Physical activity by schoolyard location : current literature, comparison with activity type, and use of a GPS-based approach

"Participation in physical activity is associated with numerous benefits, including maintenance of healthy weight status, reduced risk of chronic disease, and improved psychosocial outcomes. Promoting this behavior in young children is particularly important due to the potential for life-long benefits and for tracking of this health behavior into adolescence and adulthood. The physical environment affords specific behaviors, which in turn impacts participation in physical activity. Understanding how the environment in venues such as schools affects children's physical activity can inform the development of interventions that promote physical activity attainment and the associated short- and long-term benefits in large numbers of children. Of particular interest is the schoolyard, as outdoor time is one of the main opportunities for free-play during the predominantly sedentary school or childcare day. Prior research has suggested that certain aspects of the schoolyard environment, such as having an open, grassy field, are important for facilitating physical activity in children. The first aim of this dissertation was to systematically review the literature on how youths' physical activity varies by location on the schoolyard. Only 24 studies met our inclusion/exclusion criteria and generally supported that physical activity is not uniform across the schoolyard. There is also evidence to suggest that the locations that elicit the highest levels of physical activity vary by age group (e.g., preschool vs. adolescence) and by sex. This information could be used to inform interventions to promote physical activity participation in all youth, including those with the lowest levels of physical activity, such as adolescents and girls. However, this review brought to light a number of gaps in the literature- namely the lack of control of or consideration for other potentially impactful variables, such as the time of year or provided portable equipment. To understand the role of the physical environment, in conjunction with and separate from these other variables, a method is needed that can be used to measure physical activity by schoolyard location in a large number of schools, children, settings, and situations. Recent research has supported the use of monitor-based approaches, like Global Positioning Systems (GPS) receiver units, which collect continuous, individual-level data on a child's location, and, when paired with accelerometry, physical activity intensity. However, the majority of studies have relied on direct observation and, in preschoolers, this most often involves recording activity type and physical activity level, not location. It is unknown how comparable measuring physical activity by location is to prior research that used a physical activity by activity type-based approach. The second aim of this dissertation was to identify how similar activity type and location are in a preschool sample. We report that children participated in multiple types of activities within each location, but all locations had a primary activity type, in which children were engaged for 52.7--94.5% of their time in that location. The majority of outdoor time was spent in open spaces (type or location; 37.6--48.9%), but there was a disparity in the setting that promoted the highest activity levels between methods (teacher arranged activity type, but fixed equipment location). This finding provides information about the limitations of using a location-based methodology, like GPS plus accelerometry. A second issue when using GPS plus accelerometry is lack of consensus as to how to analyze these data. The final aim of this dissertation was to explore the use of hot spot analysis with spatiotemporal weights matrices to identify statistically significant clusters of high/low accelerometer vector magnitude counts/15-s. The location of hot/cold spots changes over the course of the outdoor period (intraperiod) and over the course of a day (interperiod), supporting our hypothesis that both location and time are important when aiming to understand children's physical activity participation during provided outdoor time. GPS plus accelerometry is a viable method for understanding both where and when children are being active, but there are methodological limitations that need to be overcome before this methodology can be used on a large scale to inform the design of and to assess the impact of schoolyard interventions."--Pages ii-iii.