Mixed-effect logit modeling of red-light violations among motorcyclists

Abstract

Abstract Red-light violations have been associated with road traffic crashes across the globe. This study was conducted to determine the rate of red-light violations among motorcyclists in the Accra metropolis, Ghana, and the associated risk factors. Observational data collected at four signalized intersections were used. Possible risk factors for red-light violation were determined using mixed-effect logistic regression model. The results showed that 64% of motorcyclists violated the red-light. The results further revealed that motorcyclists with pillion passengers were more likely to violate red-lights. Also, motorcyclists were more likely to violate red-lights in the evenings, on weekends and when the traffic cycle length was more than two minutes. The study also found that motorcyclists were less likely to violate red-lights at T-junctions and during times that other motorcyclists stop when a red traffic signal is on.