Consumer Satisfaction: Discussant Comments

H. Keith Hunt, Brigham Young University
[ to cite ]:
H. Keith Hunt (1983) ,"Consumer Satisfaction: Discussant Comments", in NA - Advances in Consumer Research Volume 10, eds. Richard P. Bagozzi and Alice M. Tybout, Ann Abor, MI : Association for Consumer Research, Pages: 262.

Advances in Consumer Research Volume 10, 1983      Page 262

CONSUMER SATISFACTION: DISCUSSANT COMMENTS

H. Keith Hunt, Brigham Young University

I will not comment on the substance of the Prakash/ Lounsbury paper. That exact paper was double-submitted to the ACR conference and the Consumer Satisfaction conference which will be held next week. The same reviewer received the paper to review for both conferences. Perhaps the reader would like to hum a few bars of "It's A Small World." Both conferences made explicit statements regarding double submissions. These authors chose to violate the rules, so their paper will not be discussed. As a point of interest, their paper was rejected for the consumer satisfaction conference and at least one reviewer recommend rejection for this conference. I suspect the paper-was accepted to fill out this session. Having previously been ACR program chairman I am only too well aware of how these decisions get made.

Now, on to the other two papers which are excellent papers and worth your careful attention.

The Oliver/Bearden paper is to be complimented on its being a natural next step in a research stream. The backgrounding in the CS/D literature is sufficient to position the study as part of a deliberate research program. Involvement is also adequately referenced to clearly show it as a likely influence on satisfaction. The literature references and the framework provided by the authors clearly show involvement to be worth study as a major influence on satisfaction.

Sometimes in our scholarly language and presentation style what we say becomes so affected by the style and jargon that the simplicity of our research question is masked. The Oliver/Bearden paper, in common language, simply studies the question of whether resulting satisfaction differs depending on how involved consumers are with the product. If you don't care a whole heck of a lot about a product, will you be as bothered if the product doesn't work out as you are when a product you really care about doesn't work out. I think we would all make the same prediction. The Oliver/Bearden paper is a scholarly questioning into whether or not our prediction is correct. One of the real values of their paper is that they discover, as is often the case in such inquiries. that it isn't all that simple.

Going on to the Westbrook/Reilly paper, we see an open questioning of the adequacy of the discrepancy model used in the Oliver/Bearden study and in almost all consumer satisfaction studies. The finding that neither the discrepancy model nor the value-percept disparity model do all that well at predicting or explaining resulting satisfaction levels substantiates the Westbrook/Reilly point that we need better constructs and models in the consumer satisfaction area.

We may learn some time in the future that involvement (a 13 Oliver/Bearden) has to be considered before specifying a model. Going out on a limb, let me speculate that the value-percept disparity approach proposed here by Westbrook/Reilly might be a better fit for high involvement matters than for low. Work satisfaction has much higher involvement than we see for most consumer products. On the other hand, for that vast array or consumer decisions which really don't matter to us, we may have to use the discrepancy approach to find anything at all. Perhaps only the seller is truly involved; we as buyers care little about the consumption process for the product.

Finally, not as a criticism of Westbrook/Reilly but rather of much research we do in consumer satisfaction, asking persons to recall days or weeks or months later their expectations before they purchased a product, is a commonsensically and methodologically risky procedure. We all know better. But the cost of a sample large enough in the pre-measure to obtain enough actual purchasers, both satisfied and dissatisfied, in the postuse-measure is too high to be feasible. It would be a massive project and would require continual frequent monitoring of each individual's cognitions and feelings toward a set of products/brands over time without contaminating those very cognitions and feelings being monitored. It is probably impossible and certainly is impractical. So we have to use the recall approach. But let's not ever fool ourselves into thinking our surrogate measure is a valid representation of what really existed in the consumer's mind. Until we find a way to do the massive study, let's keep it clearly in mind that our recall measure is fraught with problems which we ali know about but which we conveniently, selectively se; aside so we can continue our desired research projects.

Just a final comment now. A long time ago, way back in 1977, by my count, there were 7 papers published on consumer satisfaction. I just completed a consumer satisfaction/dissatisfaction and complaining behavior bibliography for the CS/D conference next week. It has 610 entries, 560 of them are main-line consumer satisfaction, dissatisfaction, and complaining behavior papers. Seldom has a research area blossomed so rapidly. Oliver and Westbrook have been key figures in the development of this research area. While we have come a long way in 10 years, their papers show us what a long way we still have to go.

----------------------------------------