences their well-being (Bachman & Schulenberg, 1993; Finch & Mortimer, 1985; Greenberger & Steinberg, 1983, 1986; Mortimer & Finch, 1986; Mortimer, Finch, Owens, & Shanahan, 1990; Shanahan, Finch, Mortimer, & Ryu, 1991; Yamoor & Mortimer, 1990). A number of these studies have found an association between teenage employment and undesirable behaviors (e.g., drug use, poor eating habits, inadequate exercise, lower levels of post-secondary education; Bachman & Schulenberg, 1993; Greenberger & Steinberg, 1986; Mortimer and Finch, 1986; Schulenberg, 1994). It is unclear, however, whether work causes these outcomes or whether teens with less optimal life experiences are more likely to seek paid employment. One of the studies has identified both positive and negative teenage employment effects (Yamoor & Mortimer, 1990). In light of the recent growth in adolescent employment, researchers need to continue to investigate how the quality of the work experience affects teenagers and the extent to which the negative aspects are a function of work-related socialization processes as opposed to self selection into the work role. REFERENCES Aldrich, J. H., & Nelson, F. D. (1984). Linear probability, logit, and probit models. Beverly Hills: Sage. Andorka, R. (1987). Time budgets and their uses. American Review of Sociology, 13, 149–164. Asmussen, L., & Larson, R. (1991). The quality of family time among young adolescents in single-parent and married-parent families. Journal of Marriage and the Family, 53, 1021–1030. 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, 220–235. Belsley, D. A., Kuh, E., & Welsch, R. E. (1980). Regression diagnostics: Identifying influential data and sources of collinearity. New York: Wiley. Blalock, H. D., Jr. (1979). Social statistics. New York: McGraw-Hill. Clark, C. L. (1980). Children’s time use in one-parent and two-parent families. Unpublished master’s thesis, Oklahoma State University, Stillwater.

192

Single-Parent Families: The Lived Experience

Cogle, F., Tasker, G., & Morton, D. (1982). Adolescent time use in household work. Adolescence, 17, 451–455. Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: The Belknap Press of Harvard University Press. Demo, D. H., & Acock, A. C. (1991). The impact of divorce on children. In A. Booth (Ed.), Contemporary families: Looking forward, looking back (pp. 162–191). Minneapolis, MN: National Council on Family Relations. ———. (1993). Family diversity and the division of domestic labor: How much have things really changed? Family Relations, 42, 323–331. Douthitt, R. A. (1991). Children’s time use in single- and two-parent families: Does household organization matter? Home Economics Research Journal, 20, 40–51. Duckett, E., Raffaelli, M., & Richards, M. H. (1989). “Taking care”: Maintaining the self and the home in early adolescence. Journal of Youth and Adolescence, 18, 549–565. Duncan, K., & Sanik, M. (1989). Time use of North American adolescents. In Proceedings of the 18th Annual Conference of the Southeast Regional Association, Family Economics–Home Management (pp. 85–94). Lexington, KY: University of Kentucky. Family time use: An eleven-state urban/rural comparison. (1981). (Bulletin VPI-2). Blacksburg, VA: Virginia Agricultural Experiment Station. Finch, M. D., & Mortimer, J. T. (1985). Adolescent work hours and the process of achievement. In A. C. Kerckhoff (Ed.), Research in sociology of education and socialization (vol. 5, pp. 171–196). Greenwich, CT: JAI. Frankel, M. (1983). Sampling theory. In P. H. Rossi, J. D. Wright, & A. B. Anderson (Eds.), Handbook of survey research (pp. 21–69). New York: Academic Press. Greenberger, E., & Steinberg, L. D. (1983). Sex differences in early labor force experience: Harbinger of things to come. Social Forces, 62, 467–486. ———. (1986). When teenagers work: The psychological and social costs of adolescent employment. New York: Basic Books. Haveman, R., & Wolfe, B. (1994). Succeeding generations: On the effects of investments in children. New York: Russell Sage Foundation. Hilton, J. M., & Haldeman, V. A. (1991). Gender differences in the performance of household tasks by adults and children in single-parent and two-parent, two earner families. Journal of Family Issues, 12, 114–130. Key, R. J., & Sanik, M. M. (1985). Children’s contribution to household work in one- and two-parent families. In Proceedings of the 14th Annual Conference for the Southeast Regional Association of Family Economics Home Management (pp. 162–165). Baton Rouge, LA: Louisiana State University. Kmenta, J. (1986). Elements of econometrics (2nd ed.). New York: Macmillan. Leone, C. M., & Richards, M. H. (1989). Classwork and homework in early adolescence: The ecology of achievement. Journal of Youth and Adolescence, 18, 531–548. Lovett, S. B., & Abdel-Ghany, M. (1988). Children’s contributions to household activities in single-parent and two-parent families. Journal of Consumer Studies and Home Economics, 12, 199–204. Lyerly, B. K. (1969). Time used for work in female-headed, single-parent families as compared with two-parent families. Unpublished master’s thesis, Cornell University, Ithaca, New York.

The Impact of Parents’ Marital Status

193

McDonald, J. F., & Moffitt, R. A. (1980). The uses of tobit analysis. Review of Economics and Statistics, 57, 318–321. McLanahan, S., & Booth, K. (1991). Mother-only families. In A. Booth (Ed.), Contemporary families: Looking forward, looking back (pp. 405–428). Minneapolis, MN: National Council on Family Relations. Mortimer, J. T., & Finch, M. D. (1986). The effects of part-time work on self-concept and achievement. In K. Borman & J. Reisman (Eds.), Becoming a worker (pp. 66–89). Norwood, NJ: Ablex. Mortimer, J. T., Finch, M. D., Owens, T. J., & Shanahan, M. (1990). Gender and work in adolescence. Youth and Society, 22, 201–224. Peters, J. M., & Haldeman, V. A. (1987). Time used for house-hold work: A study of school-age children from single-parent, two-parent, one-earner, and twoearner families. Journal of Family Issues, 8, 212–225. Robinson, J. P. (1985). The validity and reliability of diaries versus alternative time use measures. In F. T. Juster & F. P. Stafford (Eds.), Time, goods, and well-being (pp. 33–62). Ann Arbor: University of Michigan, Survey Research Center. Sanik, M., & Stafford, K. (1985). Adolescents’ contribution to household production: Male and female differences. Adolescence, 20, 207–215. ———. (1986). Boy/Girl differences in household work. Journal of Consumer Studies and Home Economics, 10, 209–219. Schulenberg, J. (1994). High schoolers’ jobs may “cost” too much. Profiles. The ISR Newsletter, 18(2), 11. Shanahan, M. J., Finch, M., Mortimer, J. T., & Ryu, S. (1991). Adolescent work experience and depressive affect. Social Psychology Quarterly, 54, 299–317. Stafford, J. P., & Evertson, C. M. (1983). Time use and activities in junior high classes. Journal of Education Research, 76, 140–147. Timmer, S., Eccles, J., & O’Brien, K. (1985). How children use time. In F. T. Juster & F. P. Stafford (Eds.), Time, goods, and well-being (pp. 353–382). Ann Arbor: University of Michigan, Institute for Social Research. U.S. Bureau of the Census. (1993). Statistical abstract of the United States: 1993 (113th ed.). Washington, DC: U.S. Government Printing Office. Wallerstein, J. S., & Blakeslee, S. (1990). Second chances. New York: Ticknor & Fields. Yamoor, C. M., & Mortimer, J. T. (1990). Age and gender differences in the effects of employment on adolescent achievement and well-being. Youth and Society, 22, 225–240.

Chapter 13

The Effects of Neighborhood and Mother-Teen Relationships on Adolescent Sexual Activity H. Harrington Cleveland and Michael Gilson*

This study considers the moderating effects of neighborhood proportion of single-parent families on the association between mother-adolescent relationships and the number of sexual partners reported by male and female adolescents. Ecological theory provides that the entirety of the social world around adolescents influences their developmental outcomes (Bronfenbrenner, 1986). In spite of this theoretical perspective, however, most research on both adolescents’ nonsexual and sexual behaviors has focused on individual-, family-, and peer-level predictors. This focus has expanded in the last 15 years, however, as a relatively small body of research has taken advantage of new data sources that allow consideration of neighborhood contexts. The general message of these studies is that, even when family-level influences are controlled for, neighborhoods effects still seem to matter (Brooks-Gunn et al., 1993; Case & Katz, 1991; Crane, 1991; but see Hogan & Kitagawa, 1985) and are not limited to inner city ghettos (Simons et al., 1996). Outcomes associated with neighborhood contexts include both nonsexual outcomes, such as school performance, employment expectations (Quane & Rankin, 1998), school dropout (Crane, 1991), conduct problems (Simons et al., 1996), and later welfare receipt (Vartanian, 1999), and sexual outcomes, such as age of first intercourse (Brewster et al., 1993), probability of early child birth (Brooks-Gunn et al., 1993; Crane, 1991), and the outcome examined herein, number of sexual partners (Ku et al., 1993).

* H. Harrington Cleveland and Michael Gilson. Effects of Neighborhood Proportion of Single-Parent Families and Mother-Adolescent Relationships on Adolescents’ Number of Sexual Partners. Journal of Youth & Adolescence 2004, 33(4), 319–329. Reprinted with kind permission from Springer ScienceBusiness Media.

The Effects of Neighborhood and Mother-Teen Relationships

195

MOTHER-ADOLESCENT RELATIONSHIPS Family level relationships, in particular mother-adolescent relationships, have been found to exert powerful influences on adolescent behavior. Adolescents who view their parents as warm and supportive and who feel that they communicate well with their parents are significantly less likely to be sexually experienced (Hogan & Kitagawa, 1985; Miller et al., 1997, 1998; for review see Miller et al., 2001). The influence of this relationship may vary by adolescents’ gender. Girls tend to be more attuned to adults, including parents, than are boys (Fagot, 1994). Thus, they might be more likely to invite parental participation in their lives. Moreover, mothers tend to be more involved in the lives of their daughters than their sons and are more likely to discuss intimate topics with daughters (Chodorow, 1978; Collins & Russell, 1991; Nollen, 1980;Youniss & Smollar, 1985).Given these gender differences in adolescent-mother relationships, the influence of mother-adolescent relationships on sexual behaviors may also differ for females and males. Other mother characteristics, attitudes, and behaviors are associated with adolescents’ sexual attitudes and behaviors. Specifically, highly educated mothers are less likely to have sexually active adolescents (Miller & Olson, 1988; Scott-Jones & White, 1990). Similarly, adolescents with ideologically conservative mothers are more likely to abstain from intercourse (Baker et al., 1988). Further, adolescents whose mothers have less permissive sexual attitudes have less permissive attitudes and behaviors themselves (Thornton& Camburn, 1987).

SINGLE-PARENTS AND NEIGHBORHOODS: AVAILABILITY OF SUPERVISION, ROLE MODELS, AND INFORMAL SOCIAL CONTROL Adolescent sexual behaviors and early single parenthood have also been found to be associated with neighborhood characteristics (Hogan & Kitagawa, 1985; Ku et al., 1996). Thus, adolescent sexual behaviors are not unlike other risk behaviors that are more frequently adopted in disadvantaged neighborhood contexts (e.g., aggression; Sampson & Groves, 1989). The specific explanations for why adolescents in some neighborhoods are more likely to adopt these deviant behaviors vary from poor adult supervision, scarcity of role models, and lack of community-level social connectiveness to poor school and police quality and neighborhood age and sex compositions that encourage delinquency (Wilson, 1987). The first three of these mechanisms, neighborhoods’ inadequate supervision, scarcity of positive role models, and lack of community-level social connectiveness, have been linked to neighborhoods’ proportions of single-parent families, which is addressed below.

196

Single-Parent Families: The Lived Experience

First, large proportions of single-parent families directly reduce the availability of adults who would normally serve monitoring roles in neighborhoods (Furstenberg, 1993). Single-parent households are normally headed by single mothers. If these mothers work outside the household, there are less parental resources available for supervision than in comparable two-parent households (Hogan & Kitagawa, 1985). Thus, as neighborhood proportions of single-parent households increase so do the number of households with relatively little or no parental supervision. This increases the neighborhood risk of deviant behavior not only for the adolescents from these households but also, by increasing neighborhood opportunities for deviance, for adolescents whose own households may be adequately supervised (Simons et al., 1996). Second, high proportions of single-parent families also reduce the number of mainstream role models, especially male role models, available to neighborhood adolescents. With few positive role models, children and adolescents in these neighborhoods may have little reason to believe that working hard and behaving prudently will lead to success (Mayer & Jencks, 1989). In place of responsible mainstream role models, children and adolescents are presented with examples of single moms, noncustodial fathers, and welfare receipt (Vartanian, 1999).Wilson suggests that the combination of these socialization experiences promotes a focus on the present, poor planning and organization, little sense of personal control, and little emphasis of school- or job related skills (Wilson, 1987). Third, the proportion of single-parent households in neighborhoods contributes to deviant behavior by increasing social isolation and limiting parents’ ability to contribute to neighborhood networks of informal social control (Furstenberg, 1993; Sampson, 1993). With more demands on their parental resources, single parents are less able to contribute to social order (Sampson et al., 1999; Taylor et al., 1984; Wilson, 1987). The deleterious effect of high neighborhood proportions of single-parent households is made clear by Furstenberg’s ethnographic comparison of parenting across two neighborhoods (Furstenberg, 1993). In one neighborhood, characterized by two-parent families and residential stability, neighbors chastised negligent parents and disciplined neighborhood youth for breaching neighborhood norms of behavior. In this neighborhood, “scrutiny of neighborhood, schools, and families” held youth in check (Furstenberg, 1993, p. 249). Parents knew each other, were confident that they shared the same norms on how neighborhood children should be raised, and felt they could count on neighbors to enforce these norms. In contrast, the neighborhood with a greater proportion of households headed by single parents was characterized by low informal social control. Parents in this neighborhood felt it was difficult enough to supervise their own offspring, let alone regulate the behaviors of other families’ children. They further believed that if they engaged in any community parenting, their

The Effects of Neighborhood and Mother-Teen Relationships

197

efforts would not be reciprocated. Thus, they did not form networks of reciprocal supervision with other parents. Instead they adopted individualistic parenting strategies (Furstenberg, 1993, p. 240), designed to insulate their offspring from the dangers of the neighborhood’s unsupervised streets, and effectively removed the safety net of community parenting.

CONTEXTUAL DATA: TRACTS AND BLOCK GROUPS Most research on neighborhood effects has used either counties or tract-level data (Sawicki & Flynn, 1996). According to Billy & Moore (1992), who compared tract- versus county-level data, there are advantages to using smaller geographic units (in their case tracts). Using tractlevel data, Brewster, Billy, & Grady (1993) found that sexual behaviors varied accordingly across neighborhoods. A tract refers to both census tracts and block numbering areas (BNAs). A census tract is a small, locally defined statistical area, normally within a county and generally having stable boundaries and relatively homogeneous demographic characteristics (for details see Billy et al., 1998). Generally, tracts are defined for metropolitan areas and other highly populated counties. They usually contain between 2,500 and 8,000 people and have an average population of 4,000. A block numbering area (BNA) is delineated cooperatively by a state and the Census Bureau for grouping and numbering blocks in areas where census tracts have not been established. Similar to tracts, block numbering areas do not cross county boundaries (Billy et al., 1998). The Census Bureau publishes the same types of data for BNAs as it does for census tracts, thereby treating them as equivalent. The use of tract-level census data as a measure of neighborhood context has been criticized for being too large an aggregate to capture the range of heterogeneity that exists in the smaller geographic units in which children live. Similarly, it is thought to be a larger aggregation than is relevant for adolescent development (Burton & Price-Spratlen, 1999). To address these criticisms, this study operationalizes neighborhoods using both tract-level and block group–level census data. Block groups are clusters of census blocks within a census tract or BNA. They are the lowest level of geography for which the Census Bureau publishes sample data. For the 1990 census, block groups averaged 452 housing units, or 1,100 people. They are small, usually compact areas, bounded by streets, other prominent physical features, or legal boundaries, and typically contain four or five census blocks (Billy et al., 1998). It may be that block groups correspond more closely to what can be thought of as neighborhoods, characteristics of which would confront individuals as they develop through childhood and adolescence.

198

Single-Parent Families: The Lived Experience

PRESENT STUDY Regardless of the mechanisms by which neighborhood contexts and mother-adolescent relationships exert their influences, both have been found to predict sexual behavior. This study examines the combined effects of neighborhood contexts and mother-adolescent relationships on adolescents’ number of sexual partners. The regression analyses will be run using both Ordinary Least Squares (OLS) and Repeated Measures procedures. By treating all cases within the same tract (or block group) as repeated observations, Repeated Measures analyses properly adjusts standard errors for the clustering of subjects within census tracts and block groups. Thus, this approach is more rigorous than using OLS regressions alone. The effects of neighborhood proportions of singleparent families will be considered using both tract and block group levels of aggregation. Considering both these aggregation levels will help determine the level of neighborhood context that may be best suited for examining the effects of neighborhood context on adolescent sexual behavior. On the basis of previous research, this study proposes several hypotheses, two main effects hypotheses, neighborhood influence and maternal-relationship influence, and a contextual interaction hypothesis. First, adolescents’ neighborhood proportion of single-parent families should positively associate with their number of sexual partners. Second, the quality of adolescents’ relationships with their mothers should negatively predict their number of sexual partners. Third, we hypothesize an interaction between neighborhood context and the mother-adolescent relationship. The relatively poor supervision, fewer role-models, and ineffective social control that have been observed in neighborhoods with high proportions of single-parent families may simultaneously create more opportunities for participating in intercourse and less motivation for abstaining from it. Thus, individual adolescents’ mother-adolescent relationships may be more important in these neighborhoods than in less disadvantaged neighborhood contexts. Finally, because prior research has found differences in mother-adolescent relationships by gender, analyses will consider differences in the above mentioned associations that may exist according to gender. METHOD Data Source: The Add Health Data The data used for this study are drawn from the first wave of the National Longitudinal Study of Adolescent Health (Add Health). The Add Health project consists of multiple data sets organized around a school sample that represents a stratified random sample of all high schools in the United States (Carolina Population Center, 1998). In the In-School sample,

The Effects of Neighborhood and Mother-Teen Relationships

199

questionnaires were collected from more than 90,000 adolescents. All students who completed an In-School Questionnaire, or who were listed on a school roster, were eligible for inclusion in the Wave I In-Home sample (N  20,745) collected between April and December 1995. Parents of the InHome adolescent respondents were asked to answer a short questionnaire about their household and their child or children in the study. Included in the project is a contextual data set, which provides demographic information on the county, tract, and block group level drawn from the 1990 census. More specific information about the Add Health research design, sampling, and data instruments is available in Bearman, Jones, & Udry (1998). The data set used for this study consisted of 12,923 cases with complete In-Home Adolescent, Parent, and Contextual data sets. Individual Level Demographic Variables The majority of respondents, 56 percent of the males and 55 percent of the females, classified themselves as white. The second most common racial/ethnic group was black/African American, with 20 and 22 percent of males and females, respectively, followed by 16 and 15 percent Hispanic, and 8 and 8 percent Asian or Pacific Islander or “other.” To distinguish between these four racial categories, three dummy coded variables were created. Males averaged 16.2 (SD  1.7) years and females 16.1 (SD  1.7) years. To complement the neighborhood measure, proportion of single-parent families, we used its individual-level analog: respondents’ intact versus non-intact household status. More males (56 percent; 3,573) were in intact households, than in non-intact households (2,786). Household distribution was similar for females, 3,538 (54 percent) were in intact households, 3,026 in non-intact households. Respondents’ family socioeconomic status was evaluated with two measures: mother’s education and annual household income drawn from Add Health’s Wave I In-Home parental interview. To obtain mother’s education, mothers were asked to provide how far they had gone in school. The eight response categories were grade 8 or less, grade 8 but did not graduate from high school, vocational school instead of high school, GED, high school, went to college but did not graduate, graduated from college or university, and professional training beyond college or university. Parental responses for this educational item were similar for the mothers of both male and female Add Health respondents. The most common single educational category for the mothers of both males and females, with 25 and 26 percent, was high school graduate. Nearly half of the sample’s mothers (45 percent of males and 44 percent of females), however, had attended or graduated from college. Of these, 11 percent and 9 percent of males’ and females’ mothers, respectively, indicated attending but not graduating. About 25 percent of both sexes reported graduating from college, and 9 percent and 10 percent reported professional training beyond college.

200

Single-Parent Families: The Lived Experience

Neighborhood Risk: Proportion of Single-Parent Families For both males and females, proportions of single-parent families within respondents’ census tract or block group ranged between 0.00 and 0.80. On the tract level, this measure was associated with other measures of neighborhood context. For example, it was correlated 0.82 with proportion of all non-intact households (including those without children), 0.65 with proportion of households living below poverty level, 0.62 with proportion of households with less than $15,000 in annual income, 0.51 with median income, and –0.45 with proportion of households with greater than $50,000 in annual income. Average proportion of singleparent families was 0.23 (0.14) for males and 0.24 (0.15) for females. Similar to the mother adolescent relationship scale, values of this variable were used for some analyses to classify participants into one of three levels of neighborhood risk: low (proportion of single-parent families less than 0.15), medium (less than 0.35 but greater than or equal to 0.15), and high (proportion of single-parent families equal to or above 0.35). Using these values, 22 percent of participants were classified into the low-risk neighborhood group, 58 percent in the medium, and 20 percent in the high-risk neighborhood context group. Mother-Adolescent Relationship A four-item scale was used to assess quality of mother-adolescent relationship. Items measured the closeness, warmth, and level of communication. Items included “how close do you feel to [name of mom]?” and “are you satisfied with the way [name of mom] and you communicate with each other?” Responses were given on a 5-point scale and reversecoded so a high score on the resulting scale represented a high-quality relationship between mother and adolescent. Alphas for the motheradolescent relationship scale were 0.89 for males and 0.86 for females. Males had a slightly higher mean (4.38) and smaller standard deviation (0.64) than females (4.23, 0.80). For some analyses, scores on this scale were used to construct a three-level variable classifying participants’ mother-adolescent relationship into good ( 5), medium (  4  5), and poor ( 4) levels. Using these values, 21 percent of participants were classified into the poor relationship group, 50 percent in medium, and 29 percent in the good relationship quality group. Outcome Variable: Number of Sexual Partners This study examines how neighborhood levels of single-parent households and mother-adolescent relationships influence adolescents’ number of sexual partners. Number of sexual partners was assessed with several items from the Wave I In-Home survey. The items asked respondents if

The Effects of Neighborhood and Mother-Teen Relationships

201

they had ever had sex with anyone and, if so, how many partners total. Not surprisingly, males reported greater numbers of partners (M  1.99, SD  6.65) than females (M  1.03, SD  2.59). Analytical Methods Most models were run separately by respondent sex. As noted above, analyses were performed at both the tract and block group level. Each set of analyses included four models. Model 1 examined the main effects of the two variables central to our hypothesis: neighborhood proportion of single-parent families (Neighborhood Risk) and mother-adolescent relationship. Models 2 and 3 added family level variables, such as individual level intact family status (Model 2) and racial and socioeconomic measures (Model 3). Model 4 added two interaction terms: neighborhood risk  mother-adolescent relationship and individual level intact family status  mother-adolescent relationship. Following these analyses, models were rerun in Repeated Measures format using SAS PROC GENMOD. This precaution was taken to adjust for potential unmeasured autocorrelation because of participants being clustered within census tracts and block groups, which could bias the estimation of standard errors within Ordinary Least Squares regression analyses. To correct for this nonindependence, Repeated Measures models were run treating respondents within each census tract as withinsubject measurements. RESULTS Male Results Results for the first model revealed that the number of sexual partners for males was positively associated with census tract proportion of singleparent families and negatively associated with quality of motheradolescent relationship. Model 2 added individual-level intact family status to the model, which was negatively associated with number of sexual partners, indicating that belonging to an intact family household was associated with lower numbers of sexual partners. Both neighborhood proportion of single-parent families and quality of motheradolescent relationship maintained their significance. Model 3 added three individual-level, dummy-coded race distinctions for white, black, and Hispanic, using the Asian/other category as the comparison group, as well as two indicators of family socioeconomic status to the model: mother’s education and family’s annual income. Of these, only being black was significantly associated with a greater number of partners—although mother’s education was nearly significant

202

Single-Parent Families: The Lived Experience

(t  1.87, p  0.062). Adding these covariates removed the significance of census tract’s proportion of single-parent families, but individual intact family status and mother-adolescent relationship maintained their significance. Model 4 added two interaction terms: mother-adolescent relationship  proportion of single-parent families in the census tract and motheradolescent relationship  intact household status of respondent. The coefficient for the first of these, mother-adolescent relationship  proportion of single-parent families, was significant and negative (3.55, t  3.69, p  0.05). The only significant main effect, excluding those involved in interactions, was being black. Again, mother’s education was nearly significant (t  1.82, p  0.069). Repeating analyses using census block group instead of census tract to define neighborhoods, the results remained substantively the same (no table provided). The only differences in findings were that mother’s education, which only approached significance in the tract-level models, was significant in models using the block group level to classify neighborhoods. Coefficients for mother’s education were 1.01 (t  2.32, p  0.05) in Model 3 and 1.00 (t  2.29, p  0.05) in Model 4. Female Results Just as they had for males, results for Model 1 revealed that the number of sexual partners for females was positively associated with neighborhood proportion of single-parent families and negatively associated with quality of mother-adolescent relationship. Unlike males, however, the addition of individual intact family status in Model 2 removed the significance of tract proportion of single-parent families. Model 3 added the three individual level dummy coded race distinctions, as well as two indicators of family socioeconomic status to the model: mother’s education and family’s income. Of these, only mother’s education was significantly associated with number of sexual partners (t  3.34, p  0.05); as mother’s education increased females’ number of sexual partners decreased. Unlike the male results, being black was not associated with a greater number of sexual partners. Adding these covariates did not remove the significance of either the mother-adolescent relationship or individual intact family status. Model 4 added the two interaction terms mother-adolescent relationship  proportion of -families in the census tract and mother-adolescent relationship  intact household status of respondent. The coefficients for both of these interactions were significant. Unlike the interaction between mother-adolescent relationship and proportion of single-parent families

The Effects of Neighborhood and Mother-Teen Relationships

203

for males, coefficients for both of these interactions were positive for females. Unlike males, switching the census unit from tracts to block groups changed the females’ results. Using block groups, the two interactions in Model 4 were not significant at the 0.05 level. Their coefficients (and t scores) were 0.167 (1.86) mother-adolescent relationship  intact household status of respondent and 0.404 (1.62) for mother-adolescent relationship  proportion of single-parent families. Examining the decrease in the coefficient from 0.181 to 0.167 for the mother-adolescent relationship  intact household status term suggests that the interaction’s loss of significance (p  0.063) in the block group model is due as much to reduction in sample size from 6,564 to 6,100 participants (block group had more missing values) as to any substantive difference in the level of neighborhood classification used. In contrast, the change in p value from  0.05 to 0.105 for the mother-adolescent relationship  proportion of single-parent families interaction appears to be because of the reduction in the magnitude of coefficient from 0.641 at the tract level to 0.404 at the block group level. This suggests that among adolescents block group levels of family composition may not be as relevant for understanding neighborhood effects of family composition on female sexual behavior as tract level classifications. Male Repeated Measures As noted previously, the nesting of participants inside census tracts and block groups creates the potential for unmeasured autocorrelation that may bias the estimation of standard errors within Ordinary Least Squares regression analyses. To correct for this nonindependence, we re-ran analyses using a Repeated Measures design using SAS PROC GEN MOD in which respondents within each census tract were treated as within-subject repeated measurements. We chose the census tract level rather than the block group level because of the generally, albeit marginally, stronger OLS results using this contextual level. In terms of the significance of particular terms the results were identical to those provided by the prior OLS regressions. Each of the 11 significant terms in the four OLS tract level regressions was significant at the 0.05 level using the Repeated Measures approach. Female Repeated Measures Unlike results for males, in which in each of the significant terms in the four OLS tract level regressions was also significant at the 0.05 level using the Repeated Measures approach, performing analyses using a Repeated Measures format changed the tract-level results for females. Differences of

204

Single-Parent Families: The Lived Experience

note include that for Model 1 the proportion of single-parent families main effect was no longer significant. Perhaps more importantly, although the magnitude of their coefficients was unchanged, Model 4’s two-way interaction terms were no longer significant. Three-Way Interactions The significance of the difference between the male and female coefficients for the two-way interaction for mother-adolescent relationship  proportion of single-parent families was tested by running the sexes together in a similar model and adding a main effect for sex of respondent and three-way interaction terms for both respondent sex  individual level intact-family  mother-adolescent relationship and respondent sex  tract-level proportion of single-parent families  mother-adolescent relationship. Again OLS regression and Repeated Measures models were run. Results from the OLS regression revealed both three-way interactions were significant, confirming that the two-way interactions between Individual-level intact-family  mother-adolescent relationship and tractlevel proportion of single-parent families  mother-adolescent relationship interactions differed by sex of adolescent respondents. Results from Repeated Measures analysis differed somewhat. The interaction between respondent sex, tract level proportion of single-parent families, and mother-adolescent relationship was still significant at the 0.05 level. The three-way interaction involving individual level intact family status, however, only reached a 0.09 level of significance. Interpreting Interactions: Tables of Means Results consistently indicate that the neighborhood risk (proportion of single-parent families) by mother-adolescent relationship interaction differed according to gender. Like any three-way interaction, the meaning of this is not immediately apparent. Accordingly, tables (not shown) of average sexual partners for males and females according to mother-adolescent relationship and proportion of single-parent families in their census tract were produced. Respondents were first grouped into good, medium, or poor levels of mother-adolescent relationship using their scores on the mother-adolescent relationship scale. Similarly, adolescents were also grouped into low-, medium-, and high-risk neighborhood groups. Using these two three-level variables, a 3  3 matrix of mother-adolescent relationship by neighborhood risk groups for males. Results reveal both the main effects of neighborhood risk as the number of partners increases from the top row of the table to the bottom row and from the left to the right. The two-way neighborhood risk  mother-adolescent relationship interaction is apparent in the patterns of means observed in bottom

The Effects of Neighborhood and Mother-Teen Relationships

205

versus top rows. In the bottom row, which corresponds to high-risk neighborhoods, average number of sex partners increases 188 percent from 1.8 to 5.2 from good to poor mother-adolescent relationships. This increase is far steeper than the corresponding increase found in the first row representing low-risk neighborhoods. In this row there is only a 66 percent increase in the number of sexual partners for good to poor mother-adolescent relationship quality. Like males, females’ numbers of sexual partners increases within neighborhood groups corresponding to decreasing quality of Motheradolescent relationship. However, unlike males, who show the greatest influence within high-risk neighborhoods, the effect of Motheradolescent relationships are the most important for females in low-risk neighborhoods. In these neighborhoods, the number of partners increases 466 percent from 0.03 to 1.7 partners from good to poor mother-adolescent relationships.

DISCUSSION Support was found for each of this study’s three hypotheses. The first hypothesis, neighborhood influence, was supported by the main effect for neighborhood risk on the number of sexual partners reported. As anticipated, adolescents living in neighborhoods with higher proportions of single-parent families reported a greater number of sexual partners. This finding was consistent across the census tract and block group levels for males and females using standard OLS regression and when using a Repeated Measures design at the tract level. Interestingly, a sex difference was found when an adolescent’s own family status was also entered into the model. Specifically, the effect of neighborhood context remained significant for males, but not for females, when the individual-level family covariate was added to the model. Thus, the processes by which neighborhood and individual risks affect adolescents’ sexual decision making appear different for males and females. For females, family-centered processes seem to insulate adolescents from neighborhood-wide risks. But for males, it appears that both neighborhood-wide and family-centered risks independently contribute to sexual behaviors. Thus, although family structure seems to provide female adolescents effective protection from neighborhood risks, similar family protections do not appear to isolate males from the risks of being in neighborhoods with high proportions of single-parent families. It is not clear why family structure is differentially protective across the sexes. It may be that males are less likely than females to have intercourse in their own families’ households, thereby reducing the effectiveness of the higher family monitoring that is possible in intact households.

206

Single-Parent Families: The Lived Experience

Results also supported the second hypothesis, maternal-relationship influence. The quality of mother-adolescent relationship was consistently associated with adolescents’ number of sexual partners. Like the main effect found for neighborhood context, this association was found across genders for both census tract and block group levels of analyses and existed independent of any neighborhood effects. This finding comports with the longstanding body of research that suggests parents have a powerful effect on adolescent sexual decision making (for review see Miller et al., 2001). No effect of mother’s education was found for males, but a negative effect was found for females. This suggests that maternal education may serve to selectively protect females from multiple sexual partners, above and beyond the quality of the maternal-daughter bond. Moreover, it seems that mothers’ characteristics may directly affect daughters in ways that they do not affect sons. The mechanism by which education affects female sexual behavior needs further exploration. With one notable exception, this study found little influence of race upon the number of sexual partners once the effect of neighborhood and the mother-adolescent relationship had been partialed out. The only exception to this was that being a black male still predicted a higher number of sexual partners. This finding is consistent with prior findings that black adolescents are more likely to report engaging in intercourse and at younger ages (Bingham et al., 1990; Christopher et al., 1993; Elliott & Morse, 1989). In contrast, the numbers of sexual partners for black females and whites and Hispanics of both sexes did differ from the Asian and other comparison group. This study’s most important finding is the consistently significant twoway interaction between neighborhood level of single-parent families and mother-adolescent relationship. Supporting the contextual interaction hypothesis, this interaction was significant in OLS models for both males and females at both census levels and for males in the Repeated Measures analysis at the tract level. This interaction reveals the dependence of family-level processes on the larger social context within which families are nested. Interestingly, this interaction took on different directions by gender, negative for females and positive for males. The significant three-way interaction term (gender  mother-adolescent relationship  neighborhood risk) confirmed that this two-way interaction differed by gender. The pattern of mean number of sexual partners by neighborhood risk and mother relationship revealed the dramatically different ways that neighborhood risk affects the association between mother-adolescent relationship and number of sexual partners for males and females. Among males, mother-adolescent relationship was increasingly important in higher risk neighborhoods, but among females the effect of motheradolescent relationship was more relevant in lower risk neighborhoods.

The Effects of Neighborhood and Mother-Teen Relationships

207

A corollary of our study’s findings is that, when considering adolescent sexual behavior, it appears that the census tract level is as efficacious as the smaller block group aggregation. Indeed, for females, the census tract level appears to be more meaningful. Researchers have found it difficult to arrive at a meaningful conceptualization of neighborhood for a developing child. Garbarino et al. (1992) contend that neighborhoods consist of the territory or “turf” of children and that the impact of neighborhoods can only be understood if the definition of neighborhood is an accurate reflection of a child’s experience. Comparing tract versus county levels of data, Billy & Moore (1992) concluded that it is better to use smaller geographic units. Perhaps this is true when examining influences on children (or when comparing counties to tracts for adolescents), but this study suggests that the impact of neighborhoods on adolescent sexual behaviors, particularly for females, is best captured at the tract level rather than the block group level. By not considering characteristics of the neighborhood beyond the narrowly defined area of a few blocks, the census level of “block group” may exclude information about adolescents’ relevant “turf.” This does not mean that the tract level should be considered adolescents’ “relevant” level of social context in every case. It is likely that level of social context—block group, tract, or county—that is most important will vary accordingly to the outcome being examined, as well as the age of population being studied.

REFERENCES Baker, S. A., Thalberg, S. P., & Morrison, D. M. (1988). Parents’ behavioral norms as predictors of adolescent sexual activity and contraceptive use. Adolescence, 23, 265–282. Bearman, P. S., Jones, J., & Udry, J. R. (1998). The National Longitudinal Study of Adolescent Health: Research Design. Retrieved from http:/www.cpc.unc.edu/ projects/addhealth/design.html Billy, J. O., & Moore, D. E. (1992). A multilevel analysis of marital and nonmarital fertility in the U.S. Social Forces, 70, 977–1011. Billy, J. O., Wenzlow, A. T., & Grady, W. R. (1998). The National Longitudinal Study of Adolescent Health: Part 1, Codebooks for the Waves I and II Contextual Database. Retrieved from http://www.cpc.unc.edu/projects/addhealth/codebk1.html Bingham, C. R., Miller, B. C., & Adams, G. R. (1990). Correlates of age at first sexual intercourse in a national sample of young women. Journal of Adolescent Research, 5, 18–33. Brewster, K. L., Billy, J. O., & Grady, W. R. (1993). Social context and adolescent sexual behavior: The impact of community on the transition to sexual activity. Social Forces, 71, 713–740.

208

Single-Parent Families: The Lived Experience

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development. Developmental Psychology, 22, 723–743. Brooks-Gunn, J., Duncan, G. J., Klebanov, P. K., & Sealand, N. (1993). Do neighborhoods influence child and adolescent development? American Journal of Sociology, 99, 353–395. Burton, L., & Price-Spratlen, T. (1999). Through the eyes of children: An ethnographic perspective on neighborhoods and child development. In A. Masten (Ed.), Cultural processes in child development (pp. 77–96): Mahwah, NJ: Erlbaum. Carolina Population Center at the University of North Carolina at Chapel Hill. (1998). Wave I and Wave II Codebooks for the National Longitudinal Study of Adolescent Health. Retrieved from http://www.cpc.unc.edu/projects/ addhealth/codebk1.html. Case, A. C., & Katz, L. F. (1991). The company you keep: The effects of family and neighborhood on disadvantaged youths. Working Paper No. 3705. National Bureau of Economic Research. Christopher, F. S., Johnson, D. C., & Roosa, M. W. (1993). Family, individual, and social correlates of early Hispanic adolescent sexual expression. Journal of Sex Research, 30, 54–61. Chodorow, N. (1978). Mothering, object relations, and the female oedipal configuration. Feminist Studies, 4, 137–158. Collins, W. A., & Russell, G. (1991). Mother–child and father–child relationships in middle childhood and adolescence: A developmental analysis. Development Review, 11, 99–136. Crane, J. (1991). The epidemic theory of ghettos and neighborhood effects on dropping out and teenage childbearing. American Journal Sociology, 96, 1226–1259. Elliott, D. S., & Morse, B. A. (1989). Delinquency and drug use as risk factors in teenage sexual activity. Youth and Society, 21, 32–60. Fagot, B. (1994). Peer relations and the development of competence in boys and girls. New Directions in Child Development, 65, 53–65. 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. 231–258). Newbury Park, CA: Sage. Garbarino, J., Galambos, N. L., Plantz, M. C., & Kostelny, K. (1992). The territory of childhood. In J. Garbarino, J. (Ed.), Children and families in the social environment (pp. 201–229). New York: Aldine de Gruyter. Hogan, D. P., & Kitagawa, E. (1985). The impact of social status, family structure, and neighborhood on the fertility of black adolescents. American Journal of Sociology, 90, 825–836. Ku, L., Sonenstein, F. L., & Pleck, J. H. (1987). Neighborhood, family, and work: Influences on the premarital behaviors of adolescent males. Social Forces, 72, 479–503. Mayer, S. E., & Jencks, C. (1989). Growing up in poor neighborhoods: How much does it matter? Science, 243, 1441–1445. Miller, B. C., Benson, B., & Galbraith, K. G. (2001). Family relationships and adolescent pregnancy risk: A research synthesis. Developmental Review, 21, 138. Miller, B. C., Norton, M. C., Curtis, T., Hill, E. J., Schvaneveldt, P., & Young, M. H. (1997). The timing of sexual intercourse among adolescents: Family, peer, and other antecedents. Youth and Society, 29, 54–83.

The Effects of Neighborhood and Mother-Teen Relationships

209

Miller, B. C., Norton, M. C., Fan, X., & Christopherson, C. R. (1998). Pubertal development, parental communication, and sexual values in relation to adolescent sexual behaviors. Journal of Early Adolescence, 18, 27–52. Miller, B. C., & Olson, T. D. (1988). Sexual attitudes and behavior of high school students in relation to background and contextual factors. Journal of Sex Research, 24, 194–200. Nollen, P. (1980). Cross-gender effect in two-child families. Developmental Psychology, 16, 159–160. Quane, J. M., & Rankin, B. H. (1998). Neighborhood poverty, family characteristics, and commitment to mainstream goals: The case of African American adolescents in the inner city. Journal Family Issues, 19, 769–794. Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94, 774–802. Sawicki, D. S., & Flynn, P. (1996). Neighborhood indicators: A review of the literature and assessment of conceptual and methodological issues. Journal of American Planning, 62, 165–183. Scott-Jones, D., & White, A. B. (1990). Correlates of sexual activity in early adolescence. Journal of Early Adolescence, 10, 221–238. Simons, R. L., Johnson, C., Beamon, J., Conger, R. D., & Whitbeck, L. S. (1996). Parents and peer groups as mediators of the effect of community structure on adolescent problem behavior. American Journal of Community Psychology, 24, 145–171. Taylor, R., Gottfredson, S., & Brower, S. (1984). Block crime and fear: Defensible space, local and social ties, and territorial functioning. Journal of Research in Crime and Delinquency, 21, 303–331. Thornton, A., & Camburn, D. (1987). The influence of the family on premarital sexual attitudes and behavior. Demography, 24, 323–340. Vartanian, T. P. (1999). Neighborhood conditions and adult welfare use: Examining neighborhood and family conditions. Journal of Marriage and the Family, 61, 225–237. Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers, fathers, and friends. Chicago: University of Chicago Press. Wilson, W. J. (1987). The truly disadvantaged. Chicago: University of Chicago Press.

Chapter 14

Religious Involvement as a Source of Support? Ann Sorenson, Carl F. Grindstaff, and R. Jay Turner*

Pregnancy is said to be a challenge in any woman’s life, but for an adolescent it adds another level of complexity to an already complex period of physical and emotional change. The importance of social support during the teenager’s pregnancy to the health status of her infant and to her own emotional well-being in the first weeks after giving birth has been described in previously published research (Grindstaff & Turner, 1989; Turner et al., 1990). While social support specifically refers to support that is the product of stable human relationships, support based in cultural and normative systems may also provide a framework for successful adaptation and coping. One such cultural system is religion. In this study we consider the influence of religious orientation and behavior on the emotional health of very young mothers. Most research suggests that religiosity has a positive association with well-being and mental health. In a longitudinal study of adults, Williams et al. (1991) describe religious attendance as presenting a classic example of stress-buffering effects. At lower levels of religious attendance there is evidence that stress exposure is associated with psychological distress. However, as attendance increases, the adverse consequences of stress are reduced, leading the authors to conclude that “religion may be a potent coping strategy that facilitates adjustment to the stress of life” (p. 1261). Other studies have suggested that religious involvement has a positive effect on subjective well-being (Ellison, 1991) and is associated with lower levels of depressive symptomatology among the elderly (Idler, 1987); that

* A. M. Sorenson, C. F. Grindstaff, and R. J. Turner. Religious Involvement among Unmarried Adolescent Mothers: A Source of Emotional Support? Sociology of Religion, 1995, 56(1), 71–81. Reprinted with permission from the Association for the Sociology of Religion.

Religious Involvement as a Source of Support?

211

a strong religious orientation benefits the mental health of adult women; and that religiously active college women appear to be less prone to anxiety (Bridges & Spilka, 1992). A number of explanations of this positive influence have been advanced. Religion is thought to provide a sense of meaning or personal identity or to enhance self-efficacy in the face of stressful life events that might otherwise induce a sense of anomy. Idler suggests that religiosity affects depression by fostering a sense of optimism, reducing fatalism, and by altering perceptions of suffering. Religion may thus provide the individual with resources for explaining or even resolving problematic situations. These kinds of explanations provide a basis for hypothesizing that religion provides a framework for successful adaptation and coping and is a positive influence on the emotional well-being of young women who have experienced an adolescent pregnancy. A religious orientation may help sustain the individual in the face of stressful life events, or the positive influence may be more indirect, by way of the social support provided by religion-based networks. However, these positive predictions may not apply in all circumstances. As Masters & Bergin (1992) observe, it “seems important that one actually behaves in synchrony with one’s religious values in order for there to be beneficial mental health consequences” (p. 228). Since the circumstances of most teen pregnancies are directly at odds with traditional religious values, it is questionable whether positive effects can be expected. Considerable variation is typically observed in the mental health of individuals who have experienced what appear to be the same stressful conditions. This variation tends to be understood in terms of differences in available coping resources. However, Pearlin (1989) argues that part of the explanation may also lie in the importance of social values that “regulate the effects of an experience by regulating the meaning and the importance of the experience” (p. 249). That is, the “threat that people experience from the circumstances they face depends to some degree on the values they hold” (p. 249). Accordingly, religious values may exacerbate rather than moderate the stress associated with pregnancy for many adolescents, resulting in a negative influence on the emotional well-being of very young mothers. This paper presents data that evaluate the competing hypotheses of positive versus negative mental health effects of religious values within this high-risk population.

DATA AND MEASURES A longitudinal study of pregnant adolescents in southwestern Ontario provides the opportunity to study the mental health consequences of religious identification or involvement in this context. Subjects were drawn

212

Single-Parent Families: The Lived Experience

between 1984 and 1986, primarily from the caseloads of family physicians and obstetricians/gynecologists practicing in southwestern Ontario. To maintain confidentiality the physician initially described the study to every teenager, regardless of marital status, as she entered the obstetric caseload. If the young woman agreed to participate, her name was released to the study. Subjects were interviewed once as soon as possible after medical confirmation of pregnancy and again approximately 4 weeks following delivery. Of the 284 respondents who completed the first interview, 261 completed the second interview as well, providing the data for this study. The dependent variable in our study is the emotional well-being of very young mothers in the first weeks after their babies were born, as measured by the Center for Epidemiological Studies Depression Scale (Radloff, 1977). Depression is thought to be especially significant as an outcome variable for this group because of its potential effects on the young mother’s ability to form and maintain relationships, to perform effectively at work or in school, and to provide the attention and nurturance required by a young child. In our sample Cronbach’s alpha for the CES-D scale is .89. The independent variables were all measured within an earlier interview conducted during the course of pregnancy. Religious preference, participation in church-related activities, and self-described religiosity were used to index the respondent’s religious orientation. Religious preference was categorized as Protestant, Catholic, Other, or no particular faith based on Wave 1 interviews. Subsequent interview data allow us to infer that the Protestant category is primarily Anglican, United Church, or Presbyterian. The Other category is primarily Pentecostal, Jehovah’s Witnesses, or other Christian denominations. Religious preference provides information about two somewhat different aspects of religious orientation. The simple expression of a religious preference can be taken as a minimal indicator of the salience of religion, and while the four categories are quite broad, they may also reflect some denominational variation in the sanctions associated with adolescent pregnancy. All religious groups are likely to share the same general view that childbearing should be confined to marriage, but there may be subtle differences in the emphasis placed on particular teachings or in the religious community’s response to the violation of these norms. Participation in church-related activities is described by a measure of frequency with ordered categories ranging from “never (or almost never)” to “more than once a week.” Participating in religious activities could mean experiencing the potential of religion for providing support or it could mean being reminded of the values that proscribe the circumstances of most adolescent pregnancies. Respondents were also asked how religious they consider themselves to be, with response categories ranging from “not at all” to “very religious.” With no reference to organized religion, responses to this item

Religious Involvement as a Source of Support?

213

could reflect rather idiosyncratic definitions of what being religious means. It is not clear that thinking of oneself as religious necessarily implied to our respondents any particular level of exposure or adherence to a coherent body of religious teachings. Several additional variables were selected based on their common association with religious factors and the emotional well-being of adolescents. For example, the likelihood of experiencing parental separation or divorce may be related to the respondent’s religious background. Adolescents from single-parent families have been shown to have higher levels of depressive symptoms than those living with two parents (Avison & McAlpine, 1992). In this study family composition is described by a dichotomous variable contrasting those who grew up in a two-parent family and those who did not. Socioeconomic status in the family of origin is assessed by parental educational attainment. A positive association is generally reported between socioeconomic status and church attendance (Greeley, 1972), and Gore et al. (1992) have observed a negative association between parental education and depression among girls. The young woman’s own marital status may also be linked to her religious background and to emotional well-being. Even for an adolescent, being married and therefore starting her family in a context that is consistent with religious norms and social expectations may be less stressful than a nonmarital pregnancy. The perception that one is cared for and supported in family relationships is measured by the Provisions of Social Relations Scale (PSR) (Turner et al., 1983). Cronbach’s alpha for the PSR scale is greater than .90. The importance of this kind of support has been demonstrated in a number of studies of depression among adolescents (Avison & McAlpine, 1992; Gore et al., 1992). This variable is relevant to our analysis of religious norms to the extent that such norms encourage supportive family relationships. These additional variables all relate to a positive influence on emotional well-being of religious values. But we have also suggested a second hypothesis that describes increased psychic costs as the values implicit in most religious teachings conflict with the circumstances of most adolescent pregnancies. The results presented below bear compellingly upon the relative plausibility of these alternative hypotheses. RESULTS Religious Involvement, Adolescent Pregnancy, and Depression Young mothers with no particular faith (and Protestants) appear to have experienced the lowest levels of depression. Participating in religious activities during pregnancy is actually associated with greater

214

Single-Parent Families: The Lived Experience

distress, with the clearest distinction between those who almost never or never attended and those who did. CES-D scores also appear to increase somewhat with the self-reported measure of religiosity. These data offer no suggestion of a positive religious influence on emotional well-being. On the contrary, there is modest evidence that religious involvement may impede the emotional adjustment of very young mothers. Whether religious factors operate directly by way of personal faith, indirectly by way of social relationships, or even through a common association with other background variables, a positive association with emotional well-being should be apparent regardless of marital status. However, if we review the logic of the second hypothesis, there is no reason to expect religious values to pose a threat to the young mothers who experienced their pregnancies within the context of marriage. If religious factors are associated with greater distress, this relationship should only be observed among unmarried women. The lower CES-D scores of the 33 married adolescents in our sample suggest that their experience was quite different than that of the other teens. The average depression score for the married respondents was only 11.0, compared to 17.32 for those living in unmarried relationships and 16.61 for single (but not cohabiting) adolescents. Furthermore, there is no pattern suggesting a negative influence of religion on the emotional wellbeing of the married teens. This is consistent with the second hypothesis, which describes conflict between behavior and religious values. Therefore on theoretical and empirical grounds the remainder of the analysis is limited to the 228 teenagers who were not married when their pregnancies were confirmed. The average depression scores of women in the more conservative Catholic and Other categories are significantly higher than the scores of Protestants and those with no particular faith (t  2.685, p  .008). Teenagers who had attended religious activities reported higher levels of depression (t  2.796, p  .006). However, there is no significant effect of self-reported religiosity on average CES-D scores. Aspects of institutional religious practice appear to be more relevant to depression than level of personal faith as we have measured it. The association between religious preference and attendance precludes any simple description of how either variable influences the emotional well-being of young unmarried mothers. Nearly 75 percent of those in the Catholic and Other categories attended church-related activities during their pregnancies, compared to only half the Protestants and fewer than 20 percent of those with no particular faith, making it unclear whether differences in depressed symptoms are due to identification with the more conservative Catholic or Other religious groups, involvement in religious activities, or some combination of both factors.

Religious Involvement as a Source of Support?

215

Because there were no significant differences observed in their depression scores or levels of religious participation, the Catholic and Other categories of religious preference were combined in a single dummy variable, CATH&OTH, scored 1 for Catholics and Others and 0 otherwise. PROTESTANT was scored 1 for Protestants and 0 otherwise. Those with no particular faith made up the omitted category. ATTEND was scored 1 for those who participated in church-related activities during their pregnancies and 0 for those who did not. The regression models confirm the previously described effects that suggest lower depression scores for Protestants and those with no particular faith and higher depression scores associated with having attended religious activities. The third model suggests that both religious preference and involvement have independent effects on CES-D scores. However, if Protestant norms and sanctions are somewhat less demanding than those of the more conservative denominations, the main effects model may not be the most appropriate. To address the possibility that religious involvement was less detrimental to Protestant teens, an interaction term, ATTEND X PROTESTANT, was computed to describe how the effect of attendance is modified if the respondent is Protestant. In the presence of the interaction term, the main effects of CATH&OTH and PROTESTANT describe the influence of religious preference among those who did not actually participate in religious activities. Likewise, the main effect of ATTEND describes the effect of religious participation among those who are not Protestant. The regression coefficients reported for Model 4 suggest an interesting pattern of effects that are consistent with the second hypothesis. Only the main effect of ATTEND and the interaction term, ATTEND X PROTESTANT, are statistically significant. For those who were not actually involved in religious activities, religious preference has no apparent influence on emotional well-being. As we shall see below, higher levels of depression are associated with religious involvement, but only among the non-Protestants. Model 5 is trimmed of the non-significant variable, CATH&OTH. The variable PROTESTANT is also nonsignificant but is retained because it contributes to the interaction term. Under the terms of Model 5, religious involvement is associated with a 7.07 point increase in depression scores, but the regression coefficient estimated for ATTEND X PROTESTANT reduces that effect by 6.60 points. This means that among the nonProtestants, CES-D scores increased an average of 7.07 points if the respondent was involved in church-related activities. Among Protestants that increase is only a slight fraction (.47) of a point. To the extent that religious factors influence the emotional adjustment of young unmarried mothers, that influence seems largely confined to the non-Protestants who were involved in religious activities during their pregnancies.

216

Single-Parent Families: The Lived Experience

Religious Factors, Background Characteristics, and Social Support In this section the common relationships of the background and family support variables to religious factors and depression are considered. Religious factors were thought to be associated with a greater likelihood of growing up in a two-parent family, higher socioeconomic status, and social support, but only one such relationship was observed in our data. Those who were involved in religious activities were more likely to have grown up in a two-parent family. However, neither a two-parent family nor higher socioeconomic status appear to have influenced emotional well-being in this sample of very young mothers. Only family support seems to have had any influence, and this variable was unrelated to religious preference, attendance, or religiosity. The respondent’s own marital status during her pregnancy was also included among the background variables, but recall that this stage of the analysis was limited to unmarried women, some of whom (nearly 20 percent) were living with a partner. If there were a religious influence on living arrangements, it would seem to discourage cohabitation, but in this sample we observed no significant relationship between religious preference or attendance and cohabitation. There was no evidence of a main effect of cohabiting on depressive symptomatology. This summary suggests that there is no significant relationship between any of the sociodemographic variables and depression scores. However, this conclusion was based on the simple bivariate relationships that are described by correlation coefficients. It remains possible that religious factors interact with background characteristics to affect the emotional well-being of these young women, but an analysis that includes all possible interactions in the absence of theory seems ill-advised.

Religious Involvement, Cohabiting, and Depression We have reason to predict an interaction between religiosity and only one background characteristic. Extending the logic of the second hypothesis, one could predict higher depression scores among those who cohabited while remaining involved with their religious faith. This particular interaction has theoretical significance in that it provides corroborative evidence relevant to the hypothesis that behavior in violation of religious norms adds to the psychological distress of respondents who remain involved in their religious faith. The conflict between living in an unmarried relationship and religious values may be more salient to respondents who are regularly reminded of those values by attending services, or their living arrangements may become a source of distress as they are witnessed (and possibly sanctioned) by fellow participants in the religious community.

Religious Involvement as a Source of Support?

217

Model 6 adds the main effect of cohabiting (COHABIT) and an interaction term describing the effect of cohabiting among those who attend services (COHABIT X ATTEND) to the variables included in Model 5. Under the terms of Model 6 the effect of COHABIT X ATTEND is statistically significant and positive. We would expect an additional 7.49 point increase in the average CES-D scores of women who lived in an unmarried relationship while continuing to attend religious activities. Among those who did not participate in religious activities, there is no effect of cohabiting on CES-D scores. It should be noted that the inclusion of COHABIT and COHABIT X ATTEND results in no substantial change in the combined effects of religious preference and involvement as they were described by Model 5.

CONCLUSIONS We have proposed two rather contradictory models of how religion could influence the emotional well-being of teenagers who have experienced a pregnancy. A religious orientation may provide direct or indirect support for a young woman in her childbearing experience, but at the same time a religious commitment may represent significant psychic costs. That religion represents a potential source of cultural or institutional support or provides a context for social support is demonstrated by the large number of studies that report a positive association between religious involvement and well-being. However, the data presented here provide no evidence that religious factors enhance emotional well-being in this sample of very young mothers. The second model of religious influence suggests that behavior that is directly at odds with traditional religious norms represents a significant threat to some young women. In his review of the sociological study of stress, Pearlin (1989) suggests that certain values, when combined with certain circumstances, are especially productive of stress. The tension between religious values and an unmarried pregnancy seems to represent that “stressful mix of circumstances and values” (p. 249) for young women with a significant religious orientation. The pattern of distress that is apparent shortly after these young women gave birth is consistent with this second hypothesis. Affiliation with denominational groups that are likely to have more conservative or clearly enunciated views is associated with elevated levels of distress, but only among those who took part in religious activities. There is no apparent influence of these religious norms among women who were not involved, as measured by attendance, with organized religion. Nonmartial fertility places our entire sample at risk of conflict with traditional religious values pertaining to reproduction and the family, but in

218

Single-Parent Families: The Lived Experience

addition to experiencing an unmarried pregnancy, some young women were also cohabiting. For young women who have retained a religious commitment, such behavior seems to represent a doubly stressful mix of circumstances and values. The even greater distress experienced by young women who lived in an unmarried relationship while at the same time taking part in religious activities is a clear replication of the pattern of religious influence associated with nonmarital fertility. While this does not provide a direct test of our explanation of how religious factors are associated with diminished well-being, these parallel findings would seem to support the notion that similar circumstances may be differently stressful, depending on the values that individuals hold. It appears that traditional religious organizations cannot claim a supportive role in the adaptation of unmarried teenage mothers in the period shortly after their babies were born. This is perhaps ironic given that the decision made by these young women to carry their pregnancies to term is consistent with the prolife position held by many of these same groups. However, much of the concern around adolescent pregnancy is for the long-term economic, social, and psychological well-being of very young mothers and their children. The influence of religious factors on measures of long-term adjustment may be very different, especially if religious affiliation and commitment are part of a framework that places young unmarried mothers in a situation that gives them access to a larger network of positive social relationships and that provides opportunities that encourage positive choices for themselves and their children.

REFERENCES Avison, W. R., & McAlpine, D. (1992). Gender differences in symptoms of depression among adolescents. Journal of Health and Social Behavior, 33, 77–96. Bridges, R. A., & Spilka, B. (1992). Religion and the mental health of women. In J. F. Schumaker (Ed.), Religion and mental health (pp. 43–53). Oxford: Oxford University Press. Ellison, C. (1991). Religious involvement and subjective well-being. Journal of Health and Social Behavior, 32, 80–99. Gore, S., Aseltine, Jr., R. H., & Colton, M. E. (1992). Social structure, life stress and depressive symptoms in a high school-aged population. Journal of Health and Social Behavior, 33, 97–113. Greeley, A. M. (1972). The denominational society. Glenview, IL: Scott, Foresman, & Company. Grindstaff, C. F., & Turner, R. J. (1989). Structural factors associated with birth complications in adolescent fertility. Canadian Journal of Public Health, 80, 214–220. Idler, E. (1987). Religious involvement and the health of the elderly. Social Forces, 66, 226–38.

Religious Involvement as a Source of Support?

219

Masters, K. S., & Bergin, A. E. (1992). Religious orientation and mental health. In J. Schumaker (Ed.), Religion and mental health (pp. 221–232). Oxford: Oxford University Press. Pearlin, L. (1989). The sociological study of stress. Journal of Health and Social Behavior, 30, 241–256. Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. Turner, R. J., Grindstaff, C. F., & Phillips, N. (1990). Social support and outcome in teenage pregnancy. Journal of Health and Social Behavior, 31, 43–57. Williams, D. R., Larson, D. B., Buchler, R. E., & Pyle, C. M. (1991). Religion and psychological distress in a community sample. Social Science and Medicine, 32, 1257–1262.

This page intentionally left blank

Part Four

Social Problems and Policy

Family changes over the past decades have led some to conclude that there is a decline in “family values” and that single-parent families are the root cause of many of our worst social problems. Alternative family structures such as mother-headed, father-headed, and stepfamilies have often been viewed as detrimental to the development of children and especially their educational attainment and employment success. Indeed, many have claimed that fatherless households are a recipe for disaster in the future as children raised without a father (most notably boys) would not have adequate economic support, and would lack a male role model and structure, thus leaving these children vulnerable. As noted earlier, single-parent families in their diverse forms appear to have become a permanent fixture on the landscape of American family life. The articles in this section explore family structure as a social problem as well as social policies needed to assist single-parent families. In what ways are these policy areas effective in assisting singleparent families? In what ways are these policy areas ineffective in meeting the needs of single parents? What should be the future direction of policy makers and community leaders with regard to single-parent families? There have indeed been vast discrepancies in the research regarding alternative family types and the effects on the child’s development, educational attainment, and employment opportunities. Biblarz and Raftery explore the concept of “pathology of matriarchy,” noting that the ideas embedded in this concept extend into many research and policy efforts regarding families with both supportive and nonsupportive findings. The

222

Social Problems and Policy

Pathology of Matriarchy view is captured in these ideas and extends into many research and policy efforts. Two bodies of research exist regarding the impact of family structure on the well-being of families. One body of research suggests that family structure (i.e., two-parent, single-parent, etc.) significantly affects families, whereas the second body of research has found that family processes— the quality of interaction—are more important to family well-being than family structure. Regardless of single-parent type of family, one area of improvement needed is child care. Child care has been, and continues to be, a very important policy issue as more women and men raise children alone or are employed and in need of child care. Single mothers’ employment has grown significantly over the past several decades, and the need for child care has grown also. Bainbridge, Meyers, and Waldfogel investigate how the expansion of child care subsidies influences the employment of single mothers. The impact of demographics and economics on child support and even policy surrounding this important area of single-parent families most often involves five interrelated factors that combine to produce a general lack of progress in this area. Case, Lin, and McLanahan assert that inflation, unilateral divorce, demographic composition, women’s economic dependence, and ineffective child support policy all contribute to the slow pace and ineffective progress of focusing on the well-being of children in these families. Finally, child support policy is addressed from a different point of view as an extremely important benefit to the well-being of single-parent families—especially low-income families. Waller and Plotnick delve into the issue of child-support reform and why it seems to not meet the needs of low-income families. Policy implications are discussed from both formal and informal levels. In many instances, informal arrangements are followed and formal regulations are seen as impeding the welfare of the children involved. This tension between formal court-ordered, agency policy or other more formal forms of policy are not only often perceived as unreasonable, but also, indeed often obstruct the more positive intent of public policy in the child support area, which has continued to grow at a more rapid rate.

Chapter 15

Rethinking the “Pathology of Matriarchy”: Family Structure, Educational Attainment, and Socioeconomic Success Timothy J. Biblarz and Adrian E. Raftery*

The effect of alternative family structures on children’s educational and occupational success has been constant over the past 30 years. Higher rates of unemployment and lower-status occupational positions could account for the negative effect of single-mother families on children’s attainment throughout the period. Children from single-father families and stepfamilies have consistently had lower attainments than children from both two-biological-parent and single-mother families. The influence of many other dimensions of children’s family background declined from the 1960s to the 1980s but has not declined since. Among six candidate theoretical frameworks, the findings are most consistent with an evolutionary view of parental investment. The “pathology of matriarchy” hypothesis that came out of the Moynihan Report (1965) is that the absence of a father is destructive to children, particularly boys, because it means that children will lack the economic resources, role model, discipline, structure, and guidance that a father provides. Moynihan (1965) focused on the African American family, but the publication came on the eve of what was to be 15 years (1965–1980) of sustained, rapid increase in the divorce rate of non-Latino whites. Some researchers have carried the “pathology of matriarchy” view beyond the African American family to the larger population.

* T. J. Biblarz and A. T. Raftery. Family Structure, Educational Attainment, and Socioeconomic Success: Rethinking the “Pathology of Matriarchy.” American Journal of Sociology, 1999, 105(2), 321–365. Reprinted with permission of the University of Chicago Press.

224

Social Problems and Policy

Popenoe (1996, p. 8), echoing Moynihan, argues that “if we continue down the path of fatherlessness, we are headed for social disaster.” Social science research has produced evidence both for and against the “pathology of matriarchy” view. Some studies using national samples show that children from single-mother families have lower attainments than children from two-biological-parent homes (Duncan & Duncan, 1969; McLanahan & Sandefur, 1994), though other studies, also using national samples, show that once other factors are taken into account, children from single-mother families do approximately as well as children from two-biological-parent families (Biblarz, Raftery, & Bucur, 1997; McLanahan, 1985). Some studies show that alternative family types— single-mother, single-father, and stepfamilies, for example—have similar, negative consequences for children (Dawson, 1991); while other studies show that children from some kinds of nontraditional families have higher attainments, on average, than children from other kinds (Amato & Keith, 1991b). This article is about searching for order in this diversity of findings. We assess whether change over time in the effects of alternative families, and differences in researchers’ decisions about which independent variables to include and leave out of models, can account for the discrepancies observed in the literature. We do this by tracking the relationship between alternative families and children’s educational and occupational success over four decades—the 1960s, 1970s, 1980s, and 1990s—using four large nationally representative surveys. For each time period, we observe how the family structure/child outcome relationship changes depending on the other dimensions of respondents’ family backgrounds that are taken into account. We find that the effect of family structure on children’s socioeconomic success has been constant over 30 years. In the 1960s and 1970s, the high rate of unemployment among single mothers could explain the negative effect of single-mother families on children’s educational and occupational attainment. In the 1980s and 1990s, single mothers’ low-status occupational positions—rather than their employment/unemployment status per se—could account for the negative effect of female headship. There were no significant differences by gender in the effects of alternative families—alternative families had the same effect on men and women. The influence of many other dimensions of children’s family background declined from the 1960s to the 1980s, but this decline did not progress further between the 1980s and the 1990s. The discrepancies found in the literature about the consequences of alternative families for children are due in part to different decisions about which variables are exogenous to the process and in part to different decisions about how to group alternative families. With or without various sets of controls, we find that over the past 30 years, children from

Rethinking the “Pathology of Matriarchy”

225

single-father families, father-stepmother families, and mother-stepfather families have consistently had lower attainments than children from both two-biological-parent families and single-mother-headed families. Among six candidate theoretical frameworks, the findings are most consistent with an evolutionary view of parental investment. BACKGROUND Why Do Children from Alternative Families Have Lower Attainments? Below we review the main theories of the effects of family structure on children and discuss the predictions made by each. Sociological theory—Almost all existing theory about the consequences of family structure for children centers around the relationship between family type and resources. Under the general rubrics of “social structure and personality,” or “social structure and psychological well-being,” sociological theory—socialization, learning, and control theory—predicts that children from alternative families get fewer economic, social, and cultural resources, which help facilitate success. Socialization theory emphasizes the essential role of parenting in shaping children’s lives (Baumrind, 1978, 1980; Parcel & Menaghan, 1994). In the case of single-mother families, father absence reduces the family’s ability to provide optimal amounts of support and control to children (Astone & McLanahan, 1991; Thomson, Hanson, & McLanahan, 1994). The emotionally distressing event of a spouse’s death or of divorce, coupled with responsibility overload, negatively impacts women’s psychological well-being (Acock & Demo, 1994; Crosby, 1987; Simons & Associates, 1996). This leads to inconsistent parenting (Hetherington, Cox, & Cox, 1978), less supervision over children (Thomson, McLanahan, & Curtin, 1992), parental authoritarianism (Bronfenbrenner, 1979), and an expectation that children mature in ways inappropriate for their age (Weinstein & Thornton, 1989). All of these undermine the healthy development of children. Learning theory views the family as a primary site where children learn about how to get along in the society when they reach adulthood (Kohn, 1969, 1983). Without a father, children will lack a male model of how to successfully achieve in market activity (Powell & Parcel, 1997; McLanahan & Sandefur, 1994). In two-biological-parent families, children learn about how authority relations are structured and how to successfully interact with authority figures (Nock, 1988). This learning facilitates children’s educational and occupational attainment. In father-absent homes, where mother-child relations run the risk of becoming more peerlike, and in stepfamilies, where stepparent-child relations may be defined more as friends than as parent-child, children will not learn these important skills.

226

Social Problems and Policy

At the same time that alternative families represent for children the removal of positive resources, they also co-vary with children’s exposure to negative ones. Events that produce alternative family structures often involve a loss or trauma—either the death of a father (or less frequently, of a mother) or the divorce of parents followed by the father (most often) moving out of the home. These events are stressful in the short term and have the potential to impact children’s long-term life course trajectories in negative ways (Amato & Booth, 1997; Glenn & Kramer, 1987; Mueller & Pope, 1977; Wallerstein, 1989). The number of changes in family configuration over the course of childhood is an important predictor of some outcomes, like the risk of having a child outside of marriage (Wu & Martinson, 1993). For other outcomes, such as children’s educational and occupational attainment, exposure to alternative family structures (even from birth) seems to be more important than the number of disruptive family events experienced or durations spent in particular kinds of families (McLanahan & Sandefur, 1994; Wojtkiewicz, 1993). Children raised in single-parent families from birth, for example, have roughly the same high school dropout rate as those who experienced a transition from a mother-father to a single-parent family following parental divorce. Problems in child socialization and parental control will occur in a variety of types of alternative families that children may experience—singleparent families, stepfamilies, and so on. In the case of single-parent families, for example, the undermining of parental control is a structural consequence of the absence of the father from the residential home. Stepparents, on the other hand, have “only a limited license to parent” (Furstenberg & Cherlin, 1991, p. 85), and the arrival of a stepparent into the residential home can create disruption and friction in intergenerational relations that can take years to resolve and adjust to (Cherlin & Furstenberg, 1994). The main prediction of sociological perspectives is that the two-biological-parent family is generally the optimal form for the successful socialization of children in modern society and that children from any kind of alternative family will, on average, do less well. Economic theory—Economic theory proposes that socioeconomic success is partly a function of human capital (Becker, 1964, 1981; Becker & Tomes, 1986). Households act as singular units to maximize collective utility. Utility comes from commodities—like children—that are produced by investments in market activity and household services. The two-parent family is among the best-functioning forms in modern capitalist society because it allows for the provision of household services by one partner and economic resources (or market goods) by the other. This is a particularly efficient system for maximizing utility and, by extension, the human capital of children. Since children’s success depends on the economic resources and equivalent services that parents provide, children who spend most of

Rethinking the “Pathology of Matriarchy”

227

their childhoods in a two-parent family (biological or stepfamily) will have the highest attainments because two parental figures are present to provide complementary resources. Single-parent families will yield less income from the market and have less time for the provision of household services. One parent cannot cover both market and nonmarket activities as successfully as two, and children from single-parent families, accordingly, will do less well. Among children from single-parent families, economic theory would predict that children from single-father families will do better than those from single-mother families because they will carry a substantial income advantage. Children who grow up in singlemother families will have the lowest attainments—mother-headed families average less than one-third the income of two-parent families and about half the income of alternative father-headed families (Meyer & Garasky, 1993; DaVanzo & Rahman, 1993). Evolutionary psychology—The evolutionary perspective on the family (e.g., Emlen, 1997) gives more weight to the role of the mother than that of the father in determining children’s fates, and it places special importance on biological relationships. The evolutionary view starts with the premise that mothers invest more of their resources in children than fathers. The survival (or, perhaps, in the modern context, the “well-being”) of a given child is of greater interest to the mother than to the father, because more of the mother’s than of the father’s potential reproductive investment is tied up in any one child (Trivers, 1972). Both parents attempt to balance investing in present children against investing in having additional children in ways that maximize their reproductive fitness. But because women’s potential for having additional children is far lower than men’s, they have a greater interest in making sure that the children they do have do well. Evolutionary psychologists sometimes depict motherhood, in part, as a strategic exercise in finding ways to secure material resources from sometimes reluctant fathers, whose reproductive calculus may be pulling them toward future children (and partners) more than present ones. Like the others, this theory would predict that children from twobiological-parent families will have an advantage over those from other kinds of families. The father’s average resource contribution to children will be less than the mother’s, but not by much because humans have high male parental investment, and so children will benefit from the presence of the biological father. But in contrast to the economic model, for example, the evolutionary view predicts that children from alternative families will do better raised by a single mother than a single father. Children from single-mother families will also have advantages over those from stepfather-biological mother families. The stepparent’s concern with his own reproductive fitness is in competition with the stepchildren for the mother’s resources, increasing the risk of abuse to children in families with a stepparent (Daly & Wilson, 1996).

228

Social Problems and Policy

Selection bias (parental competence)—One of the unanswered questions in family structure research is whether the observed negative effect of alternative families on children represents a selection effect. One variant of the argument is that people who divorce, for example, are less stable or less competent at family life. Children who experience their parents’ divorce do less well because their parents are less competent, not because of the divorce per se. Cherlin et al. (1991) found using longitudinal data that many child behavioral problems associated with divorce were actually present in the children prior to their parents’ divorce. The divorce, like the negative child outcomes, may have been a consequence of some preexisting family dysfunction (but see Cherlin, Chase-Lansdale, & McRae, 1998). Parental competence involves role performance as parent in the family, but it may also involve value orientations (Popenoe, 1993, 1996). Modern parents, particularly those who divorce, may be less “child-centered” and hold weaker “family values” than those who do not divorce (Popenoe, 1993). Alternative value systems may be another, related source of selection, causing a spurious relationship between family structure and children’s attainment. Marital conflict—Another variant of the selection hypothesis is that the main detrimental effect on children is not divorce but family conflict. Divorce is often preceded by (and sometimes followed by) high levels of conflict. Marital conflict is hurtful to children. Children of divorce have lower attainments than children from two-parent families because they have had sustained exposure to their parents’ discord (Amato & Booth, 1997; Amato, Loomis, & Booth, 1995; Glenn & Kramer, 1987; Mueller & Pope 1977). Both parental death and divorce create sadness, distress, and related problems for children in the short term. Negative effects on children’s health, self-esteem, and school performance have been observed (Dawson, 1991; Mauldon, 1990; Kline, Johnston, & Tschann, 1991; Mott, Kowaleski-Jones, & Menaghan, 1997). In the long term, children from alternative families have lower average socioeconomic achievements—in education, occupation, and earnings—than those raised in twobiological-parent families (Duncan & Duncan, 1969; McLanahan, 1985; Amato & Keith 1991b; Biblarz & Raftery, 1993; McLanahan & Sandefur, 1994; Powell & Parcel, 1997; Amato & Booth, 1997). McLanahan and Sandefur (1994) present evidence of negative effects of alternative families on selected child outcomes across four national surveys. Some scholars believe that this combination of reasonable theory and strong supporting evidence points to a scientific truth about the family— that the stable “intact” family remains the best-functioning form, at least for modern capitalist society. Glenn (1994) asserts that a plethora of evidence has led to “virtually unanimous” agreement among the best

Rethinking the “Pathology of Matriarchy”

229

social scientists that alternative families are not in the best interests of children (but see Stacey, 1996). Other findings and some theory give cause for uncertainty about the new consensus. In the classic 1973 Occupational Changes in a Generation (OCG II) survey, the negative impact of mother-headed households on sons’ occupational attainment is small and entirely a function of women’s disadvantaged employment and occupational positions (Biblarz et al., 1997). Powell and Parcel (1997) similarly find little adverse consequence of an alternative family structure for men’s education, occupation, and earnings in the Panel Study of Income Dynamics (PSID), although they do find adverse consequences for women. McLanahan’s (1985) earlier analysis of the PSID showed that father absence had no significant effect on children’s education once income is taken into account. Boggess (1998), also using the PSID, finds no effect of living with a single mother on children’s likelihood of graduating from high school. McLanahan (1985, p. 898) concluded that her results “do not support the notion that the long-term absence of a male role model itself is the major factor underlying family structure effects.” In the National Education Longitudinal Survey, holding constant other factors, there are no differences between children from two-biological-parent homes and those from femaleheaded families in the odds of dropping out of high school or attending college (Painter, 1998). Among the six family types included in Teachman, Paasch, and Carver (1997), “divorced mother” seems to be the only type to not directly increase children’s odds of dropping out of high school, holding other factors constant. Why the Discrepancy in Findings? Both the findings and conflicting nonfindings about the effects of family structure on children are based on large, national random samples, where results should be approximately the same. So why the discrepancies? One possibility is that the choice of exogenous “control” variables (e.g., race, gender, sibship size, parents’ education, residence) as well as “intervening variables” (e.g., family head’s income, employment status, occupational position, sibship size) varies across studies. Conclusions about family structure’s effect may depend on the other variables that are taken into account. For example, on average, the greater the number of siblings, the lower the attainments by children (Blake, 1989; Powell & Steelman, 1993; Steelman & Powell, 1989). If children from single-mother homes have fewer siblings than children from two-parent families, this would represent an advantage associated with the single-mother family structure. Studies that take away this advantage by controlling for number of siblings will show a stronger negative effect of single motherhood.

230

Social Problems and Policy

Studies that do not control for siblings will show a weaker effect. Conversely, the higher the socioeconomic position of parents, the greater the socioeconomic attainment of children (Blau & Duncan, 1967; Featherman & Hauser, 1978). Single mothers have lower socioeconomic positions than the fathers (and some mothers) who head two-parent families. Studies that do not take parent’s socioeconomic position into account will show a stronger effect of family structure. A second possibility is that the effect of family structure has changed over time (the change predictions of each theory are detailed below). Findings are not so much discrepant as simply pertinent to different periods. The evidence from national surveys spans at least 25 years, from about 1962 (e.g., Duncan, 1967) to 1987–1988, in the case of the first wave of the National Survey of Families and Households (NSFH). Among the cohorts represented over this time period, the main reason for families not to be two-biological-parents has shifted from the death of a parent to divorce (Bumpass & Sweet, 1989). Because most of these surveys do not contain information on cause of alternative family structure, we calculated the cause distributions for each birth cohort in the 1973–1996 General Social Surveys (GSS) and then applied the distributions to the samples listed in table 2, weighting by the number of respondents in each birth cohort (the oldest cohorts in the 1962 OCG survey were assigned the averages for the grouped “pre-1930 birth cohort” of the GSS). Over the past 30 years, divorce has come to replace death of a parent among samples of adults from alternative family backgrounds. The replacement of cause across time is striking: In 1962, 68 percent of adults from alternative family backgrounds are estimated to have experienced the death of a parent and 28 percent experienced the divorce of their parents, compared with only 33 percent experiencing the death of a parent and 62 percent experiencing parental divorce in 1992–1994. Some of the theories discussed above would predict that this shift in the cause structure accompanying cohort replacement should lead to an increase in the magnitude of the negative effect of alternative families over time (from the 1960s to the 1990s), while others would predict essentially no change. For the sociological theories, the keystone variable is family structure. Heightened risk of parental authoritarianism or neglect and inadequate models of authority relations are structural consequences of the absence of one parent, no matter what process gave rise to that absence. There should be few differences, for example, in outcomes between children from widowed-single-mother families and those from divorcedsingle-mother families because they share the same basic family structure. The sociological theories that emphasize structure accordingly would predict no change over time in the effect of single-parent and other kinds of alternative family structures, even as the process giving rise to the

Rethinking the “Pathology of Matriarchy”

231

alternative structures shifts. Other sociological perspectives—like marital conflict—that focus more on family process than on family structure would make the opposite prediction (see below). Economic and evolutionary theory also predict no change. From the evolutionary perspective, divorced and widowed single mothers have the same level of their own fitness tied up in the children, and so both types of mothers would have the same level of impetus to invest highly in their children. The presence of a nonbiological parent would negatively impact children, regardless of whether the biological father had died or the parents had divorced. The change in cause structure over time should not alter the implications for children of basic family forms. Economic theory focuses on household structure and composition. It claims, for example, that two-parent households are more efficient at maximizing utility than one-parent households and so would also predict no change over time in the effect of alternative family structures. These predictions of “no change” over time are based somewhat on the assumption “other things being equal.” Some evidence, for example, though not conclusive, shows that widowed mothers and their children have greater access than divorced mothers to certain kinds of social supports and may enjoy advantaged socioeconomic positions (McLanahan, Garfinkel, & Ooms, 1987; Acock & Keicolt, 1989; Amato & Keith, 1991a, 1991b; Holden & Smock, 1991; Biblarz & Gottainer, 1999; Sugarman, 1993, 1995). Economic theory would generally favor single-parent families with greater economic resources over those with fewer economic resources in predicting children’s human capital and subsequent attainment. But, independent of resources, economic theory makes no prediction about difference in outcomes between groups from single-parent families produced by the death of a parent vis à vis divorce, nor does evolutionary theory (although for evolutionary theory it makes an important difference whether the single parent is the biological mother or father). In contrast, the selection bias (or parental competence) perspective predicts that the magnitude of the negative effect of alternative families is growing over time, as the old cohorts who experienced a parent’s death as children leave our samples, replaced by the new cohorts who experienced their parents’ divorce. Widowed mothers would be more competent at family life than divorcees, on average, and so the children from alternative families in the earlier period should have done substantially better than those in the later period. Widowed mothers, who did not choose an alternative family structure for themselves and their children, will also have more traditional values and lifestyles than divorced parents, who did. These kinds of values will be positively functional for children’s success in society, more so than the alternative values of divorcees. Marital conflict theory makes the same prediction—the negative effect of alternative families is growing over time. Children from

232

Social Problems and Policy

alternative families in the earlier period would be expected to do substantially better than those in the later period because they will have been less likely, on average, to have had exposure to parental conflict. Children of divorce often have mixed and sometimes hostile feelings toward their fathers, for example, whereas children whose fathers died tend to develop a warm and positive inner construction of the deceased father (see, e.g., Rozondal’s [1983] “halo” effect; see also Silverman, Nickman, & Worden, 1992). Social Security survivor’s benefit payments to single widowed mothers and their dependent children tend to be substantially higher than AFDC payments to other kinds of single mothers and, unlike AFDC, they carry no requirement of an assets test to collect benefits (Sugarman, 1993, 1995). To the extent that widows with dependent children have, on average, more economic resources than their divorced and never-married counterparts, the replacement of widows with divorced and never-married mothers over time could produce a growing negative effect of the singlemother family type. Measures like the family head’s employment status and occupational position—considered later in the article—will only partly take into account the differences in material resources available across these disparate family types. If the predictions of the selection hypotheses (parental competence, marital conflict) are sustained, such a pattern of increase in the negative effect of alternative families over time would be particularly important in light of changes that have been occurring over time in the effects of other dimensions of children’s family backgrounds. A transition from ascription to achievement, or a “march toward meritocracy,” has been occurring in the United States, particularly since the 1960s. Featherman and Hauser (1978) find a modest decline in the twentieth century in the strength of the relationship between the family’s socioeconomic position and children’s socioeconomic destinations. This decline became clearly evident between 1960 and 1970 and steepened from 1970 onward (Featherman & Hauser, 1978; Hout,1984b, 1988; Grusky & DiPrete, 1990; DiPrete & Grusky, 1990). The effect of race on socioeconomic attainment has also declined substantially (Hout, 1984a), particularly at higher education levels (Hout, 1988). If the importance of family structure as a source of ascription has increased as that of the old sources has waned, the offset could produce a halt in the trend toward universalism. If the family’s socioeconomic position is making a comeback as a source of ascription in the 1990s, as some recent evidence suggests (Hout, 1997), this, coupled with a potentially growing family structure effect, could lead to a reversal of course—a trend toward growing inequality in the opportunity structure. Interactions research on the consequences of family structure for children generally explores how alternative families directly affect children’s health, psychological well-being, and socioeconomic success.

Rethinking the “Pathology of Matriarchy”

233

Two potential consequences that have been subjected to less empirical testing involve interactions, or conditional relationships. The first idea is that family disruption may reduce the intergenerational transmission of status. Drawing primarily on socialization and role modeling theory, Biblarz and Raftery (1993) argued that family disruption negatively impacts social-psychological dimensions of parent/child relations that facilitate family transmission. They found that the association between parent’s occupation (socioeconomic status, occupational autonomy, and training, from Hout [1984b, 1988]) and son’s occupation was weaker among sons from alternative families. This was extended in Biblarz et al. (1997), but both studies rely on early-1970s data (OCG II) that include only men. As far as we know, this hypothesis—that father absence may be more important as a disrupter of socioeconomic transmission than as a direct determiner of socioeconomic attainment—has not been tested further. The second idea, raised by Duncan and Duncan (1969), is that children who grow up in alternative families may have a more difficult time capitalizing on their educational accomplishments or translating their education into occupational success. Duncan and Duncan (1969) note that the returns in occupational status to each additional year of education were greater among sons raised in two-biological-parent families than among sons from alternative families. Duncan and Duncan (1969) offer no speculation about the potential process involved, nor do they statistically test the observed differences in coefficients. To our knowledge, this possible consequence of family structure has not been explored further. Fathers may be an important link to the public sphere for children. Getting a good job is partly a function of educational credentials, but it may also be a function of, among other things, parental connections and exposure to knowledge about how to secure jobs (McLanahan & Sandefur, 1994). Assuming that parents generally have the greatest knowledge about the occupational stratum in which they are located, children from female-headed families would be at a significant disadvantage, because female heads occupy substantially lower occupational positions than male heads (Biblarz et al., 1997). With less exposure to the father-as-worker model, and potentially less access to his connections, children from alternative families may find it more difficult to get the payoff to education normally achieved by children from two-biologicalparent families. Approach to the Analysis The first large-scale empirical attempt to assess the socioeconomic consequences for children of family disruption was Duncan and Duncan (1969; see also Duncan, 1967). These articles have been cited over

234

Social Problems and Policy

100 times since their publication, including recently in Powell and Parcel (1997), and Hout (1997). Duncan and Duncan (1969) found that men from female-headed families (as well as men from alternative male-headed families) had lower occupational achievements (measured as average scores on Duncan’s 1961 socioeconomic index—SEI) than men from two-biological-parent families. Duncan and Duncan (1969, pp. 284–85) interpret the five-point gap in occupational achievement between sons from two-parent and those from female-headed families as possible support for the Moynihan view: “The analyses reported above lend some support to the notion that the son raised in a family headed by a female is handicapped with respect to occupational success. The evidence in this paper obviously does not constitute ‘proof’ that the matriarchal family structure and the absence of a father are ‘pathological.’ For Negroes as for non-Negroes, however, the indication that an intact family background facilitates occupational success is quite compelling.” In the analyses that follow, we go back to Duncan and Duncan (1969) and to the 1962 Occupational Changes in a Generation survey (OCG I) in an attempt to locate the sources, or mechanisms, by which female-headed families (and other types of alternative families) may have led to a reduction in children’s occupational success. We then move forward in time, replicating the analyses on national surveys from each of the subsequent decades. Among the questions that guide our analyses are the following: 1. Do conclusions about family structure effects change substantially depending on the mix of control variables? 2. Do conclusions about family structure effects change substantially depending upon period? 3. Do the socioeconomic achievements of children from different types of alternative families vary substantially? 4. Which (if any) of the theoretical perspectives are supported in terms of both the static and change predictions that they make? 5. Do alternative families exhibit a weaker level of intergenerational socioeconomic transmission than two-biological-parent families? 6. Do alternative families reduce children’s ability to translate educational achievement into occupational success?

DISCUSSION We went into the analysis guided by six questions. In the following discussion, we assess how the empirical analysis (not shown) informs our understanding of each.

Rethinking the “Pathology of Matriarchy”

235

Do conclusions about family structure effects change substantially depending on the mix of control variables? The answer to this question is yes. The analysis shows that the effect of growing up in a single-mother family is a complex function of a set of factors that represent both risks and benefits to children’s socioeconomic success. On the side of risk, relative to sons from two-biological-parent families, sons from single-mother families have (1) the disadvantage of having a family head with a greater average likelihood of unemployment, and (2) the disadvantage of having a family head with a lower average occupational position. Across four large surveys spanning 30 years, across two dependent variables—children’s education and occupation—we find that there is no effect of growing up in a single-mother family once family head’s socioeconomic location (employment, occupation) is taken into account. This finding may mean exactly what it appears on face—that parent’s labor force attachment and occupational position are keys to understanding the effect of single-mother families on children’s socioeconomic destinations. It directs us to explore job differences between single mothers and other family heads: Where are they located in the occupational structure? How do their job conditions affect the family? What are the employment opportunities and constraints (Glass & Camarigg, 1992)? Why is parental employment good for children? In terms of policy, the findings suggest that adequate job opportunities for single mothers could go a long way toward diminishing the unfavorable consequences of single-mother families for children. Another possibility is that these variables—parent’s employment and occupation—are carrying the effects of other variables within them. One possibility is income. We may be tapping the level of economic resources available to families. Hauser and Warren (1997) entertain the possibility that the occupational index SEI may actually be a better measure of permanent income than income itself. The findings dovetail in many ways with the work of Kiernan and colleagues (e.g., Kiernan, 1997) in Great Britain. Based on the National Child Development Study, Kiernan (1997) found that differences in educational and work outcomes between children from two-biologicalparent and single-mother homes are largely a function of financial hardships. She concluded that it may be the financial advantages held by children from two biological-parent homes that propel them to higher attainments, and not necessarily the fact that their parents stayed together. We also found that children from single-mother families benefited from a good average level of origin education and low average number of siblings. These features help to offset other kinds of disadvantages associated with single-mother families. If we hold constant origin

236

Social Problems and Policy

education and siblings through statistical control, the negative effect of a single-mother family reemerges. Across all the surveys analyzed, model 3 (not shown) shows no effect of alternative mother-headed families, but model 5 (not shown) does show a significant negative effect of alternative mother-headed families. The difference in conclusion depends on the variables we decide to control statistically. Studies that treat parents’ education and number of siblings as exogenous (e.g., Duncan, 1967; McLanahan & Sandefur, 1994) will actually begin with a kind of statistically heightened “total” effect of alternative mother-headed families. In turn, whether one embraces model 3 versus model 5-like results will shape family discourse in very different ways. The former model suggests that single mothers who are able to secure adequate positions in the social structure—indicated by their employment and occupation—can offset the negative effect of the loss of the father, and their children will do approximately as well (in education) as those from two-biological-parent families. The latter model suggests something closer to a “pathology of matriarchy” perspective—pointing to some long-term legacies of father absence for children’s attainments that go beyond the loss of socioeconomic resources. Do conclusions about family structure effects change substantially depending upon period? The effects of alternative family structures on children’s socioeconomic success have remained constant over the past 30 years. Change in the effect of family structure over time probably cannot explain discrepancies observed in the family structure effects literature. The effects of other dimensions of family background have changed over the period. Race-based inequality in socioeconomic attainment is no longer on the decline—it has not been since 1987 or thereabouts—and it may even be increasing. Having a family head who was not employed became less detrimental to children over time, but this decline too stalled in the late 1980s and may be reversing course. While family head’s SEI has declined fairly steadily as a determinant of children’s success, family head’s education has not. These patterns suggest that U.S. society may indeed be hitting a “speed bump” in its progress toward universalism (Hout, 1997). The trends may have particularly grave implications for new cohorts from alternative families, who will be disproportionately more likely to face the triple threats of race, origin unemployment, and family structure. Do the socioeconomic achievements of children from different types of alternative families vary substantially? Because most studies compare the outcomes of children from twobiological-parent families with those of children from alternative families, we know less about differences or similarities in attainment among children from different kinds of alternative families. Our findings indicate

Rethinking the “Pathology of Matriarchy”

237

that, over a 30-year period, children from single-mother families consistently do better than those raised in single-father families or stepfamilies, once socioeconomic position is taken into account (see also Amato & Keith, 1991b; Amato & Booth, 1991; Hoffmann & Johnson, 1998). Single-father families and stepfather-headed families have about the same negative effect on children’s attainments, an effect that cannot be explained by socioeconomic position. That father presence gave no particular advantage to sons from alternative families may reflect the legacy of the death of the mother or of a neglectful mother. Stepfathers may have dual households to invest in, stepparent-child relations can be problematic, and stepfamilies may introduce some distance into the children’s relationship with the mother (Biblarz et al., 1997). Whatever the explanation, none of them can be neatly construed to fit with a “pathology of matriarchy” view. The “pathology of matriarchy” position is that the lack of a man in the home would be particularly detrimental to boys. The sons from alternative male-headed households in our data lived either with their biological father or a stepfather, and they were able to report the male head’s occupation. Averaging across the four surveys, male heads’ occupations were about 4–5 points higher than those of the single mothers. Yet even with the socioeconomic advantage, these sons did no better than those from single-mother families, and in some cases they did worse. If the single mothers had occupied the same socioeconomic position as the single fathers and stepfathers, their children would have had significantly higher attainments than those from the alternative male-headed families. Prompted by Wojtkiewicz’s (1993) conclusion that a dichotomous indicator of family type captures the main effects in most cases, Powell and Parcel (1997) adopted a two-parent family/not two-parent family measure, as did Biblarz and Raftery (1993). Other research treats single-mother families, single-father families, and stepfamilies together as a “one parent” group (McLanahan & Sandefur, 1994). Duncan and Duncan’s (1969) “female-headed” families include both respondents from single-mother homes and those who reported living with a female head (e.g., grandmother, aunt) but not the mother. Single, female-headed families and single-parent families are not the same as single-mother families, even though the “single female” and “single parent” terms may evoke a “single mother” image. When all alternative female-headed or single-parent families are considered together, the large group of children from singlemother families gets combined with a smaller group of children from other types of alternative families in which the negative association with children’s attainment is especially strong. This increases the negative effect associated with the combined group and so appears to reinforce a “pathology of matriarchy” perspective that may not be fully consistent

238

Social Problems and Policy

with the data. The alternative father-headed families include single-father families but also those where the father had remarried. Unfortunately, the data did not allow us to distinguish these. Which (if any) of the theoretical perspectives are supported in terms of both the static and change predictions that they make? Evolutionary parental investment theory was the only one where static and change predictions were both borne out by the data. The sociological theories faltered in that they provide no basis for our finding that children from single-mother families actually have some advantage over children from other kinds of alternative families. The static predictions of economic theory—that children from stepfamilies will do better than those from single-parent families, that children from single-father families will do better than those from single-mother homes—were entirely unsupported. The stability over time in the effect of family structure provides some evidence for its not being a selection effect. Hence, the change predictions of the parental competence and marital conflict models are not supported. Approximately two-thirds of the OCG I sample from alternative families experienced the death of a parent. In contrast, two-thirds of the NSFH2 sample from alternative families experienced parental divorce. Selection models would suggest that the former group had a more competent (widowed) parent(s) or had less exposure to marital conflict than the latter group, and so should have done better. Yet the consequences of family structure were essentially the same for both groups. The constancy of this effect may imply a fundamental family process resistant to changes in the times or the culture. The evolutionary perspective differs from the others by placing gender of parent and biological relations at the forefront of an explanation of family structure effects. It predicts that children from single-mother homes will have advantages over those from single-father homes because mothers have more of their reproductive investment tied up in their children than fathers. This prediction was supported in that, holding constant other variables, children from single-mother homes had higher attainments than those from alternative father-headed households. The evolutionary perspective also predicts that a stepparent will be of no advantage to children (stepparents have no real incentive to invest in stepchildren since stepchildren contribute). This prediction was supported insofar as children from single-mother families had higher attainments than those from stepfamilies. Finally, evolutionary theory predicts no change over time in the magnitude of the effect of family structure. Parental investment determines children’s outcomes. The experience of the death of a spouse versus divorce would not alter the custodial parents’ incentive to invest in the children. Parental investment is a function of a reproductive calculus that is part of an

Rethinking the “Pathology of Matriarchy”

239

evolved psychology developed during the foraging eons. This prediction of no change over time was also supported. While the marital conflict model is not supported by our data, further testing is required. Assume that in the earlier period only the most conflicted marriages ended in divorce, whereas in the later period families that are unhappy but less so also divorce. It is possible, e.g., that in the earlier period the pool of respondents from alternative families was made up of a large proportion that experienced parental death and a small proportion that experienced “severe” divorce of parents. In the later period, the pool from alternative families may have been composed of a small proportion that experienced parental death and a large proportion that experienced less severe parental divorce. Divorce may become less severe also as it becomes more institutionalized (Cherlin, 1978; Wolfinger, 1998). These may be offsetting effects resulting in no change in the effect of family structure over time. Do alternative families exhibit a weaker level of intergenerational socioeconomic transmission than two-biological-parent families? Do alternative families reduce children’s ability to translate educational achievement into occupational success? The finding observed in OCG II—that the effect of origin SEI on destination SEI is weaker among children from alternative families (Biblarz & Raftery, 1993)—was confirmed here using data that were both older and more recent, and that included both men and women. This supports the proposition that alternative family forms may produce a general weakening in the intergenerational transmission of socioeconomic position. The particularly interesting aspect of the finding in the present case is that family structure may be more important for children coming out of the higher than the lower end of the socioeconomic spectrum. Loosely, among children from blue-collar families, there are almost no differences in average socioeconomic attainment depending on whether one was raised in a two-parent family, a single-parent family, and so on. But there are noticeable differences in attainment by family type among children from middle and upper-middle class origins. There has been some suggestion in media and policy debates that family structure probably will not matter much for children coming out of high-resource, female-headed families. Our results suggest the opposite: it may be at the high end where family structure does matter. In a similar way, we found that children from alternative families get less occupational return to higher education. While we have no direct measure, our speculation is that the sort of “favoritism in hiring and special favors” given to families to help them launch their children, of the sort detected by Hout (1989, p. 322) in the case of Ireland, may be more accessible to two-parent families than to alternative families. Interestingly,

240

Social Problems and Policy

both of these interaction effects, one involving intergenerational transmission and the other involving intragenerational attainment over the life course, expose consequences of family structure for children that would be hidden in a more typical “direct” effects approach.

REFERENCES Acock, A. C., & Demo, D. H. 1994. Family diversity and well-being. Thousand Oaks, CA: Sage. Acock, A. C., & Keicolt, K. J. 1989. Is it family structure or socioeconomic status? Family structure during adolescence and adult adjustment. Social Forces, 68, 553–571. Amato, P. R., & Booth, A. 1991. Consequences of parental divorce and marital unhappiness for adult well-being. Social Forces, 69, 895–914. ———. 1997. A generation at risk: Growing up in an era of family upheaval. Cambridge, MA: Harvard University Press. Amato, P. R., & Keith, B. 1991a. Parental divorce and adult well-being: A metaBruce Analysis. Journal of Marriage and the Family, 53, 43–58. ———. 1991b. Separation from a parent during childhood and adult socioeconomic attainment. Social Forces, 70, 187–206. Amato, P. R., Loomis, L. S., & Booth, A. 1995. Parental divorce, marital conflict, and offspring well-being during early adulthood. Social Forces, 73, 895–915. Astone, N. M., & McLanahan, S. S. 1991. Family structure, parental practices, and high school completion. American Sociological Review, 56, 309–320. Baumrind, D. 1978. Parental disciplinary patterns and social competence in children. Youth and Society, 9, 239–275. ———. 1980. New directions in socialization research. American Psychologist, 35, 639–652. Becker, G. S. 1964. Human capital. NY: National Bureau of Economic Research. Becker, G. S., & Tomes, N. 1986. Human capital and the rise and fall of families. Journal of Labor Economics, 4, S1–S39. ———. 1981. A treatise on the family. Cambridge, MA: Harvard University Press. Biblarz, T. J., & Gottainer, G. 1999. Single-mother family structures produced by death of spouse and divorce: Why do they have different consequences for children? Journal of Marriage and the Family, 1999. Biblarz, T. J., & Raftery, A. E. 1993. The effects of family disruption on social mobility. American Sociological Review, 58:97–109. Biblarz, T. J., Raftery, A. E. & Bucur, A. 1997. Family structure and social mobility. Social Forces, 75, 1319–1339. Blake, J. 1989. Family size and achievement. Berkeley, CA: University of California Press. Blau, P. M., & Duncan, O. D. 1967. The American occupational structure. NY: Wiley. Boggess, S. 1998. Family structure, economic status, and educational attainment. Journal of Population Economics, 11, 203–222. Bronfenbrenner, U. 1979. The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

Rethinking the “Pathology of Matriarchy”

241

Bumpass, L. L., & Sweet, J. A. 1989. Children’s experience in single-parent families: Implications of cohabitation and marital transitions. Family Planning Perspectives, 21, 256–260. Cherlin, A. 1978. Remarriage as an incomplete institution. American Journal of Sociology, 84:634–650. Cherlin, A. J., Chase-Lansdale, P. L. & McRae, C. 1998. Effects of parental divorce on mental health throughout the life course. American Sociological Review, 63, 239–249. Cherlin, A. J., & Furstenberg, Jr., F. F. 1994. Stepfamilies in the United States: A Reconsideration. Annual Review of Sociology, 20, 359–381. Cherlin, A. J., Furstenberg, F. F., Chase-Lansdale, P. L., Kiernan, K., Ruane Morrison, D., & Teitler, J. 1991. Longitudinal studies of effects of divorce on children in Great Britain and the United States. Science, 252, 1386–1389. Crosby, F. J., ed. 1987. Spouse, parent, worker: On gender and multiple roles. New Haven, CT: Yale University Press. Daly, M., & Wilson, M. I. 1996. Violence against stepchildren. Current Directions in Psychological Science, 5, 77–81. DaVanzo, J., & M. Rahman, O. 1993. American families: Trends and correlates. Population Index, 59, 350–386. Dawson, D. A. 1991. Family structure and children’s health and well-being: Data from the 1988 National Health Interview Survey on Child Health. Journal of Marriage and the Family, 53, 573–584. DiPrete, T. A., & Grusky, D. B. 1990. Structure and trend in the process of stratification for American men and women. American Journal of Sociology, 96, 107–143. Duncan, B. 1967. Education and social background. American Journal of Sociology, 72, 363–372. Duncan, B., & Duncan, O. D. 1969. Family stability and occupational success. Social Problems 16, 273–85. Duncan, O. D. 1961. A socioeconomic index for all occupations. In A. J. Reiss, Jr. (Ed.) Occupations and Social Status (pp. 109–138) Glencoe, IL: Free Press. Emlen, S. T. 1997. The evolutionary study of human family systems. Social Science Information Sur Les Sciences Sociales, 36, 563–589. Featherman, D. L., & Hauser, R. M. 1978. Opportunity and change. New York: Academic Press. Furstenberg, F. F., Jr., & Cherlin, A. J. 1991. Divided families. Cambridge, MA: Harvard University Press. Glass, J., & Camarigg, V. 1992. Gender, parenthood, and job-family compatibility. American Journal of Sociology, 98, 131–151. Grusky, D. B., & DiPrete, T. A. 1990. Recent trends in the process of stratification. Demography, 27, 617–638. Glenn, N. D. 1994. The re-evaluation of family change by American social scientists. Paper presented to the Committee for the International Year of the Family of the Catholic Archdiocese of Melbourne, Melbourne. Glenn, N. D., & Kramer, K. B. 1987. The marriages and divorces of the children of divorce. Journal of Marriage and the Family, 49, 811–825. Grych, J. H., & Fincham, F. D. 1990. Marital conflict and children’s adjustment: A cognitive-contextual framework. Psychological Bulletin, 108, 267–290.

242

Social Problems and Policy

Hauser, R. M., & Warren, J. R. 1997. Socioeconomic indexes for occupations: A review, update, and critique. Sociological Methodology, 27, 177–298. Hetherington, E. M., Cox, M. & Cox, R. 1978. The aftermath of divorce. In J. H. Stevens Jr., & M. Mathews (Eds.) Mother-child father-child relationships (pp. 149–176) Washington, DC: National Association for the Education of Young People. Hoffmann, J. P., & Johnson, R. A. 1998. A national portrait of family structure and adolescent drug use. Journal of Marriage and the Family, 60, 633–645. Holden, K. C., & Smock, P. J. 1991. The economic costs of marital dissolution: Why do women bear a disproportionate cost? Annual Review of Sociology, 17, 51–78. Hout, M. 1984a. Occupational mobility of black men in the United States: 1962 to 1973. American Sociological Review, 49, 308–322. ———. 1984b. Status, autonomy, and training in occupational mobility. American Journal of Sociology, 89, 1397–1409. ———. 1988. More universalism, less structural mobility: The American occupational structure in the 1980s. American Journal of Sociology, 93, 1358–1400. Kiernan, K. 1997. The legacy of parental divorce: Social, economic and demographic experiences in adulthood. Working paper. London School of Economics, Centre for Analysis of Social Exclusion. ———. 1989. Following in father’s footsteps: Social mobility in Ireland. Cambridge, MA: Harvard University Press. ———. 1997. Speed bumps on the road to meritocracy: Occupational mobility of women and men in the United States, 1972–1994. Paper presented in the Lectures on Causality in the Social Sciences Series, University of Washington, Department of Sociology, Spring. Kline, M., Johnston, J. R., & M. Tschann, J. M. 1991. The long shadow of marital conflict: A model of children’s postdivorce adjustment. Journal of Marriage and the Family, 53, 297–310. Kohn, M. L. 1969. Class and conformity: A study in values (2nd ed.). Chicago: University of Chicago Press. ———. 1983. On the transmission of values in the family: A preliminary formulation. Research in Sociology of Education and Socialization, 4, 1–12. Logan, J. A. 1996. Opportunity and choice in socially structured labor markets. American Journal of Sociology 102, 114–160. London, R. A. 1996. The difference between divorced and never-married mothers’ participation in the aid to families with dependent children program. Journal of Family Issues, 12, 170–185. Mauldon, J. 1990. The effect of marital disruption on children’s health. Demography, 27, 431–446. McLanahan, S. S. 1985. Family structure and the reproduction of poverty. American Journal of Sociology, 90, 873–901. McLanahan, S. S., Garfinkel, I., & Ooms, T. 1987. Female-headed families and economic policy: Expanding the clinician’s focus. In J. C. Hansen & M. LindbladGoldberg (Eds.) Clinical issues in single-parent households (pp. 1–18). Rockville, MD: Aspen Publishers. McLanahan, S., & Sandefur, G. 1994. Growing up with a single parent. Cambridge, MA: Harvard University Press. Meyer, D. R., & Garasky, S. 1993. Custodial fathers: Myths, realities, and child support policies. Journal of Marriage and the Family, 55, 73–90.

Rethinking the “Pathology of Matriarchy”

243

Moss, Philip, and Chris Tilly. 1996. Raised hurdles for black men: Evidence from interviews with employers. Working paper. Russell Sage Foundation. Mott, Frank L., Lori Kowaleski-Jones, and Elizabeth G. Menaghan. 1997. Paternal absence and child behavior: Does a child’s gender make a difference? Journal of Marriage and the Family, 59, 103–18. Moynihan, D. P. 1965. The negro family: The case for national action. Washington, DC: U.S. Department of Labor. Mueller, C. W., & Pope, H. 1977. Marital instability: A study of its transmission between generations. Journal of Marriage and the Family, 39, 83–93. Nock, S. L. 1988. The family and hierarchy. Journal of Marriage and the Family, 50, 957–966. Painter, G. 1998. Family structure and the achievement of youths: The importance of parental involvement in education. Working paper. University of Southern California, School of Public Administration. Parcel, T. L., & Menaghan, E. G. 1994. Parents’ jobs and children’s lives. Hawthorne, NY: Aldine DeGruyter. Popenoe, D. 1988. Disturbing the nest: Family change and decline in modern societies. NY: Aldine DeGruyter. ———. 1993. American family decline, 1960–1990: A review and appraisal. Journal of Marriage and the Family, 55, 527–541. ———. 1996. Life without father. NY: Free Press. Powell, B., & Steelman. L. C. 1993. The educational benefits of being spaced out: Sibship density and educational progress. American Sociological Review, 58, 367–381. Powell, M. A., & Parcel, T. L. 1997. Effects of family structure on the earnings attainment process: Differences by gender. Journal of Marriage and the Family, 59, 419–433. Raftery, A. E. 1986a. A note on Bayes factors for log-linear contingency table models with vague prior information. Journal of the Royal Statistical Society, series B, 48, 249–250. ———. 1986b. Choosing models for cross-classifications. American Sociological Review, 51, 145–146. ———. 1995. Bayesian model selection in social research (with discussion). In P. V. Marsden (Ed.) Sociological Methodology 1995 (pp. 111–195). Cambridge, MA: Blackwell Publishers. Rozondal, F. G. (1983). Halos vs. stigmas: Long-term effects of parent’s death or divorce on college students’ concepts of the family. Adolescence, 18, 947–955. Silverman, P. R., Nickman, S., & Worden, J. W. 1992. Detachment revisited: The child’s reconstruction of a dead parent. American Journal of Orthopsychiatry, 62, 494–503. Simons, R. L., & Associates. 1996. Understanding differences between divorced and intact families: Stress, interactions, and child outcome. Thousand Oaks, CA: Sage. Stacey, J. 1996. In the name of the family: Rethinking family values in the postmodern age. Boston: Beacon Press. Steelman, L. C., & Powell, B. 1989. Acquiring capital for college: The constraints of family configuration. American Sociological Review, 54, 844–855. Sugarman, S. 1993. Reforming welfare through Social Security. University of Michigan Journal of Law Reform, 26, 849–851.

244

Social Problems and Policy

———. 1995. Financial support of children and the end of welfare as we know it. Virginia Law Review, 81, 2523–2573. Teachman, J. D., Paasch, K., & Carver, K. 1997. Social capital and the generation of human capital. Social Forces, 75, 1343–1360. Thomson, E., Hanson, T. L., & McLanahan, S. S. 1994. Family structure and child well-being: Economic resources vs. parental behaviors. Social Forces, 73, 221–242. Thomson, E., McLanahan, S. S., & Curtin, R. B. 1992. Family structure, gender, and parental socialization. Journal of Marriage and the Family, 54, 368–378. Trivers, R. 1972. Parental investment and sexual selection. In B. Campbell (Ed.) Sexual Selection and the Descent of Man (pp. 136–179). Chicago: Aldine DeGruyter. Wallerstein, J. S. 1989. Second chances: Men, women and children a decade after divorce. NY: Ticknor and Fields. Weinstein, M., & Thornton, A. 1989. Mother-child relations and adolescent sexual attitudes and behavior. Demography, 26, 563–577. Wojtkiewicz, R. A. 1993. Simplicity and complexity in the effects of parental structure on high school graduation. Demography, 30, 701–717. Wojtkiewicz, R. A., McLanahan, S. S., & Garfinkel, I. 1990. The growth of families headed by women, 1950–1980. Demography, 27, 19–30. Wolfinger, N. H. 1998. Trends in the intergenerational transmission of divorce. Paper presented at the 1996 meetings of the Society for the Study of Social Problems, and the American Sociological Association, New York, August. Wu, L. L., & Martinson, B. C. 1993. Family structure and the risk of a premarital birth. American Sociological Review, 58, 210–232.

Chapter 16

Child Care Policy Reform and Employment of Single Mothers Jay Bainbridge, Marcia K. Meyers, and Jane Waldfogel*

Single mothers’ employment has grown substantially in recent years, with particularly sharp increases in the 1990s. In 1991, just over one half (55 percent) of never-married mothers with children less than 13 years of age were employed; by 1996, 70 percent were in the workforce (authors’ calculations, Current Population Survey, U.S. Census). This increase is all the more striking given that the employment rate of single women without children under 13 remained nearly stable, at about 90 percent, during this same period. The growth in single mothers’ employment during the 1990s parallels a several-decades-long increase in overall maternal employment. It also parallels a number of policies enacted during the 1990s that were designed to discourage welfare use and encourage employment among low-income and single-parent families. Major changes to federal cash assistance policy began in 1988 with the passage of the Family Support Act (FSA), which imposed more stringent work requirements on welfare recipients and authorized transitional assistance for families leaving welfare for employment. Beginning in the early 1990s, states initiated other policy changes— designed to discourage women from entering the welfare system and to encourage those in the system to move from welfare to work—under waivers from the laws governing the federal Aid to Families with Dependent Children (AFDC) program. These state reforms laid the groundwork for the 1996 Personal Responsibility and Work Opportunity

* J. Bainbridge, M. K. Meyers, J. Waldfogel (2003), Child Care Policy Reform and the Employment of Single Mothers, Social Science Quarterly, 84: 4, 771–791. Copyright © 2003 Wiley. Reprinted with permission of Blackwell Publishing Ltd.

246

Social Problems and Policy

Reconciliation Act (PRWORA), which replaced AFDC with a new timelimited program, Temporary Assistance to Needy Families (TANF). PRWORA ended the federal entitlement to cash assistance for lowincome, single-mother families with children and made it very difficult for single mothers to receive cash assistance without working. During this same period, federal and state policy officials expanded a number of programs designed to support the employment of low-income parents. The expansion of child care assistance was particularly notable. Congress created a variety of categorical child care programs in the late 1980s and early 1990s and, under the 1996 PRWORA, consolidated and expanded funding through a single federal block grant (the Child Care and Development Fund). Congress also expanded the Earned Income Tax Credit (EITC) for parents with low earnings in 1991, with further increases in 1994, 1995, and 1996. The federal minimum wage was raised in 1993. Health insurance coverage for low-income children was extended through state Medicaid expansions throughout the 1990s and through the enactment of the federal Children’s Health Insurance Program (CHIP) in 1997. These and other federal and state policy changes during the 1990s created new barriers to welfare receipt and new employment incentives for low-income and single-mother families. Social policy changes in the 1990s were also notable for the extent to which they devolved authority for policy design from federal to state government. Prior to 1996, many states were actively pursuing new policy approaches under waivers from federal welfare laws. The conversion of federal welfare and child care entitlement programs into block grants under the 1996 PRWORA gave states even greater latitude for experimentation. The devolution of policy authority has increased state-to-state variation in a number of policy domains. It has also created a natural experiment for examining the contribution of policy choices to household-level outcomes. In this article, we capitalize on over time and cross-state variation to estimate the contribution of child care and related policies to the rise in single mothers’ employment during the early 1990s. We extend prior work in this area by developing new measures of child care policy that more precisely distinguish between assistance for employed and nonemployed families. We find that per capita public child care assistance grew very modestly between 1991 and 1996, with most of the growth benefiting welfarerecipient, primarily nonworking, families. After controlling for family, economic, and other policy effects, we estimate that state spending on subsidies for working-poor families had substantial and significant positive effects on the employment of single mothers with young children— dollar for dollar these effects were similar to, or greater than, those associated with tax policy changes. Per capita tax benefits increased more dramatically than per capita child care benefits, however, so tax policies

Child Care Policy Reform and Employment of Single Mothers

247

are estimated to explain a larger share of the growth in single mothers’ employment for this period. The remainder of the article proceeds as follows. In the next section, we review prior research on the contribution of social policy to single mothers’ employment outcomes and summarize key aspects of child care policy during the 1990s. In the third section, we describe our data and estimation methods. The next section presents results, and the final section discusses conclusions and implications.

BACKGROUND Prior Research A number of researchers have examined the likely contribution of public policies to the growth in single-mother employment during the 1980s and 1990s. Changes to cash assistance programs have received particular attention. Based on an analysis of data from the 1977–1995 March Current Population Survey (CPS), Moffitt (1999) concludes that state-level reforms enacted under waivers from federal welfare laws—most of which restricted access to cash benefits and/or imposed more stringent work requirements on recipients—had a positive effect on the number of weeks and hours worked by less-educated women. Using data from the 1977 to 1999 CPS, Schoeni & Blank (2000) also found that welfare waivers had a positive effect on the employment, weeks worked, and hours worked by less-educated women. These and other studies also suggest that expansion of the Earned Income Tax Credit (EITC), benefiting lower income workers, contributed to the rise in employment. Ellwood (2000), for example, used data from the 1980–1999 March CPS and found that employment rose significantly more over this period for low-wage single women with children, women who would have been most affected by the expansion of the EITC, than for higher-wage single women with children or single women without children. The impact of child care subsidies on employment during this period has received relatively less attention. A substantial body of empirical research supports the general conclusion that mothers’ labor supply decisions are sensitive to the availability and price of alternative care arrangements (see Anderson & Levine, 2000, for a recent review). Not surprisingly, the employment decisions of low-earning and single mothers are particularly sensitive to child care prices (Anderson & Levine, 2000; Han & Waldfogel, 2001; Connelly & Kimmel, 1999). Analysts who have estimated the consequences of policies that decrease child care costs predict both an increase in use of more expensive, regulated forms of care and increases in maternal employment (U.S. General

248

Social Problems and Policy

Accounting Office, 1994a; Michalopoulos, Robins, & Garfinkel, 1992; Powell, 2002; Blau & Hagy, 1998; Michalopoulos & Robins, 2000; Connelly & Kimmel, 1999). Although research directly addressing the impact of public child care subsidies on employment has been limited, studies by Gelbach (2002) on the employment effects of kindergarten (essentially a 100 percent subsidy), by Berger & Black (1992) on employment among low-income single mothers receiving or placed on waiting lists for child care subsidies, and by Meyers, Heintz, & Wolf (2002) on the employment effects of subsidy receipt among welfare recipients, all confirm the expectation that policies that subsidize the costs of substitute child care increase the probability that mothers with young children will enter employment. Despite the potentially important contribution of child care subsidies to maternal employment outcomes, relatively few studies of the effects of policy changes in the 1990s have examined child care policies in detail. In one such study, McKernan et al. (2000) used a difference-indifference-in-difference approach to compare employment outcomes for unmarried women with and without children for the periods preceding and following the 1996 PROWRA. Examining 10 areas of state policy, they conclude that eight—including one measure of child care policy (the number of months of transitional child care benefits provided to women leaving welfare for work)—significantly increased single mothers’ employment in the mid- to late-1990s. In the study closest in spirit to our own, Meyer & Rosenbaum (2001) examine the employment impact of child care expenditures, along with changes in welfare policy, Medicaid policy, job training, and the EITC, over the 1984–1996 period. Using a difference-in-difference-in-difference approach to compare the employment of single mothers to that of single women without children, they conclude that the EITC accounted for over 60 percent of the increase in single mothers’ employment during this period. They attribute much less of the change to other policies; child care expenditures, in particular, were found to account for only 4–7 percent of the increase. The modest contribution of the expansion in child care funding in the Meyer & Rosenbaum paper is puzzling, given the large body of research that links child care costs to maternal employment. In part, their result may reflect imprecision in the measurement of child care subsidy policy. Although nearly all public child care subsidies are related to employment, they do not all flow to employed parents. During the late 1980s and early 1990s, in particular, categorical eligibility rules and federal cost-sharing policies combined to direct a large share of subsidies to welfare recipients who were not in the labor force because they were engaged in short-term work-preparation activities. Aggregate measures of child care expenditures that fail to distinguish among subsidy programs are likely to

Child Care Policy Reform and Employment of Single Mothers

249

underestimate the effect of spending that was targeted directly to employed parents. Child Care Policy Reforms in the Early 1990s To better understand issues in the measurement of child care policy, it is useful to examine in some detail the structure of public child care policies during this period. Congress created four new, categorical child care programs between 1988 and 1996. All four were linked to employment, but they targeted different populations, supported different phases of the transition to work, and were financed with mechanisms that created different incentives for states to spend federal funds. The first phase of child care subsidy expansion targeted welfare recipients under the 1988 Family Support Act (FSA). The FSA created two new child care entitlements: the AFDC/JOBS child care program guaranteed assistance for welfare recipients who needed child care to participate in employment and training activities, and the Transitional Child Care (TCC) program provided assistance for parents who left AFDC due to increased earnings (but with other provisos that complicated access; see U.S. General Accounting Office, 1994b). A second round of policy reform expanded assistance specifically to low-income families outside the welfare system. The Child Care Development Block Grant (CCDBG) and At-Risk Child Care programs (passed with the OBRA of 1990 and implemented in 1991) created nonentitlement subsidy programs for lowincome working families. Along with differences in categorical eligibility, federal child care programs created different incentives for states to “draw down” their allocated funds. The AFDC/JOBS and TCC subsidies, implemented in 1990, provided uncapped federal dollars as long as states matched the spending in accordance with their Medicaid Matching Rate (MMR), which required wealthier states to pay half the new child care expenses and poorer states to pay less than a quarter of their share. The CCDBG was available to states without a cost-sharing requirement. The At-Risk program, created at about the same time, required state matching expenditures at the MMR. These funding mechanisms interacted with child care and welfare policies to shape incentives for states to draw down funds. For example, the welfare-based AFDC/JOBS entitlement imposed a cost on states (in matching funds) every time a family took up benefits. But federal law also required states to enroll a share of their caseload in employmentpreparation activities, greatly increasing the need to provide child care to welfare recipients. This created a strong incentive for states to encourage families to use AFDC/JOBS child care subsidies. In contrast, the welfarebased TCC entitlement for families leaving welfare also required states to

250

Social Problems and Policy

provide matching funds, but states had no incentive—from federal performance standards—to either place parents in employment or to encourage them to use TCC subsidies. Given this incentive structure, it is not surprising that spending rose rapidly for the AFDC/JOBS program but remained extremely low for TCC. In the case of the programs designed specifically for working-poor families outside the welfare system, state incentives to make use of funds were mixed. CCDBG funds could be spent without a state match, but the At-Risk program did impose a cost on states for every assisted family. In the absence of performance standards relating to the work effort or child care use of nonwelfare families, states had little direct incentive to spend these funds. These administrative structures help to explain why many states did not spend their full appropriation of At-Risk funds. For example, nearly half of states did not use all the federal At-Risk funds available to them in 1994. Unused funds that year ranged from as little as $67,000 in Rhode Island to as much as $19 million in California. Combined federal and state funding for child care subsidies more than doubled between 1991 and 1996. The most rapid growth occurred in the welfare-based programs. Welfare-linked subsidies through the AFDC/JOBS and TCC programs nearly tripled, growing from $600 million to $1.7 billion (in 1996 dollars). In any given year, less than a quarter of the dollars in these welfare-based programs went to families transitioning from welfare to work. During this same period, subsidies for the working poor through CCDBG, and At-Risk first grew by about 50 percent between 1991 and 1992, but then remained constant. Because spending on welfare-linked subsidies grew so much more rapidly than spending for the working poor, assistance to the working poor accounted for 59 percent of federal and state spending in 1991, but only 45 percent by 1996. Thus, although many observers describe the expansion of child care funding during this period as being for the working poor, only about half of all federal and state child care subsidy spending in the early 1990s actually flowed to employed parents.

DATA AND METHODOLOGY We use information on the structure and distribution of child care subsidy assistance to extend prior work on the impact of child care policy on the growth of single mothers’ employment between 1991 and 1996. We construct new indicators of the availability of subsidized child care that distinguish between “welfare”-based assistance that went primarily to families in short-term employment-preparation activities and “workingpoor” assistance that benefited employed parents. We expect that these indicators will provide a more precise measure of the effect of the child

Child Care Policy Reform and Employment of Single Mothers

251

care expansions of the early 1990s on the increase in single mothers’ employment. Estimation Approach Our estimation uses a difference-in-difference-in-difference approach to estimate the effect of child care policies on the probability of employment among single mothers with children aged less than 13, in comparison to employment among other single women, controlling for other policies expected to affect the employment decision. We examine the effect of changes in child care spending within states over time, differentiating between single mothers who could benefit from this assistance (those with children under age 13) and other single women who could not (i.e., those with no children or only children aged 13 and above). Our basic model is a reduced-form estimation of the probability of employment as a function of individual characteristics that affect the expected wage (and hence employment decisions) of single women: race and ethnicity, age, education, number of children aged birth to 3, number of children under 6, number of children under 13, number of children under 18, marital status, the interaction of marital status with the presence of any children, and unearned family income. To control for the local economic context, we include measures of the state unemployment rate and an indicator of whether the woman lived in a central city. To control for unobserved secular trends (common across states) and unobserved state characteristics (fixed state effects) that may be correlated with the labor supply decision, we include dummies for each state and for each year. We capture the effects of child care policies using state-level indicators of expenditures on subsidies potentially available to the single mothers with children under 13 in the sample. We include increasingly disaggregated measures of these expenditures, distinguishing between programs serving “welfare” and “working” families. To control for other policies, we include measures of relevant federal and state policies with differential benefits for mothers and other single women for each observation year (as in Meyer & Rosenbaum, 2001). These include measures of cash assistance (AFDC), benefits available for working and nonworking families, federal and state taxes and earned income tax credits, state welfare reforms affecting the probability of receiving AFDC, and investments in employment-preparation services. This model estimates the effect of a policy changing within a state in a particular year (difference-in-difference). This may not, however, give us an unbiased estimate of the effect of that policy change if other, unobserved state factors correlated with the employment of lower-skilled women changed at the same time. For instance, a state may have expanded child care subsidies for working families in the same year that

252

Social Problems and Policy

a local industry expanded employment opportunities for low-skilled workers. In this case, our estimation might erroneously attribute an increase in mothers’ employment in that state and year to the child care expansion rather than the change in employment conditions. We control for this unobserved heterogeneity across states and over time by using a difference-in-difference-in-difference approach. This approach compares the contribution of increased child care spending to state/year differences in employment among single mothers with children under 13 (the only group potentially eligible for child care subsidies) to state/year differences in employment among single women without children under 13 in that same state over time. The logic here is that if child care policies contributed to a change in the probability of single mothers’ employment, that relationship should be seen only for single women with younger children (our “treatment” group) and not for other single women (our “control” group). Thus the magnitude of the child care policy effect on women’s employment is identified by capitalizing on variability in the treatment of single women with and without children under 13 over time, as well as differences across states in the same year. We begin our analysis with simple difference-in-difference analyses of changes in mean employment rates and in social policies over time. We next estimate a probit model of single women’s employment as a function of individual characteristics, local economic factors, state tax and welfare policies, and total child care expenditures. To improve our estimate, we specify additional models in which we control for child care expenditures with greater precision, distinguishing between potential child care expenditures for “welfare” and “working-poor” families. We conclude our analysis with an accounting of the relative importance of child care reforms, and other factors, in explaining the growth in single mothers’ employment between 1991 and 1996. Data and Measures We estimate employment outcomes using microdata from the 1992–1997 March Current Population Surveys (CPS), which provides data on women’s employment, earnings, and other characteristics from 1991 to 1996. We include in our sample all unmarried women between the ages of 19 and 44 who are not in school. We code a woman as employed during the year if she reports positive earnings for that year. Women with missing data on the number of hours worked are dropped. Individual characteristics taken from the CPS include age, non-white, Hispanic ethnicity, education level, marital status, number of children less than 18, number of children less than 13, number of children less than 6, children aged 1, 2, and 3, other family income, and central city status.

Child Care Policy Reform and Employment of Single Mothers

253

We measure child care policy variation using spending data from various administrative and published sources. We construct a measure of total potential child care expenditures by state and year including federal and state expenditures for JOBS/AFDC Child Care, Transitional Child Care, At-Risk Child Care, and the Child Care Development Block Grant. We then disaggregate this measure to differentiate among funding sources that varied in their target populations and incentives for state lawmakers. (Note that by definition these measures have a value of zero for single women without children under age 13.) The first measure is the mean child care subsidy amount per single mother family with children under 13 in each state in a given year. To control for differences in cost of living (and child care), we normalize this measure by the average earnings in that state relative to earnings in the other states. We next disaggregate this measure into two separate indicators: a welfare subsidy amount, including the AFDC/JOB and TCC funds, and a working-poor subsidy amount, including At-Risk Child Care and Child Care Development Block Grant funds. Because the At-Risk Child Care funds were an individual entitlement, expenditures may be endogenous with levels of single mother employment in the state, and therefore bias our estimate of the working-poor subsidies. To isolate a more exogenous measure of child care spending, we further disaggregate this measure into separate measures of At-Risk funding and CCDBG funding (which was provided as a formula-based block grant). To control for other policies that may have affected the labor supply of single women during this period, we include several policy measures developed by Meyer & Rosenbaum (2001) and generously shared with us for this analysis. Expected tax rates are calculated as the weighted sum of income tax liabilities for a woman with her family size and ages of children, in a given state and year at various annual earnings points. This is estimated by Meyer & Rosenbaum (2001) by integrating over the wage and hours distribution of single women, with and without children. Included in the imputation of taxes are state and federal income taxes, standard deductions, and federal and state EITCs. Welfare benefits are measured three ways. Each woman with a child under 18 is assigned (1) the maximum welfare benefit, defined using AFDC and Food Stamps levels at zero-earnings levels, based on number of children, state, and year; (2) the estimated value of the welfare benefits she would receive if she were working; and (3) the probability of welfare if working, which is the weighted sum of AFDC receipt at the various annual earnings points and that captures transaction costs or stigma. State welfare reform activities (under federal waivers) are captured by two variables: a dummy for whether the state had any time limit on

254

Social Problems and Policy

welfare receipt in that year and a dummy for whether any cases had been terminated for noncompliance with work requirements in that state in that year.10 The availability of employment-preparation services in the state is measured by two variables for the JOBS program (during the state fiscal year): expenditures on activities that are educational, which includes education, postsecondary education, and self-initiated education; and expenditures on other, which includes all other JOBS activities. Expenditures are apportioned to women according to the proportion of participants who are female adults. Each training variable is divided by the number of AFDC recipients who were subject to job-preparation requirements (based on the age of the youngest child). To control for costof-living differences, expenditures are divided by the state average wage, normalized to 1996. We control for secular changes in the economy and for state-to-state differences in labor market opportunities by including indicators of the state unemployment rate for each year of observation.12 We control for individual differences in labor market opportunities by including an indicator (constructed from the CPS) for whether the respondent lived in a central city.

RESULTS Changes in Single Women’s Employment Rates Throughout the period, employment rates differed quite dramatically between single women with and without young children. Single women without children under 13 had very high employment rates throughout the period, starting at 90 percent and dipping slightly below that by 1996. Among single mothers with children under 13, a much smaller percentage worked, but the proportion rose steadily during the period, from about two-thirds in 1991 to over three-quarters by 1996. Accordingly, the difference-in-difference between those with younger children and those without dropped by 11 percentage points during the period. Employment rates and changes in the gap or difference-in-difference between these groups also varied by marital status. The most dramatic change in the difference-in-difference between single women with and without younger children occurred in the subgroup of never-married women. Never-married women with children under 13 increased their employment rates from 55 to 70 percent, while their counterparts (never married with older or no children) stayed relatively steady near 90 percent. As a result, the difference-in-difference between these groups declined by 18 percentage points.

Child Care Policy Reform and Employment of Single Mothers

255

Changes in Policy A number of state and federal policies also changed during the 1991–1996 period, with differential consequences for single women with children under age 13 and those without. The most dramatic changes are observed in tax policies. Largely as a result of expansions in the federal EITC, the average tax burden for single mothers with children under 13 changed from a $790 liability in 1991 to a $49 credit in 1996. Single mothers with children under 13 had a lower average tax burden than other single women throughout the period. As a result of changes in tax policy, this tax advantage (the difference in the difference between the groups) increased by $738.

Policies Affecting Single Women With and Without Children Under Age 13 (1991 and 1996) Changes in AFDC policy had the opposite effect, decreasing the difference in government assistance between the two groups of single women, AFDC benefits declined for both groups, and the difference in benefits between the groups narrowed by $516 for all welfare recipients, and $127 for those with employment. The percentage of single women who were exposed to new state welfare reforms—time limits on assistance and imposition of sanctions for noncompliance with work rules—increased much more dramatically for single mothers with children under 13 than for other single women, such that by 1996 the difference in the share living in states with time limits was 35 percentage points greater and the share affected by sanctions was 17 percentage points higher. State and federal child care expenditures also increased during this period, although the magnitude of the increase was much smaller than that observed in taxes. As spending rose, the differences in government assistance widened between single women who were and were not potentially eligible for child care subsidies. Considering all sources of child care funding, average spending per single mother with a child under 13 increased from $204 in 1991 to $443 in 1996 (spending on single women without children in this age group remained at zero by definition). This increased the difference-in-difference (between single women with children under 13 and those without) by $239. By way of comparison, the increase in the difference in child care benefits ($239) was about one-third of the increase in the difference in tax advantages ($738) between the two groups of single women. Disaggregating the child care measure indicates that this relatively modest increase in child care assistance disproportionately benefited nonworking, welfare-recipient families. The annual amount available for working-poor subsidies grew from $121 in 1991 to $198 in 1996 (per single mother family with children under age 13), while the annual amount for

256

Social Problems and Policy

welfare-related subsidies grew from $83 to $244 per family. Thus, the largest change—or difference-in-difference between the two groups of single women—affected those eligible for welfare-based rather than working-poor subsidies. The difference-in-difference for “working-poor subsidies” was less than $30 over the six-year period. Multivariate Results We begin our multivariate analysis by estimating the effect of total child care spending on the employment probability of single women, controlling for individual characteristics, local economic conditions, state, year, and other policies. The results are in the expected direction for individual and other policy variables. Child care expenditures per single woman with a child under 13 are associated with a small and marginally significant (p = 0.054) effect of 0.036. This effect can be interpreted to mean that a $1,000 increase in average state spending for all types of child care subsidies was associated with a nearly 4 percent point increase in employment among single mothers with children in this age group. This estimate may be biased downward because the largest increase in spending during this period affected welfare-recipient, primarily nonworking families. To obtain a more precise estimate of the effect of child care subsidies available to working families, we re-estimate the model with our disaggregated measures of child care spending. Disaggregating child care spending into the “welfare” and “workingpoor” categories reveals that these policies actually had potentially opposite results for the employment of single mothers. The coefficient for the measure of welfare-based child care, available primarily to women in work-preparation activities, is negative and does not reach statistical significance. In contrast, the marginal effect of subsidy spending for the working poor is much larger than in the prior specification and is statistically significant at the 1 percent level. When considered on its own, a $1,000 per capita increase in the average “working-poor” child care subsidy corresponds to an 11 percent increase in the employment of single mothers during the year. A further disaggregation of the measure of subsidy spending for the “working poor” suggests that the large majority of this effect can be traced to subsidies provided through the CCDBG program. Although the coefficient for At-Risk child care is small and nonsignificant, the marginal effect of the CCDBG increases is more than double that estimated in the prior model and statistically significant at the 1 percent level. For every $1,000 of per capita allocation on CCDBG assistance, we estimate a 26 percentage point increase in the employment of single mothers with children under 13.

Child Care Policy Reform and Employment of Single Mothers

257

More precise estimation of child care subsidy policy substantially alters our conclusions about the relative contribution of various policy changes during this period. Most notably, we observe estimated effects for child care subsidy spending that are approximately one-third the size of those estimated for dollar adjustments in tax liabilities and benefits. Under our preferred specifications, which disaggregate the form of subsidy spending, we estimate effects of child care spending for the working poor that are about equal to—and in the case of CCDBG expenditures, several times greater than—the effects associated with tax changes. Contribution of Policy to Changes in Employment from 1991 to 1996 Using the marginal probabilities from our preferred specification, we calculate and compare the percent of the employment increase among single women during the 1991–1996 period that is explained by changes in levels of child care subsidies and changes in other policies. A general conclusion is that in the early 1990s, child care accounted for a small relative share of the total change in the employment of single mothers with children under 13. When child care is disaggregated, we see that changes in spending in CCDBG subsidies to the working poor explain an estimated 7 percent of the increase in employment among single mothers with children under 13. Changes in welfare-based child care are associated with a 4 percent decrease in employment, which is consistent with the targeting of these subsidies on welfare recipients who are out of the workforce and preparing for employment. Child care policies are estimated to have made a larger contribution to changes in single mothers’ employment during this period than most elements of welfare policy. One state-level welfare reform—the adoption of time limits affecting eligibility—is estimated to explain approximately the same share (7 percent) of the increase in employment. Changes in AFDC benefits, sanctions for noncompliance with AFDC work requirements, and spending on JOBS-related employment services are estimated to have much smaller effects, explaining 4 percent or less of the increase in labor force participation among single mothers with children under 13. Changes in tax policies are estimated to have made the largest single contribution to changes in employment. Indeed, the growth in EITC spending over the period is estimated to account for nearly two-thirds of the increase in maternal employment among women with children under 13—nearly 10 times the contribution of child care subsidy spending. The magnitude of the contribution of tax policy changes, relative to that of the increase in child care spending, is consistent with the much larger

258

Social Problems and Policy

increase in tax benefits. The difference in tax benefits between the two groups of single women in this period grew by $738—more than 10 times the $28 per capita growth of the difference in assistance through the CCDBG program.

CONCLUSION Previous research on the effects of policy on single mothers’ employment decisions has focused primarily on the Earned Income Tax Credit, welfare benefits, and other welfare policies. When child care policy has been included in models, the effects have been modest—in part because measures have failed to capture all relevant spending or have failed to distinguish among programs that have offsetting implications for employment. In this article, we extend an existing methodology to look at the effects of child care and other policies on single women’s employment over the 1991–1996 period. We find that more complete and detailed measures of child care policy yield larger effects, controlling for individual characteristics and other policy variables relevant to the work decision. Our most important contribution is to disaggregate “welfare” child care from “working-poor” child care—our variable of primary interest. By 1996, funding for working-poor subsidies accounted for only 45 percent of all state and federal funding, down from 59 percent in 1991. This has important implications for estimating the effect of subsidies on employment, since most of the welfare child care funds are spent on families who are in education and training programs prior to employment. Isolating the working-poor child care expenditures from those to welfare families boosts the effect of child care such that a $1,000 increase in the average annual subsidy per single mother with a child under 13 translates into an estimated 11 percent increase in the probability of employment. A further disaggregation suggests that expansion of the CCDBG was associated with a 26 percent increase in employment for every $1,000 per capita investment. Our estimate of the contribution of the expansion of child care subsidies during the 1990s is modest—about the same magnitude as the estimated contribution of the adoption of welfare time limits and considerably smaller than the estimated contribution of changes in tax policy. The estimated marginal effects of working poor and CCDBG child care spending, however, suggest that, dollar for dollar, increases in child care spending were potentially more important than changes in tax benefits. This suggests that the modest estimated effect for child care is due largely to the very modest public investments in child care—particularly for the working poor.

Child Care Policy Reform and Employment of Single Mothers

259

The endogeneity of the child care measure is a source of potential concern in estimating the contribution of child care to employment. If public child care expenditures rise in response to demand, it may be difficult to determine the direction of causality between expenditures and employment. We are somewhat reassured by the fact that to the extent that endogeneity is a problem, it is more likely to be problem with welfare child care than with working-poor child care. As we saw earlier, funding for welfare child care is influenced by the number of women in the state participating in employment or training activities or leaving welfare. However, funding for working-poor child care is determined by the number of preschool-age children in the state, along with two indicators of the state poverty rate, factors that are likely to be either uncorrelated or negatively correlated with female employment in the state. Thus, while endogeneity is always a concern, it is likely to be less of a concern when it comes to estimating the impact of working-poor child care, the type of child care in which we are most interested here. These results have several policy implications. Our results are consistent with a growing body of research suggesting that public policies can encourage employment among single mothers by “making work pay” through reductions in tax burdens and child care costs. These supportive policies appear to have stronger effects than more punitive policies, such as reductions in welfare benefits and imposition of sanctions for noncompliance with welfare rules. However, total public investments in supportive assistance remain modest, particularly in the area of child care. Although public child care spending increased substantially during the 1990s, it increased from a very low base, such that relatively few low-income families are receiving subsidies and subsidies are modest— paying less than prevailing market rates in most regions of the country. There is substantial room for expanding public investments in subsidy assistance, and such investments are likely to yield benefits not only in services to children but in promotion of single mothers’ employment as well. Our analysis also suggests several directions for further research. The first is to broaden the set of child care policies considered. States are an important source of child care funding. They also play an important role in determining other aspects of child care policy, from eligibility rules to the stringency of quality regulation. Yet we know relatively little about how states are structuring child care and subsidy services or about how these policy choices influence single mothers’ employment decisions. The second is to take into account how child care policies are determined and to correct, to the extent possible, for the potential endogeneity of child care policies with maternal employment. A third direction for future research is to extend this analysis beyond 1996, when child care and welfare policies were transformed with the passage of PRWORA. State

260

Social Problems and Policy

variation in child care spending and state discretion over child care policy has always been substantial and was increased by the consolidation of federal spending into the CCDBG in 1996. This suggests the value of future work examining child care policy and single mothers’ employment in the post-PRWORA period.

REFERENCES Anderson, P., & Levine, P. (2000). Child care and mother’s employment decisions. In D. Card & R. Blank (Eds.), Findings jobs: Work and welfare reform. New York: Russell Sage Foundation. Berger, M. C., & Black, D. A. (1992). Child care subsides, quality of care, and the labor supply of low-income, single mothers. Review of Economics and Statistics, 74(4), 635–642. Blau, D., & Hagy, A. P. (1998). The demand for quality in child care. Journal of Political Economy, 106(1), 104–146. Connelly, R., & Kimmel, J. (1999). Marital status and full-time/part-time work status in child care choices: Changing the rules of the game. Paper prepared for 1998–1999 ASPE/Census Bureau Small Grants Sponsored Research Conference. Washington, D.C. Ellwood, D. (2000). The impact of the earned income tax credit and social policy reforms on work, marriage, and living arrangements. National Tax Journal, 53(2), 1063–1105. Gelbach, J. (2002). Public schooling for young children and maternal labor supply. American Economic Review, 92(1), 307–323. Han, W.-J., & Waldfogel, J. (2001). The effect of child care costs on the employment of single and married mothers. Social Science Quarterly, 82(3), 552–568. McKernan, S.-M., Lerman, R. Pindus, N., & Valente, J. (2000). The relationship between metropolitan and non-metropolitan locations, changing welfare policies, and the employment of single mothers. Revised version of paper presented at the Joint Center for Poverty Research Conference on Rural Dimensions of Welfare Reform. Washington, D.C. Meyer, B. D.,& Rosenbaum, D. T. (2001). Welfare, the earned income tax credit, and the labor supply of single mothers. Quarterly Journal of Economics, 116(3), 1063–1114. Meyers, M. K., Heintze, T., & Wolf, D. (2002). Child care subsidies and the employment of welfare recipients. Demography, 39(1), 165–179. Michalopoulos, C., & Robins, P K. (2000). Employment and child care choices in Canada and the United States. Canadian Journal of Economics, 33(2), 435–470. Michalopoulos, C., Robins, P. K., & Garfinkel, I. (1992). A structural model of labor supply and child care demand. Journal of Human Resources, 27(1), 166–203. Moffitt, R. A. (1999). Explaining welfare reform: Public choice and the labor market. International Tax and Public Finance, 6(3), 289–315. Powell, L. M. (2002). Joint labor supply and childcare choice decisions of married mothers. Journal of Human Resources, 37(1), 107–128.

Child Care Policy Reform and Employment of Single Mothers

261

Schoeni, R. F., & Blank, R. M. (2000). What has welfare reform accomplished? Impacts on welfare participation, employment, income, poverty, and family structure. Working Paper DRU-2268. Santa Monica, CA: RAND. U.S. General Accounting Office. (1994a). Child care: Child care subsidies increase likelihood that low-income mothers will work. Washington, D.C.: U.S. General Accounting Office. ———. 1994b. Child care: Working poor and welfare recipients face service gaps. Washington, D.C.: U.S. General Accounting Office.

Chapter 17

Explaining Trends in Child Support: Economic, Demographic, and Policy Effects Anne C. Case, I-Fen Lin, and Sara S. McLanahan*

When parents live with their children, they automatically share their income with the children. When parents live apart, income sharing is not automatic, and the nonresident parent often fails to provide for the children. Court-ordered child support is the mechanism through which society attempts to ensure that nonresident parents make financial transfers to their children. The importance of child support has increased dramatically during the past four decades as more and more children have become eligible for child support. In the 1950s, most children lived with both their biological parents from birth to adulthood. Today, over half of all children are expected to live apart from at least one biological parent, usually the father, before they reach age 18 (Bumpass & Lu, 2000). Thus, child support has become an increasingly important policy instrument for reducing economic insecurity among single mothers and their children. In response to growing concern about changes in family structure and increases in poverty among children, Congress began passing laws that were designed to ensure that nonresident parents pay child support. In 1975, Congress established the federal Office of Child Support Enforcement and created incentives for states to establish similar offices. In 1984, it passed a series of amendments requiring states to withhold childsupport obligations in cases of delinquency and to establish legislative

* A. C. Case, I. Lin, and S. S. McLanahan, Explaining Trends in Child Support: Economic, Demographic, and Policy Effects. Demography 2003, 40:1, 171–190. Reprinted with permission of the Population Association of America.

Explaining Trends in Child Support

263

guidelines for setting award levels. In 1988, policy makers went even further by making income withholding automatic and by making guidelines presumptive. States were also required to establish the paternity of all children who were born outside marriage. In 1996, the Personal Responsibility and Work Opportunity Reconciliation Act called for additional child-support enforcement mechanisms and required states to increase their paternity-establishment rates (Garfinkel, Meyer, & McLanahan, 1998). Despite this new legislation, the proportion of eligible children receiving child support has not changed much since the late 1970s. According to Sorensen and Halpern (1999), 30 percent of children received some child-support income in 1976 compared to 31 percent in 1997. To account for the apparent lack of progress in the receipt of child support, analysts have proposed several explanations, including high inflation, increases in women’s economic independence, shifts in the composition of motheronly families, changes in divorce laws, and ineffective policies. Each explanation seems plausible, and each has some empirical support. Nevertheless, after over 25 years of child-support reform and numerous empirical assessments, researchers still lack a clear understanding of the relative importance of economic, demographic, and political factors in accounting for child-support payments. In this study, we used data from the Panel Study of Income Dynamics (PSID) to examine trends in child-support payments over the 30-year period 1968–1997 and to assess the relative importance of the five explanations just described. No other study has investigated payments over such a long period or examined all five explanations together. We took advantage of the natural variation in policy regimes across states and over time to identify the effects of divorce and child-support laws on child-support payments. The findings suggest that political, demographic, and economic forces all exerted downward pressure on child-support payments during this 30-year period, with inflation, the shift to unilateral divorce, and declines in fertility and men’s earnings being more important during the earlier years and decreases in men’s earnings being more important during the later years. These negative forces were offset by the passage of new child-support legislation in the 1980s and 1990s, including numeric guidelines, universal withholding, and genetic testing.

TRENDS AND EXPLANATIONS The most complete information on child support comes from the April Current Population Study (CPS)—Child Support Supplement (CSS), which was introduced in 1979 and has been repeated every other year

264

Social Problems and Policy

since 1982. The supplement asks eligible mothers whether they have a child-support award, how much they are owed, and how much they actually received during the past year. Researchers have used these data to examine trends in the different components of child-support payments— award rates, award levels, and payment rates—since 1979 (Beller & Graham, 1993; Hanson et al., 1996; Robins, 1992). According to this research, award rates and payments for new cases declined between 1979 and 1989 (Hanson et al., 1996), with the greatest declines in award rates occurring between 1981 and 1983 and the greatest declines in payments occurring between 1978 and 1981. Other researchers have used data from the March CPS (which collects information on income) to examine trends in payment rates and whether mothers received any child support (Freeman & Waldfogel, 2001; Sorensen & Halpern, 1999). These researchers have found a similar pattern for the same period. To account for the overall lack of progress in child-support payments, researchers have pointed to one of five different factors: inflation, the shift to unilateral divorce, changes in marital-status composition, changes in men’s and women’s earnings, and ineffective child-support laws. Inflation Inflation is one reason why child-support payments may have declined during the past several decades. Inflation affects payments in two ways: by eroding the value of existing awards and by holding down the amount of new awards in real terms. The first problem, money erosion, occurs because child-support orders are rarely indexed to inflation. Thus, during periods of high inflation, the value of awards declines rapidly. The second problem, money illusion, may occur in times of inflation if judges, lawyers, and parents are not fully cognizant of the cost of purchasing a bundle of goods in the year in which a child-support award is set. Graham (1995) has been a leading proponent of the inflation hypothesis. According to his analysis, inflation accounted for about 90 percent of the decline in new awards between 1978 and 1985. Graham also proposed that persistent money illusion, defined as the failure to take full account of inflation over a period, was responsible for the decline, although he did not test this hypothesis directly. In contrast, Robins (1992) found that inflation accounted for only about 13 percent of the decrease in all (both old and new) child-support awards between 1978 and 1985. He also noted that although inflation was high between 1978 and 1981, it was moderate before and after that period. Finally, Hanson et al. (1996) reported that the trend in the real value of new awards between 1979 and 1990 closely (negatively) mirrored the trend in inflation rates, which is consistent with the money-illusion hypothesis, in which judges (and/or parents) fail to take account of inflation, resulting in large annual declines in the real

Explaining Trends in Child Support

265

value of new awards. They also noted that the hypothesis is difficult to test because the consumer price index (CPI) contains no cross-sectional variation and may be picking up a time trend. Unilateral Divorce The second argument for the lack of improvement in child-support awards and payments is the change in divorce laws. According to this argument, the switch to unilateral divorce, which occurred in most states during the 1970s, may have reduced women’s bargaining power in divorce and therefore their ability to obtain generous child-support awards (Peters, 1986). Under traditional family law, the partner who does not want the divorce has more power than the partner who wants it because both parties must agree for the divorce to occur. Under unilateral divorce, this power no longer exists. Assuming that women are less likely than men to want to divorce (because the economic costs of divorce are higher for women), it follows that the shift to unilateral divorce would have reduced women’s bargaining power and hence the value of their child-support awards. Several studies have examined the effects of divorce law on child support and alimony. Weitzman (1985) and Peters (1986) both found that alimony and child support were significantly lower in states with no-fault or unilateral divorce laws. In contrast, Jacob (1989) found that the effects of no-fault divorce were either modestly benign or neutral for women. Demographic Composition The third reason for the lack of improvement in child-support payments is the change in the marital status of the population of women who were eligible for child support during the 30-year period. In 1976, the vast majority of single mothers (83 percent) were divorced or separated. By 1997, the proportion was just over half (54 percent) (Sorensen & Halpern, 1999). The shift in marital status made it more difficult to obtain child-support awards and reduced the value of the average child-support award. Before a child-support order can be set, a never-married mother must first establish paternity. Therefore, child-support awards are less common among never-married mothers. Furthermore, the average unmarried father is less educated than the average divorced father, and thus his child-support order is likely to be lower. The fathers of children born outside marriage also have less incentive to pay child support because a large proportion of their children receive welfare and can keep only $50 of child support per month. In addition to changes in the maritalstatus composition of the single-mother population, family size also declined during the 30-year period as the baby-boom cohort grew up and

266

Social Problems and Policy

fertility rates decreased. Thus, we would expect the average child-support payment per household to be lower in 1997 than it was 30 years earlier. Several studies have provided empirical support for the demographiccomposition hypothesis. Beller and Graham (1993) showed that a nevermarried mother is less likely to have a child-support award than is a formerly married mother, and the amount of her award is lower. According to their analysis, changes in marital status can explain a large portion of the decline in award rates (the proportion of single mothers with awards) between 1978 and 1985. Hanson et al. (1996) also found that changes in the marital status of mothers who were eligible for child support accounted for much of the decrease in the amounts of awards between 1978 and 1989. With respect to family size, several studies have shown that declines in fertility have contributed to the decline in childsupport payments (e.g., Beller & Graham, 1993; Hanson et al. 1996), but none has highlighted the fertility component or assessed its relative importance. Women’s Economic Independence The fourth explanation for why child-support payments did not show much improvement in the period under study is women’s (mothers’) economic independence. Women’s earnings increased dramatically during the 1970s and 1980s, while the earnings of men with less than a college education declined. According to the independence hypothesis (Robins, 1992), judges responded to the relative improvement in women’s economic position by lowering their expectations about the amount of child support that nonresident fathers should be required to pay. Women’s growing economic independence may also have made men feel less obligated to support their former partners and nonresident children and made mothers more forgiving. The empirical evidence on the independence hypothesis suggests that child-support awards and payments are affected by changes in men’s and women’s earnings. Estimates of the magnitude of this effect, however, are not robust to the data used or to the specification chosen. Using macrolevel data, Robins (1992) found that the increase in women’s earnings accounted for most of the decline in child-support awards between 1978 and 1985. In contrast, Hanson et al. (1996) found much smaller effects, using microlevel data for the period 1978–1989. Graham (1995) also reported smaller effects in his analysis of new awards. Governmental Failure: Ineffective Child-Support Policies The final reason for why child-support payments have not improved is ineffective child-support policies, or governmental failure. According to

Explaining Trends in Child Support

267

this argument, states have been slow to pass or implement child-support policies, and thus the effects of the new legislation on child-support payments have been minimal. Alternatively, child-support policies may have been effective, but their effects are masked by factors such as inflation, shifts in marital status, changes in divorce laws, and the closing of the gap in men’s and women’s earnings. Several researchers have examined the effects of child-support policies on trends in payments and the different components of payments— award rates, award levels, and collection rates (Argys, Peters, & Waldman, 2001; Beller & Graham, 1993; Garfinkel & Robins, 1994). These studies have provided some evidence that child-support policies, such as wage withholding, legislative guidelines, paternity-establishment statutes, and tax intercepts, do have positive effects on payments. However, the studies did not measure the size of the effects or control for unobserved differences across states. More recently, Sorensen and Halpern (1999) used fixed-effects models to examine the effects of child-support laws on the receipt of any payment. They found that six laws—immediate wage withholding, presumptive guidelines, state income tax intercept, in-hospital paternity establishment, directory of new hires, and a $50 pass through—accounted for 58 percent of the improvement in receipt rates among never-married mothers and 29 percent of the improvement among formerly married mothers. Freeman and Waldfogel (2001) took a somewhat different approach to estimating the effects of child-support laws on payments. They argued that a particular law is not as important as the total number of laws on the books. They also noted that child-support laws are not effective unless they are actively implemented. To measure the legal environment, they constructed an index of child-support enforcement that is simply the number of laws a state has on the books. To measure implementation, they used state child-support expenditures (per absentfather family). They found that states with the most laws and highest expenditures also have the highest rates of receipt of child support.

CONCLUSION Child support is an important source of income for single mothers and their children, and federal and state laws play a major role in determining whether and how much child support a family receives. Our study (not shown) indicates that the receipt of child support is determined by multiple factors, only some of which are under governmental control. During the 1970s and 1980s, high inflation, increases in nonmarital childbearing, declines in the earnings of low-skilled men, and the passage of unilateral divorce laws all converged to exert downward pressure on child-support

268

Social Problems and Policy

payments. Until the mid-1980s, the federal government and most states treated child-support obligations as a private matter, and average real payments declined sharply. By the mid-1980s, the situation had changed. Inflation had returned to normal levels, and the federal government had begun to take a more proactive role in enforcing child-support obligations. Consequently, although increases in nonmarital childbearing and declines in the returns to low education continued to exert downward pressure, child-support payments overall began to rise. This trend continued throughout the 1990s. For never-married mothers, a major factor undermining child-support payments receipts has been the decline in the earnings capacity of lowskilled men. This decline has made it more difficult for officials to establish paternity and to collect child-support obligations. For ever-married mothers, inflation has been the major factor behind the decline in childsupport payments, followed by the changes in divorce laws. Nearly 20 years ago, Weitzman (1985) argued that no-fault divorce harmed women by affecting the distribution of marital property. Moreover, our study points out that at least three child-support policies—genetic testing, legislative guidelines, and universal wage withholding—are important determinants of child-support payments. Although these findings cannot be directly compared with those of other studies because of differences in samples, periods, and child-support measures, they are generally consistent with previous research that has shown that withholding and guidelines increase the receipt of child support among ever-married mothers (Sorenson & Halpern, 1999) and genetic testing increases child-support award rates among never-married mothers (Miller & Garfinkel, 1999). Our study extends prior work by including state fixed effects, by examining a longer time frame, and by looking at the amount of child support received as well as payment rates. Our findings also indicate that universal withholding, which targets all eligible children, is more effective than automatic withholding, which is limited to families on welfare. For never-married mothers, genetic testing is the key because unless paternity is established, there can be no child-support award and hence no payment. Numeric guidelines and universal withholding are important for obtaining child support for ever-married mothers, but these policies are less effective for never-married mothers, in part because so few of the latter actually have awards. In addition to highlighting particular policies, our study suggests that inflation and the decline in the wages of low-skilled fathers are important variables in building an effective child-support system. Because inflation can erode the value of awards over a long period, indexing guidelines and awards to changes in the cost of living would make sense. Making sure that policies treat low-income fathers fairly is also important. Research

Explaining Trends in Child Support

269

has shown that fathers who perceive that the system is fair are more likely to pay child support (Lin, 2000) than are fathers who perceive that the system is unfair. Unfortunately, ethnographic studies have suggested that many low-income parents view the system as unfair (Edin, 1995; Waller & Plotnick, 1999). Negative perceptions are due, in part, to the fact that child-support guidelines do not take account of fathers’ irregular employment patterns and, in part, to the fact that child-support payments usually go to the state, rather than to the child, if a mother receives welfare. Such practices are likely to exert downward pressure on future gains in child support.

REFERENCES Argys, L. M., Peters, H. E. & Waldman, D. M. 2001. Can the family support act put some life back into deadbeat dads? An analysis of child-support guidelines, award rates, and levels. Journal of Human Resources, 36, 226–252. Beller, A. H., & Graham, J. W. 1993. Small change: The economics of child support. New Haven, CT: Yale University Press. Bumpass, L., & Lu, H. H. 2000. Trends in cohabitation and implications for children’s family context in the U.S. Population Studies, 54(1), 29–41. Castro-Martin, T. & Bumpass, L. L. 1989. Recent trends in marital disruption. Demography, 26, 37–51. Deaton, A. 1997. The analysis of household surveys: A microeconometric approach to development policy. Baltimore, MD: Johns Hopkins University Press. Edin, K. 1995. Single mothers and child support: The possibilities and limits of child support policy. Children and Youth Services Review, 17(1–2), 203–230. Folk, K .F., Graham, J. W., & Beller, A. H. 1992. Child support and remarriage: Implications for the economic well-being of children. Journal of Family Issues, 13, 142–157. Freeman, R. B., & Waldfogel, J. 2001. Dunning delinquent dads: The effects of child support enforcement policy on child support receipt by never married women. Journal of Human Resources, 36, 207–225. Garfinkel, I., McLanahan, S. S., & Robins, P. K. (Eds.) 1994. Child support and child wellbeing. Washington, DC: The Urban Institute. Garfinkel, I. McLanahan, S. S., Meyer, D. R., & Seltzer, J. A. (Eds.) 1998. Fathers underfire: The revolution in child support enforcement. NY: Russell Sage Foundation. Garfinkel, I., Meyer, D. R., & McLanahan, S. S. 1998. A brief history of child support policies in the United States. In I. Garfinkel, S. S. McLanahan, D. R. Meyer & J. A. Seltzer (Eds.) Fathers under fire: The revolution in child support enforcement (pp. 14–30). NY: Russell Sage Foundation. Garfinkel, I., & Robins, P. K. 1994. The relationship between child support enforcement tools and child support outcomes. In I. Garfinkel, S. S. McLanahan, & P. K. Robins. (Eds.) Child support and child well-being (pp. 133–171). Washington, DC: The Urban Institute.

270

Social Problems and Policy

Graham, J. W. 1995. A comment on ”Why did child support award levels decline from 1978 to 1985?” by Philip K. Robins. Journal of Human Resources, 30, 622–632. Hanson, T. L., Garfinkel, I., McLanahan, S. S., & Miller, C. K. 1996. Trends in child support outcomes. Demography, 33, 483–496. Hill, M. S. 1992. The panel study of income dynamics: A user’s guide. Newbury Park, CA: Sage. Jacob, H. 1989. Another look at no-fault divorce and the post-divorce finances of women. Law and Society Review, 23, 95–115. Lin, I.-F. 2000. Perceived fairness and compliance with child support obligations. Journal of Marriage and the Family, 62, 388–398. Miller, C., & Garfinkel, I. 1999. The determinants of paternity establishment and child support award rates among unmarried women. Population Research and Policy Review, 18, 237–260. Peters, H. E. 1986. Marriage and divorce: Informational constraints and private contracting. American Economic Review, 76, 437–454. Robins, P. K. 1992. Why did child support award levels decline from 1978 to 1985? Journal of Human Resources, 27, 362–379. Sorensen, E., & Halpern, A. 1999. Single mothers and their child support receipt: How well is child support enforcement doing? Working paper. The Urban Institute, Washington, DC. Waller, M. R., & Plotnick, R. D. 1999. Child support and low-income families: perceptions, practices, and policy. Working paper. Public Policy Institute of California, San Francisco. Weitzman, L. J. 1985. The divorce revolution: The unexpected social and economic consequences for women and children in America. New York: Free Press. Yun, K.-R. 1992. Effects of child support on remarriage of single mothers. In I. Garfinkel, S. S. McLanahan, & P. Robins (Eds.). Child support assurance: Design issues, expected impacts, and political barriers as seen from Wisconsin (pp. 315–338). Washington, DC: The Urban Institute.

Chapter 18

Effective Child Support Policy for Low-Income Single Parents Maureen R. Waller and Robert Plotnick*

The high incidence of divorce and nonmarital childbearing in the United States means that at least half of children born today will probably spend part of their childhood in a single-parent family and that many of these children will be eligible for child support. Child support policy seeks to ensure that parents who live apart from their children contribute to their financial support. This objective serves several important public policy goals: reducing poverty and financial insecurity among children and their custodial parents; preventing single-parent families from entering the welfare system; helping families on welfare leave more quickly by substituting private, parental income for public welfare; and reducing public spending for welfare. Child support policy also seeks to affirm the widely held belief that parents are morally and socially obligated to support their children. Over the past quarter century, a series of major policy reforms has been implemented in pursuit of these goals. Although these reforms have brought improvements in enforcing child support obligations, the child support system is still widely criticized for poor performance. In 1993, only 60 percent of custodial mothers in the United States had a child support order, a figure that has remained nearly constant since 1978. For never-married mothers, the chances of having an order were only 44 percent (Committee on Ways and Means, 1998, pp. 605, 608). Of parents with orders in 1993, 61 percent actually received payments, and only 30 percent received full payment. Of the total child support due,

* Maureen R. Waller and Robert Plotnick (2001), Effective Child Support Policy for LowIncome Families: Evidence from Street-Level Research. The Journal of Policy Analysis and Management 20(1): 89–110.

272

Social Problems and Policy

38 percent was not received (Committee on Ways and Means, 1998, pp. 605, 608; Office of Child Support Enforcement, 1998, Tables 32, 33). Welfare time limits and state policies geared toward diverting parents from Temporary Assistance to Needy Families (TANF) mean that child support will assume even greater importance for single-parent families. Consequently, the child support system’s poor performance becomes more problematic following the passage of TANF. This article seeks to understand why the child support system is ineffective for many lowincome families. It does so by analyzing findings from recent studies on how child support regulations affect the economic situations of lowincome, unmarried parents and their children, the relationships between custodial and noncustodial parents, and the willingness of both parents to cooperate with child support authorities. Because the various studies’ findings represent common perceptions among unmarried parents and are consistent across research sites, they provide significant insights into the difficulties parents experience in the child support system. This synthesis suggests that the reluctance of low-income parents to comply fully with the official child support system’s rules is an important part of the explanation for weak enforcement. These rules are intended to facilitate and formalize financial interactions between unmarried parents. Lowincome, unmarried parents generally endorse the concept of child support and believe that there are circumstances in which participation in the formal system is appropriate. However, many such parents prefer private, informal agreements for support and at times do not comply with child support regulations they perceive to be unfair, counterproductive, or punitive. Child support legislation was developed mainly to apply to families with divorced fathers working full-time (Sorensen and Lerman, 1997). Qualitative studies suggest that legislation framed with this model in mind often clashes with the social and economic situations of many lowincome parents. In particular, the rules of child support and welfare agencies interact so that parents often find it difficult to comply with them, even when they wish to do so.

OVERVIEW OF THE U.S. CHILD SUPPORT SYSTEM The modern era of child support policy began in 1975 with the Title IV, Part D amendment to the Social Security Act. This landmark law established the partnership in child support between federal and state governments that remains the basis of current policy. Since 1975, Congress has revisited child support policy several times. The Child Support Enforcement Amendments of 1984 required states to adopt expedited procedures for establishing paternity and support orders, to develop guidelines for setting support levels, to establish income withholding and

Effective Child Support Policy for Low-Income Single Parents

273

other means of ensuring compliance for noncustodial parents who fall behind in their payments, and to offer enforcement services to nonwelfare families. The Family Support Act of 1988 strengthened the 1984 amendments. In the 1993 Omnibus Budget Reconciliation Act, Congress required states to develop a simple administrative process for unmarried fathers to declare paternity voluntarily. The Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, although best known for its changes in cash assistance policy, included several important measures aimed at improving paternity establishment and other aspects of child support enforcement. Two broad policy objectives have animated this series of reforms. First, a better child support system has come to be viewed as a major part of a national strategy for reducing poverty and welfare use among singleparent families. Second, in response to a variety of problems attributable to locally administered child support, which was the norm before 1975, federal legislation has increasingly sought to reduce administrative discretion, improve equity and compliance, and coordinate enforcement across states (Garfinkel, 1992). Today’s child support system involves a close partnership between the federal and state governments. The major services provided by the child support system include opening child support cases, locating noncustodial parents, establishing paternity, and establishing, enforcing, and modifying child support orders. When a custodial parent starts to receive benefits from TANF, the designated agency in her state (or county) must automatically open a TANF child support case. Special rules apply for such cases: • TANF recipients must cooperate with the state in locating the noncustodial parent, establishing paternity, and obtaining support payments. Failure to cooperate in establishing paternity will result in at least a 25 percent reduction in aid and could lead to removal from the TANF rolls. • The custodial parent must assign all rights to child, spousal, or medical support to the state up to the amount of aid received. This includes all current and past-due support and continues as long as a family is receiving TANF. If the parent will not assign her rights, her TANF and Medicaid benefits will be dropped. The children will still receive TANF and Medicaid, but a payee, rather than the custodial parent, will receive the check. • Most states use the entire monthly support payment to reimburse state and federal governments for welfare payments. The payment does not help increase the family’s standard of living. In about a third of the states the monthly “pass-through” is $50. Only Wisconsin allows full pass-through.

274

Social Problems and Policy

• No credit is given for in-kind payments made directly to the custodial parent (e.g., clothes, food, school supplies, toys, etc.). As more fully discussed below, the tight linkages between welfare and child support policy created by these special rules appear to be major reasons for low-income parents’ reluctance to cooperate with the official child support system. When the child support agency opens a case, it asks the custodial parent to help locate the noncustodial parent. The agency may also obtain help from local and state agencies, local organizations, the State Parent Locator Service, the registry for all newly hired employees, and the Federal Parent Locator Service. The federal service is particularly helpful when the parents live in different states. In 1996, agency officials found 67 percent of all noncustodial parents requiring location (Office of Child Support Enforcement, 1998, Tables 35, 41). When an alleged father is located, he can acknowledge or dispute paternity. If he disputes paternity, blood and other scientific tests and, if needed, a court, will decide paternity. States also administer in-hospital, voluntary paternity programs. In 1996, child support agencies established paternity for 718,152 children born to unmarried women. This equaled 57 percent of all births to unmarried women in 1996, or about double the percentage in 1987 (Committee on Ways and Means, 1998, pp. 649, 651; Ventura et al. 1998, Table 17). A support order requires the noncustodial father to financially support his children and sets the payment. To comply with federal laws aimed at rationalizing procedures for setting orders and reducing judicial discretion, states use formal guidelines to establish each order. Some states’ guidelines impose support obligations on low-income fathers that are high percentages of their incomes. In 1997, a noncustodial father of two with earnings of $500–$750 per month could plausibly have faced a monthly support order equal to 40 percent of his income in nine states and 20–39 percent of his income in another 20 (Committee on Ways and Means, 1998, pp. 563–565). Until 1988, it was generally assumed that the noncustodial parent would make regular payments to the custodial parent or the child support agency. The Family Support Act decisively altered this approach by requiring immediate withholding for all new or modified orders, starting in November 1990. Increasingly stringent withholding led to a tripling of the support collected via this route between 1989 and 1996 (Committee on Ways and Means, 1998, p. 572). States and the federal government have many other enforcement techniques as well. Examples of these techniques include placing liens on noncustodial parents’ property; intercepting federal and state income tax refunds and unemployment insurance payments; attachment of lottery winnings; restricting or suspending driver’s, professional, and occupational licenses; hiring private collection agencies;

Effective Child Support Policy for Low-Income Single Parents

275

and bringing criminal or civil cases against noncustodial parents who refuse to pay (Committee on Ways and Means, 1998, p. 571). Nonetheless, custodial parents do not receive nearly 40 percent of the support owed them (Committee on Ways and Means, 1998, p. 608). Under current law, reviews and adjustments of support orders are no longer mandatory. If either parent asks for a review or, for TANF cases, if the state asks, the child support agency must review and adjust support orders at least once every three years. States must also inform parents of their review and adjustment rights at least once every three years. If review indicates grounds for adjustment, the state must do so. The federal Bradley Amendment has prohibited retroactive modification of past due support payments in virtually all cases. Past due payments generally cannot be forgiven even if the noncustodial parent is unemployed, in jail or otherwise unable to earn income.

EVIDENCE FROM QUALITATIVE RESEARCH Although reliable and extensive survey data exist on single mothers, survey information on nonresident fathers has been limited. Many lowincome, nonresident fathers are underrepresented in national surveys because they are not sampled. Others who are included in surveys sometimes do not report that they have children (Garfinkel, et. al., 1998, pp. 31–32). As a result, in the National Survey of Families and Households—perhaps the best data set for identifying nonresident fathers—as many as 3.6 million fathers appear to be missing from the sample (Garfinkel, et. al., 1998, p. 39). In response to this issue, studies have used qualitative methods to effectively investigate the lives of low-income, unmarried mothers and fathers. Qualitative methods allow researchers to successfully sample nonresident fathers, to identify issues that are salient to unmarried parents, and to explore the meaning of issues in greater depth than is possible through closed-ended interviews. Researchers concerned with low-income, unmarried parents’ interactions with the child support system have spoken to parents about how the child support system affects their families, work, financial situations, relationships as couples, and other aspects of their lives. These studies provide insight into why the child support system has often been ineffective for low-income families. All seven studies examine the experiences of low-income, unmarried parents living in urban areas. All investigations were also conducted in the late 1980s and early 1990s, before PRWORA passed. Because PRWORA did not change the fundamental structure of the child support system, the authors believe the problems identified by parents in these studies, and issues they raise for public policy, remain salient. African American parents are the largest group of respondents in

276

Social Problems and Policy

each study. Achatz & MacAllum (1994), Edin (1995), Johnson & Doolittle (1995), Sullivan (1992), and Waller (1996) include white parents. Edin (1995) and Johnson & Doolittle (1995) also include a small number of Hispanic parents. This racial composition reflects the fact that a disproportionate number of African American mothers are single parents. Because qualitative methods trade depth for breadth, most of these studies have fairly small samples. None of the samples was randomly selected. Four studies used multiple techniques (including institutional and snowball sampling) to generate diverse samples (Edin; Furstenberg; Sullivan; Waller). Three selected their respondents from programs designed for low-income fathers with child support obligations (Achatz & MacAllum; Johnson & Doolittle; Sherwood). Because all of the fathers in these programs had established paternity and had direct experiences with the child support system, they may differ from other low-income, nonresident fathers. About 48 percent of poor mothers and 56 percent of poor, never-married mothers did not have child support awards in 1993 (Committee on Ways and Means, 1998, p. 605). Thus, many of the fathers of their children may not have any contact with the child support system. Similarly, other studies may include respondents with characteristics that differ from the general universe of low-income, unmarried parents. “DEADBEAT” AND RESPONSIBLE FATHERS Child support policy is intended to encourage responsible behavior among unmarried fathers. The studies analyzed here suggest that parents distinguish between “deadbeat” and responsible dads but do not believe that men who make formal child support payments are more responsible than those who provide informally. A father Waller (1996) interviewed explains: A lot of dads are deadbeat dads. A lot of mothers are deadbeat mothers. But they call them deadbeat dads because they’re not paying child support to the establishment. You know, a lot of people, a lot of people don’t like paying child support. A friend of mine just got out of jail day before yesterday. Spent ten days in jail for child support, and he does everything in the world for his son. He just doesn’t like the idea of [them] taking the money (p. 301).

Qualitative studies suggest that low-income, unmarried parents hold strong, collective beliefs about paternal responsibility and endorse the principle of child support (Furstenberg, 1992; Sherwood, 1992; Sullivan, 1992; Waller, 1996). However, many parents believe that formal child support is appropriate only when private agreements cannot be established or maintained, or when fathers do not accept their responsibility voluntarily. For example, if the parents are in a romantic relationship,

Effective Child Support Policy for Low-Income Single Parents

277

particularly if they are living together and sharing expenses, they usually do not believe participation in formal child support is warranted. Similarly, if low-income fathers participate actively in their children’s lives and make a significant effort to contribute to their support, parents often prefer not to establish a formal agreement (Waller, 1996). Given that unmarried parents agree with the idea of child support, the remainder of this article explains why this preference for informal support is so strong. FINANCIAL DISINCENTIVES CREATED BY ASSIGNING CHILD SUPPORT TO THE STATE All of the studies report that parents object to the requirement that women receiving welfare must sign over their rights to child support to the state. Parents argue that their child support payments do not increase their children’s standard of living and that their needs are not met with the “pass-through.” Though parents generally understand that the father’s payment is used to offset the costs of welfare, they do not consider the regulation fair. A father in Sherwood’s focus group explains “[i]t’s the welfare. . .They take half the money that you give them. I’m paying $132 and she only get like maybe $50” (Sherwood, 1992, p. 62). Similarly, a father Waller (1996) interviewed observes: The money doesn’t go to the kid. It’s not like you’re buying the kids something. The money goes to them because they pay that girl some welfare. So all it is nothing but a payback situation. You know what I mean? You’re giving us money to pay back what we had to give her. It’s not like you say— alright, I know this $35 will buy my son some Pampers. It ain’t like that (p. 279).

The financial disincentives facing a low-income father can be substantial. If he pays $200 per month yet his children gain only $35 as a result of his contribution, the effective tax rate is 82.5 percent. Even if he only pays $100 per month, the effective tax rate is still 65 percent. A strong economic disincentive also exists for a mother if she believes her child’s father would make direct contributions greater than $50 a month. Achatz & MacAllum (1994, p. 81) found that most fathers in the program they evaluated reported spending more than $50 each month on their child. Edin & Lein (1997, p. 44) indicate that mothers on welfare reported receiving on average $39 a month in cash from fathers in addition to in-kind support. The economic disincentive created by assigning child support rights to the state sometimes leads mothers and fathers to plan cooperative arrangements to circumvent the financial penalty they perceive. Fathers said they often gave priority to their children’s concrete needs for such provisions as clothing, diapers, and food rather than to child support

278

Social Problems and Policy

payments. Therefore, some fathers tried to provide for their children’s needs while keeping child support enforcement at arm’s length by making sporadic payments to the state. Others decided to ignore the child support order and give things directly to their children when they have extra income. As one father in Sullivan’s (1992, p. 16) project stated, “She wasn’t seeing nothing. And my son wasn’t seeing nothing. So I wasn’t paying nothing.” Mothers also resisted child support regulations in order to continue receiving financial assistance directly from the father (Edin, 1995; Waller, 1996). Edin (1995, p. 214) finds that more than half of the mothers she interviewed engaged in “covert non-compliance” by lying about the father’s identity or hiding crucial information about him to avoid establishing a support order. She notes the majority of mothers who received “covert” cash or in-kind payments from the father reported doing so because they could receive more through informal support. I actually discussed with him that I really don’t want (to report him to child support, because I’m probably better off with him paying on the side.) It is better if he doesn’t pay through them. Because they only give you $50 no matter what he pays (Edin,1995, p. 216).

In addition to mothers withholding information, Waller also documents examples of mothers backing the father’s decision not to put his name on the birth certificate, to ignore notices to appear in court for the hearing, or to make informal payments. Not all women had information such as the father’s address, place of employment, or social security number with them at the time of the intake interview. Many women were in contact with the father and could presumably obtain this additional information. However, they did not attempt to report it to child support unless they were pursuing an award. Both Waller and Edin report that by withholding some information about the father, mothers could satisfy the formal requirements of the child support system while not actively seeking an order. To work out an informal arrangement for support, a mother must be convinced that the father is making a serious effort to cooperate with her and to contribute financially. Mothers who believe that fathers are being irresponsible or uncooperative use the threat of reporting the fathers to child support authorities as a “negotiation tool” to garner informal support, induce them to be more responsible, and thereby bypass the formal system (Edin, 1995). In this way, the formal system gives a woman more power in negotiating with the child’s father, even if she does not establish a formal support order. The mother pursues this unilateral strategy without the cooperation or consent of the father. After attempting to negotiate with the father, some mothers later changed their mind about child support after they saw that the father was

Effective Child Support Policy for Low-Income Single Parents

279

not willing to take care of their children in the way they had expected (Edin, 1995; Waller, 1996). [Sometimes] you have to [put them on child support], ’cause these men don’t realize that sugar is sweet [or how good they have it]. [Before I decided to put him on the system], all I did was call him and say, “Can you help me out this week?” But [he took advantage of the situation] and kept telling me: “I don’t have to do this and that.” I said, “I need a big box of Pampers to take to the baby-sitter. He said, “Yeah, all right.” I found out that he didn’t do it. So I said, “That’s that. That’s the end,” and I put his papers in for child support (Edin, 1995, p. 219).

Men recognize women have the power to pursue child support and take their threats to do so seriously. One father observed that women who withhold information can contact their caseworker at any time to pursue child support: You know if you don’t live up to your expectations from the agreement you and this lady made, first thing she gonna do is run down there and say “I know so and so, I know where he work at” (Waller, 1996, p. 314).

Achatz & MacAllum (1994) similarly report that to keep mothers happy and deter them from seeking formal child support orders, fathers attempt to maintain friendly relationships and make voluntary financial contributions. Like I just stopped dealing with her and she be goin’ around telling all her little girlfriends what she gonna do to me. . . And the [friends] told me well just go talk nice to her. . .[and] pretend like we getting back together and stuff, you know, so she don’t turn me in (Achatz & MacAllum, 1994, p. 84–85).

Mothers who do not want the father involved with them or their children may also pursue unilateral strategies to evade cooperation with the child support authorities. Edin found that almost half of mothers who did not cooperate with child support mentioned issues such as fear of reprisal from abusive fathers, desire for exclusive control of the child, and beliefs that they did not have a legitimate claim for support from the father. Waller also found that women were reluctant to identify fathers if they believe contact with them would be detrimental to their families. FAMILY CONFLICTS CREATED BY RULES MANDATING COOPERATION Although the formal system gives mothers a “negotiation tool” to exert leverage against fathers, that tool can be a double-edged sword. The following example illustrates how conflicts between parents develop when men believe women are using the formal child support system as a weapon to get back at them. A father Waller interviewed explains that

280

Social Problems and Policy

some men interpret being called into court for child support as a hostile gesture on the part of mothers. They don’t know that once you do that, that puts a whole distance between you and the baby’s father. Now the baby’s father say, “So, you want to go that route? O.K. Then I’ll give them $35 a month, but you can’t get another dime from me for nothing.” Now you never know, this guy might come across this amount of money doing this or this amount doing that. Instead of giving it to you and your child—well here’s your $35. You know what I’m saying? So sometime the girl don’t know and it hurts them more than it helps (Waller, 1996, p. 296).

Many parents suggest that child support rules can pit mothers against fathers and create or exacerbate conflict in their relationships (Achatz & MacAllum, 1994; Sullivan, 1992; Waller, 1996). These conflicts can make already difficult parenting arrangements more antagonistic and may lead to their dissolution. The welfare system requires women on welfare to establish paternity for their children and initiate the process of collecting support from lowincome fathers and limits the pass-through. One might think that such legal limits to discretion on the part of women would reduce interpersonal conflict. Why in fact, does conflict arise? From the accounts of parents in Waller’s (1996) study, conflict seems to develop for the following reasons based, in part, on inadequate information and feelings of “gender mistrust” (Furstenberg, 1992, 1995). First, a minority of men do not understand that women are required to identify them as a condition of receiving welfare. Second, fathers may blame mothers for applying for welfare and creating their obligation to the state. Similarly, mothers may attribute their reliance on welfare to the failure of the father to support his children. Third, parents may not understand that the state keeps all but $50 of the child support payment. This leads some women to believe that men are only paying $50 and some men to believe that women receive the full payment and spend the money on themselves. Fourth, women sometimes have room to maneuver within the child support system and can decide how vigorously to pursue child support. When faced with a child support order, fathers may choose to withhold support or restrict contact. Finally, the economic demands put on poor fathers by child support, particularly when they build up large arrearages, add additional strain to their relationships. FORMAL PAYMENTS VERSUS DIRECT OR IN-KIND PAYMENTS Although parents sometimes did not comply with child support regulations, these studies suggest that they strongly endorse the belief that fathers have an obligation to support their biological children and be

Effective Child Support Policy for Low-Income Single Parents

281

involved in their lives. All studies indicate that communities recognize inkind contributions such as diapers, toys, clothing, and shoes as valid expressions of this paternal obligation. Achatz & MacAllum (1994, p. 76) argue that most fathers preferred to purchase items for their children because “they are visible symbols of responsible fatherhood in the community” and because “they are tangible and gratifying.” In addition, the authors argue that direct support makes fathers feel they have more control over how the money is spent. The issue of control is particularly important for fathers who distrust their child’s mother. Parents also describe strong emotional grounds for preferring direct support. A father in Waller’s study points out that children have difficulty comprehending child support, particularly when it is an add-on to the mother’s welfare check. It would be a whole lot better for the kid, for the dad, for the mother, if the money was coming straight from him. . . The child would understand, ’cause eventually the child as it gets older knows what child support is. Knows that his father ain’t been around. So it’s like, “Damn. My father don’t buy me nothing, but he pay child support.” Who wanna say that? What kid wanna grow up knowing—well my father pay child support, but he don’t buy me nothing. . . A child would rather have his father bring him five pair of jeans and some sneakers and some shirts than a check in the mail. . .the material stuff, [at] a child’s age is a whole lot more than paper. . . It would show more love (Waller, 1996, p. 283).

To some fathers, formal child support represents a “forced” payment rather than an authentic expression of paternal love (Sullivan, 1992; Waller, 1996). A father in Sullivan’s (1992, p. 16) study argued “[m]y child should not have to grow up with something in the back of his mind: Somebody had to force dad to give me. If only he would have freely given.’“ Fathers sometimes use an anti-state rhetoric to express opposition to compulsory payments. A father Waller (1996, p. 284) spoke with asked, “Why do I need the government to tell me that I should take care of my child when I know for a fact that I need to?” And a father in Achatz & MacAllum’s study stated, “My girl don’t need to take me to court so a judge could say I gotta kick up some cash for my child. . .There’s no need for that because I’m doin’ it on my own. I don’t understand why the system do that” (1994, p. 75). Sullivan notes that statements such as these may reflect a general belief that the courts should not interfere in their families. Some women agree with this assessment of child support, regarding “forced” child support as both tainted and unreliable because it does not derive from an emotional bond. One mother Waller interviewed said that several people had tried to encourage her to contact the child support office to expedite the process.

282

Social Problems and Policy

But that’s my baby. And when I do that I feel as though I’m forcing him to take care of his child, forcing him to love his child. And I’m not going to force him. I got love for everybody for that girl. And for me the way I feel to go through the system to force him to take care of his child is like he don’t love her (Waller, 1996, p. 281–282).

Parents may jointly seek to avoid participating in the formal system because of their belief that child support enforcement undermines their efforts to establish cooperative parenting arrangements based on an emotional commitment. A mother in Edin’s study described her reasons for withholding information about the father. It’s nice to have them contribute financially but if they’re only going to contribute financially and they are not going to be a father [then] you are not winning. If he gives you money for a child and he’s not going to be a father, the child is losing. If [he] can’t give, you is still getting a father, you know what I’m saying? The parenting part of it. And [with] Public Aid, that is something they take away from you (1995, p. 221).

Similarly, a mother in Waller’s study who lived with the father before he was sent to prison explains why she prefers his involvement with the children to pursuing child support: [In many cases] it’s a good idea. But in many cases, it’s not a good idea ’cause it will cause a conflict between you and the baby’s father. Right now welfare is trying to take him to court for child support. But what it all boils down to is if he’s going to be here with me, I’m getting more out of him being with me. It might not be exactly financially, but as far as raising the kids, you can’t put a price on that. So, I’m getting more out of him being here, than not being here and trying to pay child support (Waller, 1996, p. 294).

The compulsory nature of cash child support fosters resentment and distrust that may be directed toward the other parent instead of the state. In such cases, unilateral responses are observed rather than bilateral strategies of nonparticipation. Mothers mention that conflict may arise when the father begins to pay child support and stop supporting them informally or doing the “extras” for their children. Fathers may see the situation from a different perspective. For example, one father Waller spoke with described his dilemma. If he pays child support, he cannot afford to buy things his children request. Therefore, he feels guilty, and avoids spending time with them. He also believes his child’s mother may spend the money on herself rather than on their child. But, if he provides in-kind support instead of making child support payments, he risks arrest. He claims he has already been arrested three times for failure to pay support. Recounting a conversation with his mother and sisters about this, he explains: My mom and them are like this:

Effective Child Support Policy for Low-Income Single Parents

283

“Why don’t you stop buying them stuff? The courts can handle that. Whatever you decide to buy them, take that money down there to the courts.” But it still gets to the point of: What about my kids? ’Cause, you know, kids can talk. Kids can walk up and say, “Dad, can you buy me this?” And they know if I can afford it, they know I’m supposed to get it for them. But then if I be like “Uh uh, I can’t get that. Your mom’s supposed to take care of that”. . . Then the kid be kind of upset and then it distracts you. It makes you feel bad to tell one of your childs [that] when you know you can get it for them (Waller, 1996, p.284).

Finally, some evidence indicates that requiring fathers to make cash payments reduces their motivation to work in the formal sector. A father Achatz & MacAllum (1994) interviewed describes why in-kind contributions may increase men’s desire to work and to buy things for their children while formal child support may reverse those incentives. When the baby first came, I didn’t have no job. . .no income, so I wasn’t a good provider at first. But then when I got me a job and started buying things for the baby, like little outfits and stuff, and I seen the baby wearing them, well, that really got to me. . .I wanted to work even harder so I could give her more. . .But then when the government started taking all my money for the so-called child support, I didn’t have any change left over after paying all my bills to buy anything for my child—and like it just made me feel less like a father. . .My daughter don’t know that I pay the government for her support. . .She gonna grow up thinkin’ I’m a deadbeat father because I don’t buy her nothing. . .After I found out how that child support worked. . .it just made me feel like not working harder because the government just takes that extra money for themselves. . .That system they got. . . don’t make any sense to me (Achatz & MacAllum, 1994, p.78).

PROBLEMS CREATED BY ENFORCEMENT PRACTICES After low-income parents become involved with the formal system and a support order is established, concerns about how the system enforces support orders come to the fore. Mothers often perceive it as ineffective in enforcing their rights to support. Fathers become frustrated with the system’s insensitivity to their changeable economic circumstances and its use of criminal sanctions to enforce compliance. These perceived problems with the enforcement process are likely to contribute to poor parents’ reluctance to participate in the formal system in the first place. Ineffective Enforcement from Women’s Viewpoints Studies that document the responses of low-income women suggest that they often view the child-support system as ineffective and unresponsive (Edin, 1995; Furstenberg, 1992; Waller, 1996). Furstenberg

284

Social Problems and Policy

found that women commonly objected to the inefficiency of the childsupport system. Problems these mothers mentioned include the difficulty of filing a claim, the inability of the system to collect payments from men, the inability or unwillingness to pursue men who evade the system, and the impersonal nature of the child support agency. Mothers were frustrated that they could not talk or write to anyone in the child support agency about these concerns and had little confidence that they would be addressed (Furstenberg, 1992, pp. 52–53). For example, one mother said she had written a letter to the child support agency six months ago and felt frustrated that nothing had happened: “I know he works. . . He owes it to my child. But like you [the other respondent] said, the system sucks because they are not doing anything” (Furstenberg, 1992, p. 52). Edin (1995) and Waller (1996) also found that mothers expressed frustration about their cases being held up in the child support system. Inappropriate for Poor Fathers’ Economic Circumstances In interviews and focus groups, fathers say that a major problem they face is the system’s inability to recognize or respond to their economic circumstances. Many unmarried fathers of children receiving welfare have low skills, lack stable employment, and may not have sufficient income to pay child support without further impoverishing themselves or their families (Garfinkel et al., 1998; Mincy & Sorensen, 1998; Sorensen & Turner, 1998). Child support legislation was developed on the model of a divorced father with full time employment. Sorensen & Lerman (1997, p. 4) observe “child support policy relies on enforcement tools that assume all noncustodial fathers can afford to pay child support” but are unwilling to do so. The data raise serious questions about the validity of this assumption. The qualitative studies indicate that fathers have problems with paying regular support when they have irregular employment. Because their jobs were often part-time, temporary, or low-paying, they find it hard to make child support payments and meet their own basic expenses at the same time. Unemployed fathers argue that the child support system is least understanding of their circumstances, asking rhetorically, “How are you expected to survive?” Fathers were often incredulous that they could accumulate large debts while being out of work (Johnson & Doolittle, 1995; Waller, 1996). In some cases, fathers’ orders represent a high proportion of their income. A father Sullivan spoke with explained, “They sent me a court order to pay like $600 a month. I don’t even make that much every two weeks, and I wasn’t planning on paying something I don’t have” (Sullivan, 1992, p. 15). Many fathers also faced large arrearages on their

Effective Child Support Policy for Low-Income Single Parents

285

debt as well as the interest that had accrued on these arrearages during periods of nonpayment. Fathers who could not make their payments and had no hope of paying off their arrears felt they were in an impossible bind. A father Sherwood spoke to said the arrears on top of high payments “is what’s killing us.” Child support, you know, it’s still high. . .we’re not working. Why don’t they work with us. . . If they want to work with us so much, why don’t they try and lower that so we can start work and it won’t be so hectic. It’s still going up (Sherwood, 1992, p. 62).

When fathers are unemployed, awards may be based on imputed income, assuming full-time work at the minimum wage. Furthermore, rather than being set when paternity and the child support order are established, child support awards are often set retroactively, which may be years after the child was born and received assistance. An award usually does not take into account direct support given to the child before the award was set or the father’s income at that time (Roberts, 1999; Sorensen & Lerman, 1997; Sorensen & Turner, 1998). As a result, fathers may have child support debts for periods in which they lived with their child’s mother and helped support the household (Roberts, 1999). A father in Johnson and Doolittle’s study described how he accumulated substantial arrearages during a period in which he was supporting his children and they received welfare: I owed $28,000 [arrearage]. When they first started garnishing my wages, they’re telling me, they don’t care nothing about why are you owe them, it’s just that they’ve been paying your kids and paying your ex-girlfriend for your kids, or whatever (Johnson & Doolittle, 1996, p. 23).

Interviews with fathers suggest that child support enforcement practices that assume fathers are absent from the family—either because they are divorced or do not live with their children—may also undermine relationships between unmarried couples. A father Waller (1996) interviewed explained that in his last run-in with the child support system, he was working a night shift at a local hospital when he was arrested for an outstanding warrant for child support. This father says that he fell into arrears during a time when he lived with his children and most of his income went to purchase things for them and to pay for household expenses. Do you know that judge stood in front of my face and told me “I don’t care where you live at, you better move back in with your mom, ’cause I’m taking half your money.” He got no right to tell me that. You know what I’m saying? If I’m living with my kid’s mother, I got a roof over my kid’s head. You should have said, “Damn, if you’re still with her and you’re living with

286

Social Problems and Policy

her and helping your kids out, well, OK”. . .’Cause you’re not making no money like a doctor or lawyer so that somebody can take half your net pay. Heck no, come on man (Waller, 1996, p. 286).

Another important issue fathers mentioned frequently, but one not part of the enforcement process, is concern over visitation rights. As Achatz & MacAllum (1994, p. 88) note, fathers did not understand that establishing a child support order does not formalize their rights to see their children. Johnson & Doolittle (1996, pp. 27–28) observe that many fathers perceived the court to be extremely biased toward women in these matters. The fathers they spoke with suggested that they were given less access when they could not offer economic resources. Furstenberg (1992), Sullivan (1992), and Waller (1996) also note problems men report with mothers blocking access to their children. Fathers pursue various unilateral responses to the enforcement of regulations they perceive to be inflexible and unfair. Fathers who feel intimidated or overwhelmed by child support enforcement may ignore child support orders and accumulate substantial arrearages (Sullivan, 1992; Waller, 1996). When fathers receive notice of their child support obligation or begin to pay it back, they are often thousands of dollars in debt. Sullivan recounts the experience of one father who built up arrearages while he was in jail. When I was doing time, there was no other means for my wife with the kid. At the time, to get support was to go to welfare, and that’s what she did. A certain amount of time went by, ten years or better, I got a letter from welfare stating I owed them so much money. I never answered the letter. It was a couple of thousand dollars and until today I haven’t gotten bothered yet. When it does happen, I don’t know what to do (Sullivan, 1992, p. 27).

Waller (1996), Johnson & Doolittle (1996), and Furstenberg (1992) report that some fathers quit their job when they discovered how much of their wages were garnished. Achatz & MacAllum (1994) also found that some fathers who reduced their work in the formal economy tried to generate more income in the underground economy through under-thetable jobs, selling drugs, stealing, and gambling. It don’t seem like it make any difference whether I get a good job or not. . . the government just be so greedy. It seems like the government is just forcing me to hustle because that’s the only way I can take care of my needs plus the needs of my child. . .Whatever I hustle is mines. . .The government can’t tell me how to spend it because they don’t even know about it (Achatz & MacAllum, 1994, p. 91).

Fathers’ second major objection to the enforcement process is the practice of treating them like criminals when they fall behind on their

Effective Child Support Policy for Low-Income Single Parents

287

payments. Fathers believe that heightened enforcement efforts have been directed at nonresident fathers indiscriminately, regardless of their effort to support or be involved with their children. Some fathers believe that the system is more likely to penalize fathers who work in the regular economy than those who have gone underground. Others believe the state is targeting low-income, African American fathers for imprisonment rather than prosecuting higher income fathers who can support their children. A father Waller spoke to who previously lived with his children and had a cooperative parenting arrangement with their mother, resented being pursued for formal child support saying, “It’s ridiculous. . . I wouldn’t say that ’cause they cracking down on these fathers that are not supportive. They cracking down on the fathers that are supportive” (Waller, 1996, pp. 305–306). Sherwood (1992), Johnson & Doolittle (1996), and Waller (1996) document concerns about using imprisonment as an enforcement tactic. A father in Sherwood’s focus group who was $8,000 in arrears for child support, explained: I’m just tired of getting locked up every so often, every eight months or so. I don’t have no bad record, no record at all. But I just keep getting locked up for child support, that’s the main thing (Sherwood, 1992, p. 62).

A father Waller interviewed remarked: It’s crazy. You got 5,000 men and only 1,000 of them got jobs. . .That’s entrapment if you ask me. You know he gonna come back to jail ’cause he ain’t got the money to pay you. He might get himself locked up and then he might get out. Within a couple months he gotta come right back (Waller, 1996, p. 291).

Despite these reports, Johnson & Doolittle (1995, p. 27) note, “The perceived importance and likelihood of jail often is greater than justified in a statistical sense by its frequency.” Exacerbating these problems with the enforcement process is that low-income fathers usually do not have much knowledge of child support regulations, do not have legal representation, and do not feel they have a chance to tell their side of the story in court (Achatz & MacAllum, 1994; Furstenberg, 1992; Johnson & Doolitttle, 1995; Sullivan, 1992). Johnson & Doolittle (1995, p. 27) explain why the court proceedings often make low-income, noncustodial parents (NCPs) feel powerless: Very rarely do these men come in with any legal representation. They usually stand alone before the judge and bailiffs, opposite an attorney for the state (who, in the NCP’s eyes, is representing the CP [custodial parent], who may or may not be present). From the very start, the NCP is on the defensive . . . .The proceedings are quite brief (usually about five minutes)

288

Social Problems and Policy

and the discussion of evidence is cursory, so the NCPs often do not feel they “had their day in court.” In many of the sites in PFS (Parents Fair Share), the majority of low/no income NCPs who come before judges are found in contempt and ordered to do something (pay, get a job and pay, or go to jail until a purge payment is made).

Without sufficient information about how the child support process works, many fathers assume that part of their debt can be reduced if they document the prior, informal contributions to their child. Fathers sometimes come to their hearings “with a shoebox full of receipts in tow,” believing that the court will recognize these to offset the debt (Achatz & MacAllum, 1994; Johnson & Doolittle, 1996; Waller, 1996). One father who said he and his family had provided direct support, described how he was constrained from introducing his prior support as evidence during the court proceedings: It’s real intimidatin,’ man. My daughter, she’s like only three years old, and they told me I owe her like, man, it was a lot, I don’t know, it was in the grands. . .And where am I gonna get that kind of cash from? That’s what I want to know. How they expect me to come up with that amount of money just like that? And it ain’t fair cuz like during those three years me and my family did for her. . .And my child’s mom sat there and lied that I didn’t help out her and the baby. . .And when it was my turn, and I wanted to tell about how she was lyin cuz like whenever I got a job I would buy food, bags of Pampers, outfits. . .And all what my mom done put out. . .But the man just say, “Quiet, son. We don’t wanna hear about that. We just want you to answer these questions we got here on our paper” (Achatz and MacAllum, 1994, p. 86).

Fathers pursue unilateral responses to their experiences of “criminalization.” Fathers who have not been able to make their support payments have said they faced the choice between generating income in the underground economy or being “caught” by the child support enforcement and, possibly, imprisoned. Child support can have the unintended effect of pushing fathers into the illegal economy. One father who came to court with a box of money orders and receipts for items he gave his children says that more than half of his income is now being garnished. Unable to pay his rent, he says he has considered trying to get fired, changing his name, or selling drugs. In desperation, he says he has even considered suicide. I’m not trying to run from it. You know, I’m here. But they’re making me run. It’s like they’re giving me no other opportunity but to say: “O.K., fine. I’m gonna be a criminal.” Now I look at some of the guys out there selling drugs, like I know why they’re doing it. . .I’m trying my best, but it’s not working out (Waller, 1996 p. 292).

Effective Child Support Policy for Low-Income Single Parents

289

PROBLEMS WITH THE MODIFICATION PROCESS The employment situation of nonresident fathers often appeared to be unstable and changeable. Fathers said they needed more flexibility when they were out of work, when their income decreased, and when they were incarcerated. Yet, fathers often do not even know that their orders can be modified downward or do not know how to do this (Johnson & Doolittle, 1995, p. 36). Given the complexity of the child support system, fathers are often unaware that they should report changes in income or employment (Achatz & MacAllum, 1994, p. 94). Furthermore, child support orders are not often modified because of the expense and time involved in going through the court proceeding (Sorensen & Lerman, 1997, p. 6). Some fathers indicated that it was difficult to modify child support awards when they lost their jobs or changed employment. A father in Sullivan’s study talked about the difficulty of modifying the child support award after he changed jobs and his income decreased: I am making less money than the first time. I went and said: “Can you cut it down?” I showed them papers. “These are my expenses.” All they said was: “You still have to give us this amount of money”. . . It could have been difficult to eat, and these people knew exactly what I was making, but they still wanted the money (Sullivan, 1992).

CONCLUSIONS AND IMPLICATIONS FOR POLICY Child support is intended both to meet children’s economic needs and to ensure that unmarried and divorced parents accept financial responsibility for their children. To achieve these outcomes, the state attempts to regulate negotiations between unmarried or divorced parents, negotiations that otherwise occur privately between married parents. Although the objectives of mothers and fathers conflict at times, the case studies indicate that many unmarried, low-income parents prefer to negotiate private agreements for support. Parents believe that participating in the formal child support system detracts from their children’s well-being, exacerbates conflict between parents, and may harm poor fathers. Because of this, they typically endorse state-regulated child support only to force irresponsible fathers to take care of their children after private arrangements break down. The case study evidence provides three main insights into why the child support system is not more effective for families whose children are receiving welfare: • Many low-income, unmarried parents prefer informal arrangements of support, including in-kind contributions from the father, because they believe it benefits their children and their families.

290

Social Problems and Policy

• These parents often do not comply with child support laws and regulations they perceive to be unfair, counterproductive, or punitive. Among these laws and regulations are the assignment of rights to child support to the state, the small pass-through, large awards relative to the noncustodial parent’s income, and the threat of imprisonment. • Many such parents often face social and economic realities, such as low wages and unstable employment, or have inadequate information about how the child support system works. These conditions make it difficult for them to comply with existing policy, even when they wish to do so. These findings suggest that there is a mismatch between the premises and goals of child support policy and what low-income parents desire from the system. This mismatch impedes low-income parents’ willingness and ability to participate. The authors believe the formal child support system would gain greater compliance and legitimacy in the eyes of low-income parents if they perceived it as more beneficial to their children and supportive of their efforts to negotiate economic agreements. If this conclusion is valid, however, parents’ beliefs about how to best support their children and their informal practices for doing so conflict with important objectives and political realities of a child support policy based more on the experiences of middle-class families. Consider the $50 pass-through and the disincentive for compliance it creates. Parents experiencing persistent economic insecurity believe that noncustodial parents’ contributions should further increase their children’s standard of living. In contrast, the practice of retaining child support payments to recoup welfare payments made to the family rests on the principle that parents are responsible for supporting their children and that the welfare system should step in only when parents’ contributions do not come up to the state’s minimum standard. Child support payments, therefore, are viewed as a way to offset public welfare spending. The practice of retaining child support payments to recoup welfare costs also respects the principle of horizontal equity. Welfare recipients with earnings have their benefits cut. Hence, not cutting benefits when child support is received would favor this source of income over earnings. As states have tended to reduce the implicit tax on earnings to encourage welfare recipients to work, a reduction in the implicit tax on support payments might also be in order. Other conflicts include the following: • Parents want in-kind support and involvement to be taken into account for emotional reasons and because they believe these are important for their children’s well-being. The tradeoff is the administrative difficulties and increase in administrative costs entailed by doing so.

Effective Child Support Policy for Low-Income Single Parents

291

• Low-income parents want to negotiate private agreements because they believe it encourages cooperative parenting and paternal involvement. Policy demands formal agreements and tough enforcement mechanisms to ensure that noncustodial parents do not shirk their responsibilities to their children or the state. • Low-income, noncustodial fathers need a child support enforcement process that permits flexibility and discretion to take into account job loss, daunting arrearages, and other difficulties. Fathers do not want to be treated as criminals by the child support system. Policy has moved toward reducing administrative and judicial discretion, establishing uniform rules for all families, and bureaucratizing the system. The kinds of conflicts identified in this article mean that reform will be difficult. Any isolated policy change will have limitations. Policymakers have the opportunity to review the goals and enforcement practices of child support policy in light of the evolving welfare policy regime following passage of PWRORA. The challenge they face is to develop policies that honor the broad public interest in effective enforcement, adhere to consensus policy goals, and acknowledge and respect the social realities that shape low-income parents’ responses to the current system. REFERENCES Achatz, M., & MacAllum, C. A. (1994). Young unwed fathers: Report from the field. Philadelphia: Public/Private Ventures. Committee on Ways and Means, U.S. House of Representatives. (1998). Green Book: Background material and data on programs within the jurisdiction of the Committee on Ways and Means. Washington DC: U.S. Government Printing Office. Edin, K. (1995). Single mothers and child support: The possibilities and limits of child support policy. Children and Youth Services Review, 17(1–2), 203–230. Edin, K., & Lein, L. (1997). Making ends meet: How single mothers survive welfare and low-wage work. New York: The Russell Sage Foundation. Freeman, R., & Waldfogel, J. (1998). Does child support enforcement affect male labor supply? In I. Garfinkel, S. McLanahan, D. Meyer & J. Seltzer (Eds.), Fathers under Fire: The revolution in child support enforcement, (pp. 94–127). New York: Russell Sage. Furstenberg, F. F., Jr. (1992). Daddies and fathers: Men who do for their children and men who don’t. In F. Furstenberg, Sherwood, & Sullivan (Eds.), Caring and paying: What mothers and fathers say about child support. New York: Manpower Demonstration Research Corporation. Retrieved from http://fatherhood .hhs.gov/pfs92/ch3.htm. ———. (1995). Fathering in the inner city: Paternal participation and public policy. In W. Marsiglio (Ed.), Fatherhood: Contemporary theory, research, and social policy (pp. 119–147). Thousand Oaks, CA: Sage.

292

Social Problems and Policy

Gallagher, L. J., Gallagher, M., Perese, K., Schreiber, S., & Watson, K. (1998). One year after federal welfare reform: A description of state Temporary Assistance for Needy Families (TANF) decisions as of October 1997. Washington, DC: The Urban Institute. Garfinkel, I. (2000). Assuring child support: An extension of the Social Security System. New York: Russell Sage Foundation. Garfinkel, I., & McLanahan, S. (2000). Fragile families and child well-being: A survey of new parents. Focus, 21(1), 9–11. Garfinkel, I., McLanahan, S. S., & Hanson, T. L. (1998). A patchwork portrait of nonresident fathers. In I. Garfinkel, S. McLanahan, D. Meyer & J. Seltzer (Eds.), Fathers under fire: The revolution in child support enforcement (pp. 31–60). New York: Russell Sage. Garfinkel, I., McLanahan, S. S., Meyer, D. R., & Seltzer, J. A. (Eds.). (1998). Fathers under fire: The revolution in child support enforcement. New York: Russell Sage. Johnson, E. S., & Doolittle, F. (1996). Low-income parents and the parents’ fair share demonstration: An early qualitative look at low-income noncustodial parents (NCPs) and how one policy initiative has attempted to improve their ability and desire to pay child support. New York: Manpower Demonstration Research Corporation. Klawitter, M. (1994). Child support awards and earnings for divorced noncustodial fathers. Social Service Review, 68(3), 351–368. Mincy, R. B., & Sorensen, E. J. (1998). Deadbeats and turnips in child support reform. Journal of Policy Analysis and Management, 17(1), 44–51. Office of Child Support Enforcement, U.S. Department of Health and Human Services. (1998). Twenty-first annual report to Congress for the period ending September 30, 1996. Retrieved from http://www.acf.dhhs.gov/programs/cse/ annrptc.htm. Roberts, P. (1999). Setting support when the noncustodial parent is low income. Center for Law and Social Policy (February 8). Retrived from http://www .clasp.org/pubs/childrenforce/supaward.htm. Sherwood, K. E. (1992). Child support obligations: What fathers say about paying. In Furstenberg, Sherwood, & Sullivan (Eds.), Caring and paying: What mothers and fathers say about child support. New York: Manpower Demonstration Research Corporation. Retrieved from http://fatherhood.hhs.gov/pfs92/ ch4.htm. Sorensen, E. (1997). A national profile of nonresident fathers and their ability to pay child support. Journal of Marriage and Family, 59, 785–797. Sorensen, E., & Lerman, R. (1997). Welfare reform and low-income noncustodial fathers: New constraints and opportunities. Paper presented at the 19th Annual Research Conference of the Association for Public Policy Analysis and Management, Washington DC. Sorensen, E., & Turner, M. (1997). Barriers in child support policy: A review of the literature. National Center on Fathers and Children. Retrieved from http://www.upenn.edu/gse/ncoff/litrev/sblr.htm. Sullivan, M. L. (1992). Noncustodial fathers’ attitudes and behaviors. In Furstenberg, Sherwood & Sullivan (Eds.), Caring and paying: What mothers and fathers say about child support. New York: Manpower Demonstration Research Corporation. Retrieved from http://fatherhood.hhs.gov/pfs92/ch2.htm.

Effective Child Support Policy for Low-Income Single Parents

293

U. S. Bureau of the Census. (1995). Child support for custodial mothers and fathers. current population reports (Series P60–187, 1991). Washington DC: U.S. Government Printing Office. Ventura, S. J., Martin, J. A., Curtin, S. C., & Mathews, T. J. (1998). Report of final natality statistics, 1996, Monthly Vital Statistics Report, 46(11)(suppl.). Washington DC: National Center for Health Statistics. Waller, M. R. (1996). Redefining fatherhood: Paternal involvement, masculinity, and responsibility in the other America. Princeton University, doctoral dissertation. Waller, M. R., & Plotnick, R. D. (1999). Child support and low-income families: perceptions, practices and policy. San Francisco: Public Policy Institute of California.

This page intentionally left blank

Index

Beckerman, A., 88 Beller, A.H., 127 Block group data, 197 Blumenthal, Karen, 88 The Bradley Amendment, 275

Child support, 128, 161–162, 169–173, 262–269 in-kind payments, 280–283 modification, 289 The Child Support Enforcement Act, 172 Child Welfare League of America, 88 Childbearing, 10–13 Children, achievement, 61 Children’s Health Insurance program, 246 Consumer Expenditure Survey, 129 Coparenting, 101–102, 108–111 Criminal offense, 91 Current Population Survey (CPS), 125, 247, 252, 263 Custodial parent, 101–102, 108–11,164 sexual behavior, 165–167

Center for Epidemiological StudiesDepression Scale, 106, 212 Childbearing, 10 Cohabitation, 7–10, 29–30, 166, 216 Child Care Development Block Grants (CCBG), 249 Child care subsidies, 247–248 Child custody, 162 Child development, 54, 77 Child placement, 87

Deadbeat dads, 276 Delinquency, 64, 73–82 Demographic changes, 3–18, 90, 265 Demographic composition hypothesis, 266 Demographic transition theory, 3–4 Depression, 46, 146–147, 211, 216–217 Discrimination, 158–161 Divorce, 265 consequences of, 6, 42–43, 62

Acculturation, 105 Aid to Families with Dependent Children (AFDC), 63, 245 Adolescent development, 45, 54, 77, 195 Adolescent time use, 184–186, 188–191 Adoption and Safe Families Act (ASFA) of 1997, 87 Adult supervision, 196 American Housing Survey (AHS), 130 At-Risk Child Care programs, 249

Index

296 Domestic relations law, 161–162 Doubling–up, 30–31, 40. See also Living arrangements Drug involvement, 91 Duncan, G.J., 124 Earned Income Tax Credit (EITC), 39, 246–247 Ecological fallacy, 77 Ecological theory, 194 Economic issues, 32, 122 consequences, 124–125 xtrain, 140, 149 See also Financial issues Economic theory, 226, 231, 266 Education, 40, 65, 94, 108 Employment, 12–15, 29, 32, 127, 254 Ethnicity, 102–103, 126, 147 Evolutionary perspective, 63, 227, 231 Expenditures, 129 Factor analytic approach, 77 Familisim, 103 Family background, 90 Family structure, 62, 73–74, 90–91, 185 Family Structure Model, 62 Family Support Act (FSA), 249, 273–274 Federal Parental Locator Service, 274 Financial issues, 122–133 assets, 133 disincentives, 277–279 inflation, 264 liabilities, 133 strain, 140, 146–153 taxes, 126, 174 Fronczek, P.J., 130 Gang Resistance Education and Training (GREAT), 79 General Social Survey (GSS), 66 Government assistance, 31, 126 Aid to Parents with Dependent Children (AFDC), 245 single fathers, 39–40 Government failure, 266–267 Graham, J.W., 127

Grandparents, 93, 100–115 custodial, 108 coparenting, 108 well-being, 113 Hafstrom, J.L., 130 Hoffman, S.D., 124 Homosexual parents, 165–166 Horton, S.E., 130 Household Economics Model, 62–63 Housework, 140–141, 187 Income, 124–126 black families, 126 divorced, 126 never-married mothers, 126 predivorce, 124 postdivorce, 124 single parent families, 126–127 widowed, 126 white families, 126 Individual level research, 75–76 Individualist perspective, 137–139 Inflation, 264 Institutional changes, 3–18 Intergenerational relationships, 15–16, 109 Interracial relationships, 168 Jendrek, M.P., 101 Joint custody, 168–69 Learning theory, 225 Legal context, 159–160 Lesbian mothers, 43–59 Lino, M., 129–130 Livelihood strategies, 28–31 Living arrangements, 30–31, 88 single-father, 40 children’s, 92 foster care, 92 grandparents, 92 See also Doubling-up Loeber, 74 Marital Conflict Model, 65, 67, 228, 231 Marital disruption, 5. See also Divorce

Index Marital relationships, 13 Marital stability, 6 Marriage, 6–10 McGowan, Brenda, 99 Medicaid Matching Rate (MMR), 249 Minkler, M., 101 Mother-child relationship, 48 195, 200, 205–207 Moynihan Report, 74, 223 Multi-level research, 76–79 The National Conference of State Legislatures, 87 The National Longitudinal Study of Adolescent Health, 198 The National Survey of Families and Households (NSFH), 5, 28, 145, 230 The National Survey of Family Growth (NSFG), 32 Neighborhood context, 194–207 Risks, 200 Nonmetro single mothers, 27–37 Occupational Changes in a Generation Survey, 229 Office of Child Support Enforcement, 262 Panel Study of Income Dynamics (PSID), 229, 263 Parental competence perspective, 228, 231 The Parental Fitness Model, 64–65, 67 Parental incarceration, 86–97 consequences of, 87 Parental mobility, 164 Pathology, 159, 223 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), 27–28, 245–246, 263 The Positive and Negative Affect Scale (PANAS), 107 Poverty, 27, 34, 126, 159 Psychological well-being, 42, 109, 113, 210 Putative fathers, 163–164

297 Religion, 142, 146–153, 167–168, 210–211, 217–218 Residential stability, 196 Roe, K.M., 101 Role models, 196 Savage, H.A., 130 Schools, 78–80 Sexual behavior, 165–167, 194–207 Single-father families, 38–40 Social capital theory, 184 Social control, 74–75 Social disorganization theory, 75–76 Social isolation, 196 Social policy, 174–176 child care, 245–260 child support, 271 Social support, 15–16, 141–142, 146–153, 210 Socialization theory, 225 State Parent Locator Service, 274 Stirling, K.J., 124 Stouthamer–Loeber, 74 Stress, 106, 112, 211, 217 Structural disadvantage, 80–81 Structuralist perspective, 137–139 Survey of Consumer Finances, 132 Survey of Income and Program Participation, 130 Survey of Inmates in State and Federal Correctional Facilities, 89 Survey of Inmates in State Correctional Facilities, 89 Temporary Assistance to Needy Families, 64, 246, 272–273 Theoretical perspectives demographic composition hypothesis, 266 demographic transition theory, 3–4 ecological theory, 194 economic theory, 226, 231, 266 evolutionary perspective, 63, 227, 231 Family Structure Model, 62

Index

298 Theoretical perspectives (continued) Household Economics Model, 62–63 individualist perspective, 137–139 learning theory, 225 Marital Conflict Model, 65, 67, 228, 231 parental competence perspective, 228, 231 Parental Fitness Model, 64–65, 67 social capital theory, 184 social disorganization theory, 75–76 socialization theory, 225 structuralist perspective, 137–139 Tract-level data, 197 Transitional Child Care program, 249

Uniform Child Custody Jurisdiction Act (UCCJA), 165 Uniform Reciprocal Enforcement of Support Act (URESA), 173 U.S. Department of Justice, 87 Well-being grandparent, 113 parental, 136–154 pregnant teen, 212 Weitzman, L.J., 124 Widowed single mothers, 61, 126 Visitation rights, 286 Zeitz, Dorothy, 88

About the Editors Dr. ANNICE D. YARBER is a professor, researcher, activist, and public speaker who challenges conventional ideas about single-parent families, opting instead to offer an expanded perspective to discourse on single-parent families. Professor PAUL M. SHARP has taught, conducted research, and published in a wide array of areas within the field of sociology with special emphasis on crime, youth violence, and public policy.