To evaluate several psychometric properties of a questionnaire designed to assess college students' self-efficacy to employ 21 cognitive-behavioral strategies intended to reduce the amount and/or frequency with which they consume marijuana, we recruited 273 marijuana-using students to rate their confidence that they could employ each of the strategies. Examination of frequency counts for each item, principal components analysis, internal consistency reliability, and mean interitem correlation supported retaining all 21 items in a single scale. In support of criterion validity, marijuana use-reduction self-efficacy scores were significantly positively correlated with cross-situational confidence to abstain from marijuana, and significantly negatively correlated with quantity and frequency of marijuana use and marijuana-related problems. In addition, compared with respondents whose use of marijuana either increased or remained stable, self-efficacy was significantly higher among those who had decreased their use of marijuana over the past year. This relatively short and easily administered questionnaire could be used to identify college students who have low self-efficacy to employ specific marijuana reduction strategies and as an outcome measure to evaluate educational and skill-training interventions.