Toward a comprehensive model of analogy in persuasion: A test of the persuasive roles of analogy

Bryan Burton Whaley, Purdue University

Abstract

Rhetorical tradition has long hailed analogy as an effective persuasive device. While there are many theoretical explanations of how analogy functions in persuasion, empirical support for analogy's persuasive impact is limited, as is evidence for the explanations for analogy's influence. This dissertation tested analogy as thought facilitator. One hundred and forty female and male undergraduate subjects participated in a three-factor study measuring two variables (need for cognition and verbal ability) and manipulating one variable, analogy. Subjects read a six argument persuasive message describing various side-effects of using AZT in combating the AIDS virus. The messages for the three experimental conditions varied in the number of explanatory analogies that explicated corresponding message arguments. The Two Analogy condition had analogies placed at the end of arguments two and five. The Four Analogy message had analogies conclude arguments one, two, four and five. The Six Analogy condition had analogies after each of the six arguments. Control condition subjects read the analogy absent version of the message. After reading the message subjects completed various dependent variable measures. Responses to dependent measures suggested nonsignificant findings for the hypotheses and research questions posed. There was no main effect for analogy on attitude change, nor were there significant effects of the interactions between each of the measured independent variables and the analogy manipulation on attitude change. A nonsignificant main effect for analogy on the number of message-relevant thoughts and a nonsignificant main effect for need for cognition on the number of message-relevant thoughts was revealed. Nonsignificant effects were determined for the interactions between need for cognition and the analogy manipulation, and between verbal ability and the analogy manipulation, when the dependent variable for both predicted interactions was the number of message-favorable thoughts (an unpredicted main effect surfaced for verbal ability on the number of message-favorable thoughts). Finally, there were no differences between conditions on argument recall, expressed effort in processing the message, and adjective ratings of the message. Factors that may have contributed to the findings are discussed.

Degree

Ph.D.

Advisors

Babrow, Purdue University.

Subject Area

Communication

COinS