Consumer Response to Advertising: Implications For Copy Testing and Copy

Mary Jane Schlinger, University of Illinois, Chicago
ABSTRACT - AND INTRODUCTION In a 1976 "state of the art" review, Jacoby identified several trends in research on consumer behavior. The first and most important was the attempt to apply and, more significantly, develop theory. Another was the move toward searching for cause-effect relationships. The three papers included in the session titled "Consumer Response to Advertising" exemplify these trends; they are theoretically oriented and search for explanations. There also have practical implications for the development of copytesting research, and it is those implications that are discussed in this overview.
[ to cite ]:
Mary Jane Schlinger (1981) ,"Consumer Response to Advertising: Implications For Copy Testing and Copy", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 432-433.

Advances in Consumer Research Volume 8, 1981      Pages 432-433

CONSUMER RESPONSE TO ADVERTISING: IMPLICATIONS FOR COPY TESTING AND COPY

Mary Jane Schlinger, University of Illinois, Chicago

ABSTRACT - AND INTRODUCTION

In a 1976 "state of the art" review, Jacoby identified several trends in research on consumer behavior. The first and most important was the attempt to apply and, more significantly, develop theory. Another was the move toward searching for cause-effect relationships. The three papers included in the session titled "Consumer Response to Advertising" exemplify these trends; they are theoretically oriented and search for explanations. There also have practical implications for the development of copytesting research, and it is those implications that are discussed in this overview.

GUILT AROUSAL

Ghingold's paper titled "Guilt Arousing Marketing Communications: An Unexplored Variable" claims that guilt appeals are widely used in advertising and suggests that guilt arousal as persuasive device should be studied. On a broad level, this paper raises the question of how emotions in general relate to consumer responses to advertising.

"Emotion" is one of those hypothetical constructs that fall through the definitional and theoretical cracks in the jigsaw puzzle of psychology. "All experimenters know that emotion is a topic about which there is no agreement .... " stated Murray in 1938. Nearly forty years later, Zimbardo and Ruch (1977) point out that psychologists still have failed to develop a precise definition of emotions, which variously are seen as motives, traits, bodily changes, and subjective feelings experienced and reported by the individual.

Lists of emotions, like lists of motives, differ both in length and content. Among those mentioned in the literature are admiration, affection, amusement, anger, anxiety, confidence, disgust, enjoyment, fear, friendliness, frustration, gratitude, guilt, hate, irritation, jealousy, love, passion, pity, pride, shame and surprise (Davitz 1969).

Although psychologists do not agree about what emotions are, there is a substantial body of research on the topic that primarily focuses on: (1) the physiology of emotional arousal; (2) the judgment of emotion in others; and, (3) the negative effects of emotion on behavior and health.

In contrast to the psychologists, consumer behavior researchers tend to have ignored the construct, except for the research on fear appeals. Ghingold's paper is interesting because it suggests the possibility and potential usefulness of measuring not only guilt arousal but also a wide range of emotional, nonrational, impressionistic responses to advertising. At the present time, such reactions either are not measured, or the range is limited, e.g., to liking, amusement, empathy and irritation.

With regard to guilt, Ghingold's paper prompts three observations. First, it is important to distinguish guilt as an appeal from shame. Guilt arises when the individual goes against his conscience. Shame, on the other hand, occurs when a cultural norm is violated in the presence of a real or imaginary audience. Advertising about ring-around-the collar, bad breath, underarm odor, spots on dishes, smelly carpets, and household dirt seems more likely to threaten the viewer with shame than guilt.

Guilt may be more appropriate and more often used for relatively involving products, such as low tar cigarettes, or for nonprofit and social advertising, such as anti-alcohol abuse campaigns and fund solicitations.

Second, it is a fairly common strategy for advertising to try to allay guilt about using products such as convenience goods. The purpose is to help the consumer avoid or reduce guilt arousal. One typical approach is to suggest or imply that the time saved by using the advertised brand can be devoted to more important matters --family activities, exercise, and so on. One wonders if the guilt reduction effect of such messages is assessed.

Finally, any discussion of arousing guilt or other negative emotions must consider the moral and ethical implications. Advertising is criticized for manipulating emotions, both positive and negative, when the appeal is seen as exploitive, debasing or insulting. Rokeach (1970) has urged advertisers to avoid promoting products via destructive emotions or beliefs, i.e., negative self-conceptions and primitive fears about rejection, self-worth, self-identity and self-competence.

CELEBRITY ENDORSERS

Nearly every advertising textbook includes a typology of commercial executional styles or structural designs such as slice of life, humor, product demonstrations, celebrity endorsers, etc. (Hefzallah and Maloney 1979). The latter design is the subject of the paper by Mowen and Brown: "Effectiveness Measures of Celebrity Endorsers."

It is estimated that from five to ten percent of television commercials feature celebrities (Shimp 1979, McCollum and Spielman 1980). This is a substantial proportion, considering the disadvantages: stars can be expensive and risky for the advertiser both in terms of money and loss of control, i.e., live stars, unlike Tony the Tiger, may get involved in scandal, lose their star status, die, etc. On the other hand, the advantages that celebrities potentially can offer include credibility for the claims, attention value and memorability for the commercials, and prestige and positive affect that may be generalized to the brand.

Recently McCollum and Spielman (1980) analyzed data from hundreds of celebrity commercial tests conducted by the company over the past twelve years. Among their findings: fewer then half of the celebrity endorser commercials scored above product category norms on either clutter awareness or attitude change measures. The implication is that there is another serious risk in using celebrities: the risk that they may not "work."

Given that the use of stars is both common and chancy, it is not surprising that advertisers are keenly interested in finding ways to evaluate their potential effectiveness. Mowen and Brown approach this problem via a study that is theoretically and operationally complicated, i.e., it is based on an integration of Heider's balance theory with Kelley's attribution theory, and the resulting experiment utilizes multiple manipulations and several dependent variables. The fact that the predicted results were not obtained in the experiment may reflect inadequacies or misinterpretation of the theories (cf. Kiesler, et. al. 1969, Eagly and Himmelfarb 1978, Kelley and Michela 1980) or inadequacies in the research design and execution.

With regard to implications for evaluating and selecting celebrity presenters, Mowen and Brown's study and the few other studies that have been published on the subject suggest that a multifaceted and multiple criteria approach will be required. Among the factors that need to be considered are the celebrity's likeableness, credibility (expertise and trustworthiness), attention value, compatibility with the brand and message, and distinctiveness with respect to endorsing one versus multiple products and brands. There definitely is a need for additional theoretically oriented research on celebrity endorsers, but the research may need to aim at theory building rather than the testing of theories borrowed from other disciplines.

COGNITIVE PROCESSES

"Linear Effects of Cognitive Response to Advertising" by Percy and Lautman examines reactions to television commercials in an information processing framework. Mere specifically, it illustrates the practical application to copytesting of a theory, i.e., that cognitive processes mediate affective and conative reactions to advertising, and a measurement technique, i.e., the quantification of cognitive reactions by content analysis of verbatim protocols (cf. Wright 1973, Olson and Muderrisoglu 1979, and Mitchell 1980).

It is common for advertising agencies to utilize nondirected free elicitation procedures to obtain verbatim protocols from subjects who have viewed an advertisement of commercial. Typical open-ended questions are "what went through your mind as you watched the commercial?" (or looked at the ad) and "Tell me in your own words what the advertisement said and showed." It also is common for the coding or content analysis of such protocols to be simplistic and atheoretical. For example, an analyst might count the number of respondents who: mentioned various executional elements, actions or characteristics; played back messages, stated the brand name, and made evaluative comments about the brand or execution. In copytesting research, the codes and coding procedures usually are not carefully defined, and the reliability is not measured.

In short, the content analysis of open ended verbatims generated from copytests could be, should be and probably will be improved in the future. The paper by Percy and Lautman suggests one approach, i.e., the coding of support arguments. There are, of course, many other analytic possibilities not only for coding of different types of mediating cognitions but also for linguistic analysis and the measurement of nonverbal factors such as voice pitch and response time.

There is an unanswered question about the Percy and Lautman data that is relevant to the implications of the study for copytesting. Respondents were shown a commercial for a well known brand which they had been told to watch very carefully and then asked to relate all of their thoughts while viewing the commercial. The number of support arguments elicited for each respondent was counted, and it was found that as the number of support arguments offered by a respondent increased so did her positive attitude toward the brand and her intention to buy. It is not clear what influence a viewer's prior exposure to advertising, product experience, and pre-existing knowledge, attitude and behavioral intentions with regard to the brand had on her cognitive processing of the commercial and the number of support arguments that she produced. For example, preexisting favorable intentions to buy may have led both to more support arguments and to a high post-viewing intention to buy rating. If copy testers are to evaluate and compare advertisements for established brands in a cognitive processing framework, then the influence of prior experience will need to be controlled or sorted out from the influence of the advertisements.

REFERENCES

Davits, Joel R. (1969), The Language of Emotion, New York: Academic Press.

Eagly, Alice H. and Himmelfarb, Samuel (1978), "Attitudes and Opinions," in Annual Review of Psychology, Vol. 29, eds. M. Rosenzweig and L. Porter, Palo Alto, California: Annual Reviews Inc., 517-574.

Hefzallah, Ibrahim M. and Maloney, Paul (.1979), "Are There Only Six Kinds of TV Commercials?" Journal of Advertising Research, 19 (August), 57-62.

Jacoby, Jacob (1976), "Consumer Psychology: An Octennium," in Annual Review of Psychology, Vol. 27, eds.

M. Rosenzweig and L. Porter, Palo Alto, California: Annual Reviews, Inc., 331-358.

Kelley, Harold H. and Michela, John (1980), "Attribution Theory and Research," in Annual Review of Psychology, Vol. 31, eds. M. Rosenzweig and L. Porter, Palo Alto, California: Annual Reviews, Inc., 457-501.

Kiesler, Charles, Collins, B. E. and Miller, N. (1969), Attitude Change, New York: John Wiley & Sons.

McCollum/Spielman & Company (1980), "Starpower," Topline, 2 (August).

Mitchell, Andrew A. (1980), "The Use of an Information Processing Approach to Understand Advertising Effects," in Advances in Consumer Research, Vol. 7, ed. J. Olson, Ann Arbor: Association for Consumer Research, 171-177.

Murray, Henry A. (1938), Explorations in Personality, New York: Oxford University Press, 88.

Olson, J. C. and Muderrisoglu, A. (1979), "The Reliability of Responses Obtained by Free Elicitation: Implications for Measuring Attribute Salience and Memory Structure," in Advances in Consumer Research, Vol. 6, ed. W. Wilkie, Ann Arbor: Association for Consumer Research, 269-275.

Rokeach, Milton (1970), Beliefs Attitudes and Values, San Francisco: Jossey-Bass, Inc.

Sampson, Edward E. (1971), Social Psychology and Contemporary Society, New York: John Wiley & Sons.

Sheth, Jagdish N. (undated manuscript), "The Surpluses and Shortages in Consumer Behavior Theory and Research."

Shimp, Terence A. (1976), "Methods of Commercial Presentation Employed by National Television Advertisers," Journal of Advertising, 5 (Fall), 30-36.

Wright, Peter L. (1973), "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, 10, 53-62.

Zimbardo, Philip G. and Ruch, Floyd (1977), Psychology and Life, 9th edition, Glenview, Illinois: Scott, Foresman and Company.

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