CONFIRMATORY FACTOR-ANALYSIS AND ITEM RESPONSE THEORY - 2 APPROACHES FOR EXPLORING MEASUREMENT INVARIANCE

This study investigated the utility of confirmatory factor analysis (C FA) and item response theory (IRT) models for testing the comparabilit y of psychological measurements. Both procedures were used to investig ate whether mood ratings collected in Minnesota and China were compara ble. Several issues were addressed. The first issue was that of establ ishing a common measurement scale across groups, which involves full o r partial measurement invariance of trait indicators. It is shown that using CFA or IRT models, test items that function differentially as t rait indicators across groups need not interfere with comparing examin ees on the same trait dimension. Second, the issue of model fit was ad dressed. It is proposed that person-fit statistics be used to judge th e practical fit of IRT models. Finally, topics for future research are suggested.