In addition to considering gender and race/ethnicity in the mediational model, our study makes several other contributions to the literature. First, we examined relationships among 9th and 10th graders, during a developmental period in which substance use has typically already been initiated but has not yet reached its peak (Johnston, O'Malley, Bachman, & Schulenberg, 2010). Second, we examine within our mediational model adolescents’ use of a variety of substances, including alcohol, cigarette, marijuana, inhalant, ecstasy, and illicit prescription drug use; previous studies have often examined only a single substance (e.g., Ryan, Jorm, & Lubman, 2010). Third, while much of the empirical literature has reported relationships between parental and adolescent substance use based on frequency counts of parental substance use, we examined adolescents’ perceptions of problematic parental substance use (i.e., the consequences associated with parental use), which may be a better predictor of both poorer parenting and greater subsequent adolescent substance use (Barnard & McKeganey, 2004). Fourth, we examined whether the proposed mediational model differed for adolescents who lived in a two- versus single-parent household, as previous studies suggest the latter is associated with greater risk for substance use (Breivik & Olweus, 2006).
