The objectives of this study were to compare daily physical activities, and activities performed according to a structured protocol, measured with tri-axial accelerometers (Tracmor-4), between lean and overweight children. Fourteen overweight children (59.8+/-9.5 kg) and fifteen lean matched controls (47.2+/-8.7 kg) wore the Tracmor-4 daily, during 12+/-1.3 h, for one week in their home environment. Of these, 24 children participated in a sports afternoon, where they performed activities according to the same structured protocol. In addition, physical activity was estimated using a modified Baecke questionnaire. Body composition was determined. Total mean Tracmor counts/day were significantly lower for the overweight children than for the lean (overweight: 46.1+/-6.9 vs. lean: 54.4+/-11.2 kCounts/day, p=0.02), while reported activities (Baecke score) were similar. When performing activities according to the structured protocol, there was no difference in mean Tracmor counts between the two groups (overweight: 36.3+/-6.9 vs. lean: 34.7+/-6.6 kCounts, p=0.6). Daily physical activities were inversely related to percentage body fat (r(2)=0.29, p<0.01); structured activities were not. As compared to lean children, overweight children moved less without being aware of it; yet exerted the same movements per activity. We conclude that in overweight children daily physical activities were reduced and structured activities performed according to instructions were not. In order to prevent progressive overweight or obesity, overweight children should take part in as many as possible structured and scheduled sports activities throughout the week, and be encouraged to behave physically active in daily life.