Help for marketing researchers formulating questionnaires

When constructing a scale or selecting multi-item measures for use in a questionnaire, marketing researchers face a dilemma: Should they include reversed items, or should they rely exclusively on items that are all worded in the same direction?

Marketing researchers frequently struggle with the issue of whether or not to use reverse-coded items in questionnaires: for example, “I am very cautious in trying new or different products” (in measuring consumer innovativeness). Reversed items have important advantages: e.g., control of acquiescence, disruption of non-substantive responding, better coverage of the domain of content of a construct. But they can also lead to measurement problems: e.g., low measure reliability, complex factor structures.

Recommendations from a Review of the Literature

Marketing researchers frequently struggle with the issue of whether or not to use reverse-coded items in questionnaires: for example, “I am very cautious in trying new or different products” (in measuring consumer innovativeness). Reversed items have important advantages: e.g., control of acquiescence, disruption of non-substantive responding, better coverage of the domain of content of a construct. But they can also lead to measurement problems: e.g., low measure reliability, complex factor structures.

In an effort to help researchers construct more valid measurement instruments, Professors Bert Weijters and Hans Baumgartner report a comprehensive review of the literature that has examined the psychological mechanisms underlying misresponse to reversed and negated items. Then, based on their literature survey, they provide recommendations for reducing misresponse in questionnaires.

We advocate the continued use of reversed items in survey research, but caution researchers to use them with care.

Prof Bert Weijters, Assistant Professor of Marketing, Vlerick Business School & Prof Hans Baumgartner, Smeal Professor of Marketing, Department of Marketing, Smeal College of Business, Pennsylvania State University

In particular, their nuanced discussion of how to construct reversed and negated items should help researchers enhance the validity of their questionnaires while avoiding the problems created by misresponse to reversed and negated items. The authors take it as a given that, except for special circumstances, multiple items are needed to capture the full meaning of a construct. Acknowledging that responding to reversed items can be error-prone, eliminating them from questionnaires and wording all questions in one direction does not solve the problem and may create a false sense of security by masking non-substantive responding. Furthermore, apparent response inconsistencies do not always indicate a misresponse on the part of the respondent but may instead signal a problem with the questionnaire item itself.

Related article

“Misresponse to Reversed and Negated Items in Surveys: A Review” by Bert Weijters and Hans Baumgartner. To be published in the Journal of Marketing Research.

Accreditations
& Rankings

Equis Association of MBAs AACSB Financial Times