Assessing the factor structure of the Behaviour of Young Novice Drivers Scale (BYNDS)

Abstract

The Behaviour of Young Novice Drivers Scale (BYNDS) is a self-report instrument to assess risky driving behaviours among youth population. Although previous studies have offered evidences of reliability and validity for the BYNDS scores, results are not conclusive as regards its factor structure. The aim of this research was to assess the factor structure of the BYNDS via Confirmatory Factor Analyses (CFA). The BYNDS was administered to a sample of 832 drivers aged from 18 to 25 years (Argentina = 270, Colombia = 350, and Mexico = 212). Six models were examined: three derived from previous literature, one emerged from an exploratory factor analysis, a unidimensional model, and a bifactor model. Results showed that none of the models analysed offered a completely acceptable fit to the data. However, the original model hypothesizing five orthogonal factors showed the best fit indices, with all of the items loading onto the associated factor. Results of the bifactor analysis suggested that three of the subscales (transient violations, fixed violations, and misjudgements) share a strong common basis and their scores would not offer much information above and beyond the total score. On the contrary, the items in the subscale of risk-exposure are clearly influenced by a specific domain factor, contributing to the multidimensionality of the BYNDS. In line with previous studies, males scored higher than females in driving violations, fixed violations, and risk-exposure. This study contributes to the psychometric refinement of a novel measure of risky driving in youth population. This self-report could be a valuable tool in the evaluation of road safety performance.