Facts and Artefacts in the Comparison of Attitudes Among Ethnic Minorities. A Multigroup Latent Class Structure Model with Adjustment for Response Style Behavior.

There is a growing interest in cross-cultural comparative research of attitudes. In many studies it is assumed that attitudes are comparable across cultures. Often this assumption is not tested. In this paper we argue that before comparing attitudes among different cultural groups two methodological issues need to be investigated. First, any measurement of latent attitudes should exclude the possibility that response style biases measurement. Secondly, cross-cultural comparisons are only valid if construct equivalence is tested. In this article we demonstrate the flexibility of a multigroup latent class factor approach in detecting extreme response bias and in investigating construct equivalence. Using data from the MHSM surveys among Turkish and Moroccan minorities in Belgium we detected a tendency toward extreme response style behavior in answering Likert type attitudes toward gender roles and perceived discrimination. Furthermore, a model with several covariates proved to be partially homogeneous. The results indicated that the partially homogeneous model with direct effects of ethnic group and correcting for response style behavior allowed for `fine-tuning' the estimated effects of the covariates on the latent content factors. [ABSTRACT FROM AUTHOR]