Measurement equivalence of a safety climate measure among Hispanic and White Non-Hispanic construction workers

• Published in: Safety science Vol. 54; pp. 58 - 68
• Main Authors: Cigularov, Konstantin P., Lancaster, Paige G., Chen, Peter Y., Gittleman, Janie, Haile, Elizabeth
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier India Pvt Ltd 01.04.2013; Elsevier
• Subjects: Biological and medical sciences; Construction; Measurement equivalence; Medical sciences; Miscellaneous; Occupational medicine; Public health. Hygiene-occupational medicine; Safety climate; Construction; Safety climate; Measurement equivalence; Human; Buildings; Latinamerican; Safety culture; Ethnic group; Construction industry; Work safety; Race; Worker; Caucasoid; Hispanic; Comparative study; Public health
• ISSN: 0925-7535; 1879-1042
• DOI: 10.1016/j.ssci.2012.11.006

► We examined cross-ethnic validity of a multidimensional safety climate scale. ► We tested measurement equivalence across White and Hispanic construction workers. ► Equivalent factor patterns and factor loadings were found across groups. ► Group differences in error variances and intercepts were also identified. ► Measurement equivalence should be established before group mean comparisons. Research continues to expose ethnic disparities in safety and health outcomes, making comparative studies of work-related factors that may explain these disparities increasingly important. Such studies raise issues about the cross-ethnic validity of the measures used to assess the factors of interest, such as safety climate. The current study is the first to examine the measurement equivalence of a multidimensional safety climate scale. A multi-group confirmatory factor analytic approach was used to assess the equivalence of the measure across White English-speaking, Hispanic English-speaking, and Hispanic Spanish-speaking construction workers. Results indicated that the same pattern of factors and equivalent factor loadings adequately represented the safety climate items across groups. However, other differences in item parameters were identified, including non-equivalence of some error variances and intercepts. This study highlights the importance of establishing measurement equivalence before proceeding with mean comparisons among groups. Future research should continue to investigate why differences in safety-related perceptions across ethnicities might exist.