Subjective driving skill is commonly assessed with self-reports and many studies have found that drivers have overly positive beliefs in their own skill. The purpose of this review was to examine how subjective driving skill has been measured in different studies. Secondly, the aim was to discuss the methods used from a measurement perspective, i.e., to discuss the reliability and the validity of the measures and methods used. The findings from the review indicated that studies about subjective driving skills can be divided into three different domains with respect to the methodology used. In two of the domains, subjective driving skill is measured by comparing the individualÆs own skill to internal criteria: the skill of the average driver and specific aspects of driving skill. In the third domain, the subjective skill is compared with an external criterion, i.e., the actual skill in order to determine if drivers have an accurate perception of their own skill. The conclusion of the review was that there are several methodological problems with the reference to the average driver that can result in biased assessments. Moreover, this methodology cannot be used to conclude whether driversÆ are overconfident or not. In order to obtain indicators of reliability and validity, the measurement of subjective driving skill should incorporate judgments of specific aspects of driving skills. By the use of this methodology, subjective driving skill can be validated through comparison with actual driving skill. (A) Reprinted with permission from Elsevier.
Abstract