The skyrocketed prevalence of childhood obesity is one of the most serious health concerns in the U.S. It is well known that pediatric obesity is associated not only with increased risk for physiological problems but also with increased risk of psychological and psychosocial problems. Children with higher BMI are more likely to be obese adults who are susceptible to chronic illnesses and decreased life expectancy. Various childhood obesity programs are targeted at schools such as more hours of physical education, healthier school lunch menus, and elimination of soda machines. Previous research studies examine the effectiveness of those interventions in schools. However, the fundamental question on whether schooling itself leads to obesity among children is not yet clearly answered. The question we attempt to answer is not about whether higher human capital accumulated by more schooling reduces obesity or not. Rather, our question is about whether more years of rigid and regulated school life, especially in public schools, affects children’s obesity and health outcomes.  

Children’s exposure to formal school life is endogenous because it is parental choice and is aligned with children’s physical growth. Hence, it is not easy to examine the impact of schooling on obesity outcomes of children with observational data. To deal with this endogeneity, we use a natural experiment to examine the effect of schooling on children’s obesity outcomes. School districts have set different cutoff dates for children’s admission to kindergarten (i.e. turn-five cutoff date is based on the child’s birthdate). Using different turn-five cutoff dates, we compare students whose birthdays are three months before the cutoff date to those whose birthdays are three months after the cutoff. These cutoffs serve as a natural experiment where students belonging to the former group have one more year of education compared to those in the latter group though they are of the same age. Using this natural experiment and regression discontinuity design (RDD), we propose to investigate whether having more formal education leads to more or less obesity. To address our research question, we use data from the Early Childhood Longitudinal Study-Kindergarten Class of 1998-1999 (ECLS-K), a nationally representative cohort of Kindergarteners in the U.S. The ECLS-K study is appropriate for our research since it follows about 20,000 children from kindergarten in the fall of 1998 to the eighth grade in 7 waves of surveys and contains detailed information on children and parent, teacher, and school characteristics to analyze, physical health and growth such as objective measurement of height, weight and BMI at all waves. The turn-five cutoff date for entering kindergarten is recorded by school principals in the data.