Advances in Consumer Research Volume 16, 1989 Pages 529-534
THE REPETITION/VARIATION HYPOTHESES CONCEPTUAL AND METHOLOGICAL ISSUES
David W. Schumann, University of Tennessee
D. Scott Clemons, University of Tennessee
Of crucial importance to both the marketing and advertising disciplines is the need to understand the impact of extended advertising campaigns on the consumer. How often must consumers be exposed to product or brand advertisements before information relevant to the consumers' purchase decision is encoded, stored in memory, and made available for retrieval? Furthermore, at what point in an advertising campaign do consumers form opinions about a product that may influence their future purchase decisions? Related to these questions is the crucial task of determining the point at which consumers will tire of seeing and hearing repeated advertisements for the same product.
Media strategists have the difficult assignment of determining the appropriate number of exposures a consumer should receive. Under-exposing the consumer to product advertisements may result in having an insufficient effect on the target consumer. On the other hand, over-exposing the consumer to the advertisements may result in the generation of negative feelings toward the advertisement, the advertised product, or the even company itself. An investigation of the specific conditions under which repeated advertisements are most likely to lead to positive consumer reactions would be immensely valuable to advertising professionals.
In conjunction with the concerns about consumers' sensitivity to repeated advertising, cost concerns also exist. Many companies develop and employ multiple exposure ad campaigns to try to familiarize consumers with their products and brands, hoping that some aspect of the advertisements will persuade consumers to purchase their product. However, media time is expensive. A minute of television commercial time may cost advertisers as much as $1 million for the sponsorship of a major television event (Advertising Age, 1987a). Rates for color print ads in major publications can cost in excess of $100,000 (SRDS, 1987). In 1986, each of the top 100 leading advertisers had advertising expenditures in excess of $80 million (Advertising Age, 1987b). Of these, the top three (Proctor & Gamble, Phillip Morris, and Sears, Roebuck & Company) incurred advertising expenditures of over a billion dollars each. Such high media related costs lower the profit margins of companies thereby potentially diminishing their ability to sustain positive growth in a competitive environment.
It is apparent that the consumer must receive multiple exposures if product or brand advertisements are to be effective (Stanky, 1988). Also apparent is the likelihood that the cost of media will increase as an increasing number of information sources compete for the consumer's attention. In this cluttered environment, messages targeted toward the consumer are likely to take more exposures to be effective (Webb and Ray, 1979). Therefore, the need for finding appropriate levels of advertising exposure which positively influence consumers while avoiding wasted expenditures (and possible negative effects on consumers) is of paramount importance to those developing media strategies.
The purpose of this article is to address this need by discussing advertising variation methods and the conditions of their optimal effectiveness. Two hypotheses are proposed that stem directly from the theoretical framework of the Elaboration Likelihood Model of persuasion (Petty & Cacioppo, 1979, 1981, 1986).
REPETITION - VARIATION
A recognized drawback of using repeated advertising is a negative effect on consumers termed "wearout" (Craig, Sternthal and Leavitt, 1976). A number of studies attempting to discover the impact of repeated messages have found a tedium reaction as people tire of being exposed to the same message over and over again. Several of these studies have noted the emergence of a curvilinear effect, whereby an initial positive effect resulting from a moderate number of exposures is followed by a negative reaction as the number of exposures is increased (Appel, 1971; Cacioppo & Petty, 1979; Calder & Sternthal, 1980; Gorn & Goldberg, 1980; Grass & Wallace, 1969; Winter, 1973). For example, while testing the effects of cogent messages that advocated a position either consistent with or contrary to the attitude initially held by the subject, Cacioppo and Petty (1979) found an increase in agreement with an attitudinal position in the one and three exposure conditions, but by five exposures a decreasing trend became apparent.
Such curvilinear effects have also been found to occur in situations of repeated advertising. Under laboratory conditions, Calder and Sternthal (1980) were able to show that a moderate amount of exposures led to a slight increase in product i evaluation, while increasing the repetition of exposures led to a dramatic decrease. In a more natural setting in which children were exposed to an ice-cream advertisement in the context of a cartoon program, Gorn and Goldberg reported that children voiced a greater preference for the product after receiving three ad exposures (moderate exposure) compared to children receiving either one or five exposures.
Emerging from this evidence of a tedium effect is an important question that warrants investigation. Can this tedium effect be forestalled or perhaps even avoided? If so, what variables might increase the probability of this happening? Different variables thought to perform this function have been investigated including individual differences in product knowledge (Alba & Hutchinson, 1986) and an individual's expectation of support or counter argument production (Batra & Ray, 1986). Perhaps the most intuitive way in which to avoid tedium is to vary the ads used in a multiple exposure ad campaign. Several researchers have, either explicitly or implicitly, found that the act of varying the ads within a particular ad campaign results in some avoidance of negative consumer reaction.
Evidence supporting the use of ad variation to forestall or avoid tedium is provided by several previous studies. Employing an experimental methodology whereby consumers were exposed to slightly varied print advertisements, McCullough and Ostrom (1974) reported that as repetition was increased, liking for the product also increased. In the previously mentioned study with children (Gorn & Goldberg, 1980), the results suggested that varying the ads had a significant positive effect on the preferences and behavior of the children. Grass and Wallace (1969) employed the COMPAAD technique to study the effects of ad variation. The technique involved a method by which the subject could control and maintain the clarity of a presented picture by pressing a foot pedal. The frequency in which the viewer pressed the pedal was recorded automatically and interpreted by the researchers as a measure of interest. High-(low) frequency was interpreted as high (low) interest. Using this measure, subjects rapidly indicated decreased interest when exposed to the same commercial six times (no variation) in a program context. However, upon receiving six different ads during the program, no significant loss of interest was found on the part of subjects.
Though these studies clearly support the argument that ad variation in multiple exposure ad campaigns forestalls or avoids the effects of tedium, two additional studies report less positive results. Burnkrant and Unnava (1987) employed the "encoding variability hypothesis" (Melton 1967; Madigan 1969) to explain why varying the copy of an ad campaign would be more effective than multiple exposures to a single ad with respect to brand recall and attitude measures. Though brand recall was significantly better under varied conditions, ad variation led to no significant differences with respect to attitude. Further, Mitchell and Olson (1981) failed to find any influence, positive or negative, by varying repeated print ads with respect to liking for the product.
In sum, to date some studies have demonstrated a potential for repetition - variation strategies, but the true effects of such strategies are still unclear. Little or no work has been done to investigate either the characteristics of varied ads or the situations in which varied ad campaigns are most effective. It is apparent that ad variation can occur in several different forms. For example, ads may be varied by changing the central aspects of the ad (i.e. arguments or messages), by changing more peripheral aspects of the ad (i.e. format, illustrations, or print style), or the variation could be of a type falling at some point in between these two ends of the continuum. For the purpose of this discussion, the two types of variation representing the ends of the variation continuum will be labeled substantive and cosmetic variation respectively.
Substantive variation represents changes from ad to ad that are central to the product being advertised. Examples of substantive variation include changes in the message or arguments and/or the copy found within the ad. To illustrate this point, consider a varied ad campaign for a brand of soft drink. An example of substantive variation in this context would be to promote the brands taste in one ad while promoting price in a second and perhaps convenient packaging in a third.
Cosmetic variation includes changes more peripheral to the product or advertisement. Such changes might include changing the appearance of the ad using color, print style, format, or characterizations. Consider again the soft drink example. Here, cosmetic variation might entail changes in the background colors, illustration of the ad, or context of the ad (e.g., sandy beach or office scene). Such changes are not central to the product but rather exist in the advertising environment. Given the different strategies available with respect to varying ad campaigns, the obvious question is when will these different strategies be most effective.
THE ELABORATION LIKELIHOOD MODEL
The Elaboration Likelihood Model (Petty and Cacioppo, 1979, 1981, 1986) has its roots in the social psychology literature, but appears to be helpful in providing a basis for predicting certain consumer phenomena. In this case, the Elaboration Likelihood Model (ELM) addresses how repetition - variation strategies might affect the formation or change in attitude toward an attitude object, and the nature of the persuasion produced by these strategies. The ELM proposes two different routes to persuasion. The "central route" is employed when individuals are motivated and able to think or elaborate on an attitude object (an advertised product in this case). In the situation of "central route" processing, Petty, Cacioppo, and Schumann (1983) stated that "attitude change is viewed as resulting from a person's diligent consideration of information s/he feels is central to the true merits of a particular attitudinal position" (i.e. substantive information). It is postulated that attitude changes induced via the central route are relatively enduring and predictive of behavior (Cialdini, Petty, & Cacioppo 1981; Petty & Cacioppo 1980).
A second route to persuasion inherent to the ELM is referred to as the "peripheral route." In this instance, attitude changes occur because the attitude object (a product in the case of advertising) is associated with positive or negative cues, or because the individual makes inferences about the merits of an attitude object based on the cues associated with it in the persuasion context. In an advertising context, the peripheral route is illustrated when a consumer forms a positive attitude toward the advertised product due simply to the presence of an "attractive" characteristic in the ad (or in the immediate surrounding). The positive evaluation of the ad characteristic is shared with the product with which it is associated rather than considering the true merits of the product. The peripheral route makes possible a consumer's formation of an attitudinal position without engaging in any extensive thought (elaboration) about product relevant arguments. In contrast to the central route of persuasion, attitude change induced via the peripheral route is thought to be relatively temporary and not very predictive of behavior.
Immediately one can see how the ELM might be used to aid in understanding the effects of advertising in the context of repetition - variation advertising campaigns. Likewise, repetition variation advertising provides a fitting context to test the positions of the ELM. When applied to such a context, the principles of the ELM suggest that the likelihood of elaboration will moderate the effectiveness of the two types of variation strategies (substantive and cosmetic variation strategies) discussed previously. The following hypotheses are offered as a way of predicting when each form of variation is likely to have the most positive impact on the consumer.
H1 Substantive variation in repeated ads will have greater impact on measures of advertising effectiveness when there is high likelihood of elaboration than under conditions of low likelihood- of elaboration.
If the consumer is highly motivated and able to process substantive information found in the ad, then varying this information over several ads should obtain more positive results by retaining the consumers interest thereby delaying possible tedium effects (see high likelihood of elaboration in Figure 1). Conversely, if the consumer is neither motivated nor unable to process the substantive information, then no effects are expected to result by varying ads with resPect to the information provided.
H2 Cosmetic variation in repeated ads will have greater impact on measures of advertising effectiveness when there is little- likelihood of elaboration than under conditions of high likelihood of elaboration.
Since the unmotivated consumer is not likely to process information central to the product, peripheral aspects of the advertising environment should have a greater impact on the consumer (see low likelihood of elaboration in Figure 1). Furthermore, substantive variation is expected to have no significant effect when the consumer's elaboration likelihood is low. These two predictions are consistent with those of the ELM and state the conditions under which tedium is expected to be forestalled.
METHODOLOGICAL ISSUES IN TESTING THE REPETITION VARIATION HYPOTHESES
It has been pointed out by Petty, Cacioppo, and Schumann (1983) that the accumulated research on persuasion indicates that neither the central nor peripheral approach alone can account for the diversity in the observed results of attitude change (see Petty, Cacioppo, & Schumann, 1983 for a discussion of this research literature). Initial tests of the above hypotheses will have to consider three important factors.
Motivation to Process:
What determines the magnitude of a consumer's elaboration likelihood? A proposed determinant is the degree to which the consumer views the product as being relevant to him/her personally. If the consumer views the product as being highly relevant, then the likelihood of extended elaboration on the part of the consumer is expected to be high. Conversely, if the consumer views the product as having low personal relevance, then the likelihood of elaboration is expected to be low.
The personal relevance of an attitude object can be discussed in terms of how important the attitude object is to the consumer. If the attitude object is made more important to a person (high personal relevance), then substantive variation strategies are predicted to be more effective. Similarly, if the attitude object is made to be less important (low personal relevance), then cosmetic variation is predicted to be a more effective strategy
Several previous studies have successfully manipulated product relevance (e.g., Wright, 1975; Petty, Cacioppo and Schumann, 1983). However, the laboratory environment may provide conditions that enhance relevance of a different nature. Under laboratory conditions, task involvement (personal relevance of the task) may be at such a level that processing of all presented stimuli may be quite high. This may result in a product relevance manipulation that, if successful, provides relative differences at the high end of the elaboration likelihood continuum rather than differences representing the ends of the continuum.
A second factor to consider when testing the repetition - variation strategies hypotheses is the operationalization of the respective variation conditions. As noted earlier, many different aspects of advertisements can be varied to represent changes which are either central or peripheral to the consumer's processing. Substantive variation must consist of varying those aspects of the ad that are central to the viewers decision process (i.e. arguments, messages, product features). Cosmetic variation must be limited to those ad aspects which are peripheral to the viewer (i.e. ad colors, endorsers, layout). Prior to designing the variation strategies, much thought is needed to determine the aspects of the advertising which are considered by the viewer to be central or peripheral.
The effects of variation within different media should also be investigated. Different forms of media might perhaps differ characteristically on the amount of elaboration inherent to each. Krugman (1967) suggested that television is a "low involvement" medium and that viewers tend to be passive processors (low likelihood of elaboration). Given this, cosmetic variation might have a greater impact when employed with television as the medium. On the other hand, the use of print media might result in more active processing (greater elaboration) suggesting that substantive variation may have a greater impact. Other media forms using audio only (i.e radio) or visual only (i.e. outdoor advertising) techniques may also differ on the elaboration dimension, warranting the investigation of the inherent differences of each. It is apparent that the investigation of variation strategies may be quite complex, and require careful consideration with respect to conceptualization and methodology.
REPETITION x VARIATION PREDICTIONS
The third factor requiring consideration deals with the operationalization of repetition. In testing the proposed hypotheses, low, moderate, and high levels of repetition must be manipulated in order to test the effects of each. Achieving these relative levels of repetition within a research paradigm requires a great deal of care and consideration. Whether the repetition levels are considered by the subject to be low, moderate, or high is dependent on the context provided. For instance, 1, 3, and 5 exposures might be perceived by the subject as low, moderate and high repetition in a 30 minute program. However, increasing the program length may reduce the perceived level of repetition. Likewise, decreasing the length might lead to an increased perceived level.
The perception of repetition is also dependent on the number of competing objects in the programming context. If the target ads are placed among a high number of other ads competing for the subject's attention, then the perception of repetition might be reduced. One the other hand, the perception of repetition might increase when target ads are place among a low number of competing ads. Furthermore, spacing between the target ads might dictate the perception of repetition.
A further issue to be deliberated on is that of generalizability. Repetition within programs (which, with few exceptions, is most often employed in laboratory research on repetition), may be conceptually different from repetition between shows and/or exposure across days or weeks. These various issues concerning repetition must be considered in determining what depicts low, moderate, and high levels of repetition in the research context.
This paper has sought to present a discussion of the conditions under which different strategies of advertising variation might be optimally effective. Testing of the repetition - variation hypotheses should follow, but first, certain methological issues should be considered before such an undertaking is initiated.
Advertising Age, (1987a) "Super Bowl-buster," September 24, Chicago: Crain Communications, Inc., p. 1.
Advertising Age, (1987b) "Ad Growth Edges Up," December 7, Chicago: Crain Communications, Inc., p. 1.
Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411454.
Appel, Valentine (1971), "On Advertising Wearout," Journal Of Advertising Research, 11, (February), 11-13.
Batra, Rajeev and Michael L Ray (1986), "Situational Effects of Advertising Repetition: The Moderating Influence of Motivation, Ability, and Opportunity to Respond," Journal of Consumer Research, 12, (March), 432-445.
Burnkrant, Robert E. and Hanurnantha R. Unnava (1987), "Effects of Variation in Message Execution On the Learning of Repeated Brand Information," Advances in Consumer Research, Vol. 14, ed. Melanie Wallendorf and Paul F. Anderson, Provo, UT: Association For Consumer Research, 173176.
Cacioppo, John T. and Richard E. Petty (1979), "The Effects of Message Repetition and Position On Cognitive Response, Recall, and Persuasion," Journal of Personality and Social Psychology, 37, 97-109.
Cacioppo, John T. and Richard E. Petty (1980), "Persuasiveness of Communications is Affected by Exposure Frequency and Message Quality: A Theoretical and Empirical Analysis of Persisting Attitude Change," in Current Issues and Research in Advertising, eds. James H. Leigh and Claude R. Martin, Ann Arbor, MI: University of Michigan, 97-122.
Calder, Bobby J. and Brian Sternthal (1980), "Television Commercial Wearout: An Information Processing View," Journal of Marketing Research, 17 (May), 173-186.
Cialdini, Robert B., Richard E. Petty and John T. Cacioppo (1981), "Attitude and Attitude Change," Annual Review of Psychology, 32, 357-404.
Craig, C.S., Brian Sternthal and Clark Leavitt (1976), "Advertising Wearout: An Experimental Analysis," Journal of Marketing Research, 13, 365-328.
Gorn, Gerald J. and- Marvin E Goldberg (1980), "Children's Response to Television Commercials,'' Journal of Consumer Research, 6 (March), 421-24.
Grass, Robert and Wallace H. Wallace (1969), "Satiation Effects of Television Commercials," Journal of Advertising Research, 9 (September), 38.
Krugman, Herbert E. (1967), "The Measurement of Advertising Involvement," Public Opinion Quarterly, 30 (Winter), 583-596.
Madigan, S. A. (1969), "Intraserial Repetition and Coding Process in Free Recall," Journal of Verbal Learning and Verbal Behavior, 8, 828-835.
McCullough, J. L. and Thomas M. Ostrom (1974), "Repetition of Highly Similar Messages and Attitude Change," Journal of Applied Psychology, 59, 395-397.
Melton, A. W. (1967), "Repetition and Retrieval From Memory," Science, 158, 532.
Mitchell, Andrew A. and James C. Olson (1981), "Are Product Attribute Beliefs The Only Mediator of Advertising Effects On Brand Attitude?" Journal of Marketing Research, 18, 318-332.
Petty, Richard E. and John T. Cacioppo (1979), "Issue Involvement can Increase or Decrease Persuasion by Enhancing Message Relevant Cognitive Responses," Journal of Personality and Social Psychology, 37, 1915-1926.
Petty, Richard E. and John T. Cacioppo (1980), "Effects of Issue Involvement On Attitudes in an Advertising Context," in Proceedings of the Division 23 Program, eds. Gerald G. Gorn and Marvin E. Goldberg, Montreal, Canada American Psychological Association, 75-79.
Petty, Richard E. and John T. Cacioppo (1981), Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, Iowa: William C. Brown.
Petty, Richard E., John T. Cacioppo and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.
Petty, Richard E. and John T. Cacioppo (1986), Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York: Springer Verlag.
SRDS (1988), New York: Standard Rate and Data Service.
Stankey, Michael J., "Using Advertising Media More Effectively," Business, 40 (April-June 1988), 20-27.
Webb, Peter H. and Michael L. Ray (1979), "Effects of TV Clutter," Journal of Advertising Research, 19 (June), 7-14.
Winter, F.W. (1973), "A Laboratory Experiment of Individual Attitude Response to Advertising Exposure," Journal of Marketing Research, 10 (May), 130-40.
Wright, Peter L. (1973), 'The Cognitive Process Mediating the Acceptance of Advertising," Journal of Marketing Research, 10 (February), 53-62.