Abstract
Central to every pedestrian network analysis is the mapping of the network. Many studies have used street networks (SNs) to analyze connectivity. Recent studies suggest that analyses based on informal pedestrian networks (e.g., park trails, greenways, or shortcuts) may provide a more robust network representation, however, no studies have examined whether switching to a network consisting of formal pedestrian facilities (FPFs) (e.g., sidewalks and crosswalks) would provide more effective representation of real pedestrian movement and perceptions. Does it yield different outcomes compared to an SN? When does an SN still remain a good proxy, and when should a FPF network (FPFN) be adopted? How worthwhile is it to switch to a FPFN? To address these questions, this paper examines the difference between SNs and FPFNs and how the differences affect the outcomes of network measures. Four U.S. neighborhoods were selected to represent different types of networks. Four classic connectivity measures and four new measures related to crossings were evaluated using both SNs and FPFNs. Comparisons were conducted and the following conclusions drawn: (i) using FPFNs yields greatly different outcomes from using SNs, (ii) an SN can broadly describe the morphological features of a network, especially for a gridiron pattern with narrow streets and small blocks full of sidewalks and crosswalks; however, an FPFN performs much better in describing features which are invisible when using SNs; and (iii) using FPFNs can provide opportunities to develop new measures of performance, such as safety or convenience.
Get full access to this article
View all access options for this article.