Megan M. SweeneyDepartment of SociologyRutgers, The State University of New Jersey
Presented at the NICHD panel,"Visions of the Future: A Town Meeting on New Directions in Population Research"1Annual Meeting of the Population Association of AmericaMarch, 2000
This is a fascinating time to study the family, largely because change in this social institution has been so rapid and dramatic. The story is now familiar to demographers. Young people are delaying entry into first marriage, more years of life are being spent in unmarried cohabitation, a larger proportion of marriages are expected to end in divorce, fewer children are growing up in married-couple households, and fertility is lower than it was a generation ago. While the end-point of this trajectory of family change remains unclear, current trends suggest some important directions for future population research.
I will center my comments on two sets of issues. First, we should work towards a better understanding of the nature and meaning of recent trends in family life, both to better understand present family patterns and to gain some insight into what families might look like when these trends eventually stabilize. Much attention has been paid to the fact that people are spending fewer years of their lives in marriages and more years in unmarried cohabiting relationships than was true a generation ago. Researchers disagree as to whether cohabitation is most appropriately viewed as a stage in the marriage process or an alternative to marriage, and also as to the mechanism(s) underlying the observed positive association between cohabitation and divorce. In addition to considering the meaning of these trends for adults, we should continue current efforts to investigate the changing contexts of marriage, cohabitation, and singlehood as environments for bearing and rearing children. Second, we should work towards a better understanding of the implications of current family trends for population health and well-being. Research is particularly needed to identify causal mechanisms underlying observed associations between family structure and the well-being of adults and children. I envision several directions for future research that would help to achieve both of these goals. While these are specifically discussed in the context of the family, they also have applications in many other areas of population research.
In overviews of existing research on cohabitation, both Pamela Smock (forthcoming) and Judith Seltzer (forthcoming) argue that researchers face difficulty in understanding cohabitation because its meaning is unstable over time. The growing importance of cohabitation in people’s lives may similarly indicate that marriage is most appropriately viewed as a "moving target." Although relatively long-term intimate relationships also exist outside of marriage or cohabitation, we know remarkably little about these other arrangements. Classic strengths of qualitative research include the investigation of rapidly changing situations and situations we know little about. Combining analyses from large-scale surveys with research based on focus groups or intensive interviewing would greatly improve our understanding of the meaning of current family patterns to the people living within them. Such research will also improve our ability to pose the most important questions, phrased in the most appropriate ways, on future survey instruments (see also Smock, forthcoming).
We need to better understand the ways in which the family is changing over time, which requires a dynamic approach to the study of behaviors such as childbearing, union formation, and union dissolution. Relatively few past studies are designed to investigate the nature of historical change in the determinants of family transitions, thus limiting our understanding of the roots and meanings of contemporary family patterns (Modell 1999). For example, detailed and representative data on cohabitation has only been available since the mid-1980s (Smock, forthcoming), precluding a large-scale historical analysis of the ways in which the correlates and meaning of cohabitation have been changing over time. As the duration of available data on cohabitation expands, we should explicitly design analyses to investigate the ways in which cohabitation, and thus also marriage, have been changing over time. Again, identifying the direction and determinants of change provides insight into what families will look like when the pace of family change stabilizes.
Theory broadens our understanding of the population processes we study, and improves our ability to make predictions about the future. Yet many of our current theoretical models offer insufficient explanations for contemporary family patterns. For example, although large race and ethnic differences have been identified in family patterns, our major theories of family life still remain largely race-neutral (Cancian and Sweeney 2000). More explicit attention to racial, ethnic, gender, and socioeconomic variation in family experiences is needed. Use of both qualitative and quantitative techniques, as well as attention to life course and historical variation in family experiences, will further improve our ability to develop more appropriate theoretical models of family life. Development of new theory will also benefit from international comparisons of the nature and pace of family change in various political, economic, and cultural settings.
We tend to study both marriage and cohabitation as early adulthood issues, giving less attention to variation in the nature and context of these transitions over the life course. This focus on the behaviors of young people is not surprising, given that young adulthood is a "demographically dense" period, involving many inter-related life transitions in a relatively brief span of time, and because young people are often the engines behind major social change (Rindfuss 1991). Yet there is great reason to expect the economic and social context of family decisions to vary with age, and some empirical evidence suggests that the correlates of first marriage do indeed change over the early life course (e.g. Goldscheider and Waite 1986; Oppenheimer and Lew 1995; Sweeney 1998). Additional research into the nature of such variation would improve our understanding of the determinants of family transitions. Such research would also improve our ability to assess the impact of factors that alter marriage timing -- such as a changing economy or shifting attitudes towards premarital sex -- on the nature of family arrangements.
We might further expect family experiences to vary with positions along marriage (or relationship) "careers". For example, Cherlin and Furstenberg (1994) report that one in three Americans will marry, divorce, and remarry during their lifetimes. Approximately one in four children born in the early 1980s are expected to spend some time living in a stepfamily (Bumpass, Raley, and Sweet 1995). Despite the large numbers of adults and children experiencing remarriage, remarkably little attention has been devoted to understanding higher-order marital transitions.
We have also tended to study cohabitation as an early life-course issue, yet cohabitation is more common before second marriages than before first marriages (Bumpass and Sweet 1989: Table 2). There are several reasons to expect post-marital cohabitations to differ in meaningful ways from cohabitation among the never married. For example, post-marital cohabitors are more likely to have children in the household than are never married cohabitors, making the study of post-marital cohabitation particularly important for investigating children’s home environments. Children living with ever-married parents are further substantially more likely to be the biological offspring of only one partner, and these post-marital cohabiting households are three times as likely to have a youngest child over the age of ten (Bumpass, Sweet, and Cherlin 1991: Tables 5 and 6). Post-marital cohabitors are less likely to have definite plans to marry their partners and are less likely to expect to ever marry (again), and thus it is not surprising that post-marital cohabitations tend to be of somewhat longer duration than pre-marital cohabitations (Bumpass, Sweet, and Cherlin 1991). Although there are numerous reasons to expect post-marital cohabitation to differ in meaningful ways from cohabitation in never-married populations, remarkably little research has taken a life-course perspective to the study of cohabitation.
As we consider the nature and implications of contemporary family patterns, we should continue to examine the ways in which individuals and families are embedded in larger communities and social structures (Coleman 1988; Portes 1998). Examinations of the extent to which local economic and demographic characteristics influence family transitions offer one example of research in this direction. More attention is needed, however, to identifying the most appropriate level of measurement for local-area characteristics in these studies. For example, for some area-characteristics (such as economic opportunities), measures at the level of local labor market areas seem appropriate. For other types of area-characteristics (such as the availability of potential partners), it seems that indicators ideally ought to be measured at the level of the specific occupational, educational, or neighborhood settings in which research subjects experience their most frequent social interactions. The quality of relationships with neighbors, teachers, and peers, as well as the quality of these neighborhoods and schools themselves, are likely to be particularly useful in understanding the ways in which social environments influence children's well-being.
Several social scientists argue that the nature of "new" family life will be determined in important ways by decisions made by men (Goldscheider and Waite 1991; Nock 1998) highlighting the importance of gathering data on the family experiences and attitudes of men as well as women. Yet recent research reveals that men are less likely to be included in survey samples and data from male survey respondents is found to be of lower quality than that from women (Bumpass, Martin and Sweet 1991; Rendall et al. 1999). In fact, Rendall and colleagues (1999) report that one-third to one-half of men's non-marital births and births in previous marriages are missed in estimates from retrospective histories. We know men tend to be better reporters of fertility histories when they are married to their children's mother, but know less about the quality of reporting among men who live with children but are not currently married to their mother. We need to develop ways to improve survey coverage of men, and to increase accuracy of reporting by male survey respondents in all family structures. To better understand both the meaning of current family arrangements and the likely direction of future change, we also need more qualitative examinations of men’s orientations and attitudes towards family life.
A growing literature on the effects of marriage on the well-being of women and men suggests that marriage may cause people to have better heath and health behaviors, to be happier, and to be better off economically (for reviews see Nock 1998; Waite 1995). Research is needed to disaggregate how much of these effects are due to self-selection into marital statuses, and also to better understand the nature of any causal mechanisms which do underlie these observed associations between marriage and relatively higher levels of well-being. If people are spending more years of their lives unmarried, and being unmarried is indeed unhealthy, current family trends have serious implications for population health in the coming decades. To the extent that substantially altering marriage behavior itself may not be a viable public policy option, identifying the mechanisms through which marriage benefits people, and the ways in which these needs may be met in other ways, will help us maintain high levels of population well-being even as marriage becomes less common. Interdisciplinary research efforts among sociologists, psychologists, and medical researchers would be particularly beneficial in this area of inquiry.
We also need to better understand the extent to which the benefits of marriage may change over the life course, may not be equally shared by women and men, and may vary across racial, ethnic, and socioeconomic sub-populations. For example, Kathryn Edin's (1999) qualitative research in Chicago, Charleston, and Camden, suggests that one reason many low-income women have negative attitudes towards marriage is that they fear exposure to domestic violence. Steven Nock (1998) finds that while first marriage improves men's labor market achievement -- as indicated by income, labor force participation, and occupational prestige -- remarriage reduces men's achievement in these areas (Nock 1998: 67-69). More research is needed to discern whether the variation in "marriage benefits" by marriage order points to a strong role of selection in producing the observed association between marriage and well-being, or whether there are some mechanisms that can be reasonably expected to operate only in first marriages.
We should continue efforts to identify mechanisms underlying observed associations between family structure and children’s well-being, and intensify efforts to understand sub-population variation in these relationships. Continuing current efforts to understand how family processes influence children’s well-being (processes such as parenting style, family conflict, the quality of family relationships), and how the nature of these processes may vary across family structures, is important to this end. We also need to better understand the role of economic factors in producing observed associations between family structure and children's well-being. A key question here is whether children's access to family resources varies across family structures. To what extent do stepparents and parents’ cohabiting partners share their income with the children living in their households? Answering these questions will greatly improve our understanding of the implications of current family patterns for the economic well-being of children, and is particularly critical given that half of the disadvantage associated with growing up with a single parent is thought to be explained by economic deprivation (McLanahan and Sandefur 1994).
In order to make such analyses possible, however, we need to include improved indicators of spending that directly benefits children in our large and federally-funded data sets. Because of the centrality of economic deprivation to theoretical models of the ways in which family structure impacts children's well-being, it is also important to obtain longitudinal data on the level and sources of family income over time, as well as a wide (and longitudinal) array of child outcomes. Finally, we need to improve the quality of our measurements of children’s family structure histories. For example, we should make greater efforts to differentiate between single-parent families (and stepfamilies) formed after a non-marital birth or after marital dissolution (Bumpass, Raley, and Sweet 1995). It is also important to continue our efforts to differentiate between families formed through marriage and cohabitation. Children's complete family structure histories should be gathered, with the aim of improving our understanding of both the short and long-term effects of living in families of varying forms.
Although by no means comprehensive, the initiatives outlined here aim to improve our understanding of the nature and direction of current trends in the family, and to identify the consequences of these trends for the health and well-being of adults and children. The preceding discussion highlights the need for continued resources directed towards exploratory, descriptive, and explanatory approaches to population research. We should continue efforts to document current population trends, and also attempt to identify the trajectories of change that population processes are likely to follow in the future. It is further important to understand mechanisms underlying observed associations between demographic trends and population health and well-being. For example, it is important to recognize that divorce is associated with relatively lower well-being among adults and children, but also to apply sophisticated methodological techniques to better understand the ways in which this observed association may (or may not) reflect a true underlying causal relationship. Population research should both document demographic trends and identify the mechanisms underlying these patterns. Such explanatory work is particularly important given the potentially large role population researchers can play in informing policy decisions.
1 I wish to thank Larry L. Bumpass for comments on an earlier version of this manuscript.
Bumpass, Larry L., Teresa Castro Martin and James A. Sweet. 1991. The Impact of Family Background and Early Marital Factors on Marital Disruption. Journal of Family Issues 12:22-42.
Bumpass, Larry L., R. Kelly Raley and James A. Sweet. 1995. The Changing Character of Stepfamilies: Implications of Cohabitation and Nonmarital Childbearing. Demography 32:425-436.
Bumpass, Larry L. and James A. Sweet. 1989. National Estimates of Cohabitation. Demography 26:615-624.
Bumpass, Larry L., James A. Sweet and Andrew Cherlin. 1991. The Role of Cohabitation in Declining Rates of Marriage. Journal of Marriage and the Family 53:913-927.
Cancian, Maria and Megan M. Sweeney. 2000. The Importance of Economic Prospects for Assortative Mating. Paper prepared for the 2000 Population Association of America annual meetings, held in Los Angeles, CA.
Cherlin, Andrew J. and Frank F. Furstenberg. 1994. Stepfamilies in the United States: A Reconsideration. Annual Review of Sociology 20:359-381.
Coleman, James S. 1988. Social Capital in the Creation of Human Capital. American Journal of Sociology 94:S95-121.
Edin, Kathryn. 1999. What Do Low-Income Mothers say about Marriage? Unpublished manuscript.
Goldscheider, Frances Korbin and Linda J. Waite. 1991. New Families, No Families? The Transformation of the American Home. Berkeley, CA: University of California Press.
Goldscheider, Frances Korbin and Linda J. Waite. 1986. Sex Differences in the Entry into Marriage. American Journal of Sociology 92:91-109.
McLanahan, Sara and Gary Sandefur. 1994. Growing Up with a Single Parent: WhatHurts, What Helps. Cambridge, Mass.: Harvard University Press.
Modell, John. 1999. When History is Omitted. Pp. 226-233 in Transitions to Adulthood in a Changing Economy: No Work, No Family, No Future? Edited by A. Booth, A. Crouter, and M. Shanahan. Westport, CT: Praeger.
Nock, Steven L. 1998. Marriage in Men’s Lives. New York: Oxford University Press.
Oppenheimer, Valerie and Vivian Lew. 1995. Marriage Formation in the Eighties: How Important Was Women’s Economic Independence? Pp. 105-138 in Gender and Family Change in Industrialized Countries, edited by K. O. Mason and A. Jensen. Oxford: Clarendon Press.
Portes, Alejandro. 1998. Social Capital: Its Origins and Applications in Modern Sociology. Annual Review of Sociology 24:1-24.
Rendall, Michael S., Lynda Clarke, H. Elizabeth Peters, Nanlini Ranjit and Georgia Verropoulou. 1999. Incomplete Reporting of Men's Fertility in the United States and Britain: A Research Note. Demography 36:135-144.
Rindfuss, Ronald R. 1991. The Young Adult Years: Diversity, Structural Change, and Fertility. Demography 28:493-512.
Seltzer, Judith A. Forthcoming. Families Formed Outside Marriage. Journal of Marriage and the Family.
Smock, Pamela J. Forthcoming. Cohabitation in the United States: An Appraisal of Research Themes, Findings, and Implications. Annual Review of Sociology.
Sweeney, Megan M. 1998. Gender, Race, and Changing Families: The Shifting Economic Foundations of Marriage. Ph.D. dissertation, Department of Sociology, University of Wisconsin, Madison, WI.
Waite, Linda J. 1995. Does Marriage Matter? Demography 32:483-507.