The demerit points system (DPS) was introduced in Victoria in 1970 with a view to identifying drivers with high offence rates. The DPS is nominally still in operation, but without the inclusion of speeding and red light offences due to the introduction of the owner onus provision of the photographic detection devices legislation. In all 63,449 standard drivers who committed a second demerit point offence, within three years of the first offence, between March 1982 and February 1985 were studied for the evaluation of the DPS. A statistical model for the evaluation of the effectiveness of the DPS system in reducing traffic offences is presented. The model is based on the assumption that the waiting time between offences follows an exponential distribution. Since traffic offences are relatively rare events, the length of the study period was such that censoring occurs. Under these assumptions, maximum likelihood estimates of the waiting time between offences are determined and the corresponding test statistic is derived. The empirical results of the study showed that in every case, except for drivers with 3-4 years of experience, the interval between the second and third offence was statistically longer than that between the first and second offence. It was concluded that the DPS was responsible for most of the deterrent effect reflected in the increased mean time to the third offence subsequent to a first and second offence detected less than three years apart (A).
Samenvatting