Testing measurement invariance across groups: applications in cross-cultural research.

TLDR

Researchers compare groups on psychological variables assuming instruments measure the same construct across groups; if measurement invariance holds, comparisons are valid, otherwise interpretations are unreliable. The paper reviews the importance of equivalence in psychological research and discusses theoretical and methodological issues of measurement invariance in confirmatory factor analysis. The authors provide a step‑by‑step empirical example of measurement invariance testing and LISREL syntax for fitting such models. The example illustrates measurement invariance testing and supplies LISREL syntax for practitioners.

Abstract

Researchers often compare groups of individuals on psychological variables. When comparing groups an assumption is made that the instrument measures the same psychological construct in all groups. If this assumption holds, the comparisons are valid and differences/similarities between groups can be meaningfully interpreted. If this assumption does not hold, comparisons and interpretations are not fully meaningful. The establishment of measurement invariance is a prerequisite for meaningful comparisons across groups. This paper first reviews the importance of equivalence in psychological research, and then the main theoretical and methodological issues regarding measurement invariance within the framework of confirmatory factor analysis. A step-by-step empirical example of measurement invariance testing is provided along with syntax examples for fitting such models in LISREL.