Background Insufficient physical activity (PA) among adolescents is a major global public health concern. While theoretical frameworks, such as the Social Ecological Model, highlight multilevel influences—individual, familial, and socio-environmental—on PA behaviors, empirical evidence establishing associations, especially regarding structured PA among high school students in China, remains limited. Methods This study used a 1:1 matched case-control design to recruit 222 students from three high schools in Taiyuan City, including 111 participants in the sufficient PA group and 111 matched controls. PA levels in the sufficient group were partly objectively verified using ActiGraph GT3X+ accelerometers. Standardized questionnaires were administered to both groups to assess variables at five levels: individual, interpersonal, organizational, community, and policy. Multivariable logistic regression analyses were performed to identify independent predictors of PA. Results At the individual level, self-efficacy was the strongest predictor of PA, with an odds ratio (OR) of 39.45. At the interpersonal level, perceived support from physical education teachers was a significant predictor (OR = 7 .96). At the organizational level, adequate access to school PA facilities independently predicted higher engagement in PA (OR = 5 .15). Finally, at the community level, multi-channel access to PA information was also a prominent predictor (OR = 11 .31). Conclusion Self-efficacy, support from physical education teachers, access to adequate school facilities, and multi-channel availability of PA information have been identified as key factors potentially associated with PA engagement among high school students. Therefore, promoting PA in this population may require a comprehensive strategy—one that fosters self-efficacy at the individual level, enhances the support provided by physical education teachers, improves access to and quality of school sports facilities, and strengthens the dissemination of evidence-based PA information through diverse community channels.
Li et al. (Wed,) studied this question.