Exploring relationship between perceived risk and risky driving behavior of motorcyclists in Bangkok, Thailand

• Exploratory factor analysis extracts risk perception and behavior latent constructs. • Anti-conventional ride and risk underestimation strongly influence risky behavior. • Male risky behavior was driven by more risk factors than that of female counterparts. • Aversion to helmet use significantly predicts risky drive of commercial motorcyclists. • Peer has strong influence on risky behavior for privately-owned motorcyclists. The number of motorcycles in Thailand have been increasing rapidly in recent years. The major advantages are speed and convenience in reaching destinations, especially through heavy traffic areas of Bangkok and its vicinity. However, motorcycles involved in numerous severe traffic crashes and fatalities. According to the World Health Organization, Thailand has a traffic fatality rate of 32.7 per 100,000 population, with motorcycles involved in approximately 84 % of these fatalities. These potential harm and losses are directly related to motorcyclists and their passengers. Primary aims of this research were to explore underlying factors of risk perception and risky driving behavior of motorcyclists in Bangkok, Thailand, and assess the contribution of perception factors to risky driving behavior. A total of 450 motorcyclists were interviewed using questionnaire survey to analyze self-reported risky driving behavior and perceptions towards potential causes of traffic crashes. Exploratory factor analysis extracted four latent constructs underlying perception, explaining 60.2 % of the total variance, and five behavioral constructs which explained 58.7 % of the total variance. Path coefficient model indicated that anti-conventional riding ( β = 0.621 ) and risk underestimation ( β = 0.423 ) were the two strongest predictors positively associated with unsafe driving behavior. For commercial motorcyclists, perception on anti-conventional riding, aversion to helmet use, underestimated safety risks, and peer influence were less likely to affect risky behavior when compared to private motorcyclists. Gender differences illustrated that these four perception factors significantly predicted risky driving behavior in male drivers whereas only anti-conventional riding and peer effects were more likely to influence risky behavior for their counterpart. This study contributes new insights into underlying perceptions associated with risky driving behavior among motorcyclists with a particular focus on key predictors of behavior across distinct motorcyclist subgroups. The study also offers practical implications for developing targeted interventions and policies aimed at improving road safety and reducing risky driving behavior.