Design, Evaluation, and Analysis of Questionnaires for Survey Research, 2nd Edition

16

COPING WITH MEASUREMENT ERRORS IN CROSS-CULTURAL RESEARCH

In this chapter, we will introduce the problem of measurement errors in cross-cultural research. In particular, we will focus on comparative research across countries. In the last chapters, we established that measurement errors have strong effects on results of research. Therefore, when the effects of measurement error differ in the individual countries of a study, cross-country comparisons become quite challenging.

Two types of comparisons are most frequently made: comparisons of means and comparisons of relationships of different variables across countries. Often, comparisons based on single requests or on composite scores of the latent variables are made. In this chapter, we will add to this the comparisons based on latent variables.

The problem of such comparisons is that one can compare the results across different countries only if, in fact, the data are comparable, that is, if the measures used in the different countries have the same meaning. This topic is studied under the heading of functional equivalence or invariance of measures in different countries.

This chapter will concentrate on the procedures to determine the equivalence of measurement instruments. But, before we can introduce this topic, we have to introduce the notation for this topic. So far, we used standardized variables and concentrated on the effects of these variables on each other. However, in cross-cultural research, the means of the variables ...