Background
The physical and social environments of neighborhoods and schools likely influence patterns of participation in low and moderate intensity physical activity. Objective physical activity measures will be essential to identify the effects of ecologic context. This proposal uses ecosocial theory to examine how physical and social environments of schools and neighborhoods shape the routine physical activities of youth. Informed research must be cognizant of the type and setting of physical activities and the attributes of physical movements occurring in each ecologic context.

Specific Aims
1. Develop and assess objective measures of active neighborhoods (i.e., neighborhood accessibility, pedestrian context, open space), active schools, and local weather environments that are hypothesized to promote time- and context-specific physical activity using GIS data, site visits, archival records, and orthophotos and pictometry.
2. Define TriTrac measures of time- and context-specific physical activity including the level of physical activity performed during the journey to school, the school day, after-school, in the evening, and on weekends.
3. Test hypotheses regarding influence of active neighborhood and school environments on time- and contextspecific measures of physical activity. Example hypotheses:
3.1. Neighborhood pedestrian context (e.g., level of sidewalk completeness) will be directly associated with moderate and vigorous physical activity on weekend days.
3.2. Active school environments (e.g., greater number of sport and recreation facilities) will be associated with increased levels of moderate and vigorous physical activity during the hours of the school day.
4. Test hypotheses regarding points of interaction between active environments and local weather environment, and between active environments and individual-level variables. Example hypothesis:
4.1. The unfavorable effect of inclement weather on active journeys to school will be greater in areas with a lower pedestrian context (e.g., lower level of sidewalk completeness)
5 Document methods and disseminate findings via presentations at meetings (RWJF, ACSM), through publication, and professional associations with local and statewide networks (Metropolitan Area Planning Council, MA Coalition on Obesity Prevention and Education, New England Childhood Obesity Network, Harvard Prevention Research Center website).

Methods
This multi-level, observational study will supplement 2 million person-minutes of TriTrac accelerometer physical activity data from a stratified, random sample of 251 students in 10 neighborhoods with data characterizing neighborhoods, schools and weather. TriTrac accelerometers are reliable, valid measures of physical activity in youth. Social and objective physical characteristics of schools and neighborhoods will be assessed retrospectively by site visit, survey data, orthophotos and pictometry, and publicly available GIS data corresponding to the years of physical activity assessment. Reliability of environmental measures will be assessed. Estimated hierarchical regression models will examine associations between school and neighborhood environmental variables and relevant time- and context-specific objective and self-report physical activity behaviors.