Different Factors Predict Different Risky Driving Behaviours: A Challenge to the Assumed Generalizability of Prediction and Countermeasure

It has been shown repeatedly that adolescents are over-represented in crashes among all classes of road user, compared with other age groups. Causation of this overrepresentation is complex, with many different factors implicated from both basic and applied research perspectives. Ultimately, applications of theories of human behaviour must guide road safety studies of road user behaviour, and are critical to practical outcomes in addressing the problem of road trauma. While there is a gamut of applied research on road safety, it is typically focused on single problems (e.g. drink-driving or speeding). The present study begins the process of comparing factors involved in various problem behaviours in relation to attitudes across various risky driving behaviours. Study 1 (N=109) examined a range of possible predictors of risky driving, and investigated the nature of attitudes to risk taking for young drivers. The aim of Study 1 was to investigate whether predictive factors for various risky driving behaviours differed from behaviour to behaviour (e.g. drink-driving vs. speeding vs. non-use of seat belts). Results illustrated that different risky driving behaviours were predicted by different factors (e.g. speeding was predicted by authority rebellion, while drink driving was predicted by sensation seeking and optimism bias). These results are inconsistent with the cherished assumption in the field that the predictive factors of one risky driving behaviour can be generalized to another. Study 2 (N=115) examined the generalizability of the results from a student sample to the general driving population. Again, the common practical assumption of generalization from university students to the general population was not supported in several cases. Overall, the results clearly illustrate that different factors predict different risky driving behaviours, and future research must now focus on a multi-factor framework for each specific risky driving behaviour, rather than assuming generalizability from one risky behaviour to another. It is consistently shown that adolescent drivers are over-represented in road crash statistics, both in Australia and worldwide (Jonah, 1986; Job, 1996; 1999). Causation of this over-representation is complex (Cameron, 1985). Exposure is seen as a major contributor, with young people driving more often than older drivers, and at more dangerous times (Lee, Prabhakar & Job, 1993). The lack of driving experience in adolescents is another contributing factor, with an important consequence believed to be a lack of driving skill. However, recent literature suggests that motivation, rather than skill, is a major determinant of road safety (Naatanen & Summala, 1976; Job, 1999; Job & Hatfield, 1996). Consequently, risk taking has been proposed as a substantial contributor to the over-involvement of young drivers in road crashes (Cameron, 1985; Jonah, 1986; Job, 1996). It is believed that young people with limited experience engage in behaviours with the anticipation of reward or gain, and without fully understanding the consequences of their actions (Job, 1999). As a result, many young people continue to take risks while driving despite the possibility of injury or death. Compared with older drivers, young drivers are more likely to drive fast, tailgate, allow too little time to merge, and fail to give way to pedestrians (Jonah, 1986; Job, 1990; 1999). Given the extensive research on risk taking and its apparent role in road trauma causation, thorough investigation into young peoples’ attitudes toward driving is essential. However, the lack of valid and reliable instruments to measure risk taking attitudes makes it difficult to measure any effects of attitude change, and many selfreports of driving behaviour have not been psychometrically examined. Furthermore, most past research has focused on single driving behaviours (e.g. Harre, Brandt & Dawe, 2000). Consequently, there is a pervasive assumption that factors such as optimism bias and sensation seeking contribute to risky driving behaviours in the same way, regardless of the behaviour. Thus, it is thought that we must use the same approach in advertising to speeding, drink-driving, and not wearing seat belts. For instance, successes in drink-driving are being used to generate principles to apply to many other risky driving behaviours. This assumption, however, has never been tested. Consequently, before it can be assumed that attitude change will lead to improving road safety, it is essential to establish the nature of any relationship between attitudes and onroad behaviour, across a variety of risky driving behaviours. STUDY 1: INVESTIGATING THE RISK-TAKING ATTITUDES OF YOUNG DRIVERS Study 1 examined demographic, personality, and attitudinal factors in relation to predicting a range of risky driving behaviours, and investigated whether different factors predict different risky driving behaviours. One important theory for risky driving is Jessor & Jessor’s (1977) Problem-Behavior Theory, which has since been extended specifically to risky driving. Problem-Behavior Theory proposes an interaction of factors that arise from three major systems – the personality system, the perceived environment system, and the behavior system – and predicts a regular general tendency for deviance across a large range of behaviours, with similar predictors present for each behaviour. The vast research on driver risk taking illustrates that many different factors have been implicated as determinants of risky driving, and it is difficult to acknowledge a single factor as significant. Based upon this and Problem-Behaviour Theory, a range of possible predictors of risky driving was examined in Study 1. The greater value of the multiple testing of many factors is that it negates the problem of possibly confounded predictors that have not been directly compared. The factors tested in Study 1 are listed below: (A) Age – Research has found that there exist significant differences in risky driving behaviour between specific driver age groups (Jonah, 1990; Begg & Langley, 2001). (B) Gender – Gender effects have been prominent for optimism bias (Job, 1990b; 1996). (C) Sensation seeking – Zuckerman (1994) suggests that sensation seeking is presently the most common purpose of risky driving for young men. (D) Competitiveness – The inclusion of this factor was based on logical possibility. (E) Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism – Arthur & Doverspike (2001) have found that crashes significantly correlated with the five-factor model of personality, and, in particular, with conscientiousness. (F) Aggression – Recent research suggests that aggression is a significantly high-risk personality sub-group of young drivers (Ulleberg, 2001). (G) Psychopathy – several studies have illustrated the relevance of antisocial personality to deviant behaviour (e.g. Kuriychuk, 1992). (H) Authority Defiance – The inclusion of this factor was based on logical possibility. (I) Time saving – Research shows that people engaging in risky driving are more likely to view being on time for appointments as desirable (Adams-Guppy & Guppy, 1995). (J) General and Specific Attitudes – It is often believed that people treat different driving behaviours as the same. That is, they hold a general attitude about driving, which groups all driving behaviours together. On the other hand, people may hold specific attitudes for performing each behaviour. Consequently, all attitude questions were designed to cover general attitudes and specific attitudes. (K) Optimism Bias Attitudes – The influence of optimism bias on road trauma has been reviewed previously in this paper (c.f. General Introduction). (L) Avoidance Attitudes – The inclusion of this factor was based on logical possibility. It was also necessary to investigate a range of driver behaviours, with the following ten selected: speeding, drink-driving, driving while fatigued, red light running, aggressive driving, reckless driving, competitive driving, not wearing seat belts, not using indicators, and tailgating. (Ethics approval was granted, for both studies, by the University of Sydney Ethics Committee).