Citation

Hirsch P, Bellavance F. CIRRELT. Montréal, Canada: Interuniversity Research Centre on Enterprise Networks, Logistics and Transportation (CIRRELT), 2016.

Copyright

(Copyright 2016, CIRRELT)

The full document is available online.

Abstract

Background and Objectives: Although driver education (DE) is a controversial intervention, there is continued support for DE among policy makers. Flight simulators are effective tools for aviation training. A basic assumption of this transfer of training (ToT) study is that driving simulation can improve DE training outcomes. The three objectives of this study are to measure: perceptions of driving school students concerning the efficiency and acceptability of driving simulator-based training (DBST); the effect of DBST on driver performance on the SAAQ probationary permit road exam, and; the influence of DBST on driver behavior as reflected by infractions and crashes during unsupervised driving with a probationary permit. Method: This is a long-term, prospective cohort, naturalistic study. Learner perceptions of DBST were collected through two questionnaires, one at the start and the other at the end of the nearly one-year long driving course. For the study, one driving simulator hour could replace one hour of on-road training for up to six of the 15-hour curriculum in effect in Québec since January 2010. Driver performance was measured by SAAQ probationary permit road test results and the potential behavioral effects of DBST were measured by driving infractions and police-reported crashes during the first two years of driving with a probationary permit. Results: Four driving schools provided a convenience sample of 2,187 learner drivers, average age 17,6 years, 53,8% female and 46.2% male. In total, 95% of the study sample had fewer than five hours of DBST. The age- and sex-matched comparison group was drawn from all new Quebec drivers. Questionnaire responses indicate a strong positive appreciation for DBST. At the end of on-road training, learners perceived that 14 of the 15 driving skills listed, especially the most safety-critical ones, were taught more or equally efficiently on the driving simulator than on the road. In relation to the comparison group, pass rates on the road exam were statistically significantly higher, and, during the first two years of driving with a probationary permit, infraction rates were statistically significantly lower and, after controlling for vehicle ownership, crash rates were equivalent. Conclusion: The data indicate that DBST is well appreciated by learners and that there was positive transfer of driving skills to road exam performance. There is also encouraging evidence that during the first two years of unsupervised driving, DBST had a positive impact on driving infractions and a null effect on crash risk.

Keywords. Driver education, driving simulator, prospective cohort, naturalistic study, road safety, infractions, collisions.