The purpose of this study was to explore the associations between several social learning theory variables and self-reported vigorous exercise and to consider the implications for exercise promotion. A random sample of adults in San Diego, California, was surveyed by mail. The 2,053 respondents (response rate, 43.4%) overrepresented Caucasian, affluent, and well-educated groups. A model of 24 variables accounted for 0.27 of the variance in exercise, and results strongly supported social learning theory. The strongest correlates were self-efficacy (i.e., confidence in the ability to exercise in specific situations), perceived barriers to exercise, modeling, dietary habits, support from friends, and age. Smoking was inversely associated with exercise in men only. We encourage researchers to conduct intervention trials to test the hypotheses generated in this study.