Measuring risky risky driving behaviours among young drivers: Development of a scale for the Oman setting

Background: A large body of road safety research has focused on developing self-report measurement scales that identify the type and frequency of risk driving behaviours that lead to crash involvement. With the dearth of Arabic measurement tools, the aim of the study was to develop a modified, valid and reliable measurement tool that can be utilized among young drivers within the Oman context. Methods: A total of 1319 (27.1% female) young drivers aged 17-25 years completed a questionnaire that was distributed through a snowballing sampling technique across Oman. The survey included a range of demographic information and driving behaviours, and utilized aspects of the Driver Behaviour Questionnaire (DBQ) and the Behaviour of Novice Young Drivers Scale (BNYDS). An exploratory factor analysis (EFA) was undertaken to examine the factor structure of the modified 40-items tool. Results: A maximum likelihood and varimax rotation factor analysis revealed seven behavioural dimension comprising 39 items, which explained 49.28% of the variance in the behavioural scale of young drivers. These factors were transient violations (20.12% of the variance), mood driving (7.03% of the variance), speeding (6.59% of the variance), fatigue driving (4.36% of the variance), distracted driving (4.12% of the variance), seatbelt usage (3.55% of the variance) and close following (3.51% of the variance). The composite behavioural scale (39-items) showed an excellent internal consistency (alpha = 0.939) with transient violations exhibiting the highest internal consistency (alpha = 0.927) and close following showed the lowest internal consistency (alpha = 0.700). Crash predictability of the seven behavioural dimensions was investigated (as 39.6% of the sample reported crash involvement). Conducting logistic models between each behavioural dimensions and crash involvement adjusted for drivers' characteristics found that mood driving, fatigue driving and distracted driving were strong predictors of crash involvement among young drivers. However, consistent with previous research, the full model was not an efficient predictor of crash involvement among the sample of young Omani drivers, as distracted driving was the only significant predictor in the model. Conclusions: The modified risky driving behaviours scale exhibited appropriate psychometric properties and key aberrant driving behaviours were associated with crash involvement. This paper further outlines the key study findings and provides suggestions for future research that aims to develop effective self-report methods to identify "at risk" drivers. (C) 2018 Elsevier Ltd. All rights reserved.