In the most recently published validation test of the Extended Fishbein Model of consumer attitudes (Lutz 1977), a unique and interesting experimental approach was used to successfully generate a number of attitudinal constructs. The major conclusion was that the postulated causal flow of effects was reasonably supported. While the attempt to validate this particular attitude model using a process approach is to be applauded, this comment questions whether, on the basis of the analysis reported, the conclusion was warranted. Additional analysis and interpretation does not support the Extended Fishbein Model as operationalized in these experiments. The normative component is redundant and the measure of attitude appears to be misspecified. Alternatively, the results are spurious because of research design problems. The general implication is that marketers and behaviorists should be very cautious in their use of the Extended Fishbein Model in attitude change situations. Future research needs to confront a number of serious theoretical problems and measurement issues not addressed in these experiments.
[1]
E. Laird Landon,et al.
Order Bias, the Ideal Rating, and the Semantic Differential
,
1971
.
[2]
Michael J. Ryan,et al.
The Fishbein Extended Model and Consumer Behavior
,
1975
.
[3]
Olli T. Ahtola,et al.
Toward a Vector Model of Intentions
,
1976
.
[4]
R. Lutz.
An Experimental Investigation of Causal Relations Among Cognitions, Affect, and Behavioral Intention
,
1977
.
[5]
I. Ajzen,et al.
Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research
,
1977
.
[6]
William L. Wilkie,et al.
Issues in Marketing's use of Multi-Attribute Attitude Models
,
1973
.
[7]
J. Elashoff,et al.
Multiple Regression in Behavioral Research.
,
1975
.
[8]
H. E. Krugman.
THE IMPACT OF TELEVISION ADVERTISING: LEARNING WITHOUT INVOLVEMENT
,
1965
.