Modeling the Severity of Motor-Vehicle Accidents by Different Age Groups of the Driver using Multinomial Logit Regression

The issue of how to make roads safer for drivers in different age groups is important due to the constant change in the age distribution of drivers. With this in mind, the purpose of this study is to investigate the different characteristics of accidents and to provide suggestions on how to improve road safety for all age groups. This study uses the Road Safety Information Database (HSIS) to determine the factors influencing the severity of motor vehicle accident injuries for young drivers (25-25 years old), middle-aged drivers (64-26 years old) and elderly drivers (over 64 years old) in California between 2015 and 2017. A multinomial logit model was used to model the accidents related to each age group and to evaluate the weight of different variables predicting the severity of the driver injury. The predictor variables were classified into four characteristics: driver, road, accident and environment. The results show that rear-end collisions significantly increase the risk of death and injury for all three age groups, while clear weather and rainy weather reduce the risk of death and injury for all three age groups. In addition, the results show that there are many similarities between the predictors of accident severity for young and middle-aged drivers, but older drivers should be more careful in environmental and road conditions compared to other age groups. Young drivers improve their driving skills over time due to less driving experience, while middle-aged drivers, especially male drivers, have more driving experience and therefore tend to engage in less risky behaviors. Another result of this study is that the driving behavior pattern of older drivers have less variation than other age groups regardless of the surrounding conditions.