Despite representing a number of advances on prior research, the study has some limitations. First, the sample was predominantly White and included male adolescents only. The extent to which these findings would generalize to other ethnic groups and to girls requires testing. Second, reports of peer alcohol use were limited to reports by the adolescent and did not include assessment of volume of use. Third, although serving size was assessed for the main alcohol categories (beer, wine, and liquor) in an attempt to control for variation in ethanol levels across them, youth may not be very accurate in reporting serving sizes, and there are variations within categories in ethanol levels (e.g., strong versus regular beer). Fourth, the parental discipline measure had relatively low reliability. Fifth, predictors were not assessed every year and, therefore, some of the change score predictors spanned more than 1 year. Finally, the sample size was relatively modest for the models tested, although a strength was the repeated measurements of the dependent variables across the period. In the current study, growth in the middle and high school
