Statistical Techniques for Making Cross-Cultural Comparisons on Organizational Instruments

Abstract

Researchers tend to evaluate psychological instruments in terms of reliability (internal consistency) and construct validity (exploratory factor analysis and confirmatory factor analysis). In many instances, these instruments are used for cross-cultural comparisons such as gender and race—however, many of these studies do not provide evidence of measurement invariance or measurement equivalence. Measurement equivalence is a statistical property of an instrument that indicates that participants interpret and respond to the items similarly or that the same latent construct is being measured across observed groups of people. Partial measurement equivalence is a necessary condition for comparing latent mean differences across cultures. This area of construct validity is often neglected in the literature; therefore, this chapter aims to introduce the concept of measurement invariance. Additionally, it highlights the necessity of testing for measurement invariance when making cross-cultural comparisons on organizational leadership instruments.