Analysis of the Logistic Model for Accident Severity on Urban Road Environment

In order to identify the possible contributory factors on urban road environment to accident severity, the logistic regression model was applied to traffic accident data collected from police-reports. A total of 2117 accidents occurred in Shijiazhuang in 2002 were considered for the purpose of this paper. Accident severity (dependent variable) in this paper is a dichotomous variable with two categories: extra serious or major accident and ordinary or minor accident. Because of the binary nature of this dependent variable, the logistic regression model was found suitable. Of nine independent variables obtained from police-reports, five were found most significant associated with accident severity, namely, road cross-section, accident location, road alignment, road type and lighting condition. A statistical interpretation is given of the model-developed estimates in terms of odds ratio concept. The findings show that the logistic regression model as used in this paper provides a better understanding of risk factors related to urban road environment.