The acquiescence effect in responding to a questionnaire.

OBJECTIVE: Since the acquiescence effect can distort assessments, it is important to test techniques to quantify this effect. METHODS: The tendency of acquiescence is tested by means of a questionnaire. 2037 representatively selected subjects filled in the Multidimensional Fatigue Inventory (MFI-20), which consists of five subscales with positively and negatively formulated items. For each subject and for each subscale an acquiescence score was calculated based on the simple sum of the answers to the items of both orientations. RESULTS: Extreme acquiescence scores were rare (about 0.5%). All correlations between the acquiescence scores of the subscales were positive with mean values of 0.24, which indicates a certain degree of individual consistency in the acquiescence behavior. In the exploratory and confirmatory factorial analyses the polarity of the items was at least as meaningful as the contents of the subscales. Persons with high degrees of acquiescence were principally older and more depressed. By means of Rasch scaling procedures differences in threshold parameters for the response categories between positively and negatively oriented items were found. CONCLUSION: The acquiescence scores derived by simple addition of item values proved to be well suited to clarify amount and conditions of the acquiescence effect.