The Formation and Stability of Same-Sex and Different-Sex Relationships.

摘要

There is little demographic knowledge about same-sex relationships, despite intense debates about same-sex marriage. In this dissertation, I investigate the formation and stability of same-sex and different-sex co-resident relationships. I also evaluate the quality of retrospective data often used to study movement in and out of same-sex and different-sex relationships.;Chapters 1 and 2 describe the formation and dissolution of co-resident couple relationships in young adulthood (ages 16--34). In these chapters, 1 analyze retrospective histories of same-sex and different-sex unions from two British birth cohort studies: the National Child Development Study (NCDS) and the 1970 British Cohort Study (BCS). In Chapter 1, I investigate the sequencing, timing, and correlates of entering same-sex and different-sex unions. I find that individuals' choices of same-sex and different-sex co-resident partners are fairly stable throughout young adulthood. This high degree of stability could be due to the relatively public nature of co-residence and substantial commitments associated with co-residence. Individuals enter same-sex cohabitation at older ages than they enter different-sex cohabitation and marriage, possibly due to uncertainty about their sexual orientation. In addition, being from the 1970 cohort (rather than the 1958 cohort), growing up in London and the Southeast of England, and having a higher socioeconomic status are positively associated with entry into same-sex cohabitation. These patterns are consistent with the perspective that individuals who grow up in social contexts that are more favorable to same-sex partnering---or individuals who have the economic resources necessary to migrate away from unfavorable social contexts---are more likely to enter same-sex unions.;In Chapter 2, I investigate differences in the levels and correlates of stability among four couple types: female same-sex cohabitation, male same-sex cohabitation, different-sex cohabitation, and marriage. I find that same-sex cohabitation has higher rates of dissolution compared to different-sex cohabitation and marriage. Among different-sex couples, cohabiting unions are less stable than are marital unions. The higher rate of instability among same-sex and different-sex cohabiting unions relative to marriage is consistent with previous research and theory about the importance of legal and social institutionalization for union stability. In addition, men in same-sex cohabitation experience slightly higher rates of dissolution than do women in same-sex cohabitation. The correlates of union stability are similar across union types, suggesting that the same interpersonal stressors may affect same-sex and different-sex couples.;The final chapter of the dissertation evaluates the quality of retrospective data on non-marital cohabitation and marriage. I compare estimates of cohabitation and marriage from a comparable age-period-cohort group in the 2000, 2003, and 2005 British General Household Surveys (GHS). I find that the 2005 GHS produces rates of entry into non-marital cohabitation that are 17% lower than does the 2000 GHS for the same age, period, and cohort group. This difference in estimates of non-marital cohabitation between the 2000 and 2005 surveys is even larger for unions in which children were not born and for short-term unions. These findings are consistent with the perspective that the longer length of recall in the 2005 GHS relative to the 2000 GHS might have led 2005 OHS respondents to omit non-marital cohabitations from their reports---especially for less salient unions such as those without children and short-term unions. I also investigated the extent to which these different estimates of cohabitation affected substantive conclusions about the association between previous cohabitation experience and marital stability. Data from the three surveys produced a similar association between previous cohabitation and marital stability, suggesting that underreporting of certain types of cohabitation does not necessarily bias the estimated correlates of cohabitation. In sum, my research shows that cohabitation may be susceptible to recall errors, but these errors may be limited to less consequential types of cohabitation such as cohabiting unions without children and shorter-term cohabitations.