BACKGROUND: Demographic and psychosocial correlates of activity in domains other than recreational activity have not been well characterized and may be particularly relevant for health promotion efforts aimed at women. METHODS: Cross-sectional relationships between recreational, occupational, and household/caregiving physical activity and demographic and psychosocial factors were assessed with a mail survey in a random sample of 2,636 ethnically diverse women members of a large health maintenance organization, ages 20-65. Activity was assessed with a modified Baecke questionnaire that uses categorical responses regarding frequency of domain-specific activities to create four semicontinuous activity indices (sports/exercise, active living, occupational, household/caregiving). RESULTS: Multivariable logistic regression analysis showed that the likelihood of being in the highest quartile of the sports/exercise and active-living indices, compared with the other three quartiles, was decreased among older, nonwhite, less well educated, heavier women who had young children at home, lacked motivation to exercise, and perceived external obstacles to exercise behavior. The odds ratios (ORs) ranged from 0.38, 95% confidence interval (CI) 0.33-0.45, associated with low motivation, to 0.95, 95% CI 0.93-0.98, associated with increasing body mass index. Social support and confidence in one’s ability to continue to exercise, even when faced with other pressures and demands (termed self-efficacy), were associated with increased likelihood of high levels of sports/exercise and active living (OR = 2.34, 95% CI 1. 83-2.98 and OR = 3.96, 95% CI 2.92-5.38, respectively). In contrast, the highest quartile of household/caregiving activity was positively associated with increasing age (OR = 1.28, 95% CI 1.16-1.42), Hispanic ethnicity (OR = 1.58, 95% CI 0.55-1.01), being married (OR = 1.70, 95% CI 1.33-2.18), having young children at home (OR = 6.99, 95% CI 4.33-11), and greater time constraints as a barrier to exercise (OR = 1.55, 95% CI 1.38-1.74) and was negatively associated with employment (OR = 0.38, 95% CI 0.30-0.47). Increased likelihood of the highest quartile of occupational activity was associated with high school education or less (OR = 2.26, 95% CI 1.74-2.94) and current smoking (OR = 1.66, 95% CI 1.23-2.23), while self-efficacy regarding exercise was associated with decreased likelihood (OR = 0. 77, 95% CI 0.61-0.96). CONCLUSIONS: These findings suggest that demographic and psychosocial correlates of physical activity vary by domain and that initiatives to promote physical activity in the population need to take these differences into account.