Does safety climate predict safety performance in Italy and the USA? Cross-cultural validation of a theoretical model of safety climate

• Published in: Accident analysis and prevention Vol. 77; pp. 35 - 44
• Main Authors: Barbaranelli, Claudio, Petitta, Laura, Probst, Tahira M.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.04.2015
• Subjects: Adult; Attitude; Climate; Cross-cultural; Cross-Cultural Comparison; Equivalence; Female; Humans; Invariance; Italy; Male; Mathematical analysis; Mathematical models; Measurement equivalence; Models, Theoretical; Organizational Culture; Organizations; Perception; Reproducibility of Results; Safety; Safety climate; Safety performance; United States; United States; Italy; Cross-cultural; Safety climate; Measurement equivalence; Safety performance
• DOI: 10.1016/j.aap.2015.01.012
• ISSN: 0001-4575

•We test the measurement equivalence of a widely used safety climate scale.•Invariance of the Griffin and Neal (2000) safety model was found in Italy and US.•There are practical and theoretical implications for cross-cultural safety research. Previous studies have acknowledged the relevance of assessing the measurement equivalence of safety related measures across different groups, and demonstrating whether the existence of disparities in safety perceptions might impair direct group comparisons. The Griffin and Neal (2000) model of safety climate, and the accompanying measure (Neal et al. [NGH], 2000), are both widely cited and utilized. Yet neither the model in its entirety nor the measure have been previously validated across different national contexts. The current study is the first to examine the NGH measurement equivalence by testing whether their model of safety climate predicting safety performance is tenable in both English speaking and non-English speaking countries. The study involved 616 employees from 21 organizations in the US, and 738 employees from 20 organizations in Italy. A multi-group confirmatory factor analytic approach was used to assess the equivalence of the measures across the two countries. Similarly, the structural model of relations among the NGH variables was examined in order to demonstrate its cross-country invariance. Results substantially support strict invariance across groups for the NGH safety scales. Moreover, the invariance across countries is also demonstrated for the effects of safety climate on safety knowledge and motivation, which in turn positively relate to both compliance and participation. Our findings have relevant theoretical implications by establishing measurement and relational equivalence of the NGH model. Practical implications are discussed for managers and practitioners dealing with multi-national organizational contexts. Future research should continue to investigate potential differences in safety related perceptions across additional non-English speaking countries.