FAMILY-STRUCTURE AND THE RISK OF A PREMARITAL BIRTH

The positive association between growing up in a nonintact family and the risk of a first premarital birth has been interpreted by researchers as consistent with three hypotheses: (1) a childhood socialization hypothesis - that women who grow up in a mother-only family during early childhood are socialized in ways that result in a high risk of a premarital birth; (2) a social control hypothesis - that the supervision of adolescents is more difficult in single-parent families than in two-parent families; and (3) an instability and change hypothesis - that a premarital birth is a response to the stresses accompanying changes in a woman's family situation. Although these hypotheses imply distinct behavioral mechanisms, adjudicating between them has proven difficult, in part because researchers have relied on static measures of family structure. We use data from the National Survey of Families and Households and continuous-time hazard models to investigate the effects on premarital births of dynamic family measures that reflect a woman's family situation between birth and age 19. Our findings are consistent with the instability and change hypothesis, but provide little support for the socialization hypothesis and the social control hypothesis.