The impact of response styles on the stability of cross-national comparisons

Abstract

Response style effects are a source of bias in cross-national studies, with some nationalities being more susceptible to particular response styles than others. While response styles, by their very nature, vary with the form of the stimulus involved, previous research has not investigated whether cross-national differences in response styles are stable across different forms of a stimulus (e.g., item wording, scale type, response categories). Using a quasi-experimental design, this study shows that response style differences are not stable across different stimulus formats, and that response style effects impact on substantive cross-national comparisons in an inconsistent way.