Older African Americans and Latinos tend to exercise less than older Whites and are more likely to have chronic diseases that could benefit from exercise. Measurement of self-efficacy of exercise and exercise outcome expectations in this older population is required if exercise is to be monitored carefully and enhanced in this population. The purpose of this study was to test the reliability and validity of the Self-Efficacy for Exercise Scale (SEE) and Outcome Expectations for Exercise Scale (OEE) in a sample of African American and Latino older adults. A total of 166 individuals, 32 males (19%) and 134 females (81%) with an average age of 72.8 +/- 8.4 years participated in the study. The SEE and OEE scales were completed using face-to-face interviews. There was evidence of internal consistency for both scales with alphas of .89 and .90 for the SEE scale and .72 and .88 for the OEE scale. There was some evidence of validity for both scales based on confirmatory factor analysis and hypothesis testing, because factor loadings were greater than .50 in all but two items in the OEE, and there were significant relationships between self-efficacy and outcome expectations and exercise behavior at all testing time-points. The measurement models showed a fair fit of the data to the models. The study provided some evidence for the reliability and validity of the SEE and OEE when used with minority older adults, and it provides some guidelines for future scale revisions and use.