Determinants of bicyclist injury severities in bicycle-vehicle crashes: A random parameters approach with heterogeneity in means and variances

2017 ◽  
Vol 16 ◽  
pp. 35-47 ◽  
Author(s):  
Ali Behnood ◽  
Fred Mannering
Keyword(s):  
Vehicle Crashes ◽  
Random Parameters ◽  
Injury Severities

Pedestrian Injury Severities in Pedestrian-Vehicle Crashes and the Partial Proportional Odds Logit Model: Accounting for Age Difference

2019 ◽  
Vol 2673 (5) ◽  
pp. 731-746 ◽  
Author(s):  
Yang Li ◽  
Wei (David) Fan
Keyword(s):  
Logit Model ◽  
Injury Severity ◽  
Parameter Estimates ◽  
Age Difference ◽  
Age Group ◽  
Vehicle Crashes ◽  
Ordered Logit ◽  
Proportional Odds ◽  
Pedestrian Injury ◽  
Injury Severities

This study investigates factors that significantly contribute to the severity of pedestrian injuries resulting from pedestrian-vehicle crashes. Multinomial logit (MNL) models, mixed logit (ML) models, and ordered logit/probit models have been widely used in modeling crash injury severity, including pedestrian injury severity in pedestrian-vehicle crashes. However, both MNL and ML models treat injury severity levels as non-ordered, ignoring the inherent hierarchical nature of crash injury severities, and the data used in ordered logit models need to be strictly subjected to the proportional odds (PO) assumption. In this study, a partial proportional odds (PPO) logit model approach is employed to explore the issues of pedestrian safety associated with each age group: young (aged under 24), middle-aged (aged 25–55), and older pedestrians (aged over 55). Data used in this study are police-reported pedestrian crash data collected from 2007 to 2014 in North Carolina. A variety of motorist, pedestrian, environmental, and roadway characteristics are inspected. Results from likelihood ratio tests statistically show the better performance of developing separate injury severity models for each age group compared with estimating a single model utilizing all data. Relevant parameter estimates and associated marginal effects are used to interpret the results, followed by recommendations made in the concluding section.

Sign in / Sign up

Export Citation Format

Share Document