Abstract
BACKGROUND: The schoolyard environment provides key opportunities to promote physical activity and socioemotional development for children. Schoolyards can also serve as a community park resource outside of school hours. We aimed to: (i) implement and evaluate reliability of the System for Observing Outdoor Play Environments in Neighborhood Schools (SOOPEN), (ii) assess schoolyard use by children during recess and community members of all ages outside of school hours, and (iii) investigate relationships of schoolyard and children´s group characteristics with physical activity levels and prosocial interactions. METHODS: In this cross-sectional study, we observed student and community visitor behavior using SOOPEN at three urban elementary schoolyards in Tacoma, Washington, USA, prior to renovations intended to expand each facility's use as a community park in neighborhoods with poor park access. We assessed interrater reliability using intraclass correlation coefficients and described current levels of schoolyard use (at the group level), physical activity, and prosocial behavior. Physical activity was assessed on a five-point scale and dichotomized to indicate moderate-to-vigorous physical activity (MVPA). Social interactions were coded as prosocial, antisocial, or neutral. We examined associations of selected schoolyard features and group characteristics with group MVPA and prosocial behavior during recess using modified Poisson regression to estimate prevalence ratios (PR) and 95% confidence intervals (CI). RESULTS: We observed a total of 981 activity-defined, informal groups in the schoolyards, and achieved good to excellent interrater reliability using SOOPEN. Community use of the schoolyards during evenings and weekends was limited (n = 56 groups). During 26, 25-50 min recess periods (n = 833 groups), 19% of groups were engaged in MVPA. Schoolyard areas with paved surfaces were associated with more MVPA (PR = 1.52, 95% CI: 1.04, 2.23) compared to field/grass areas; supervised groups were associated with less MVPA than groups not directly supervised by an adult (PR = 0.59, 95% CI: 0.36, 0.96). Schoolyard characteristics were not associated with prosocial behavior. Mixed-gender groups were associated with more MVPA and more prosocial behavior. CONCLUSIONS: Our study using SOOPEN, a reliable new activity observation tool, highlights the multi-dimensional dynamics of physical activity and social interactions in schoolyards, which could be leveraged to promote healthy behaviors during and outside of school hours.