The measurement equivalence of a safety climate measure across five faultlines.
Additional Document Info
This study examines the appropriateness of comparing safety climate survey responses across multiple faultlines-hypothetical dividing lines that split a group into subgroups based on one or more attributes. Using survey data from 8790 employees of a multinational chemical processing and manufacturing company from 76 work sites nested within 19 different countries, we examined the multilevel measurement equivalence of a safety climate measure across cultural dimensions, survey languages, organizational hierarchy, employment arrangements, and work environments. As simulation studies support the faultline at the individual-level requires measurement equivalence tests that are different from the faultline at the country-level, we used multi-group multilevel confirmatory factor analyses for the Level-3 faultline, and multilevel factor mixture models for known classes for the Level-1 faultlines. The results demonstrated that faultlines can prevent safety climate measurement equivalence, which prohibits the aggregation of individual-level scores to higher levels and making comparisons across faultlines. This first study on multilevel safety climate measurement equivalence serves as both a warning to safety climate researchers and practitioners regarding the importance of faultlines and reminds us to consider the level of the faultlines when testing measurement equivalence with multilevel data.