The identification of individual differences in safety performance : development and validation of a safety values instrument

Abstract

In the past century, there has been little to no attention focused on an individual’s values towards safety. In this thesis, I have developed and validated a safety values scale with the purpose of investigating the extent to which employees’ safety values are related to safety performance. In study 1, nine subject matter experts identified six items that represent the safety values domain. In study 2 (N= 182), the factor structure of the SVS was examined using an exploratory factor analysis (EFA) and was affirmed through a confirmatory factor analysis (CFA) in study 3 (N =410). The EFA and CFA supported the unidimensional structure of the SVS, χ2 (9, N = 410) = 20.88, p = .01, CFI = .99, RMSEA = .06 (90% CI =.025, .089), SRMR = .02. The internal consistency of the SVS was α=.85. Results from a hierarchical multiple regression supported that the SVS contributed significant incremental validity over the Big-Five personality traits and safety climate for safety performance and injury metrics. Practically, the SVS has the potential to be utilized in a selection or training context.