The authors compared 3 measures of readiness to change alcohol use commonly used in clinical research and practice with adolescents: the Readiness Ruler, the SOCRATES (subscales of Recognition and Taking Steps), and a Staging Algorithm. The analysis sample consisted of 161 male and female adolescents presenting for intensive outpatient alcohol-abuse treatment who reported current alcohol use at the initial assessment. Evidence for concurrent validity was assessed by computing simple correlations of each readiness measure with the other 3 and of each readiness measure with drinking behavior (percentage of days abstinent [PDA] and drinks per drinking day [DDD], respectively, in the last 30 days) at the start of treatment and at the 6-month follow-up assessment. Evidence for predictive validity was based on percentage of independent variance accounted for by each of the readiness measures in predicting drinking behavior at 6 months from the start of treatment, and then in predicting drinking behavior at 12 months from the readiness assessment at 6 months. The results showed that all but Recognition had good concurrent validity, the Readiness Ruler score showed consistent evidence for predictive validity, and the Staging Algorithm showed good predictive validity for DDD at 6 and 12 months. For the 82 participants with an alcohol-use disorder diagnosis, the findings for the Ruler and Recognition were similar, but the Staging Algorithm had poorer prediction of DDD at 12 months, and Taking Steps was a better predictor of 6- and 12-month PDA and DDD. Research and clinical implications of the findings are discussed. (Contains 1 footnote and 7 tables.)