| dc.description.abstract | Given the relevance and importance of safety climate to workplace safety in organizations worldwide, researchers and practitioners recognize the utility of measuring and tracking safety climate, especially in high-reliability organizations. However, sample characteristics or faultlines including national culture, language, hierarchical position, employment arrangement, and work environment create meaningful differences between groups of respondents that may make it inappropriate to compare safety climate scores across groups. Differences were expected to emerge for numerous reasons including the construct relevance of item content, response sets or tendencies to use response scales in a particular manner, the relative strength of item endorsement, and/or the frame of reference used. The purpose of this study was to examine the measurement equivalence of a safety climate measure across five faultlines within an archival dataset containing survey responses from 8,790 multinational chemical processing and manufacturing employees. In order to take the multilevel nature of the data into account, the measurement equivalence of the safety climate measure was examined by using multilevel multi-group CFAs for the Level-3 (national culture) faultline and multilevel factor mixture model for the Level-1 faultlines. Scalar (intercept) equivalence was not established across all five faultlines, whereas metric equivalence held for hierarchical position and employment arrangement. These results suggest that the same safety climate instrument may not be used across different contexts. Results have important practical implications for benchmarking safety climate ratings across studied faultlines. | en |
