The Effects of the All-Red Clearance Interval on Intersection Accident Rates in Indiana

Abstract

Seven years of accident data were obtained to assess the accident rates for two randomly sampled groups of intersections. One group was treated with the all-red interval and the other was not. Each comparison group intersection was paired with one in the treated group, based on entering volume, angle of intersect, and approach speed limits. From traffic volume counts, the estimated exposure rate at each intersection was calculated. Rates for four different accident types (left turn, rear end, right turn, and right turn angle) and for the total number of accidents were determined. Several statistical methods were used to analyze the resulting accident rates. In addition, three of the early all-red interval accident studies were reproduced with Indiana data to see if the results would be the same. The results from the statistical analysis contradict those of previous studies. The Wilcoxon Signed Rank Test yielded few differences between the treatment and comparison intersection accident rates. The Student's t-test yielded a difference in right angle accident rates at the treated intersections one year before and after the installation of the all-red interval. All other Student's t-test comparisons resulted in no difference. The Chi-square test results showed few differences in accident rates over four years. Thus, the all red clearance interval did not cause a reduction in accident rates at the intersections studied in Indiana. Possible explanations and suggestions for further research are presented.