Advances in Consumer Research Volume 17, 1990 Pages 766-769
SITUATIONAL PRODUCT RELEVANCE AND ATTITUDE PERSISTENCE
Curtis P. Haugtvedt, Ohio State University
Alan J. Strathman, Ohio State University
Attitudes of subjects exposed to an advertisement for a bicycle were assessed immediately and two days later. Based on Petty and Cacioppo's (1981; 1986) Elaboration Likelihood Model, it was predicted and found that product attitudes formed under a condition of relatively high personal relevance showed less decay than product attitudes formed under conditions of relatively low personal relevance. In addition, it was predicted and found that individuals exposed to the critical ad under conditions of high personal relevance chose to spend more time thinking about the ad than individuals-exposed to the same ad under low personal relevance conditions.
In a recent ACR paper, Haugtvedt and Petty (1989) discussed how individual differences in need for cognition (NC) may be associated with different levels of attitude persistence. Because NC was developed as a dispositional operationalization of the motivation to process information construct in Petty and Cacioppo's (1981;1986) Elaboration Likelihood Model, Haugtvedt and Petty (1989) predicted and found that initial product attitudes of individuals high in need for cognition (HNC) decayed less over a two day period than did the attitudes of individuals categorized as low in need for cognition (LNC). That is, because HNC individuals tend to put more effort into carefully evaluating information and tend to be more influenced by the quality of arguments contained in a persuasive appeal (see Haugtvedt, Petty, Cacioppo, & Steidley, 1988), it was predicted that the attitudes formed by such individuals would be more durable. On the other hand, because individuals low in need for cognition tend to be less influenced by the quality of message arguments and more influenced by factors such as the attractiveness or expertise of an endorser, it was predicted that their attitudes would exhibit higher rates of decay. The underlying assumption was that attitudes would be more likely to persist if a broader cognitive base existed to support the attitude (multiple associations and cognitive responses for HNC individuals) than if attitudes were based on a lesser number of associations or thoughts (the single fact that an endorser was attractive or an expert for LNC, for example).
THE PRESENT STUDY
Despite the conceptual parallels between NC and the effects of situational variables that enhance message processing, no consumer study to date has examined the effects of situational motivations on the persistence of newly formed or changed attitudes. Thus, the present study was conducted to extend the Haugtvedt and Petty (1989) study by examining the influence of the situational variable of personal relevance on the persistence of attitudes formed toward a consumer product.
Situational manipulations of personal relevance have been shown to affect the nature and amount of message processing in past research in both social psychology and marketing (see Petty, Cacioppo, and Goldman, 1981; Petty, Cacioppo, and Schumann, 1983 for examples). Existing studies have shown that under conditions of relatively low personal relevance, attitude change tends to be based on the quantity rather than quality of information, (Petty & Cacioppo, 1984); celebrity status of product endorsers (Petty, Cacioppo, & Schumann, 1983); and the mere number of people supporting a position (Harkins & Petty, 1981). On the other hand, under conditions of relatively high personal relevance, attitude formation or change tends to be based on the quality of message arguments (see Petty & Cacioppo, 1986 for a thorough review).
In the only study to date to examine the effects of personal relevance on attitude persistence, Petty, Cacioppo, Haugtvedt, & Heesacker (1986, experiment 1) exposed undergraduates to the strong argument version of a message advocating the adoption of senior comprehensive exams (attributed to an expert source) under high and low levels of personal relevance. Based on previous research it was predicted that attitudes changed under conditions of high personal relevance would be based on evaluations of the arguments whereas attitudes changed under conditions of low personal relevance would be based more on the fact that the source of the message was an expert. Consistent with predictions, it was found that attitudes changed under high relevance conditions decayed less over a ten day period than did attitudes changed under low relevance conditions.
In addition to studying the effects of situational relevance on attitude persistence, another goal of the present study was to examine the idea that individuals presented an advertisement under high relevance conditions would, if given the opportunity, choose to spend more time thinking about or "reflecting on" an ad than individuals exposed to the same ad under low relevance conditions. This idea stemmed in part from Tesser's (1978) suggestion that the amount of time individuals spend engaged in what he has labeled 'mere thought" can lead to greater attitude polarization. In research in this area, Clary, Tesser, and Downing (1978), for example, found that individuals who were given 90 seconds to reflect on their attitudes toward individuals were more polarized in their attitude than were individuals who were not given time to think. Whereas Tesser's research has examined the effects of assigning individuals various amounts of time to think, we were interested in whether the personal relevance of externally paced advertisement would influence the amount of time individuals would choose to reflect on an advertisement. That is, we predicted that individuals exposed to an externally paced ad under conditions of relatively high personal relevance would choose to spend more time thinking about information contained in an ad after exposure than individuals exposed to the same ad under conditions of relatively low personal relevance.
Eighty-six undergraduate students from an introductory marketing class participated in groups of two to six for extra credit. The students were informed that the study involved the computer presentation and evaluation of print advertisements. They were further informed that because of the number of ads to be evaluated, the study would involve two sessions. The ads were presented on one of six Macintosh SE computers using the application software HyperCard. A series of six advertisements were presented. An informational "card" containing the brand name of the product, where it was manufactured and where the product was to be introduced, appeared immediately before each ad. Importantly, the amount of time the information card and each advertisement remained on the screen was under computer control. After the presentation of each ad, a card (screen) with the statement "take a few seconds to reflect on the ad you just viewed. Click the button below when you are ready to move on" appeared. The amount of time between the removal of the ad and when the subject clicked the button served as a measure of "reflection time." Ads were presented for amounts of time varying from 15 to 22 seconds. The information card for the bicycle ad was presented for 10 seconds and was followed immediately by the bicycle ad. The bicycle ad was presented for 20 seconds. Personal relevance of the product was manipulated by informing subjects via the information card that the product would soon be available in the local or a distant market. The ad was specifically developed to contain strong arguments and positive cues (expertise of the endorser and sheer amount of information). After exposure to all of the ads, subjects were presented a series of questions via the same computer about the ads and products viewed. Among these were questions assessing their opinions of the products contained in the ads.
At the end of the first session, subjects were told to return at the same time two days later. In the second session, subjects' attitudes toward the products viewed in the earlier session were again assessed. They were then debriefed, asked not to tell other students about the nature of the study, and dismissed.
Analysis of attitude toward the bicycle assessed by agreement with the statement "Speedtour Bicycles are good" on a seven point disagree-agree scale revealed comparable attitudes in the low and high relevance conditions in the first session. The mean attitude score of subjects in the low relevance condition was 5.06 and the mean attitude score of subjects in the high relevance condition was 4.72. Analysis of agreement with the statement "The bicycle advertised in session 1 is good" obtained in the second session revealed a significant difference in the attitudes of persons in the low and high relevance conditions (F 1,85 = 5.37, p<.02). Mean delayed attitude of subjects in the low relevance condition was 3.68 whereas the mean delayed attitude of subjects in the high relevance condition was 4.42. As indicated by this pattern, a 2 (low vs. high relevance) X 2 (immediate vs. delayed attitude) mixed design ANOVA revealed the predicted interaction (F 1,84 = 7.71, p<.006).
Analysis of reflection time revealed that subjects in the high relevance condition waited longer after the presentation of the critical ad (F 1,85 = 5.37, p <.02) before moving on to the next ad than did subjects in the low relevance condition (mean reflection time in the high relevance condition = 6.38 seconds; mean reflection time under low relevance= 4 46 seconds).
Both the attitude and reflection time results are supportive of our hypotheses. Attitudes formed or changed under conditions favoring greater elaboration decayed less over a two day period than did attitudes formed or changed under conditions less favorable to argument elaboration. Aside from the Petty et al. (1986) study and the Haugtvedt and Petty (1989) study, very few studies have been reported on the persistence of communication-induced attitude change (see Cook and Flay, 1978, for a review).
Understanding the processes by which attitude formation or change situations influence the durability of attitudes would seem to have important theoretical and practical implications. While there has and continues to be interest in econometric studies on the duration of advertising effects on sales (e.g., Clarke, 1976) and the analytical modeling literature on advertisement scheduling (e.g., Balakrishnan & Hall, 1989), such approaches provide little insight as to the psychological processes at the individual level that may mediate the observed effects. We believe studies of the kind presented here have the potential to build a body of literature that might provide such understanding.
As discussed in the introduction, the Haugtvedt and Petty (1989) study provided evidence that the individual difference variable of need for cognition may be associated with different levels of attitude persistence and the Petty et al. (1986) study provided evidence that situational manipulations of personal relevance for a college exam can have an impact on attitude persistence. There are, however, a number of important differences to consider between the kind of message and situational manipulation used in the Petty et al. (1986) study and the current test of the basic principles in an advertising context. Students, for example, tend to hold very negative attitudes toward the issue of senior comprehensive exams prior to message presentations and are thus very motivated in the high relevance conditions to engage in counter-argumentation during message exposure. It is only when they are unable to successfully counter-argue the very strong arguments contained in the message that attitude change takes place under the high relevance condition. On the other hand, most consumers would not hold such extreme negative views toward an unknown brand of consumer product portrayed in an advertisement. Additionally, the extent of personal relevance or strength of a personal relevance manipulation is likely to be very high for an issue like a college graduation requirement when compared to a personal relevance manipulation like the fact a product will soon be available in a local or distant market. Results of the present study suggest that even with the more subtle manipulations in an advertising context, situational personal relevance can have an important influence on attitude persistence.
The finding of differential reflection time is consistent with our idea of a subject's desire for time to "consolidate" their attitudes after exposure to an externally paced advertisement. Such a finding is also consistent with studies that have examined the distracting nature of rhetorical questions under high relevance conditions in persuasive audio messages (e.g., Petty, Cacioppo, & Heesacker, 1981) and past research on cognitive responses (see Wright, 1981). That is, previous studies have suggested that restricting an individual's ability to digest information from an externally paced presentation can influence cognitive responses. To our knowledge, however, the present study is the first to show that individuals under different situational motivation will take different amounts of time to think about an advertisement after it has been presented.
The impact of providing "reflection time, under appropriate conditions, is interesting and potentially very informative with regard to advertising practice. One might predict, for example, that ads likely to be processed under conditions of high personal relevance might have greater long term impact if individuals are given the opportunity to think about them after they have been removed. Having another ad follow too closely in time might disrupt thoughts and attitude "consolidation" presumably occurring during this time, ultimately reducing the impact of the ad. It should be noted that this suggestion may be quite different than one of simply of keeping an ad in front of an individual longer. There may be some important advantages to having a person think about the ad without having the luxury of its presence. For one, such a process may be more effortful and lead to memorial advantages. Secondly, allowing one's own tendencies or schema's to influence or bias the information might lead to evaluatively more polarized and/or "stronger" attitudes (Sadler & Tesser, 1973). Future research designed to address this issue is currently being planned.
Although the results of the present study are consistent with predictions, little data at the process level of analysis was collected---largely do to constraints of the design. For example, one would ideally like to have collected cognitive response data from persons in the high and low relevance conditions to assess differential message processing. However, collection of such data immediately after the presentation of the advertisement may have caused individuals in the low relevance conditions to engage in more elaborative processing than they might "naturally" ---potentially artificially increasing the persistence of their attitudes. Additionally, previous research by the authors suggests that the collection of cognitive response data at the end of the second session (two days removed) yields little. The examination of such process measures, however, is being addressed in continued research on the topic of attitude persistence.
In addition to the study of attitude decay, fruitful future research might examine the effects of situational manipulations or dispositional tendencies on the extent to which an individual's newly formed/changed attitude will resist change when exposed to a later counter-persuasive appeal. One might, for example, present subjects who initially processed an advertisement under conditions of high vs. low personal relevance with a communication challenging the validity of their attitudes and beliefs. Individuals having a broader base of support (high relevance conditions) for their initial position would likely marshal that support to defend their opinions. Only when the "counterattack" is judged more cogent than their own supportive arguments, would such individuals change. On the other hand, individuals whose support of an initial position is based mainly on some simple cue association may be very susceptible to the influence of other strong cues associated with positions or ideas counter to their own. In any case, it is clear that much might be learned from an examination of attitude change processes over time as well as in multiple message contexts.
Balakrishnan, P.V. (Sundar) and Nicholas G. Hall (1989), "The optimal scheduling of advertisements." Paper presented at the Marketing Science Conference, Duke University, March.
Clarke, Darrel (1976), "Econometric measurement of the duration of advertising effect on sales," Journal of Marketing Research, 13, 345-57.
Clary, E.G., Abraham Tesser, and L.L. Downing (1978), "Influence of a salient schema on thought induced cognitive change," Personality and Social Psychology Bulletin, 4, 39-43.
Cook, Thomas D., and Brian Flay (1978), "The temporal persistence of experimentally induced attitude change: An evaluative review," In L. Berkowitz, (Ed.), Advances in Experimental Social Psychology., Vol. 11.
Haugtvedt, Curtis P., Richard E. Petty, John T. Cacioppo, and Theresa Steidley (1988), "Personality and ad effectiveness: Exploring the utility of Need for Cognition," Advances in Consumer Research, 15, 209-212.
Haugtvedt, Curtis P. & Richard E. Petty (1989), "Need for Cognition and Attitude Persistence," Advances in Consumer Research, Vol. 16, 33-36.
Petty, Richard E., and John T. Cacioppo (1981), Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, I.A.: Wm. C. Brown.
Petty, Richard E., John T. Cacioppo, and Rachel Goldman (1981), "Personal involvement as a determinant of argument-based persuasion," Journal of Personality and Social Psychology, 41, 847-855.
Petty, Richard E., John T. Cacioppo and Martin Heesacker (1981, "The use of rhetorical questions in persuasion: A cognitive response analysis," Journal of Personality and Social Psychology, 40, 432440.
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, 134-148.
Petty, Richard.E., and John T. Cacioppo (1984), "'The effects of involvement on responses to argument quantity and argument quality: Central and peripheral routes to persuasion," Journal of Personality and Social Psychology, 46, 69-81.
Petty, Richard E., and John T. Cacioppo (1986), Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York: Springer-Verlag .
Petty, Richard E., John T. Cacioppo, Curtis P. Haugtvedt, and Martin Heesacker (1986). Consequences of the route to persuasion: Persistence and resistance of attitude changes. Unpublished manuscript, University of Missouri, Columbia, MO.
Sadler, O., and Abraham Tesser (1973), "Some effects of salience and time upon interpersonal hostility and attraction during social isolation," Sociometry, 36, 99-112.
Tesser, Abraham (1978), "Self-generated attitude change," In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 11.
Wright, Peter (1981), "Cognitive responses to mass media advocacy," in R.E. Petty, T.M. Ostrom and T.C. Brock (Eds.), Cognitive Responses in Persuasion, Erlbaum: Hillsdale, New Jersey.