Validation of Malaysian Driving Style Self-Assessment with Observational Road Study

Abstract

Along with the increasing application of a self-assessment questionnaire to recognize the driving style, questions have been raised about the possibility of reporting bias because of the driver’s misjudgment. A hazy reference point was the cause since drivers lack objective input directly tied to their driving behaviour when answering a self-assessment questionnaire. This study aims to validate the results of a driving style self-assessment questionnaire with road studies for Malaysian drivers. A total of 20 drivers’ driving styles were recorded and evaluated by the two designated observers. The driver completed the driving self-evaluation questionnaire (18 items) after finishing the two designated routes (familiarization and experiment route). In comparison, the observer evaluated the driver by using two forms: 1) a driver evaluation questionnaire (18 items) after the experiment and 2) an on-road driver observation form (25 items) during the experiment. Inference statistics analyzed the data using regression, Pearson correlation, Wilcoxon z-value, Analysis of Variance (ANOVA), and reliability test. For the reliability test, the questionnaire illustrates the consistency and reliability of the set questionnaire used in this study, ranging from acceptable to good. However, the result shows no significant correlations between driver and observer with on-road driver observation forms for error/violation during driving activities. Besides, there were significant differences between the driver and observer regarding driving style evaluation. The over-positive appraisal was higher among drivers with higher error/violation scores and with the ones that the observer evaluated. The theoretical and practical significance of the self-driving questionnaire is addressed.