ABSTRACT: The study aimed to develop and evaluate a multicomponent school and home based physical activity (PA) intervention in children in grades 3-7 (aged 8-13 years) and determine the psychological variables that influence PA; 10 × 1 h school-based training sessions, a home-based activity program and 4 × 1 h lifestyle workshops for parents. PA was assessed at an intervention and nearby control school using accelerometers and self-report at 3-time points: baseline, post intervention and 10-week follow-up. Self-efficacy, self-management strategies, enjoyment, perceived barriers to PA, outcome-expectancy and social support were evaluated. The study showed 73% of the children with complete data sets at the intervention school (n = 27) did not increase device measured moderate to vigorous PA (MVPA) in the after-school period (3 p.m. to 6 p.m.) or over the whole day or during school break time immediately following the intervention or at follow-up, as compared to 70% of children with complete data sets at the control school (n = 35; p > 0.05 for all). Overall, 59% of boys attained more than double the recommended 120 min of MVPA each day compared to 42% of girls (p = 0.013). At the baseline, children's self-reported PA in the intervention school positively correlated with: outcome expectancy (R = 0.240, p = 0.015), enjoyment (R = 0.339, p < 0.001), self-efficacy (R = 0.399, p < 0.001), self-management (R = 0.617, p < 0.001), social support at home (R = 0.406, p < 0.001), and social support at school (R = 0.407, p < 0.001). Similar relationships were observed after the intervention and at follow-up. Focus groups with the children, parents and interviews with teachers identified areas for improvement of the intervention. In conclusion, while the multifaceted approach to improve PA was ineffective over the time span of the study, important predictors of PA in this sample of disadvantaged children were identified.