Analyses of 94 Fund for the Improvement of Post-Secondary Education (FIPSE)-sponsored drug-prevention programs and their outcomes used the Core Survey to identify 34 institutions where college students' binge drinking increased (M = 5.44%) and 60 institutions where it decreased (M = -4.59%) during 2 years of program operation. The authors used an inductively derived taxonomy of prevention program elements, student variables, student substance use, use-related variables, and institutional variables to compare the 2 groups of institutions. Only prevention program elements discriminated between groups. Factor analysis of discriminating elements identified 8 prevention factors that improved base-rate prediction of institutional decrease in binge drinking by 28.1%. Factor synthesis yielded a 3-construct binge-drinking prevention model based on student participation and involvement strategies, educational and informational processes, and campus regulatory and physical change efforts. This model improved base-rate prediction of decreased binge drinking by 33.2%.