Numerous alcohol and drug abuse prevention trials have included social resistance training as a strategy for reducing early-stage adolescent alcohol use. Evaluations of these trials has shown them to be moderately effective, although the precise impact of the resistance training in comparison to other programmatic features has not been clearly identified. The current study examined the extent to which assertiveness and related social skills, personal competence (perceived cognitive mastery), and refusal efficacy predict alcohol involvement. Males were at greater risk for poor refusal skills and reported higher alcohol involvement. Cross-sectionally, youth characterized by poor social skill development reported lower refusal efficacy, lower grades, poor competence, and more alcohol use. Poor refusal efficacy was associated with more risk-taking, lower grades, less competence, and more alcohol use. Longitudinally, both poor refusal skills and risk-taking were associated with higher alcohol use. High personal competence was associated with lower alcohol use in both the eighth and tenth grades, but had no long-term effects on alcohol use. Findings highlight the close interplay between perceived competence and refusal skill efficacy, both of which should be included as essential components of school-based prevention strategies.