Modeling results for the frequency of any drinking and drinking to intoxication are presented in Figures 2 and 3, respectively. In these models, significant stability of educational attainment and drinking behaviors across adolescence was observed. The cross-lagged effects indicated that drinking without adults at age 12 predicted lower school performance at age 14, and the frequency of any drinking and drinking to intoxication at age 14 negatively predicted student status at age 17. In contrast, school performance at ages 12 and 14 did not predict later drinking behaviors. Student status at age 17 positively predicted the frequency of any drinking and drinking to intoxication in young adulthood. The models had an excellent fit to the data (drinking: χ2[7]=4.52, p=.718, CFI=1.000, RMSEA=.000; intoxicating: χ2[7]=6.74, p=.456, CFI=1.000, RMSEA=.000).
