Introduction

It is important to understand the relationship between the built environment and pedestrian safety to promote walking. Nevertheless, current literature insufficiently studied the impact of the built environment on pedestrian safety. There is few research considering spatial autocorrelation effects. This study aims to expand the investigation of conventional built environment correlates of pedestrian safety using the D variables from the walkability literature to further the discussions of the association of the built environment with both safety and walkability.

Methods

Data on pedestrian involved injuries were collected from the New York State Department of Transportation from 2012-2016. The results from the global Poisson regression (PR) models and the geographically weighted Poisson regression (GWPR) models were compared to show the spatially varying effects of built environment characteristics on pedestrian involved injuries. Data were analyzed in 2019.

Results

The results from the GWPR models were more pronounced than the global PR models. The findings confirmed the association of pedestrian safety with the D variables. The number of intersections, Walk Score, and population density showed conflicting impacts on safety and walkability. The statistically significant association of Walk Score with pedestrian safety in the study enriches current research on the relationship between built environment and pedestrian safety.

Conclusions

This study expanded the traditional built environment variables used in pedestrian safety research with the D variables in walkability research to help understand the role of built environment characteristics for building walkable and safe neighborhoods. The findings also shed light on the significance of distance in environmental modification for safe and walkable communities.

中文翻译：

---

建设适合步行且安全的社区：评估纽约州布法罗市行人安全的建筑环境特征

介绍

了解建筑环境与行人安全之间的关系以促进步行非常重要。然而，目前的文献没有充分研究建筑环境对行人安全的影响。很少有研究考虑空间自相关效应。本研究旨在使用步行性文献中的 D 变量扩大对传统建筑环境与行人安全相关性的调查，以进一步讨论建筑环境与安全性和步行性的关联。

方法

2012 年至 2016 年从纽约州交通部收集了有关行人受伤的数据。对全局泊松回归 (PR) 模型和地理加权泊松回归 (GWPR) 模型的结果进行了比较，以显示建筑环境特征对行人受伤的空间变化影响。数据在 2019 年进行了分析。

结果

GWPR 模型的结果比全球 PR 模型更明显。调查结果证实了行人安全与 D 变量的关联。十字路口的数量、步行得分和人口密度对安全性和步行性的影响相互矛盾。研究中步行得分与行人安全的统计学显着关联丰富了当前关于建筑环境与行人安全之间关系的研究。

结论

本研究将行人安全研究中使用的传统建筑环境变量扩展到可步行性研究中的 D 变量，以帮助理解建筑环境特征对构建适合步行和安全的社区的作用。研究结果还阐明了距离对安全和适宜步行社区的环境改造的重要性。