Valid Measurement in Marketing

It is generally assumed that when people fill out a questionnaire, their answers are based on the substantive meaning of the items to which they respond. However, it has been known for a long time that people’s responses are also influenced by content-irrelevant factors, such as the rating scale associated with an item. The key sources of non-content-based responding are response styles and social desirability bias. The term response style refers to a person’s tendency to systematically respond to questionnaire items regardless of item content (i.e., what the items were meant to measure). Common examples of response styles are acquiescence, extreme responding, and the use of the middle response categories on ratings scales. Response styles are a source of concern in both domestic and international marketing research (as the bias may differ systematically between countries) because they threaten the validity of empirical findings by contaminating respondents’ answers to substantive questions. Related to response styles is socially desirable responding, which can be defined as the tendency for people to present themselves favorably according to current cultural norms. Norms are important determinants of socially desirable behavior as they determine what constitutes a favorable impression in a given situation (Nederhof, 1985). This may cause the respondent to overreport or underreport, depending on the situation. While response styles are largely independent of item content, this is not the case for socially desirable responding.

Current research focuses on the occurrence of response style bias and socially desirable responding in domestic and international marketing, their sources and effects, and how to purge response styles from different types of questionnaire data (agree/disagree scales, importance scales, dichotomous scales) to accurate and comparable measurement across countries.

Our published and under construction work on Valid Measurement in Marketing.