Premarital Cohabitation and Marital Disruption Among White, Black, and Mexican American Women

We use data from the 1995 National Survey of Family Growth (N 1⁄4 4,547) to investigate racial and ethnic differences in risk factors for marital disruption, with a particular emphasis on premarital cohabitation. We find that the nature and strength of the estimated effects of several risk factors for disruption differ across groups. In particular, premarital cohabitation is positively associated with subsequent marital disruption among non-Hispanic White women but not among non-Hispanic Black or Mexican American women. Little of the observed gaps between groups in levels of disruption, however, appears to be attributable to differences in premarital cohabitation. In addition to improving our understanding of marital disruption, this research contributes to a growing literature emphasizing heterogeneity across groups in the meaning and function of cohabitation. A large body of research points to racial and ethnic variation in levels of marital disruption in the United States. Recent estimates indicate that whereas 32% and 34% of all first marriages involving non-Hispanic White and Hispanic women, respectively, end in separation or divorce within the first 10 years, the figure is substantially higher for non-Hispanic Black women, at 47% of all marriages (Bramlett & Mosher, 2002; see also Raley & Bumpass, 2003). Rates of disruption are further shown to vary considerably across Hispanic subpopulations and by nativity (Bean, Berg, & Van Hook, 1996; Bean & Tienda, 1987; Frisbie, 1986). These differences are important to identify and understand, as marital disruption is associated with negative economic, emotional, and health outcomes for adults and children (Amato, 2000; Holden & Smock, 1991; Waite, 1995). Although racial and ethnic differentials in levels of marital disruption are consistently documented, we know less about whether risk factors for disruption vary across groups. There is growing evidence, however, that the effects of important risk factors for disruption do in fact vary by race and ethnicity. For example, several studies indicate that the association of marital disruption with age at marriage and with premarital childbearing differs for White women and Black women (e.g., Castro Martin & Bumpass, 1989; Heaton & Jacobson, 1994; Department of Sociology, Institute for Health, Health Care Policy, and Aging Research, Rutgers University, 54 Joyce Kilmer Avenue, Piscataway, NJ 08854 (jphillips@sociology. rutgers.edu).