The research reported in this article focuses on the role of cohabitation in premarital childbearing among U.S. women. Using data from the National Survey of Families and Households and the New York Fertility, Employment and Migration Survey, we examine the influence of cohabitation on the likelihood of premarital pregnancy and the decision to marry between premarital conception and birth. Our analyses show marked racial and ethnic differences in the role of the cohabiting union in family building. Although cohabitation increases the rate of premarital pregnancy for all women, its effect is much greater among Puerto Ricans than among non-Hispanic Whites and African Americans. Cohabitation accelerates the transition to marriage among premaritally pregnant White women, but has no effect among Blacks and has a strong negative effect among Puerto Ricans. We interpret our findings in terms of long-standing family patterns and cultural traditions within each group. A common practice in demographic studies is to treat nonmarital childbearing as equivalent to single motherhood in spite of the fact that births to unmarried women are increasingly occurring within cohabitation. In the early 1980s, about one-quarter (27%) of nonmarital births in the United States were to women living in cohabiting unions (Bumpass & Sweet, 1989). There is considerable variation in this phenomenon by race and ethnicity. Among Puerto Ricans, over half (59%) of nonmarital births occurred within informal unions during this period (Landale & Hauan, 1992), in contrast to 40% among Mexican Americans, 29% among non-Hispanic Whites, and 18% among African Americans (Bumpass & Sweet, 1989). Moreover, births within cohabitation account for differences in the proportion of children born outside of marriage among Puerto Ricans, Mexican Americans, and non-Hispanic Whites (Bumpass & Sweet, 1989; Landale & Hauan, 1992). Despite the growing prevalence of cohabitation prior to marriage, few studies have considered the role of informal unions in premarital childbearing. Studies conducted in other countries (e.g., Balakrishnan, 1989; Cooper, 1991; Haskey & Kiernan, 1989; Leridon & Villeneuve-Gokalp, 1989) indicate that cohabitation is associated with a higher likelihood of childbearing among unmarried women. Although this linkage has not been examined in the U.S. context, a similar association is likely. Studies based on U.S. data show that never-married cohabiting women expect to have children sooner, have more frequent intercourse, and are more likely to approve of unmarried childbearing than their noncohabiting single counterparts (Bachrach, 1987; Rindfuss & VandenHeuvel, 1990; Sweet & Bumpass, 1990). Furthermore, cohabiting women are more likely to have greater commitment to their sexual partners than are noncohabiting women. The research reported here focuses on the role of cohabitation in childbearing among U.S. women who have never been married. Both the numbers presented above and research on the meaning of cohabitation for various racial and ethnic groups (e.g., Landale & Fennelly, 1992; Loomis & Landale, 1994; Manning, 1993) suggest potential group differences in the effect of cohabitation on the fertility of never-married women. Thus, the central questions we address are whether and to what extent cohabitation differentially influences entry into premarital motherhood by race and ethnicity. Our analyses focus on two key components of the process; leading to premarital motherhood: (a) the occurrence of a premarital pregnancy that results in a live birth and (b) the decision to marry between premarital conception and the birth of the child. Comparisons are made between cohabiting and noncohabiting African American, Puerto Rican, and non-Hispanic White women. We restrict our analyses and discussion of Puerto Ricans to those living in the mainland United States. For ease of presentation, w refer to this group as Puerto Ricans without further qualification throughout the text. …