3.3 Description of the two samples with regard to transport

Results are in tables 2 and 3.

As one would expect, the course takers and the controls did not have the same number of points left on their licence. For both genders, the course-takers drove much more, both in connection with their occupation and for personal reasons. They spent much more time on the road. They reported committing more violations than the controls. A more detailed analysis (not shown) of the 21 reported violations shows that course-takers committed all the road traffic violations, as opposed one particular violation, more frequently than control drivers. They more often paid an insurance surcharge, and more often had powerful cars. They drove by themselves more frequently than the controls. The car was more frequently considered to be an extension of their office (all other modalities have been grouped, being not different between the two groups), and they more frequently described their driving as sporty. They had a stronger fear of losing their driving licence than the controls. Two-thirds of the course-takers considered the risk of being arrested after committing a violation to be at least 1 in 10, compared to 42% for the controls.

Some differences between the course-takers and controls were only significant among the men: the course-takers less frequently had third-party insurance and owned their vehicle less often. They also had more damage-only and injury accidents in the 5 previous years. Thus, the raw odds-ratio among crash-involved men of being a course-taker was 1.7 for injury accidents and 1.3 [1.1; 1.7] for at least one damage-only accidents in the 5 previous years compared with individuals who had not been involved in an accident. The course-takers counted more crash-involved individuals among their close relations.

It is noteworthy that male course-takers and controls differed with regard to 19 of the 22 transport-related variables, the exceptions being the insurance surcharge, self-esteem on the road and their opinion of the behaviour of other individuals on the road. For women, the comparison between course-takers and controls revealed a difference with regard to 9 of the 22 variables. The differences with regard to crashes were not significant, but the crude odds-ratio of course-takers vs controls was similar to that for men. It is then only a statistical power matter.