Attitude intensity, importance, and certainty and susceptibility to response effects.

Abstract

Changes in attitude question form, wording, and context have repeatedly been shown to produce change in responses. It is often assumed that such response effects are less pronounced among individuals whose attitudes are intense, personally important, or held with great certainty. We report the results of 27 experiments conducted in national surveys designed to evaluate this hypothesis. Measures of attitude intensity, importance, and certainty were found not to differentiate individuals who show response effects from those who do not. We discuss possible explanations for these counterintuitive findings. It is now well documented that people's reports of their own attitudes are influenced by a host of factors in addition to the attitudes themselves. Variations in the wording of an attitude question, the formal properties of its structure, or the context in which it is asked all alter responses in systematic ways (Schuman & Kalton, 1985). Such variations in attitude reports are referred to in the survey methodology literature as