Pedestrian Environments and Transit Ridership

This paper explores how the quality of the pedestrian environment around transit stops relates with transit ridership. The primary hypothesis tested is that transit tripmaking is higher in urban environments that are more conducive to non-motorized travel, given that bus transit systems are most frequently accessed via walking or biking. A secondary goal is to contribute to an improved understanding of the measurement of the built environment in geographic information systems (GIS). A composite measure of walkability—incorporating land use mix, density and street patterns—was developed for all transit stops in San Diego’s Metropolitan Transit Systems service area and used as a measure of the built environment. Findings indicate a small but significant, positive relationship between the walkability of the built environment and transit ridership. Introduction This research assesses the relationship between transit ridership and the quality of the pedestrian environment near bus transit stops. Academic and professional planners have theorized about the importance of the built environment in shaping an individual’s travel behavior (Ewing and Cervero 2001; Handy 1996; Frank, Engelke, and Schmid 2003; Krizek 2003). The current research tests the nature of this relationship using bus ridership and built-environment data from the San Diego region. The primary hypothesis of this research is that transit trip-making is Journal of Public Transportation, Vol. 12, No. 1, 2009 40 higher in urban environments that are more conducive to non-motorized travel, given that bus transit systems are most frequently accessed via walking or biking. A secondary goal is to contribute to an improved understanding of the measurement of the built environment in geographic information systems (GIS). Literature Review Several important strands of literature examine the relationship between various human behaviors and the built environment. Previous studies have assessed associations between the built environment and a range of human behaviors, including travel behavior, community capacity-building behaviors, criminal behavior, and health-promoting behaviors. Highlights of this research are summarized in the following paragraphs, with a focus on delineating how researchers from different disciplines approach measurement of the built environment, especially in regard to methods and spatial units of analysis. Travel Behavior Much of the relevant travel behavior literature focuses on improving our understanding of how the built environment influences an individual’s travel mode choice, specifically the decision to drive versus walk, bike, or use public transit (Zhang 2004). New urbanists and smart growth advocates suggest that changes in the built environment could lead to more frequent decisions to use non-motorized modes of travel (Duany, Plater-Zyberk, and Speck 2000; Calthorpe 1993; Frank et al. 2004). Aspects of the built environment thought to influence travel mode choice include population or residential density, land use mix, and characteristics of the transportation system (Cervero and Kockelman 1997). The travel behavior literature employs a range of built-environment measurements where the unit of analysis varies from the street block at the smallest level up to the metropolitan area in its entirety. Recent research by Boarnet et al. (2005), for example, employs an environmental assessment survey where field observers walked from street block to street block within a particular study area noting specific attributes of each block, such as sidewalk widths, presence of trash, and presence of tree canopy. Krizek (2003), Song and Rodriquez (2005), and Frank et al. (2005) employ an artificial grid composed of small cells (ranging from 150m x 150m to 1km x 1km) laid out across an urban area and then utilized to calculate measures of the built environment. Frank et al. (2004; 2006) measured the built environment within a one-kilometer road network distance of where people live.