[Questions about questions: cognitive survey of questionnaire development]

Abstract

Objectives: Traditional pretests are used in the development of survey items to identify technical and comprehension problems. Cognitive processes involved in answering survey questions are not the object of this kind of test.

Methods: Cognitive survey methods were used here to test a questionnaire screening for rehabilitation needs in people suffering from back pain. Essential techniques of cognitive testing (think-aloud, probing, confidence ratings) are outlined. We applied these techniques to 20 patients suffering from either acute or chronic back pain in order to test the survey.

Results: The main goal, i.e., identifying problems in item formulation by means of cognitive testing, was achieved. Almost one third of the survey questions were rephrased according to the results of the study. Some of the improvements of the questionnaire are illustrated.

Conclusions: The increased effort required to perform cognitive testing as compared to traditional pretesting pays off. The two methods have specific pros and cons and cannot replace one another.