Probleme der kulturvergleichenden Umfrageforschung

Abstract

This paper calls attention to important methodological problems in cross-cultural survey research and questionnaire construction. Different sorts of bias limit the comparability of data: respondents give socially desirable rather than true answers, their cultural responsesets show different tendencies toward acquiescence and extreme-response checking, they have different denotations and connotations of key expressions in mind and find some items ambiguous or perceive them as doublebarrel questions. The most severe limitations to comparative validity lie in the difficulty of finding equivalent translations for questionnaires. Scalar equivalence is related both to translation and cultural responsesets. Adequate sampling is an additional challenge to comparative cross-cultural studies. Cross-cultural survey researchers recommend a wide range of validation procedures, but a unified and secured methodology has yet to be developed. Until that time, statements about cultural differences should not be accepted uncritically by researchers or the public.