Childhood obesity has become a defining health concern in the U.S. with the robust evidence of its adverse outcomes. Policymakers and health professionals have come up with diverse plans to reduce and reverse the problem of excess weight in children. Intrigued by the fact that many children do not engage in recommended levels of physical activity, provision of more opportunities to engage in physical activity in schools (through Physical Education (PE)) as a policy instrument has received a tremendous amount of attention from policymakers and media.

            Accordingly, the research community has dedicated itself to offering evidence that 1) PE promotes children’s moderate-to-vigorous physical activity during PE class; 2) PE contributes to children’s daily physical activity levels; 3) physical activity is negatively associated with children’s Body Mass Index (BMI). However, a limited number of studies have examined the effect of PE on children’s weight, but findings from the previous studies are inconsistent. For instance, while Datar & Sturm (2004) found that additional PE instruction time reduced BMI, Cawley et al. (2012) reported that PE time did not lower BMI z-score and the probability of being obese. This discrepancy might due to the omission of diverse factors which may affect an effective PE implementation. Recently, the growing body of literatures has begun to discuss about why and how schools offer PE class using state PE requirements. However, evidence pertaining to the PE implementation is limited and mostly based on anecdotal evidence.

To fill the gap in the existing literature, this study examines in what context PE would be more effective in reducing childhood obesity, using Early Childhood Longitudinal Study Kindergarten Cohort (ECLS-K) and School Health Policies and Practice Study (SHPPS) 2000 data, sponsored by the National Center for Education Statistics (NCES). Specifically, this study investigates whether the impact of PE on childhood obesity varies depending upon state’s supports on school PE implementation, such as the existence of clearly specified PE goal and regulation on PE assessment and feedback, PE student-to-teacher ratio, and PE teacher qualification and development, etc. This study also examines whether school management in each school, such as sufficient school facility, retaining qualified and well-trained PE teachers, and collaboration with parents, may influence PE impact on weight status.

The preliminary analysis using Ordinary Least Square regression indicates that PE reduced BMI z-score and probability of being overweight of first graders, when they were offered by recommended amount (at least 30 minutes per day and more than 3 days per week) of PE. However, the beneficial impact of PE on weight related health outcome was more prominent when schools and states were supportive in PE implementation, especially when PE class had clearly defined goals, when PE assessments were shared with parents, when PE was offered by certified teachers, and when schools had sufficient PE related facilities and equipment. These findings would provide policymakers and health professionals with valuable information in developing effective PE to ease current childhood excessive weight problem.