Search

In two experiments, the relationship between attitudinal affect and cognition, in terms of Fishbein's (3, 4, 5) model of attitude structure, was investigated, including attempts to delineate some of the factors that might affect the strength of any such relationship. These factors included the source of beliefs (whether freely elicited from the subjects or standard beliefs supplied by the experimenter), the number of beliefs, and the type of belief statement. When wide variation in the level of correlation across attitude objects was noted in the first experiment, a final factor, the type of concept, was also included. hi general, the results suggested that the relationship between affect and cognition is more complex than previously thought. Only the type of belief statement showed a straightforward effect, indicating a difference between the two types of statements used where no difference had been anticipated. Of major importance was the type of concept, which was involved in interactions with the source of beliefs and the number of belief s in the second study. The two-way interaction between concept type and belief source showed that, in terms of highest correlation with the criterion measure of affect, standard beliefs gave the highest correlations for abstract concepts, while free beliefs gave the best for both concrete issues and person concepts. Number of beliefs was also part of a triple interaction with these variables. In this case, six beliefs gave higher correlations for the abstract and concrete concepts, while three beliefs gave the best for person concepts. Using six standard beliefs gave as good a correlation with criterion, though, as did six free beliefs. [ABSTRACT FROM AUTHOR]