The prevalence of obesity among children is increasing in the United States (Ogden, Flegal, Carroll, & Johnson, 2002) and has been tied to increased risks of adult obesity and cardiovascular disease (Whitaker, Wright, Pepe, Seidel, & Dietz, 1997). Physical activity, which is defined as "any bodily movement produced by skeletal muscles that results in caloric expenditure," increases energy expenditure and can help prevent obesity (Haapanen, Miilunpalo, Pasanen, Oja, & Vuori, 1997). The socio-ecological perspective suggests that attention to intrapersonal, interpersonal, environmental, and policy factors are necessary to effectively prevent and then reduce the incidence of childhood obesity through physical activity promotion. We posit that systematic attention to the structure and design of recreation programs is a first step for determining best practices in youth activity promotion during recreation times. The current study examined the physical activity outcomes associated with four common facilitation strategies used in recreation programs. In a sample of 74 predominantly low income youth aged 6 to 11, four types of recreation facilitation were adopted including: (a) skills and drills, (b) scrimmage, (c) modeled play, and (d) free play. Physical activity was objectively measured using Actigraph accelerometers. Results indicated that as a group, youth were most active during modeled play in the presence of college-aged role models. When participant age, gender, and body mass index were considered, age and gender demonstrated significant relationships to facilitation style and activity outcomes, while body mass index did not. Boys were more active during both free play periods, whereas girls had equal levels of activity in all facilitation settings. Youth of all ages were most active in play settings, and younger boys and girls were least active during scrimmage sessions. Results allow us to make suggestions for recreation program leadership. First, when given an appropriate activity environment and social support, children are intrinsically motivated to be physically active during play. When schedules allow, recreation programmers may provide children with sports or fitness equipment and give them a chance to be active without the structure that constrains play during classroom hours, physical education lessons, homework time, dinner time, etc. Also, whenever possible, pairing younger children with play leaders and older activity role models should be undertaken to facilitate this play period and avoid a drop in activity levels, which occurred after 15-20 minutes of activity. Next, scrimmages do not provide an ideal mechanism for physical activity promotion among young children. Finally, we recognize that children must learn and practice the skills of specific sports and games if they are to continue to participate in these activities as they grow and mature. To maximize activity levels during the minimum necessary skill sessions, we recommend teaching only primary activities involving gross motor movements to young children, providing equipment equal to the number of participants, and giving directions in short, more frequent updates to minimize lengthy activity lags that undermine the benefits of youths' rise in heart rate. Additional limitations, directions for future research, and applications are discussed.