Researchers suggest using Google Street View and Walk Score to conduct pedestrian safety studies

By Chris McCahill
Pedestrian safety and the effects of environmental features are important considerations in cities everywhere. However, studies on the subject typically rely on two uncommon sources of information: pedestrian volume counts and street audits. Researchers at Columbia University suggest a way to overcome this challenge by using readily available information from Google Street View and Walk Score. Their research was just published in the American Journal of Public Health.
The research team sampled crash data at 532 New York City intersections, then conducted virtual street audits using the Computer Assisted Neighborhood Visual Assessment System (CANVAS), developed and employed at Columbia. Auditors evaluated each location for about 10 minutes in order to determine the presence of crosswalks, sidewalks, curb cuts, pedestrian signals, pedestrian refuges, traffic calming, bus stops, billboards, sidewalk condition, and road condition. According the authors, field audits would take roughly 36 times longer to conduct.
The team estimated pedestrian volumes using a New York-specific model that incorporates census data, commercial zoning data, and transit ridership data. They then compared those estimates to values from Walk Score and found that the two were highly correlated, suggesting that Walk Score could be a useful surrogate in places without pedestrian counts or models.

Relation Between Estimated Pedestrian Count and Walk Score for 532 Intersections in New York City in 2013 (Source: Mooney et al. 2016)
Relation Between Estimated Pedestrian Count and Walk Score for 532 Intersections in New York City in 2013 (Source: Mooney et al. 2016)

Similar to prior studies, the researchers found the number of crashes per pedestrian is lower at intersections with more pedestrian traffic. However, they observed higher crash rates at intersections with marked crosswalks, pedestrian signals, bus stops, and billboards. They urge caution in interpreting these findings. Crosswalks, for example, may give people a false sense of security, but they are also often installed at inherently dangerous crossings. Similarly, bus stops and billboards may cause more conflicts and distractions, but they are also often located in high-traffic areas.
Using CANVAS, Google Street View, and Walk Score, they conclude, is a “promising mechanism to not only reduce costs of, but also increase geographic scope of, location-based studies of pedestrian injury risk.”
Chris McCahill is a Senior Associate at SSTI.