Do low levels of confidence suppress the tendency to violate? Frank P. McKenna Department of Psychology, University of Reading, Reading RG6 7BE, UK article info Article history: Received 12 June 2017 Received in revised form 22 January 2018 Accepted 1 March 2018 Keywords: Driver Behaviour Questionnaire Novice drivers Driving skill abstract It is well known that the tendency to violate is associated with accident involvement. What is paradoxical is that across the early development of driving, violations are increasing at the same time that accident liability is decreasing. It is hypothesised that at the early stage of driving low levels of confidence suppress the tendency to violate. As confidence is increased over time the tendency to violate is gradually released. Consistent with this hypothesis an analysis of several thousand new drivers found that as confidence increased so also did the tendency to violate. Implications for testing and training are considered. Ó 2018 Elsevier Ltd. All rights reserved. 1. Introduction In driving it has been known for some time that the tendency to violate is associated with accident involvement (de Winter & Dodou, 2010; Little, 1966) such that higher levels of violations are associated with higher levels of accident involve- ment. What is more surprising is that the tendency to violate increases across the early months of driving at the same time that accident liability is decreasing (Roman, Poulter, Barker, McKenna, & Rowe, 2015; Waller, Elliott, Shope, Raghunathan, & Little, 2001). Another surprising feature of the violation developmental trend is that the factors that might underlie viola- tions, such as sensation seeking, have a quite different developmental trend. Sensation seeking peaks in the early teens and decreases thereafter (Waylen & McKenna, 2008), whereas the tendency to violate increases from the mid to late teens. One possible explanation for the violation developmental trend is that the expression of the tendency to violate is being sup- pressed in the early part of driving by low levels of confidence. A more detailed exposition of the suppression hypothesis is as follows. First, we might reasonably expect (and will test) that confidence levels will be low at the start of the driving career and that, with experience, confidence levels increase. It has been demonstrated for some time that drivers have a relatively high opinion of their driving skill (McKenna, Stanier, & Lewis, 1991; Svenson, 1981) but it might reasonably be anticipated that such a perception would develop over time. Second, people vary in their tendency to violate (Reason, Manstead, Stradling, Baxter, & Campbell, 1990). Findings from Waylen and McKenna (2008) indicated a wide variation in future tendency to driving violations with sex differences and correlations with sensation seeking already established even before novices learn to drive. Third, and critically, it is proposed that low levels of confidence will suppress the tendency to violate. In other words, at the early part of driving we might expect new drivers to express a range of confidence levels with most expressing low levels of confidence. The proposal is that those who report low levels of confidence will be less inclined to express any tendency to violate. As drivers move from low to higher levels of confidence we might expect the expression of the tendency to violate to be gradually released. In order to examine the relationship between confidence and the tendency to violate the UK Department for Transport’s Cohort II study was examined. This large-scale study tracked new drivers across the first three years of driving. Within the https://doi.org/10.1016/j.trf.2018.03.004 1369-8478/Ó 2018 Elsevier Ltd. All rights reserved. E-mail address: f.p.mckenna@reading.ac.uk Transportation Research Part F 55 (2018) 136–140 Contents lists available at ScienceDirect Transportation Research Part F journal homepage: www.elsevier.com/locate/trf