Exploring motorcytclist injury severity at T-junction in the UK using the ordered probit/logit models

There exist several studies examining influential factors on the occurrences of different crashes between various motorised vehicles but surprisingly very little has empirically investigated the relationship between various types of crashes, occurring at a specific type of junction, and different motorcyclist injury levels. The complex interaction of factors, for example, different crash types, traffic control measures and the involved vehicles, which influences motorcyclist injury severity resulting from a junction-type accident, is not well explored and understood. In order to untangle the contributions of crash configurations, traffic control measures and the involved vehicles, this study aims at investigating crash consequences conditioned on crash occurrence at T-junctions in the UK, while controlling for environment, vehicle and demographic factors. The statistical modelling techniques applied are the ordered- 7 discrete choice models (i.e., the ordered probit/logit models) using the data extracted from the UK STATS 19 accident injury database (2001-2004), measured on three injury levels: fatal, serious and slight. The modelling results uncover crucial determinants of injury severity among motorcyclists at T-junctions in the UK, and offer an insight into potential prevention strategies that could be undertaken to lessen motorcyclist injury levels. The implications of this study and findings are then discussed.