Cross-validation of an employee safety climate model in Malaysia.

PROBLEM Whilst substantial research has investigated the nature of safety climate, and its importance as a leading indicator of organisational safety, much of this research has been conducted with Western industrial samples. The current study focuses on the cross-validation of a safety climate model in the non-Western industrial context of Malaysian manufacturing. METHOD The first-order factorial validity of Cheyne et al.'s (1998) [Cheyne, A., Cox, S., Oliver, A., Tomas, J.M., 1998. Modelling safety climate in the prediction of levels of safety activity. Work and Stress, 12(3), 255-271] model was tested, using confirmatory factor analysis, in a Malaysian sample. RESULTS Results showed that the model fit indices were below accepted levels, indicating that the original Cheyne et al. (1998) safety climate model was not supported. An alternative three-factor model was developed using exploratory factor analysis. DISCUSSION Although these findings are not consistent with previously reported cross-validation studies, we argue that previous studies have focused on validation across Western samples, and that the current study demonstrates the need to take account of cultural factors in the development of safety climate models intended for use in non-Western contexts. IMPACT ON INDUSTRY The results have important implications for the transferability of existing safety climate models across cultures (for example, in global organisations) and highlight the need for future research to examine cross-cultural issues in relation to safety climate.