Adolescent physical inactivity has risen to an alarming rate. Several theoretical frameworks (models) have been proposed and tested in school-based interventions. The results are mixed, indicating a similar weakness as that observed in community-based physical activity interventions (Baranowski, Lin, Wetter, Resnicow, & Hearn, 1997). The theoretical models were decontextualized, thus are unable to address issues central to adolescentsʼ physical activity behavior. In this article, we propose using a theoretical model derived from school-based research on learning behavior change. We review related research on adolescentsʼ physical activity to demonstrate the relevance of using the model to study the dynamic impact of personal, school curriculum, and community variables on adolescent physical activity. We also translate the conceptual model into empirically testable cross-sectional and longitudinal latent growth models and propose concrete steps researchers can take to design empirical studies to examine them. We believe that research studies guided by the proposed conceptual and empirical models will provide useful data for us to better understand the mechanisms of adolescent physical activity motivation and behavior change.Physical inactivity in adolescents has risen to an alarming rate. It is a consensus that the school-centered, especially physical education-based, interventions are the most appropriate avenue for all school-age children (Corbin, 2002). Several theoretical frameworks (models) have been proposed and clinically tested for effectiveness in schools. Results are quite mixed in terms of their actual effects on changing childrenʼs long-term physical activity behavior (Stone, McKenzie, Welk, & Booth, 1998). The mixed results may indicate a similar weakness as that observed in community-based physical activity interventions (Baranowski et al., 1997), that the theoretical models were decontextualized and were unable to address issues central to adolescentsʼ physical activity behavior. Most theoretical models derived from outside school physical education and were designed for treating adult obesity or physical inactivity (Chambliss, 2004), rather than nurturing and sustaining the desired behavior of physically active living. In addition, the application of the Made available courtesy of Human Kinetics: http://www.humankinetics.com ***Reprinted with permission. No further reproduction is authorized without written permission from Human Kinetics. This version of the document is not the version of record.Figures and/or pictures may be missing from this format of the document.***
356Chen and Hancock models is tailored mostly for adult populations with high risk, rather than seemingly ordinary but risk-behavior-prone children and adolescent population. The usefulness of the models for school-age children and adolescents can be limited (see Chambliss 2004 for a description of some models).Studies on the models seldom measure school curricular and instructional variables. Thus, it is diffi cult to apply them d...