This study investigates techniques for studying the incidence of unsafe driving actions (UDAS) in both the accident and normal driving population in an attempt to develop an overall plan for measuring relative crash risks of UDAS. Stage 1 reanalysed the Adelaide in depth study coding driver behaviour with definitions from previous studies. During this procedure it became apparent that an improved system for assigning UDAS was required. This led to the development of a flow chart technique which was found to be both reliable and sufficiently detailed. Stage 2 applied the flow chart technique to police crash report forms. This resulted in too many 'unknown' responses to flow chart decisions. A pilot study suggests that additional information held at police headquarters may be the most successful way to reattempt stage 2. Stage 3 piloted several techniques for observing udas in the normal driving population (A). This report is also Road Traffic Authority Report No: SR/87/4; and Australia Office of Road Safety ; Report No CR 58.
Abstract