Patterns of motor vehicle crash involvement by driver age and sex in Hawaii

The purpose of this paper is to develop a statistical description of patterns of motor-vehicle crash types among drivers of different age and sex in order to identify underlying differences in behavior and ability. Using techniques of categorical data analysis and comprehensive data on crashes in Hawaii during 1991 and 1992, we relate crash type (e.g., sideswiper, sideswiped, rearender, rearended, etc) to driver age and sex categories. We also examine interactions between driver age and sex, crash type, and vehicle type. By fitting a loglinear model, we find that different crash types are associated with different driver groups in a way that suggests systematic differences in driver behavior and ability In particular, young drivers have much greater frequency of rollovers and of being the rearender or headoner, whereas older drivers have much higher frequency of being rearended, sideswiped, or broadsided. The most notable effect of vehicle type is the much higher frequency of rollover for pick-up trucks, a vehicle much more frequently driven by young male drivers. We conclude with a discussion of the implications of our findings for appropriate interventions and future research.