Abstract: To analyze the factors affecting the injury severity of motorcycle crashes, Nested Logit and Random Parameters Logit models were developed. Motorcycle crashes data from 2013 to 2015 in Indian were collected with a total of 1947 single-vehicle motorcycle crashes occurring during that time period. The parameters of Nested logit and Random Parameters Logit models were estimated under full information maximum likelihood (FIML) method and Monte Carlo method separately. Two model estimation results all show that female, age, helmet use, alcohol use, speeding, run off of road, passenger on vehicle, vehicle age over 10 years old, wet pavement, horizontal curve with slope, intersection, speed limits over 50mph, months of April and July, nights without street lights, rural area, fixed collision objections (guardrail, tree, wall, curb, pole, culvert) are significantly related to motorcycle injury severity. By compared with the AIC value and BIC value of Nested logit and Random Parameters Logit models, showing that the overall model fit of Random Parameters Logit model is better than Nested logit model. The paper can provide a reference and guidance for further analysis of crash injury severity in domestic motorcycle crashes.

Key words: traffic safety, motorcycle crash, crash injury severity, Ｒandom Parameters Logit model, Nested Logit model