Heroes, Villains, Wildcats and Marketing Science Discussion of New Directions in Consumer Behavior and Consumer Research

Thomas C. O'Guinn, University of Illinois
[ to cite ]:
Thomas C. O'Guinn (1989) ,"Heroes, Villains, Wildcats and Marketing Science Discussion of New Directions in Consumer Behavior and Consumer Research", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 426-428.

Advances in Consumer Research Volume 16, 1989      Pages 426-428


Thomas C. O'Guinn, University of Illinois

Two of these papers could be said to be indicative of the "new directions" taken by ACR. Although in point of fact, they are, outside of ACR, not at all new. The fact that we label them as such is testament to the need for ACR members to get off the block, and discover things beyond marketing and psychology. The third paper suggests that it is precisely this "new direction," that is precipitating the "divorce" of consumer researchers (as represented by ACR) and marketing scientists (as represented by ORSAJTIMS). As discussant I offer these few comments, criticisms and suggestions.

Heroes and Villains

Belk's paper, "Effects of Identification with Comic Book Heroes and Villains of Consumption on Materialism Among Former Comic Book Readers," suggests that identification with certain comic book consumption heroes in childhood and adolescence produced less materialistic adults. Identification with consumption villains was less clear in its effect.

The general topic of this paper, the effect of identification with mass mediated consumption archetypes, could prove a fruitful area of consumer behavior inquiry. Most consumer research involving comic books has treated them as cultural artifact, interpretation of which tells us of the consumption values of cultures and subcultures. As valuable as that type of analysis is, this paper takes quite a different approach. It explores the effects of identification with certain types of comic book characters. Comics seem particularly appropriate as an initial foray into this area of inquiry because comic book characters are generally so archetypal, mythic and, therefore, rarely ambiguous. They can thus be typed more easily for analysis, and may make the case for effects somewhat easier to make and defend.

BeLk's work is certainly a departure from previous studies and deserves serious consideration. The consumer behavior field has, other than for advertising, generally ignored the effects of mass mediated communication. There is a lot of mass media beyond advertising out there. We can safely assume that exposure to mass media programming, much of it explicitly involving consumption, has some effect on the behavior of consumers, perhaps even greater than that of advertising. This would certainly seem a reasonable pursuit for those interested in consumer socialization (see Faber and O'Guinn 1988 for a more thorough treatment of this issue). The field's singular focus on advertising as the only mass media material worthy of research attention is difficult to understand. Belk deserves credit for moving consumer research in this "new" direction.

I offer only two cautionary notes to this paper: (1) the case for "effect" is always a tough one to make; and (2) the finding that the overall effect of exposure to such consumption portrayals is generally positive or neutral must be carefully qualified. Thirty-five years of research on mass media has generally failed to demonstrate clear effects of exposure or even identification. The communications field has generally moved away from the exposure "effects" paradigm, and more to a limited effects, cultivation, or "uses and gratifications" paradigm. The later seems particularly appropriate in this instance since it could very well be the case that early childhood socialization or individual differences in personality or motivational factors lead the child or adolescent to seek out and identify with certain types of mass media characters for a wide variety of uses and derived gratifications. As a meta-theory it offers more in terms of a broader definition of "effects," as well as a much more active view of mass media audiences. I am not disputing the association BeLk reports, but offer caution as to the nature of the association and related process.

Another point made by Belk is also worth amplifying, and may represent a very promising path of future research. Belk suggests that it may be the contextual and incidental learning that is the most important effect of identification with comic book consumption characters, thus drawing on the ideas of long overlooked Riesman and Rosenborough (1955). This is an excellent point. It doesn't have to be a direct modeling effect to be meaningful. It may be that by identifying with certain consumption heroes that important associations and inferences are made about how the world works for certain types of people. "Good" rich people have certain things, and things work a certain way for them. The process and it effects may be much more subtle that some straight "effects" model would suggest.


Holbrook offers seven suggestions for those "wildcats" who choose to study consumption symbolism through the semiology of works of art. Because this is a largely prescriptive paper, I have few criticisms, but will offer a few comments. All seven of Morris's suggestions are good advice. While I don't really disagree with any of them, I would offer a comment with respect to two more global comments made by the author.

Holbrook (p. 3) takes the position that "it's not what semiology can do for marketing and consumer research, but rather what marketing and consumer research can do for semiology." While this approach is perfectly legitimate and valuable, the analysis of symbol systems is ultimately an interactive process. It seems that semiology is at its most powerful level when it operates in both directions, by being self-reflexive, asking what each system in turn say about the other. When one is "reading" a film by focusing on consumption one can uncover another layer of meaning, thus making it a richer consumption experience. This is because consumption is such a central component of ours, and many other cultures, that art reveals more meaning through direct consideration of consumption. You can, however, just as well see the film as providing evidence of the way cultures feel about themselves, including their consumer values, ideals, and norms. Frankly, I find the two perspectives inseparable and the distinction somewhat artificial.

I would also like to amplify another point made by Morris. He says that being a wildcat is both "dangerous and difficult." He's right on both scores. Semiological analysis is by its nature difficult. But I think more important is his assertion that it's also dangerous for one's career. This may be true, although I do not understand why this has to be. Furthermore, continually issuing these dire warnings runs the risk of creating a self fulfilling prophecy and a disinhibition and legitimation of prejudice against semiology. I would prefer to think choosing semiology as a primary research tool is at present riskier than some other course of action, but I still would under no circumstances try to dissuade someone who really wants to use this analytical system to better understand consumer behavior. I suppose if you approach an academic career with the prime imperative being "strategy," becoming a semiologist may be a bad idea. If, however, things like happiness and intellectual integrity matter to you, study what you want to know. Be passionate in your pursuit. If you want to be safe, the advice is simple: don't take chances. Study what everyone else studies and do it the same safe way they do it. You'll have a nice safe career, and when you get to the end of it you can say, "wasn't that fun; wasn't that fulfilling." If you are going to spend your brief time on this planet being a camp follower, there are plenty of other domains in which the hours are shorter, the pay much better, and the expectations lower (i.e. politics). Morris really wants you to be a wildcat. He's just telling you the cost. I would ask you to think of the cost of not pursuing what you desire to pursue.


Let me preface my comments on John Rossiter's paper by saying that I realize that it represents a heartfelt and absolutely sincere appeal on the part of Rossiter for what he sees as the good of the organization and the discipline. I, however, respectfully disagree with some, but not all, of his arguments. I do, regrettably, find the paper's tone troubling.

Rossiter says that "Consumer research is losing touch with marketing science;".. and that this "now almost complete divorce from marketing science by consumer behavior theorists is extremely dangerous." He rests his case on three illustrations. First, Rossiter asserts that scanner data in marketing science "threatens to empirically rewrite consumer behavior theory." His argument is that scanner data is "primarily the province of marketing scientists," and that we "can't argue with the data, because they are real." Confrontation with these "real data" is going to force consumer behavior theorists to "radically revise" theory. His suggestion is that consumer researchers draw upon the largely inductive work of the "scientists," so that consumer researchers (by inference, non-scientists) can reformulate theory and then deductively test it. I find this phrasing odd and disturbing. The explicit assertion is that marketing equals science and consumer research is something less than that. I was under the impression that consumer researchers were also scientists. I suppose if you equate behaviorally atheoretic empiricism wit} science, then many of us would plead guilty as charged. I realize, it's all a matter of how you see it but what modelers call stochastic error consumer researchers call human process.

There is, nonetheless, some truth in what Rossiter says. As error is reduced by better collection methods, theorists often have to rework and revise. This is neither unusual or damning. It is simply par of the practice of science. Just as theoretical physicists have had to revise their theories as a result of data gained through advances in radio astronomy and inter-planetary probes, so too will consumer behavior theorists be challenged by advances in scanner data. On this much we agree. Still, I would also suggest that what scanner data really do best is improve dependent measures and a few of the more gross independent ones. It does little to speak to process or even higher level description. It's undeniably valuable, particularly to managers, but to suggest that it's "real" and cannot be "argued with" i strong rhetoric. Scanner methodology is not without problems and is perfectly capable of producing error You are essentially right in asserting that it has more validity to the ecology than a laboratory setting, but that's essentially because most experimental research in consumer behavior is examining micro processes which could not be appropriately examined through scanner technology. You may see the gross effects z many simultaneously operating micro processes in gross scanner measures, but I can't imagine how you would hope to unravel, for example, the way specific cognitive processes operate within the individual consumer. We are talking about different levels of analysis, different process, and different phenomena.

Still, you are right that some consumer behavior theories could to a greater extent be tested by field experiments, but only certain levels and type of theory. This very well might be the case of the two examples you cite (Scott 1976; Dodson, Tybout and Sternthal 1978). Still, even in these cases, would it really test the underlying behavioral theory of why these particular patterns would be expected? Maybe what it's really testing is the veracity of the associated and claimed managerial implications of these theories. If that was your point, I think you would have a much stronger and defensible one.

John's second point is that ACR members are ignorant of "the behavioral facts of consumer purchase behavior discovered by marketing scientists." Again there is some truth here, particularly with respect to American researchers' undeniable geocentrism, and it certainly wouldn't hurt anyone to be more aware of some of the patterns that appear in aggregate data. But again, consumer researchers could just as easily say that modelers are largely ignorant of behavioral processes. Responsibility for the failure to integrate must be shared by both camps.

John's second thesis is clear in this quote (p. 12).

Their (behavioral facts') importance is unarguable and central: the regularities of purchase behavior in the product category constitute the "norms" the we as consumer researchers--must "beat" using our psychological theories of how to better produce responses to marketing stimuli.

I guess I don't quite understand how knowing these norms will make us any more likely to produce better marketing stimuli, or more importantly, build better theory. The paper never gets around to explaining just exactly how knowing normative information like "what the average new brand can expect" will improve our theories of process. As you point out, the--marketing "science" models have no consumer behavior theory- in them. Having normative consumer data is valuable for managers. No one denies that. It also helps academics sell their ideas as managerially relevant (to journals and to consulting clients). But beyond that, I don't see it playing a terribly significant role in behavioral theory development. I certainly could be wrong, but I've been unable to think of a single example where that would clearly be the case.

Rossiter's third and final point is that ACR members have failed "to recognize that stochasticity may be a cause of consumer behavior." John says that consumers "go stochastic," and engage in random choice behavior. This is a very provocative idea, much like Bass' (1974) model which requires "no cognition, no memory of past choices." The difference is that Rossiter says that it is voluntary and dependent on consumers developing "an initial purchasing pattern for the product category which indicates when to repeat in a brand loyal sense and when to "go stochastic" and simply selects a brand probabilistically off the shelf" (p. 14). You say people "voluntarily go stochastic." Doesn't it seems just the least bit contradictory to say that consumers use cognition and memory in order to not use cognition and memory in making a choice? Something drives consumer choice: a preference for a box, a color, previous advertising, or something in the mind, but not stochasticity. There is no cognitive tabula rosa. Your assertion flies in the face of an enormous body of psychological literature. Furthermore, what you offer as evidence is no evidence at all. The presence of stochasticity at the aggregate level is no evidence of the absence of psychological process. The only thing it is evidence of is ecological fallacy.

Admittedly, across enough consumers many individual level processes appear to be random or near random. The problem is that if you take a gross dependent measure like brand switching, there are so many cognitive, social, cultural, and situational processes involved that while each of those may be explained in terms of its role and effect on the final outcome, that for some processes that are essentially low risk and multi-factored that there is the appearance of randomness at the aggregate level. That illusory condition is called the ecological fallacy. You are taking aggregate level data and making inferences about individual level process. When some behavioral outcome of admittedly minor consequence involves the use of many inputs which themselves are variously distributed throughout large populations, invoking a large enough N will give you the illusion of stochasticity as cause. But, it is only an illusion.

To conclude, Rossiter blames ACR's "new directions" for all of this non-science, and asserts that we are "missing the boat" in not attending ORSAMMS. Interestingly, he never explains just how the "new direction" of ACR has done this, by what mechanism. He suggests that we should invite them (scientists) back, reform, change our ways and repent. We are in danger of a divorce if we don't mend our ways.

As far as I know, everyone is invited to ACR. I certainly welcome the interaction you propose. I too think consumer researchers should be challenged and confronted by data from advancing technology. I also think it is healthy for us both.


Bass, Frank M. (1974), "The Theory of Stochastic Preferences and Brand Switching," Journal of Marketing Research, 11 (February), 1-20.

Belk, Russell W. (1988), "Effects of Identification with Comic Book Heroes and Villains of Consumption on Materialism Among Former Comic Book Readers," paper presented at the Association for Consumer Research, October, Honolulu.

Dodson, J.-A., Alice M. Tybout and Brian Sternthal (1978), "Impact of Deals and Deal Retraction on Brand Switching," Journal of Marketing Research, 15, (February), 72-81.

Faber, Ronald J. and Thomas C. O'Guinn (1988), "Expanding the View of Consumer Socialization: A Nonutilitarian Mass-Mediated Perspective," in Elizabeth C. Hirschman and Jagdish N. Sheth (eds.), Research in Consumer Behavior, v. 3, Greenwich: JAI Press, pp 49-77.

Holbrook, Morris B. (1988), "Seven Routes to Facilitating The Semiological Interpretation of Consumption Symbolism and Marketing Imagery in Works of Art: Some Tips for Wildcats," paper presented at the Association for Consumer Research, October, Honolulu.

Riesman, David and Howard Rosenborough (1955), "Careers and Consumer Behavior," Consumer Behavior, V. 3, Lincoln H. Clark, ed., New York: New York University Press.

Rossiter, John R. (1988), "Consumer Research and Marketing Science," paper presented at the Association for Consumer Research, October, Honolulu.

Scott, Carol A. (1976), "The Effects of Trial on Repeat Purchase Behavior," Journal of Marketing Research, 13 (August), 263-269.