AIM: To describe how movement behaviours (sedentary time, light-intensity physical activity, moderate-to-vigorous physical activity [MVPA]) and parent-reported community physical activity participation vary across levels of environmental supportiveness (home, school, community) in children with cerebral palsy (CP). METHOD: This was a secondary cross-sectional observational study using pooled data (n = 141) from four studies of children with CP aged 6 to 16 years (mean = 10 years 1 month [SD = 2 years 5 months]; Gross Motor Function Classification System level I = 71, level II = 41, levels III-IV = 29). Children wore hip-mounted triaxial accelerometers; validated machine-learning models classified sedentary time, light physical activity, and MVPA. Parents reported community physical activity participation and environmental supportiveness using the Participation and Environment Measure for Children and Youth. Linear models were adjusted for wear time and study. We estimated marginal mean differences (MMDs) in time spent in movement behaviours between the 25th and 75th centiles of supportiveness scores. RESULTS: Greater community (MMD = -30.1 minutes, 95% confidence interval [CI] = -53.4 to -8.1), home (-31.4 min, 95% CI = -51.5 to -11.3), and school (MMD = -54.1 min, 95% CI = -76.0 to -32.2) supportiveness were associated with less sedentary time. Higher home supportiveness was associated with more time in light physical activity (MMD = 23.7 minutes, 95% CI = 11.0 to 36.3), whereas school and community supportiveness were not associated with light physical activity. No environmental supportiveness measures were associated with time in MVPA. INTERPRETATION: Greater environmental supportiveness across all contexts was associated with lower sedentary time, while greater home supportiveness was associated with more time in light physical activity. Other factors may have stronger associations with MVPA time.