The Interview Effect in Polling

Abstract

Each year approximately one million Americans are confronted by interviewers and asked to answer questions about current issues. Do these interviews have an effect on the respondents? If so, what is it, and is it socially beneficial or the reverse? This article presents data to show that there is an effect on the respondent, and suggests that this effect is in part socially beneficial. In addition, however, the respondent may become biased in one of a number of directions which are clearly undersirable. Opinion surveying organizations have it in their power to minimize the biasing effect, while conserving the beneficial factors of intellectual stimulation and clarification of view which interviews also tend to induce in respondents.