Reid Ewing, William Greene, Amir Hajrasouliha, Kathryn M. Neckerman & Marnie Purciel-Hill
Ewing, R. , Greene, W. , Hajrasouliha, A. , Neckerman, K. M. & Purciel-Hill, M. (1). Streetscape Features Related to Pedestrian Activity. Journal of Planning Education and Research, 36(1), 5–15. http://dx.doi.org/10.1177/0739456X15591585
By measuring twenty streetscape features and numerous other variables for 588 blocks in New York City, we were able to identify variables that explain pedestrian traffic volumes. We found significant positive correlations between three out of twenty streetscape features with pedestrian counts after controlling for density and other built environmental variables. The significant streetscape features are the proportion of windows on the street, the proportion of active street frontage, and the number of pieces of street furniture. This study provides guidance for streetscape projects that aim to create walkable streets and pedestrian-friendly environments.
This study seeks to explain pedestrian counts on 588 block faces in New York City in terms of D variables— development density, land use diversity, street network design, destination accessibility, distance to transit, and one demographic variable, household size—plus micro streetscape features. The study finds that the proportion of active uses along the block face, number of pieces of street furniture and other street items, and proportion of first floor with windows are three significant streetscape variables in combination with the standard D variables. Additionally, the three density measures (buffer FAR, buffer population density, and block FAR) and block length are directly and significantly related to pedestrian counts. The research demonstrates that urban design generally, and streetscape design features (the proportion of active uses along the block face, number of pieces of street furniture and other street items, and the proportion of first floor with windows), in particular, have a significant influence on pedestrian activity.
Description of method used in the article
The study drew on characterizations of D variables from Ewing and Cervero (2010) and Ewing et al. (2011) for control variables. Secondary data used were publicly available Geographic Information System (GIS) data for the study area from the New York City Department of City Planning, including DCPLION (the street centerline GIS file from City Planning), street segment centerlines, and MapPluto™ parcel layers, Census 2010 SF1 100% and Tiger 2010 Census Block shapefiles to calculate roadway network, land use, and demographic variables. The full set of variables were analyzed using negative binomial regression which was dictated by the distribution of the dependent variable, the average pedestrian count for four passes up and down each block face rounded to the nearest integer.
Of practical use