Institute of Transportation, MOTC

       The objective of this study is to evaluate the impact of sidewalk design on pedestrian safety. As the most promising study to pedestrian accident reduction is to try to accommodate those road users by treating their environment. However, studies usually focused on pedestrian facility at intersection compared to the roadway approach. Moreover, they didn’t separate the accidents from intersection and roadway approach. Meanwhile, building suitable sidewalk and sidewalk width isn’t random, which means it could be endogenous by sample selection bias. Therefore, this study focused on roadway pedestrian facility and coded the accident scene which is at the roadway approach or not. Then, use probit regression with endogenous variables which is the structural equation model to treat the endogenous, including built sidewalk width and suitable sidewalk (this study defined the suitable sidewalk as a continuous and a compliance of regulatory sidewalks). Due to high fatality rates and more complete sidewalks, this study utilizes crash data, accident scene data, traffic flow data, roadway geometric design data including sidewalk data and socioeconomic data from Taipei during 2017 to 2021. The results show that built suitable sidewalk and sidewalk width are endogenous. The counterfactual model counted average treatment effect of built suitable sidewalks could reduce 8.81% pedestrian accident rate compared to non-suitable sidewalks. Finally, pedestrian accident decreases with sidewalk width at a rate such that if the rate were constant, pedestrian accident rate would decrease by 19.32% if sidewalk width increased by 1 meter. To sum up, sidewalk continuity and sidewalk width would lead to pedestrian safety, then this study can be better understanding how to precisely improve pedestrian safety.