Attitudes, Values, and Opinions: Models and Dynamics
Stockholm School of Economics
It is commonly assumed in attitude research that attitudes are causally related to beliefs and values. In this chapter, I suggest that a different type of psychological dynamics is involved, and that the seemingly good fit often obtained by expectation models of attitude is due to the existence of a common component in attitudes, values, and beliefs: an underlying image. The image dynamics is revealed in belief-value correlations, which measure thought constraints; high belief-value correlations, often observed, show that beliefs and values are frequently not independent. Three studies will be presented. In Study 1, imagery data are shown to be related to attitude and to belief-value correlations. In Study 2, the concept of negative beliefs, called for in Fishbein scaling, is pursued and psychologically explained. In Study 3, the concept of salient beliefs is challenged. It is found that improbable properties also add importantly to the explanation of attitudes. The chapter is concluded with a discussion of consequences for attitude formation models.
Fishbein and Ajzen ( 1975) gave an exhaustive overview of the field of attitude measurement. They pointed out that all the traditional measurement methods could be conceptualized in the same manner in the sense that they required participants to indicate agreement or disagreement with statements about belief-value combinations of the type "Object X has the property Y." The property Y needs to have value implications, of course. Thus, all attitude measurement calls for observing to what extent people attribute value to the attitude object, to what extent they find it good or