Development of a Measure of Corporate Safety Culture for the Transportation Industry

This project was designed to develop the Corporate Safety Culture Scale, a unique measure of corporate safety culture that is supported by a strong theoretical foundation. Items were developed conceptually, discussed in cognitive interviews, and submitted to expert review. Survey items were administered to employees of a large public transportation organization. Principle components analysis reduced the data and identified an empirical model. The empirical model held two domains identified as: 1) Values, and 2) Behavior. Confirmatory factor analysis compared the fit of the empirical model with that of a modified empirical model, and the hypothesized model, which held the following three domains: 1) Behavior, 2) Values, and 3) Meaning Systems. Goodness of fit indices (Chi-Squared, RMSEA, CFI, ECVI) identified the modified empirical structure as the model with best fit. The similarities between the hypothesized and modified empirical models suggested that the CSCS carries strong theoretical support. The measure\u27s reliability was tested using Cronbach\u27s Alpha. The full CSCS and each identified domain were shown to have strong Alpha values. One-way between groups analysis of variance, was used to evaluate the relationship between scores on the full measure and a categorical safety behavior item. Findings suggested that high scorers on the CSCS were safer workers than those with lower scores on the measure. A post hoc analysis using Tukey\u27s HSD showed that scores on the full CSCS, and the Behavior domain were significantly related to safety behavior. The analysis of variance and the post hoc results indicated that a relationship exists between the CSCS and safety behavior. This finding added to the validity of the measure. The CSCS was shown to be a theoretically supported, valid and reliable measure that can be used to evaluate corporate safety culture