Resumen
Post-license advanced driver training addresses different categories of road users such as: novice drivers, professional drivers, company employers and recidivists. These training courses can be carried out on-track or on the road. On-track courses allow participants to gain knowledge on driving physics and experience limits in a safe road environment. On-road courses are more focused on hazard perception and situation awareness. Although extensive research has been done in this field, knowledge of the effects of these courses on road accident risk remains unclear. Previous evaluation of on-track courses did not always show a positive effect on crash rate. For example, post-license training focused on mastery of driving skills can lead to an increase of accident risk, especially on young males. However, research identified several factors that may enhance the effectiveness of driving training. In Europe a new framework for driver education and training has been proposed based on a safe driver hierarchical model (the GADGET model) and the development of a strategy for continuous learning. According to this framework, an evaluation study of on-track post-license advanced driver training has been undertaken in Italy with the main goal of assessing the safety effects of these courses and identifying training aspects to be improved. Besides crash rate, the study aims at assessing also driver behavior, knowledge of risks, self-evaluation and training quality. This paper presents the results of the possible effects of advanced driver training on driving behavior, considering in particular the number and type of violations. For each driver, data on age, gender and driving violations history were extracted from the platform and the national violations database. Three cases were addressed through a before-after analysis with control group. Case 1 considers all drivers who attended an ADT course. Case 2 aimed at understanding the effects of the courses on a specific target group: the traffic violators. Case 3 is similar to Case 2, however the control group was selected in a way that drivers characteristics and the violation rate was similar to the violation rate of the treatment group in the before period. The significance of the differences highlighted was assessed through appropriate statistical tests (i.e. paired t-test and the Wilcoxon signed-rank test). The study showed in general a higher propensity to commit traffic violations after attending an ADT course. These results are in contrast to what expected and show the necessity to diversify the training classes according to the different needs of participants.