Portal de Periódicos da CAPES

The Theory of Cognitive Dissonance and Experimental Research

1968; SAGE Publishing; Volume: 5; Issue: 4 Linguagem: Inglês

10.1177/002224376800500412

1547-7193

Sadaomi Oshikawa,

Cultural Differences and Values

Experimental research on cognitive dissonance has been hindered by methodological problems. As Chapanis and Chapanis pointed out, some experiments were designed and manipulated to ensure outcomes predicted by the theory [2]. In such cases it is impossible to determine whether predicted outcomes should be accepted as supporting evidence for the theory. Whenever alternative explanations occur, further research is needed to eliminate alternative explanations and to show that the predicted outcomes are postulated by the theory [1]. This article illustrates this methodological problem by examining a published study and suggests a method to eliminate an alternative explanation. The theory of cognitive dissonance suggests that an individual has cognitive elements (or knowledges) about himself, his past behavior, his beliefs and attitudes, and his environments. If one cognitive element follows from another, they are said to be consonant. If one does not follow from another, they are said to be dissonant and arouse a psychological tension called cognitive dissonance. Because cognitive dissonance is psychologically uncomfortable, the individual changes the dissonant cognitive element to reduce the dissonance. The stronger the cognitive dissonance, the more strongly motivated he is to reduce dissonance by changing the cognitive element.1 If a person, given a choice between two equally desirable products, chooses one and rejects the other, the cognitive elements about the rejected product's favorable attributes are counterparts of the cognitive elements that caused him to reject the product. Since this condition arouses high dissonance, the person would, among other alternatives, try to evaluate the chosen product more favorably and the rejected product less favorably to reduce cognitive dissonance. If, given his choice between a product he likes very much and a product he dislikes, he chooses the first and rejects the second, he would not experience post-choice cognitive dissonance, nor would he try to evaluate the chosen product more favorably or the rejected one less favorably. In their study [3], LoSciuto and Perloff had subjects rank nine phonograph records in preferential order. To arouse strong dissonance, one group of subjects was given a choice between the third-ranked and fourthranked albums; another group was given a choice between the third-ranked and the eighth-ranked albums. It was found that high-dissonance subjects tended to rerank the chosen phonograph records as more desirable and the rejected records as less desirable, but low-dissonance subjects did not show this tendency as strongly as did high-dissonance subjects. The experimenters labelled as divergent change the reranking of the chosen albums as more desirable and the rejected albums as less desirable, and reranking of the chosen albums as less desirable and the rejected albums as more desirable was called convergent change. According to the cognitive dissonance theory, a greater proportion of highdissonance subjects will show divergent changes, and a greater proportion of low-dissonance subjects will show convergent changes. A chi-square test of the frequency of divergent and convergent changes supported this prediction at the .001 level of significance and was judged as evidence for the theory.2 I investigated the possibility that changes in desirability ratings of phonograph records after choice reflected the experimental design and manipulation. Study of the experimental design indicated that the experiment was set up to make it easier for low-dissonance subjects to show convergent changes and more difficult to show divergent changes. Since they chose between the thirdand eighth-ranked albums, the third albums had two places to move up and the eighth albums had one place to move down, totalling three places. Because the two albums had 16 possible movements, 3 of 16 movements * Sadaomi Oshikawa is assistant professor of marketing, University of Washington.