Consumer Response to Advertising: a Discussion of Three Studies

Robert L. King, Virginia Polytechnic Institute and State University
ABSTRACT - Three papers related to consumer response to advertising are discussed. Two papers report results of field investigations, one concerning linear effects in cognitive response to advertising, and the other concerning the effectiveness of celebrity endorsers. The third paper proposes a conceptualization of guilt-arousing communications. Some questions are raised concerning methodology and design of the first two studies, and some suggestions are made for extending the scope of the third paper.
[ to cite ]:
Robert L. King (1981) ,"Consumer Response to Advertising: a Discussion of Three Studies", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 449-451.

Advances in Consumer Research Volume 8, 1981      Pages 449-451

CONSUMER RESPONSE TO ADVERTISING: A DISCUSSION OF THREE STUDIES

Robert L. King, Virginia Polytechnic Institute and State University

ABSTRACT -

Three papers related to consumer response to advertising are discussed. Two papers report results of field investigations, one concerning linear effects in cognitive response to advertising, and the other concerning the effectiveness of celebrity endorsers. The third paper proposes a conceptualization of guilt-arousing communications. Some questions are raised concerning methodology and design of the first two studies, and some suggestions are made for extending the scope of the third paper.

PERCY AND LAUTMAN'S STUDY OF LINEAR EFFECTS IN COGNITIVE RESPONSE TO ADVERTISING

The authors' report of their empirical investigation of linear effects in cognitive response to advertising is both interesting and relevant to the concerns of a substantial part of its Association for Consumer Research audience. Further, the study's design is simple, appropriate for the present investigation, and essentially well executed. Finally, the authors' make good use of recent relevant research literature, and they provide support for existing theory in this field of investigation.

If the paper contains disappointments for this reader, they center on two points. First, early in the design of their study the authors made some critical decisions which had potential for influencing their results significantly, but they did not share the rationale for these decisions with their audience. Second, the reader who seeks a relatively full discussion of the applications and implications of this research remains disappointed. While the value of parsimony may be exceeded only by its rarity in professional dialogues such as ours, in the present situation neither the authors nor the audience seems well served by the brevity of the report in these regards.

Several examples may serve to illustrate the first concern, i.e. inadequate explanation of some basic decisions related to the universe which was studied, and to the testing devices which were developed for data collection purposes in this study. A fuller explanation of issues such as the following would be helpful, particularly to prospective users who have a marketing practitioner orientation:

First, why did the authors select frozen food as the product subject for this investigation? Is it conceivable that their findings might have been substantially different had they chosen a consumer durable product such as a videotape recorder? Might the degree of product diffusion among the general population alter the nature of the responses? What are the implications of selecting a well-known brand of the product, as opposed to a lesser known brand or even a newly introduced brand of the same product type?

Second, what was the basis of the authors' decision to feature "a couple" in one experimental commercial, while using "several people" in the other? More specifically, do the people in these to advertisements represent similar or different age groups, marital statuses, social classes, and life styles, or do they reflect essentially different populations? What are the potential consequences of these similarities and differences to the responses which are forthcoming?

Third, is the definition which is suggested for the product's target market acceptable (i.e., female household heads, aged eighteen or older, with family incomes over $15,000, who have had experience with the product category)? Are such women actually indicative of the product's target market? Or are there substantial regional, urban/ rural, racial, religious, and other demographic differences in the actual target market which may not be accounted for in either the study's definition of the target market, or in the three "widely distributed" geographic areas from which respondents were selected, or both of these factors? More specifically, what does "experience with the product category" imply? Is there a measure of brand usage experience? And perhaps more important from the marketing practitioner's perspective, is attention given to differences in responses from heavy and light users of the product type?

This reader's second concern relates to the paucity of the authors' remarks about applications and implications of their findings. Specifically, how projectable are their findings across various product categories? What are the implications for promotional program development and evaluation generally, and for advertising pre-testing specifically? On the other hand, have the authors possibly generalized excessively when they refer to the broad area of "cognitive response"? Their data refer essentially to "support arguments," which are in fact a subset of "cognitive response."

In conclusion, the paper is a sound one which adds credence to existing theoretical and empirical work. However, a fuller discussion of some of the decisions underlying the study's design, and of possible applications of the findings, would have proved extremely useful.

MOWEN AND BROWN'S STUDY ON EXPLAINING AND PREDICTING THE EFFECTIVENESS OF CELEBRITY ENDORSERS

According to legend, the first recorded instance of a modern athlete leasing his name to a commercial sponsor occurred in 1905 when Honus Wagner, shortstop for the Pittsburgh Pirates, gave the J. F. Hillerich and Son Company permission to use his name on its Louisville Slugger bats for a consideration of $75. However, the non-smoking Wagner sued the American Tobacco Trust, which put his name on a Sweet Caporal baseball card without his permission. In 1907 Wagner appeared with Ty Cobb (who had a cigarette and a candy bar named for him) in Coca-Cola's "The Great National Drink of the Great National Game" advertising campaign ("Tinker To Evers to JWT," 1978).

Two issues which were central to the situation of 75 years ago are addressed by Mowen and Brown, who properly remind us that definitive answers have not yet been found to the questions:

(1) Does endorsement of multiple products tarnish a celebrity's effectiveness?

(2) Is it advantageous to the sponsor to use a number of different endorsers to recommend a given product?

These questions take on a new urgency in view of contemporary promotional practice, which is characterized by increasing use of celebrity endorsers, especially in television commercials. The substantial fees paid to celebrity endorsers also give these questions a new urgency. For example, shortly before his death John Wayne filmed several commercials for Great Western Savings and Loan, a regional advertiser, under a three-year contract which paid Wayne an estimated $1.5 million. Recent trade estimates suggest that some athletes are paid $25,000 or more for making a single national commercial (Forkan 1979, and Morris 1979).

The authors are due appreciation, then, for raising relevant and interesting issues, and for attempting to extend previous related research. Similarly, they have provided us with a useful review of the limited literature in this area of research. In retrospect, however, one may wonder whether the cause would not have been better served had they chosen to investigate only one of these related but difficult questions, rather than both, and had they structured the design of that more limited inquiry much more tightly. More specifically, it appears that the authors have relied unduly on assumptions produced by logic as the basis of much of their study's design, when a "factual" basis could have been obtained relatively painlessly. The latter approach would have given the report's audience greater confidence in its results. Several examples are offered to illustrate this point.

First, reference is made to the possibility of respondents having perceived risk when allowing advertising to influence their perceptions of relatively costly television sets in their earlier study. In fact, was consequential risk perceived by those respondents? Why was "a new ballpoint pen" selected as the product-subject in the present study? Is its comparatively low price sufficient evidence that respondents would not perceive consequential risk if they allowed advertising to influence their perceptions of it? In short, in conducting this replication of their earlier study, did the researchers objectively measure the risk levels associated with letting advertising influence one's perceptions of various product-subjects?

Second, products are classified by status level in the report. But, are the products which are characterized as relatively "low status" (motorcycles, cigars, cologne, and rums) and "higher status" (Hilton hotels, Chivas Regal scotch, and Mercedes-Benz automobiles) in fact significantly different in their status identifications among respondents? Were these differences tested? And in any event, how does product status level relate to the research problem, as it is defined in this study?

Third, the researchers' operational measure of "relevancy" seems open to question. Does the use of a script in which the endorser of the ballpoint pen discusses its performance in signing autographs establish "greater relevancy"? Or does the authors' concern, which was expressed about respondents' perceptions of the three actors' relevancy in the television set study, call for a more objective measure of relevancy in the present study?

Fourth, similar questions can be raised concerning the operational measures of "consensus" (endorsement by multiple sources) and "distinctiveness" (endorsement by the endorser of a single product only). Also, it seems that "who" the endorser(s) is (are) must be examined. That is, did Steve McQueen represent essentially the same image to all of the respondents? Did George C. Scott? What is the impact on responses of this particular mix of endorsers? Does this mix possibly introduce additional complexities into the study?

Fifth, direct attention needs to be given to the issue of "credibility" of individual celebrity endorsers in order to understand the phenomena under investigation here. A substantial professional and trade literature exists to help us in this regard. (See Middleton 1974; Eagly and Chaiken 1975; Dholakia and Sternthal 1977; and Morris 1979, for example.) Unfortunately, this literature is not well represented in this study's "References" section.

Similar questions could be raised concerning other uncontrolled sources of variation, the absence of reported manipulation checks, and the rather thinly disguised experimental conditions in the research. One would hope that the authors will continue their investigation of this timely and interesting subject after reconsideration of some aspects of their operationalizations and manipulations. More rigorous testing would yield much needed guidance for business academicians and practitioners alike.

GHINGOLD'S STUDY OF GUILT AROUSING MARKETING COMMUNICATIONS

Whereas the previous papers reported results of field investigations, the present paper is essentially concerned with conceptualization of a construct for exploring an infrequently investigated phenomenon, guilt-arousing marketing communications. The author has very adequately met his three stated objectives. He introduces this largely new area of research interest for marketing communicators, acknowledging the distinction as well as the interrelationship which exists between "fear" and "guilt." His review and synthesis of the literature dealing with guilt-arousing persuasive communications, along with the accompanying bibliography, should be particularly welcomed by concerned marketing communicators. Finally, his "first draft" conceptualization of guilt-arousing communications is a fresh, thought provoking statement, although it is subject to possibly extensive revision.

Even so, some concerned academicians and practitioners in the field of marketing communications will be dissatisfied with, and less well served than was necessary by, several characteristics of the paper. First, the paper may appear to be too truncated: the first one-third of an empirical paper, omitting the methodology and analysis sections. Certainly, the paper would have been much stronger if it presented empirical findings, however tentative they might be. Failing this, it would be useful to know the author's more detailed thoughts concerning measurement approaches. Probably the major shortcoming of the paper in the eyes of its critics is its lack of a suggested research design for testing the recommended hypotheses.

Second, the reader is offered insufficient guidance in identifying specific guilt-arousing marketing communications, and in measuring the incidence of their use by marketing communicators. The author perceives guilt approaches to be in "fairly widespread use" by marketing communicators, but his perception seems to be untested in the market place. An expanded listing of contemporary examples of guilt-arousing commercials, according to the author's definition, would be quite helpful. In fact, how common are such appeals in marketing communications? We would be assisted in identifying examples if we had an improved or expanded definition of guilt-arousing communications which specifically reflected a marketing communications context. Presently we must ask whether "lowering of self esteem, or feeling of lessened personal worth" and "a need to make retribution for the transgression" (Wolman 1973) result from exposure to advertisements such as the following:

(1) Wisk's "ring around the collar" advertisements, suggesting the housewife's distress upon finding her laundry efforts publicly rancid;

(2) Shower-to-Shower powder advertisements, showing passengers piling into the rear seat of an automobile in order to avoid unpleasant body odor of the person in front;

(3) Church of the Latter Day Saints advertisements, asking whether one has hugged his children today;

(4) Crest toothpaste advertisements directed toward parents, reporting reduced incidence of tooth decay among using children of "cavity-prone" age; and

(5) CARE and Save the Children Foundation advertisements, showing the lean faces and hungry eyes of undernourished children.

Third, in view of the author's necessarily heavy reliance on non-marketing literature, reinforced by the paucity of references to the conceptualization's specific marketing communications implications and applications, one may wonder why he elected to develop and present his work within the context of marketing communications. With a minimum of editing, the report would seem equally applicable to a variety of essentially non-marketing communications, such as those between candidate and voter, office holder and citizen, supervisor and worker, pastor and parishioner, teacher and student, and parent and child. While it is not surprising that applications are more generalized than area-specific in the "first draft" conceptualization in a less explored field, the paper does seem unduly brief in setting forth specific marketing communications examples, implications, and applications.

The paper raises a vide range of additional issues, each of which is worthy of fuller discussion. For example is guilt research's "piggybacking" on fear research really so "dubious"? Is the inappropriateness of cognitive dissonance and information processing for understanding guilt possibly more apparent than real? What is the relevance to the marketing communicator of such distinctions as high vs. low self-esteem, copers vs. avoiders, and high vs. low inherent guilt? Might not the suggested hypotheses take on increased relevance and be more finely tuned to the paper's title if they were expressed specifically in a marketing communications context? Might not an additional hypothesis, reflecting the possible dissipation of guilt arousal with the lapse of time, be appropriate? Could/ should tolerance and intolerance thresholds of aroused guilt be incorporated into the "model"?

In conclusion, the author has developed a thought-provoking conceptualization. But one recalls his earlier statement that "a single experimental investigation could be sufficient as a preliminary test of the proposed paradigm." One would agree, while regretting that the author did not include the results of such an investigation as part of his paper, but hoping that he will continue his interesting efforts and share those results with us in the near future.

REFERENCES

Dholakia, Ruby Roy and Sternthal, Brian (1977), "Highly Credible Sources: Persuasion Facilitators or Persuasive Liabilities?," Journal of Consumer Research, 3, 223-232.

Eagly, A. and Chaiken, S. (1975), "An Attribution Analysis of the Effect of Communication Attractiveness," Journal of Personality and Social Psychology, 32, 136-144.

Forkan, James P. (1979), "You Can Reach for the Stars, Talent Agents Say," Advertising Age, April 23, 1979, 52-53.

Middleton, Patricia N. (1974), Social Psychology and Modern Life, New York: Knopf.

Morris, Bourne (1979), "Will a Personality Sell a Product Better? Pros & Cons," Advertising Age, February 5, 1979, 43-44.

"Tinker to Evers to JWT" (1978), Media Decisions, November 1978, 124 ff.

Wolman, B. B. (1973), Dictionary of Behavioral Science, New York: Van Nostrand Reinhold Company.

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