Abstract
This article demonstrates a latent growth curve methodology for analyzing longitudinal data for adolescent alcohol use by combining information from different overlapping age cohorts to form a single developmental trajectory. Hypotheses concerning the form of growth in alcohol use, the extent of individual differences in the common trajectory over time, and covariates influencing both initial status and the form of growth were tested. Utilizing five separate age cohorts each measured over the same 4-year period, results suggested a common trajectory existed across the 8 years represented by the cohort-sequential analysis, with alcohol use increasing more rapidly during the adolescents' transition to high school. Family cohesion and peer encouragement for alcohol use were hypothesized to influence both initial status and the trajectory of alcohol consumption during adolescence. While family cohesion served to suppress initial levels of consumption delaying the upward trajectory of alcohol use, peer encouragement was related not only to initial, and elevated, levels of use, but was predictive of those changes that occurred during adolescence. Discussion involves the importance of family and peer influences in the development of adolescent alcohol use and the utility of the cohort-sequential approach in the analysis of longitudinal data.