A test of a model for reducing measured attitude-behavior discrepancies

Identifies 4 sources of attitude-behavior discrepancy and examines the capacity of Fishbein's model to cope with 2 of these sources: inadequate sampling of potential antecedents; and inadequate conceptualization and measurement of attitudes. A survey of intentions regarding 6 kinds of medical transplant donation was conducted among 195 adults. The model's 3 components (attitude toward the act, personal normative beliefs, and social normative beliefs), explained an average of more than 50% of the variance in intentions, with personal normative beliefs consistently the strongest and social normative beliefs the weakest contributor. Regression weights of the model's components remained relatively stable across kinds of transplant and across classifications of Ss by demographic, experience, and personality characteristics. The model was insufficient, however, to obviate the need for sampling further from the multitude of behavioral antecedents. With regard to attitude measurement, attitude toward the act proved superior to the traditional type of index. Data from a validation study suggested that when the model was used to predict volunteering to become a bone marrow donor months later, it was not superior either to 1 of its components (personal normative beliefs) or to 1 alternative model for predicting behavior. (27 ref.) (PsycINFO Database Record (c) 2006 APA, all rights reserved).