ABSTRACT. Objective: This study extended previous research on the association of substance use with family formation behavior by assessing the effects of the type and extent of adolescent substance use in a competing risks model. Substance use was expected to increase the likelihood of nonmarital family formation overall and differently by gender. Method: Longitudinal data from home interviews with the 14-to 16-year-old respondents to the fi rst wave of the National Longitudinal Survey of Youth 1979 (N = 4,011) were used in multinomial logistic regressions estimating the odds that fi rst cohabitation, parenthood, or remaining single occurred before fi rst marriage among fi ve types of substance users compared with nonusers. Full sample analysis preceded separate analyses of women (n = 1,946) and men (n = 2,065). Results: Illegal drug use and concurrent substance use increased the likelihood that cohabitation, as opposed to marriage, was the fi rst family type. Concurrent use of three types of substances had the largest effect on family formation behavior. The effects of singular marijuana use mattered only for men. The effects of substance use on parenthood as the fi rst family type were signifi cant only for women and increased the likelihood that marriage occurred fi rst. Conclusions: The effects of substance use depended on the type(s) of substance(s) used, singular or concurrent use, and gender. Previous research regarding cohabitation was supported and extended. Assumptions that substance use leads to teenage or unwed parenthood based on the relationship of substance use to pregnancy or its predictors should be re-examined. (J. Stud. Alcohol Drugs, 71, 938-949, 2010) Received: February 18, 2010. Revision: June 14, 2010. *This research was supported by National Institute on Drug Abuse grant 1R26DA021318-01.† Correspondence may be sent to Dr. Andrea Kay Ryan at the above address or via email at: akr31@pitt.edu. T EENAGE OR PREMARITAL COHABITATION, preunion parenthood, or not forming an adult family relate to later poverty, educational failure, unstable labor force participation, criminality, depression, stress, partner violence, and poor physical health (Brown, 2000;Crane, 1991;DeKlyen et al., 2006;Frye et al., 2007;Hagestad and Call, 2007;Knight et al., 1977;Popenoe, 2007). Substance use is an important behavior associated with family formation. For instance, young adults signifi cantly reduce their frequency of substance use when they marry; however, frequency declines only modestly among people who remain single (Bachman et al., 1997;Leonard and Rothbard, 1999; Substance Abuse and Mental Health Services Administration, 2004). A longstanding explanation of the "marriage effect" is that the cultural norms of adult family roles in marriage discourage substance use (Bachman et al., 1997(Bachman et al., , 2002Coleman, 1978;Duncan et al., 2006;Homish and Leonard, 2005).Assumptions that marriage changes all types of substance use behavior may not be justifi ed, however. In this study, illegal drug use was de...