Multiattribute Measurement Models and Multiattribute Attitude Theory: A Test of Construct Validity

A distinction is drawn between the multiattribute attitude model as a measurement device and as a theory of attitude formation and change. Using an analysis of variance paradigm to investigate the underlying multiplicative and summative assumptions, Fishbein's multiattribute theory is found to demonstrate reasonably high construct validity. Individual differences in attribute combination rules are identified, and the issue of cognitive averaging vs. cognitive summation is raised.