Effects of paved shoulders on accident rates for rural Texas highways

The shoulder is one of the most extensively studied roadway elements; however, its effectiveness in reducing accidents has been the subject of much debate. A study is described in which accident rates and characteristics were compared for three different types of rural Texas highways: two-lane roadways without paved shoulders, two-lane roadways with full-width paved shoulders, and four-lane undivided roadways without paved shoulders. Approximately 30 roadways of each type were selected for the study. A rigorous screening procedure was developed to ensure that each site was a typical Texas roadway. A detailed three-year accident history was obtained for each site. More than 1250 km of highway and 16,000 accidents were included in the study data base. For each roadway type, accident rates increased as traffic volume increased. Two-lane highways without paved shoulders had the highest accident rates and were the most sensitive to changes in traffic volume. Two-lane highways with paved shoulders had the lowest accident rates until the traffic volume reached 7500 vehicles/day. At that point, four-lane undivided highways without paved shoulders were safer. Based on the study findings, it was concluded that full-width paved shoulders are effective in reducing the accident rate on rural highways. It also appears that the presence of full-width paved shoulders may reduce the number of rural intersection accidents. (Author)