The study, a longitudinal cohort-sequential survey, involved a nationally representative sample of Norwegian junior and senior high school students (n = 5,308). The study used the following measures: alcohol consumption, number of intoxications, school dropout, conduct problems, drug use, friends' problem behavior, truancy, attitude toward school, educational aspirations, grades, homework, youth centrism, leisure time activities, religiosity, depression, academic self concept, and attachment to parents. Intoxication did override the effect of alcohol consumption and predicted dropout. This effect was confounded by parental attachment, whereas truancy and associating with peers high in problem behavior mediated the effect of intoxications. Frequent alcohol intoxications increase the risk for school dropout by means of increased truancy and differential association with counternormative peers. Table, references