The Effect of Socio-economic and Traffic Factors on Pedestrian Crossing Behaviors: Inclusion of Latent Variables in Multinomial Logit Models
Zhuping Zhou,Gang Ren,Wei Wang,Yang Cheng,Liteng Zha,Haiyuan Li
2012-01-01
Abstract:Traffic accidents involving pedestrians make up a large proportion of road casualties, and they are more likely to occur at signalized intersections. The primary objective of this paper is to study the factors that influence the behaviors of regular users, late starters, sneakers, and partial sneakers. Behavior information was observed manually in field study. After obtaining the behavior data, the survey team distributed a questionnaire to the same participant who has been observed, in order to acquire detailed demographic and socioeconomic characteristics as well as attitude and preference indicators. In total, 1878 pedestrians were surveyed at 16 signalized intersections in Nanjing, China. First, correlation analysis is performed to analyze the effect of each factor separately. Then, five latent variables including safety, conformity, comfort, flexibility and fastness are obtained by structure equation modeling (SEM). Based on the results of SEM, a multinomial logit model with latent variables is developed to describe how the factors influence pedestrians' behavior choices. The following conclusions are drawn from the model: (1) the presence of oncoming cars, crosswalk length and green time are the most significant factors when pedestrians choose to be late starters, (2) gender, oncoming cars and trip purpose have the most significant effects on the pedestrians to be sneakers, and (3) for the choice of being partial sneakers, age, gender and oncoming cars are the most important factors.