|The aim of this study was to investigate the relationship between vehicle performance (defined by the number of cylinders and the power-to-weight ratio of the vehicle) and the risk of serious injury crash involvement among Western Australian drivers, particularly for those drivers aged 17-19 years of age (young novice drivers).
Western Australian crash data were analysed in two main ways. Firstly, the rates of crashing by young novice drivers were compared across categories of vehicle performance. The objective was to see if young drivers in high powered vehicles crash more often than young drivers in lower powered vehicles. Crash rates for older drivers were also calculated.
Secondly, the ratio of single-vehicle to multiple vehicle crashes was calculated for various categories of driver-vehicle combinations. This ratio is designed to be a sensitive indicator of crash risk on the basis that as riskiness increases, the number of single vehicle crashes increases more than the number of two-vehicle crashes. This second analysis used multiple logistic regression, which is a method that allows various other factors that are likely effect the ratio to be accounted for.
Specially constructed datasets were used for the study, as additional data was required on the driver and the performance of each crash involved vehicle. To this end, linkages were made between crash data and TRELIS, the WA registration and licensing database, and with automotive data maintained by RL Polk Australia. The former database provided information on the driver and owner of the vehicle, while the latter provided information on the make, model, performance and other characteristics of each crash involved vehicle.
Crash rates were expressed as crashes per registered vehicle. Hence, a snapshot of the registered vehicle fleet was obtained from Licensing division of Main Roads WA. Additional information on these vehicles was also provided by RL Polk Australia.
A secondary aim was to characterise the WA passenger fleet in respect of performance levels and levels of safety. The registered sample was used for this purpose as well.
The process of data extraction and data linkage is described in more detail in section 2. Section 3 describes the rate of crashing of various driver and vehicle combinations, and section 4 examines how driver age and vehicle performance affects the ratio of single- and two-vehicle crashes. Finally section 5 describes the availability of safety features across categories of vehicle performance.
The study was approved by the Institutional Human Ethics Committees of Monash University, Curtin University and the University of Adelaide.