Objective Data

Neighborhood observations

Existing built environment data from GIS-based land use databases should be supplemented by pedestrian infrastructure measures obtained from observational instruments. The direct observations should be conducted on representative street segments and intersections measure features such as sidewalk availability and quality, buffers between streets and sidewalks, street lighting, trees, aesthetics such as views, gathering spaces, sidewalk furniture, and architectural form. We have piloted an audit tool in the NQLS adult project and items demonstrate good inter-rater reliability. Several urban form attributes from existing GIS data, such as land use type, street network type, building age, and bus stop presence, should be used to select representative street segments and intersections to be surveyed. All these data can be easily viewed using GIS software to display the attributes and to select and map the sample of streets to be observed. A complete evaluation of all street segments and intersections in study neighborhoods is not feasible, so 10-15% of street segments and intersections in each neighborhood should be evaluated. You could observe all commercial street segments within the neighborhoods and a random sample of residential street segments because of greater uniformity. If neighborhoods do not have commercial areas within their boundaries, observe the closest shopping center. Generalization of attributes from the evaluated street segments and intersections to all others within a neighborhood can be done based on neighborhood typologies defined by macro-scale attributes of neighborhood walkability. With latent class analysis, classes of micro- and macro-scale variables which covary have been determined in the NQLS project (unpublished data). Using these classes, unsurveyed streets and intersections can be assigned pedestrian infrastructure attributes based on their macro-scale attributes.

Microscale Audit Tools

Environment Crime Weather Physical activity Neighborhood observations