Despite well-documented health benefits of engaging in regular physical activity (PA),  school children do not meet recommended PA guidelines to promote health (NASPE, 2004; USDHHS, 2000). To increase children's activity levels and achieve public health goals, there is an emerging trend calling for social ecological approaches that incorporate individual (self-efficacy), social environmental (social support from parents, friends, and teachers), and physical environmental (equipment accessibility and neighborhood safety) factors to complement the conventional research paradigm of individual-centered influences on PA (Sallis & Owen, 1997, 2002). Although several researchers have used the social ecological approach as a framework to study adults' PA, the influence that environmental variables have on school children's PA has not been studied. The purpose of this study, therefore, was to investigate how individual, social, and physical environmental variables contribute to PA participation. Participants were 285 middle school students (143 girls, 142 boys; M age = 13.36 years) recruited from a southeastern public school. They completed questionnaires assessing their perceived physical environment (Motl et al., 2005), social support from parents and friends (Prochaska et al., 2002) and from P.E. teachers (Daigle, 2003), self-efficacy (Motl et al., 2000), and self-reported PA (PAQ-C, Kowalski et al., 1997). Data were analyzed using bivariate correlations to examine the relationships among the variables and a hierarchical regression analysis to determine relative contributions of individual, social, and physical environmental variables in predicting levels of PA. Correlation analyses revealed a pattern of positive correlations among the individual, social, and physical environmental factors, and PA (r's ranging from .15 to .61, p < .01). The hierarchical regression analysis, entering the self-efficacy variables first, followed by the social, and then physical environmental variables, showed that self-efficacy, friends' support, P.E. teachers' support, parents' support, and equipment accessibility, were significant positive correlates of PA (b = .32, p< .01; b = .17, p< .05; b = .11, p< .05; b = .12, p< .05; b = .11, p< .05), accounting for a total of 38.2% of the variance. Consistent with previous research, this study underscored that self-efficacy was the strongest correlate of PA. These findings also highlight the unique contributions of specific social environmental, and physical environmental variables on PA which can be used in guiding effective PA interventions. To better understand individual, social and physical environmental variables for promotion of PA for children, future studies should assess the potential direct or indirect pathways through which these variables influence PA.