Abstract: This study identifies and quantifies the effects of different explanatory variables that increase the severity of drivers' injuries related to single-vehicle collisions involving light-duty vehicles. The research is based on utilizing logistic regression to analyze records of all traffic collisions that occurred in North Carolina for the years from 2007 to 2013. The study also investigates temporal stability of the identified explanatory variables throughout the analysis period. The identified explanatory variables include those related to the roadway, vehicle, driver, and environmental conditions. The explanatory variables related to the roadway include whether the roadway is divided or undivided, and whether it is in an urban or rural area. The explanatory variables related to the vehicle include vehicle's age, travel speed, and the type of the light-duty vehicle. The explanatory variables related to the driver include driver's age, gender, influence by alcohol or illicit drugs, and the use of seatbelt. The explanatory variables related to the environmental conditions include weather, lighting, and road surface conditions. Three of the investigated explanatory variables were found to be temporally unstable with significantly varying effects on the severity of drivers' injuries. Those temporally unstable variables include the travel speed, the type of the light-duty vehicle, and the age of the driver. All other investigated variables were found to be consistently significant throughout the analysis period. The findings of this research have the potential to help decision makers develop policies and countermeasures that reduce the severity of drivers' injuries by focusing on explanatory variables that consistently exhibit significant effects on the severity of drivers' injuries. The findings of this research also provide quantitative measures that may be used to determine the feasibility of implementing those countermeasures in reducing the severity of drivers' injuries related to single-vehicle collisions. Recommendations for future research are also provided.