Entity context and format induced instability in multiattribute ratings of country image

Abstract

Multiattribute ratings of country‐image are commonly obtained to study the influence of country‐of‐origin on product preference and purchase decisions. Usually, such ratings are obtained for products made in different countries in order to make comparisons across countries. However, recent research evidence indicates that, when individual respondents rate multiple entities (e.g., multiple countries), the ratings of a particular country can be affected by the other “contextual” countries included in the questionnaire. The purpose of this study is to examine, via a controlled experiment, the issue of measurement context effects in the measurement of country images. The hypotheses tested concern the stability of attribute‐based and entity‐based country image scaling and potential methods of reducing the vulnerability of multi‐entity scaling to entity context ‐‐ the use of an anchor entity and an insulator question set to increase country image measurement stability.