Conceptual Implications of Item Bias

Abstract

Any comparison presupposes a tertium comparationis, i.e. an invariant standard or measurement scale in terms of which the comparison can be made. In the natural sciences scale identity can be formally defined in terms of measurement procedures. In psychology where interactions occur between the measurement instruments and the objects of measurement, this identity cannot be taken for granted. This is especially so in research where the results of culturally divergent groups are compared. Various approaches to the analysis of scale equivalence of comparability in cross-cultural studies have been developed (Poortinga, 1983). Of these the analysis of item bias is the most important.