Grant: Project Grant

There is growing interest in how the neighbourhood built environment (or "walkability") shapes human behaviour, including physical activity. Studies have found positive associations between walkability and physical activity, and in particular recreational and transportation walking. Despite promising findings, most studies to date have estimated associations between walkability and walking from cross-sectional data. Cross-sectional data however, cannot provide causal evidence regarding the temporal direction of associations. Thus, based on current evidence it is difficult to determine if the estimated associations reflect higher walkability causing more physical activity or if the estimated associations just reflect the non-random process of already physically active adults seeking out neighbourhoods that have high levels of walkability (i.e., residential self-selection). Longitudinal residential relocation studies (a type of natural experiment) can provide causal evidence regarding the temporal direction of associations between the built environment and physical activity. We will use existing longitudinal data (n=10,330) collected from a random cross-section of Albertan adults (35-69 years) to investigate the extent to which changes in objectively-determined walkability are associated with changes in transportation and recreational walking following residential relocation. Evidence from this study will provide causal evidence that will be useful for urban form and transportation planners and policymakers in creating walkable neighbourhoods in Canada and elsewhere.