Data from a randomized clinical trial were used to examine the extent to which a health promotion intervention affected changes in psychosocial constructs and if so whether these in turn explained changes in physical activity (PA). PA and psychosocial data on 878 adolescents (ages 11—15) recruited through primary care providers (age M = 12.7 years, SD = 1.3; 58% white non-Hispanic) were measured at baseline, six and 12 months. Parallel process latent growth curve analyses found positive relationships between the growth trajectories of behavior change strategies, self-efficacy, family support, peer support and the growth trajectory of PA. However, mediation analyses did not reveal statistically significant intervention-mediated effects.