NCFR Family Focus articles often address the policy implications of research findings, but few address the data implications. The articles are more about what we know, somewhat less about what we need to know, and even less about what data we need to collect to be able to address unanswered questions. Statements about what we know are useful, but to move forward we must also determine what we still need to know and how we can know it. Existing data have a big effect on what we study, especially for those of us who use survey data. It is a chicken-and-egg problem. We collect data to address important research questions, but existing data then constrain our ability to address the new questions we discover.

I believe that we need new U.S. data to answer the following questions: When do parents and adult children help each other? Why do they help? How do they help? And what difference do intergenerational ties make for individual family members? There are two reasons we need new data to study intergenerational relationships in the United States. First, the demography of family life has changed. Increased life expectancy means that there are more three- and four-generation families than ever before. An adult child is increasingly likely to have an older parent and an even older grandparent. Low fertility and increased churning of couple relationships as marriages and cohabitations form and dissolve have changed the context of child-rearing in ways that exacerbate class differences and are likely to have lasting effects on individual welfare. These demographic changes alter what a family is and may change what a family does.

Second, the Great Economic Recession has increased the need for the family safety net, while at the same time, family members have become less able to act as a safety net because parents and children are often exposed to the same disadvantages, such as unemployment. The public safety net is also fraying, thereby increasing the need for kin support and simultaneously reducing the resources available to share among kin.

There are surely U.S. datasets that we can use to begin to answer these questions. Examples include: The Health and Retirement Study (HRS), the National Survey of Families and Households (NSFH), the National Longitudinal Surveys (NLS), the National Longitudinal Study of Adolescent Health (Add Health), the Panel Study of Income Dynamics (PSID), and the Wisconsin Longitudinal Survey (WLS). (See Bianchi, Evans, Hotz, McGarry, & Seltzer, 2007, for an assessment of existing data on intergenerational relationships.) NCFR researchers have used these datasets well to help us anticipate the implications of demographic change for intergenerational relationships.

Analyses of the HRS and NSFH data suggest that the disruption of family ties due to divorce and remarriage may disadvantage nonresident fathers and stepparents in later life if they need care from adult children (Lin, 2008a; Pezzin, Pollak, & Schone, 2008b; Shapiro, 2003). From the NSFH data we know that, compared to those who are married, individuals in cohabiting unions are less likely to exchange help with parents and less likely to consider parents to be part of their emergency support system (Eggebeen, 2005). But there is much that we do not know about how the demographic changes in couple relationships in both the parent and child generations affect intergenerational relationships. Do parents think they are part of their cohabiting child's emergency support system? In periods of economic hardship are parents as likely to share a home with an adult child and cohabiting partner as with a married adult child and spouse? Census data can tell us whether or not intergenerational households include cohabiting or married adult children, but they cannot tell us how likely coresidence is for these two groups because we do not know the denominators or what demographers call the populations "at risk," that is, who has cohabiting or married children who live elsewhere and might need parents to help them with a place to live.

New types of family relationships formed by cohabitation and remarriage create challenges for data collection that were not concerns when the primary datasets for studying inter-generational relationships were originally designed. For instance, the HRS relies on one spouse to report about all of his or her own and the spouse's children and to characterize the type of relationship. This is an increasingly difficult task as family networks have become more complex and couples have yours, mine, and our children. Knowing about all of an individual's parents and children is an important step toward answering new questions about intergenerational relationships because it tells who might need help or who might give help. This information can help adjudicate competing explanations. We know, for example, that adult stepchildren and parents are less likely to live together (Pezzin et al., 2008b). This may be due to tensions associated with remarried family roles or because the stepchild is living with his or her other biological parent. Without information about what parents are available, researchers cannot distinguish these two explanations.

Knowing who is available is just the start. We also need to know about the resources and needs of family members. A parent-child dyad is part of a broader constellation of relationships. Most parents have at least two children, and siblings may coordinate or act strategically in caring for elderly parents (Matthews, 2002; Pezzin, Pollak, & Schone, 2008a). Studies that focus on a single caregiving child underestimate the cost of having a parent who needs care because siblings may contribute different things or take turns and because even children who are not primary caregivers suffer negative mental health effects when they have a disabled parent (Amirkhanyan & Wolf, 2003). Earlier in life, at the transition to adulthood, the help that parents provide one child affects what they can provide to other children in the family.

It is no accident that much of what we know about intergenerational relationships in adulthood comes from research that focuses on the transition to adulthood and old age. These are periods when children need extra help from parents to go to college or set up an independent household, or when parents must call on children for help with transportation, household tasks, money, or shared housing. What happens in these life stages and whether or not family members share resources affect individuals' well-being. Policies also affect how individuals fare at the transition to adulthood (employment opportunities, financial aid programs), and when old age brings serious health problems (Medicare). But what about the things that happen in between? Parents and children help each other at other times. Usually parents help children more than children help parents (Bianchi, Hotz, McGarry, & Seltzer, 2008). Parents who are better off may help with a house down payment and grandmothers of all socioeconomic statuses may help with child care. But probably the most important way that parents and children help each other in between is that they form a safety net for each other that insures against risk. Knowing that there is someone who would help in an emergency, if illness strikes, or when money is tight may improve individuals' mental health and even allow them to take risks they would not otherwise consider, such as choosing a job without health insurance over one with benefits. Eggebeen's (2005) finding that cohabiting couples are less likely to include parents as a source of emergency help is important for two reasons. First, it reminds us that family members bail each other out when times are difficult. Second, this finding implies that the family safety net is not as secure when it includes new types of family relationships. Despite the theoretical importance of the insurance value of kin ties (Wong, 2008), most major national surveys do not ask about latent support but instead restrict attention to actual transfers.

The safety net function of parents and children may be the hardest to address in new data collection because "being there" may depend on the kind of help a family member needs and why they need help, i.e., whether or not they did something to cause the poor circumstances. Vignettes with a factorial design are one way to learn when parents and children think they should help each other. Ideally, questions about the family safety net would include information from survey respondents about whether they would expect a parent or child to help them, as well as respondents' own willingness to provide help if a parent or child needed it. Information from parents as well as children in the same families is important, as we know from comparisons of reports about actual transfers from donors and recipients. Donors report that they give more than recipients say that they receive (Lin, 2008b). Researchers do not know if similar differences occur for reports about potential help from family members, but it is important to know if parents and children share the same expectations about responsibility for economic and other help because shared attitudes indicate a more durable safety net.

This is a difficult economic period in which to make a pitch for new data collection. Any new data would be a valuable public good for the research community. But as much as we would welcome new data, no one dataset can address all of the theoretical and practical concerns of researchers who study inter-generational relationships. Nevertheless, there are good policy reasons to know more about how families with a history of non-union childbearing, divorce, cohabitation, and remarriage manage in difficult economic times. Compared to conventionally married, two-biological-parent families, can disrupted families count on the same informal support that keeps older adults out of nursing homes and gives food and shelter to young adults who have trouble finding a job? Only time, and new data, will tell.  

References

• Amirkhanyan, A. A., & Wolf, D. A. (2003). Caregiver stress and noncaregiver stress: Exploring the pathways of psychiatric morbidity. Gerontologist, 43, 817-827.
• Bianchi, S., Evans, V. J., Hotz, V. J., McGarry, K., & Seltzer, J. A. (2007). An assessment of available data and data needs for studying intra- and inter-generational family relationships and behavior. Working Paper no. 2007-20, California Center for Population Research. Los Angeles: University of California. Retrieved from http://papers.ccpr.ucla.edu/papers/PWP-CCPR-2007-020/PWP-CCPR-2007-020.pdf
• Bianchi, S. M., Hotz, V. J., McGarry, K., & Seltzer, J. A. (2008). Intergenerational ties: Theories, trends, and challenges. In A. Booth, A. C. Crouter, S. M. Bianchi, & J. A. Seltzer (Eds.) Intergenerational caregiving (pp. 3-43). Washington, DC: The Urban Institute.
• Eggebeen, David J. 2005. "Cohabitation and Exchanges of Support." Social Forces, 83, 1097-1110.
• Lin, I.-F. (2008a). Consequences of parental divorce for adult children's support of their frail parents. Journal of Marriage and Family, 70, 113-128.
• Lin, I.-F. (2008b). Mother and daughter reports about upward transfers. Journal of Marriage and Family, 70, 815-827.
• Matthews, S. H. (2002). Sisters and brothers/daughters and sons: Meeting the needs of old parents. Bloomington, IN: Unlimited Publishing.
• Pezzin, L. E., Pollak, R. A., & Schone, B. S. (2008a). Family bargaining and long-term care of the disabled elderly. In A. Booth, A. C. Crouter, S. M. Bianchi, & J. A. Seltzer (Eds.), Intergenerational caregiving (pp. 257-275). Washington, DC: Urban Institute.
• Pezzin, L. E., Pollak, R. A., & Schone, B. S. (2008b). Parental marital disruption, family type, and transfers to disabled elderly parents. Journal of Gerontology: Social Sciences, 63B, S349-S358.
• Shapiro, A. (2003). Later-life divorce and parent-adult child contact and proximity: A longitudinal analysis. Journal of Family Issues, 24, 264-285.
• Wong, R. (2008). Are we asking the right questions on intergenerational ties? In A. Booth, A. C. Crouter, S. M. Bianchi, & J. A. Seltzer (Eds.), Intergenerational caregiving (pp. 45-51). Washington, DC: Urban Institute.