Comparison of field and online observations for measuring land uses using the Microscale Audit of Pedestrian Streetscapes (MAPS)

Land use mix reflects the availability of diverse destinations providing opportunities for active transportation. Online mapping platforms (e.g. Google Maps) show promise for measuring neighborhood features due to their ease of use and accessibility. The purpose of this study was to evaluate the validity and reliability of an online version of the Microscale Audit of Pedestrian Streetscapes (MAPS) tool by comparing Google's Aerial View (AV), Street View (SV), and a composite of the two views (Composite) to in-person field audits.Raters evaluated residential routes and commercial clusters (n=120) using the online MAPS tool in low and high walkability and socioeconomic status (SES) neighborhoods of two U.S. regions. Thirty land uses were evaluated for each route and cluster. Concurrent validity between field and online audits was examined using percent agreement and weighted kappa statistics (κ) for individual items and intra-class correlation coefficients (ICCs) for item subscales. Reliability was examined using ICCs.Most field-to-online land use item comparisons showed moderate to perfect agreement (κ=0.41-0.88). SV (10 out of 19 comparisons) outperformed AV (7 out of 19) and Composite (9 out of 19) compared to field audits. Subscale comparisons were fair to almost perfect (ICC 0.27-0.89). AV (7 out of 10 subscales) outperformed SV compared to field audits. Reliability between AV and SV was substantial to perfect (ICC 0.61-1.00) for most items (15 out of 21 eligible items) and substantial to almost perfect (ICC 0.61-0.96) for 9 of the 10 subscales. Online approaches did not differ after stratifying routes by SES.Audits using online MAPS were an acceptable alternative to evaluating land uses in the field. The Street View method is recommended based on better performance with individual land uses and equal performance in high and low socioeconomic status neighborhoods.