The effect of response scale type on cross‐cultural construct measures

Abstract

PurposeMethodological advances in cross‐cultural scale development have addressed many concerns regarding the development of valid scales. However, several issues remain to be examined – including the potential problems of using language to measure communication phenomena using self‐reported studies and addressing the effect of response scale type on the validity of resultant measures. The purpose of this paper is to expand the cross‐cultural measurement paradigm by comprehensively examining these issues and suggesting a new response scale type that may potentially produce more valid cross‐cultural measures of communication‐based phenomena.Design/methodology/approachMeasures of Hall's concept of context were developed using three types of response scales – Likert, semantic differential, and conceptual metaphoric. The last response scale type is developed within this research. Samples were gathered in 23 countries using existing scale development procedures. The response scales were compared for psychometric properties and validity based on reliability, metric invariance, response styles, and face validity.FindingsOverall all three response scale types adequately measured the construct of context. The newly developed conceptual metaphoric scale performed marginally better on most comparative metrics.Practical implicationsInternational marketers measure a host of variables related to culture for many purposes. The new response scale type may provide slightly better measures to more accurately reflect communication based constructs – many of which are central to marketing.Originality/valueThe findings indicate that the new conceptual metaphoric response scale type may overcome some existing biases inherent in standard response scale types. In addition, this research provides the first viable and parsimonious measure of Hall's concept of context.