Bayesian Approach to Assess Factors Affecting Driving Behaviour: An Attitude-Behavioural DBQ for Abu Dhabi

The aim of this study was to identify the socio-demographic features and attitudes that influence driver behaviour. A specially developed tri-construct Driver Behaviour Questionnaire (DBQ) was used to investigate drivers’ behavioural choices and included violations, errors, and lapses. The analysis framework, based on the Theory of Planned Behaviour, and a Bayesian Belief Network (BBN) modelling approach, was applied to responses from the Emirate of Abu Dhabi, a region with a diverse expatriate population (n=1,792). Results indicate that driver attitude is a stronger predictor of aberrant driving behaviour than socio-demographic features. Additionally, the study explored the influence of demographics (e.g., age, nationality, education level), on driving behaviour. It was observed that young drivers had a higher likelihood of committing violations, while older drivers were more associated with safer driving habits. These findings suggest that traffic safety interventions should focus on attitude variables within the DBQ to identify drivers more likely to engage in risky behaviours. Findings highlight the importance of understanding interactions among drivers from different cultures with diverse attitudes and socio-demographic backgrounds to predict complex behavioural patterns. The insights are crucial for stakeholders to design more effective traffic safety strategies.