An investigation on the risk factors associated with driving errors under the influence of alcohol using structural equation modeling
Objective: Driving errors contribute to traffic crashes and thus the causal factors associated with driving errors are of great interest to the road safety researchers. The present study attempts to identify and quantify the effects of various risk factors that are likely to influence driving error with the application of structural equation modeling (SEM). These risk factors include blood alcohol concentration (BAC) levels (0%, 0.03%, 0.05%, and 0.08%), different driving environments (rural, urban), and driver attributes (such as age, gender, education).
Method: Eighty-two licensed drivers participated in a driving simulator experiment. They completed driving under the influence of 4 BAC levels in the driving environments within the framework of a full-factorial within-subjects design. Driving error was modeled as an unobserved latent variable based on several driving simulator indicators. An SEM approach was utilized to examine the influence of BAC level, driving environment, and driver attributes on the latent variable pertaining to driving error.
Results: The findings suggest the suitability of an SEM approach in the investigation of driving error. The results revealed that all 3 positive BACs (0.03%, 0.05%, and 0.08%) had a significant positive influence on driving error compared to 0% BAC, and the tendency toward driving error increased with increasing BAC (factor loadings for 0.03%, 0.05%, and 0.08% BAC were 0.22, 0.31, and 0.37, respectively). Moreover, driving in an urban environment led to more driving errors compared to a rural environment, including sober drivers. Among the driver attributes, gender and awareness about drink and drive laws were the only factors influencing driving error.
Conclusion: This study highlights a novel approach to investigate driving error by modeling it as a latent variable instead of modeling individual performance measures. The successful execution of SEM in alcohol impairment research may serve as a significant step in the human factors field moving from piecemeal analysis to a combined analysis where interrelationships among numerous risk factors and driving error can be established. The study outcomes may serve as a reference while developing strategies to enhance road traffic safety where special emphasis can be given to the critical risk factors influencing driving error identified in the study.
