The results of this study support the hypothesis that social variables are related to exercise behavior and satisfaction. Exercise frequency was related to having friends at the health club and socializing outside the club with people met at the club. Frequency of skipping workouts was negatively associated with having friends at the club. The results shown in Table 2 suggest that of the three social variables, socializing outside the club with people met at the club was the best predictor of exercise frequency, having friends at the club was the best predictor of frequency of skipping, and exercising with a friend was the best predictor of exercise satisfaction. The three social variables explained 11% of the variance in exercise frequency, 8% of the variance in skipping frequency, and 4% of the variance in satisfaction. These findings suggest that friendships that involve exercising together and the social contacts that result from exercising in public places such as health clubs may motivate exercise behavior. A large segment of the population may derive social, as well as physical, benefits from exercise, and these social benefits may encourage them to adhere to their exercise programs. The significant interaction of marital status and socialization with people met at the club in the prediction of exercise frequency suggests that social interaction may be a more important motivator of exercise behavior for single people than for people who are married or have significant others. Exercise programs involving social interaction may be especially effective for single, divorced, or widowed people.