The relationship between cluster-analysis derived walkability and local recreational and transportation walking among Canadian adults

Gavin R McCormack, Christine Friedenreich, Beverly A Sandalack, Billie Giles-Corti, Patricia K. Doyle-Baker, Alan Shiell

We investigated the association between objectively-assessed neighborhood walkability and local walking among adults. Two independent random cross-sectional samples of Calgary (Canada) residents were recruited. Neighborhood-based walking, attitude towards walking, neighborhood self-selection, and socio-demographic characteristics were captured. Built environmental attributes underwent a two-staged cluster analysis which identified three neighborhood types (HW: high walkable; MW: medium walkable; LW: low walkable). Adjusting for all other characteristics, MW (OR 1.40, p < 0.05) and HW (OR 1.34, approached p < 0.05) neighborhood residents were more likely than LW neighborhood residents to participate in neighborhood-based transportation walking. HW neighborhood residents spent 30-min/wk more on neighborhood-based transportation walking than both LW and MW neighborhood residents. MW neighborhood residents spent 14-min/wk more on neighborhood-based recreational walking than LW neighborhood residents. Neighborhoods with a highly connected pedestrian network, large mix of businesses, high population density, high access to sidewalks and pathways, and many bus stops support local walking.