Measurement Equivalence and Extreme Response Bias in the Comparison of Attitudes Across Europe

Abstract

It is generally accepted that both measurement inequivalence and extreme response bias (ERB) can seriously distort measurement of attitudes and subsequent causal models. However, these two issues have rarely been investigated together. In this article we demonstrate the flexibility of a multigroup latent-class factor approach in both analyzing measurement equivalence and detecting ERB. Using data from the European Values Survey from 1999/2000, we identified an ERB in answering Likert-type questions on attitudes toward morals of compatriots. Furthermore, we found measurement inequivalence in the form of direct effects of countries on the attitude items. The model that included both these issues resulted in quite distinct findings regarding country difference in the latent attitude compared to the models that only included one of these effects – either measurement inequivalence or extreme response. It is suggested that the all-inclusive model provides the more valid estimates of country differences in the latent attitude.