International migration over the past half century has increased the racial and ethnic diversity of countries in North America and Western Europe. My dissertation highlights two ways in which intergenerational relationships can be studied in light of increasing population heterogeneity. One approach is to assess whether immigrants and their descendants adopt "mainstream" attitudes, norms and behaviors related to parent-child relationships over time. A second approach is to explore ways in which increasing population diversity changes the influence that parents and offspring have over one another and the ways in which they interact.

In my first chapter, I adopt the first approach and ask how social contexts influence immigrants' attitudes towards family obligation and in particular, the difference between attitudes of immigrants and the native born towards family support. This chapter examines nativity differences in intergenerational obligation across two social contexts: Germany and the Netherlands. Intergenerational obligation is defined as the extent to which parents and children feel a sense of duty to assist one another and to take into account the needs and wishes of each other when making decisions. The paper focuses on first- and second-generation Turkish immigrants only and compares them to their native counterparts in Germany and the Netherlands. By comparing immigrants and their descendents from the same sending country to the "native" population of two countries, a main obstacle that commonly hinders cross-national migration research - comparing immigrants from different countries across contexts - is addressed.

I use data from the Generations and Gender Survey and apply structural equation models in the main analysis. After demonstrating consistency in the measurement and meaning of intergenerational obligation across groups, I find that immigrants have stronger family ties than natives in both countries. However, the nativity gap is much smaller in Germany compared to the Netherlands. In addition, the overall level of family obligation is lower among Turkish immigrants in the Netherlands compared to their counterparts Germany. I explain these differences from a policy perspective: More generous social welfare supports for families as well as multicultural policies that help immigrants retain their cultural identity in the Netherlands, compared to Germany, shed light on these findings. Importantly, the results suggest that attitudes towards family obligation are not fixed upon arrival; rather, they vary depending on the contexts into which immigrants settle. The findings speak to previous research that often describe differences between immigrants and natives as if immigrants' characteristics are fixed, ignoring the role that the receiving country context plays in altering immigrants' behaviors and attitudes.

My second chapter asks how increasing population diversity affects the type of partner individuals choose to marry and whether parent-child relationships influence these decisions. I apply a linked lives approach by exploring the connection between parent-child ties and when and whom offspring marry. Parental resources and parent-child relationships are well-known factors influencing children's family formation behaviors. Parents shape when offspring marry, whether they cohabit before marriage, when they have children and the number of children they have. However, far less is known about how parent-child relationships affect who children marry. Growing population diversity and changing patterns of race/ethnic segregation provide individuals with more opportunities to meet partners of a different race/ethnic background than their own. Although recent research asserts that parental influence on children's marital behaviors is waning, parents may still influence who children choose to marry.

In this chapter, I use data from the National Longitudinal Study of Adolescent Health to investigate how parent-child relationships during adolescence affect the timing and type of marriage young adults choose. I ask whether strong parent-child relationships are more likely to lead to marriage, rather than remaining single and whether they are positively associated with entry into a same-race, rather than cross-race exogamous unions. Finally, I ask whether the association between parent-child ties and offspring's union formation vary across race/ethnic and nativity groups. The results from this chapter suggest that individuals with closer ties to the family of origin start families of their own at younger ages. Yet the effect of parent-child relationships on offspring's marital timing is moderated somewhat by the respondent's background. In addition to influencing when children marry, strong emotional support across generations tends to increase the probability of entering into a same-race union, rather than a racially exogamous union.

The third and final chapter of my dissertation examines how marriage and intermarriage in particular affects young adults' ties to parents. Family scholars today argue that modern marriage privileges self-fulfillment and a reliance on partners to fulfill emotional and social needs that did not characterize marriage among earlier generations. An emphasis on couple quality and the time and resources needed to maintain such partnerships may have negative consequences for ties to parents.

I use data from the most recent wave of National Longitudinal Study of Adolescent Health and ask how marriage, and exogamous versus endogamous marriage in particular, are associated with ties to mothers. I also ask whether the association between offspring's union type and intergenerational ties are stronger for some groups compared to others, and whether the specific race of the partner matters. Findings from this chapter suggest that married children are not completely detached from mothers; rather, they occupy a middle ground. Married children tend to live nearby, but not close to mothers and tend to visit weekly, but not daily compared to those who are single. Who offspring marry also affects relationships with mothers. Children who married across race/ethnic lines are less likely to live near mothers compared to those who married within race/ethnic lines. Because of the geographic distance, these individuals are also less likely to visit or talk to mothers frequently compared to those who married within race/ethnic boundaries. However, the consequences of intermarriage are particularly detrimental for some groups compared to others. Hispanics, Asians and children of immigrants tended to have worse relationships with mothers following intermarriage compared to Whites and children of U.S. natives. These results highlight how intermarriage - a commonly understood mechanism that at the population level decreases the distance between groups - may in fact detrimentally affect ties among family members involved.

Using pooled data from the 1995 and 2002 NSFGs, we compare the timing and type of first union, fertility behavior in cohabitation and marriage, and the duration and outcome of first cohabiting unions for White, Black, U.S.-born Mexican American, and foreign-born Mexican American women. We find that the most pronounced differences in cohabitation are between foreign-born Mexicans and women born in the United States. Although the behavior of most foreign-born Mexicans favors marriage over cohabitation, cohabitation may substitute for marriage for a small number of foreign-born Mexicans. Patterns of cohabitation among U.S.- born Mexican Americans are consistently between those of foreign-born Mexicans and U.S.- born Whites, suggesting that assimilation changes union behavior. For Whites, our results suggest that cohabitation is a stage in the courtship process leading to marriage; whereas for Blacks, cohabitation is a highly unstable union that appears to substitute for marriage. Much of the variation by race, ethnicity and nativity status is accounted for by group differences in socioeconomic characteristics. Remaining variation may be attributable to group differences in the value of marriage and the obligations of partners in consensual unions.

This paper addresses the question: Does joint legal custody increase child support payments? It describes differences in formal child support payments for those with and without joint legal custody among divorce cases. It examines legal custody differences in the short-term after divorce as well as in the intermediate term, through the sixth year after divorce, to assess whether any economic benefits of joint legal custody endure through a significant part of childhood.

To the extent that legal custody differences in payments exist, we ask whether family and case characteristics, such as parents' incomes, number of children, and amount of child support orders

account for these differences. Finally, the paper uses statistical methods that adjust for the fact that unmeasured characteristics may affect both the adoption of joint legal custody as well as higher child support payments. Ignoring these unmeasured differences may overstate the benefits of joint legal

custody for child support payments. The paper aims to provide a less biased estimate of the effects of joint legal custody on child support payments than is available from most previous studies.

Divorce is costly for parents because of the challenges of meeting children’s economic and socio-emotional needs after separation. Using the National Survey of Families and Households (N = 1,935), we investigate whether expected economic and parenting costs deter divorce. Mothers expect higher economic costs than fathers, whereas fathers expect more parenting difficulties. The majority of mothers and fathers, however, expect high economic and parenting costs. In a large minority of families mothers and fathers differ in their expected costs. Furthermore, only parenting costs deter divorce, not economic costs, and mothers’ parenting concerns are a greater barrier to divorce than fathers’ concerns when parents disagree. Finally, expected parenting costs are more of a barrier to divorce for unhappy than happy couples.

Coresidence is of enduring importance across an individual’s life and a recurring feature of kinship. Most individuals will live with at least one person related to them through biological or legal ties for a majority of their lives. Given that demographic processes, social change, and economic development during the 20th century have resulted in increasingly diverse families, along economic, race/ethnic, geographic, and cultural lines, studying coresidence among contemporary American families is crucial. This research will inform the development of social and economic policies for the well-being of an increasingly heterogeneous American population. In this dissertation, I expand our current understanding of kinship relationships by studying coresidence among diverse American families. Recognizing that the prevalence of coresidence and the individuals with whom one lives will vary by life stage, I take a life course approach, studying coresidence at different developmental stages: middle childhood (Chapter 1), young adulthood (Chapter 2), and late middle-age (Chapter 3). Moreover, I build on the foci of prior work at each developmental stage, but shift attention to the aspects of households and families that are of particular relevance today: multigenerational households and immigrant families, the role of geographic opportunities and constraints in becoming residentially independent, and racial differences in support to aging parents. Across the three chapters, the results underscore the reliance on family members for help. Thus, variations in living arrangements are reflective of the hardships distributed unevenly across the population that contribute to the growing economic inequality characterizing the United States. Given that large swaths of the American population continue to experience financial insecurity, this research may inform the development of social and economic policies for the well-being of an increasingly heterogeneous American population.

This paper uses new data from the Los Angeles Family and Neighborhood Survey (L.A. FANS) to examine how neighborhood norms shape teenagers’ substance use. Specifically, it takes advantage of clustered data at the neighborhood level to relate adult neighbors’ attitudes and behavior with respect to smoking, drinking, and drugs, which we treat as norms, to teenagers’ own smoking, drinking, and drug use. We use hierarchical linear models to account for parents’ attitudes and behavior and other characteristics of individuals and families. We also investigate how the association between neighborhood norms and teen behavior depends on: (1) the strength of norms, as measured by consensus in neighbors’ attitudes and conformity in their behavior; (2) the willingness and ability of neighbors to enforce norms, for instance, by monitoring teens’ activities; and (3) the degree to which teens are exposed to their neighbors. We find little association between neighborhood norms and teen substance use, regardless of how we condition the relationship. We discuss possible theoretical and methodological explanations for this finding.

We use data from a sample of divorced parents in Wisconsin (N = 1,392) to examine how parents describe their children's living arrangements. Parents are likely to say that the children live with them as opposed to the other parent, controlling for the number of nights children spend with the mother. When the children spend substantial time in both parents’ homes, both parents are less likely to use the phrase live with to describe living arrangements. Together these findings suggest that family researchers no longer can rely on simple questions to capture complex living arrangements. We need clearer and more careful question wording and, in some instances, follow-up questions to accurately describe where children live.