Injury severity analysis for car, pickup, sport utility vehicle and minivan drivers: Male and female differences.

摘要

This research explores how factors affecting the probability of injury severity differ between male and female drivers in single-vehicle accidents involving a car, pickup, sport utility vehicle, or minivan; and in two-vehicle accidents involving a car vs. pickup, sport utility vehicle, or minivan. Multinomial logit probability models are estimated to predict the probability of four injury severity categories: no injury (property damage only), possible injury, evident injury, and fatal/disabling injury; conditional on an accident occurring. Separate models are estimated for male and female drivers for the different vehicle types.; The gender-specific models are compared statistically for each vehicle type, showing significant differences between them. For example, there is a greater probability of more severe injury for sport utility vehicle/minivan drivers on wet roads while the opposite is true for car drivers, suggesting sport utility vehicle/minivan drivers do not compensate enough in speed and caution on wet roads. Curved road increases the probability of fatal/disabling injury for female pickup drivers in single-vehicle accidents, and defective tires or wheels are associated with higher probability of more severe injury for female sport utility vehicle/minivan drivers in single-vehicle accidents, suggesting behavioral differences between the genders when responding to emergencies. The results indicate that drivers of pickups, sport utility vehicles, and minivans need to show more caution when merging or changing lanes. There is a notable difference between male and female drivers of cars that collided with a barrier or guardrail as a part of a collision with another vehicle. The probability of fatal and disabling injuries increases for females; conversely, the probability of no injury or possible injury increases for males.; The results show significant differences between the genders when considering conditional models of injury severity, suggesting important behavioral and physical differences. The behavioral differences need to be studied with experiments and addressed with additional driver education. The effects of the physical differences should be studied using crash-test dummies that represent both genders when crash-testing vehicles against each other, barriers, and guardrails, since vehicle and roadway safety features affect the genders differently.