Appeals to Image and Claims About Quality: Understanding the

Journal of Personality and Social Psychology1985, Vol.«, No. 3, 586-597

Copyright 1985 by the American Psychological Association, inc.0022-3514/85/S00.75

Appeals to Image and Claims About Quality:Understanding the Psychology of Advertising

Mark Snyder and Kenneth G. DeBonoUniversity of Minnesota

In three investigations we examined the evaluative and behavioral reactions of highand low self-monitoring individuals to two advertising strategies: appeals to a prod-uct's image and claims about a product's quality. High self-monitoring individualsreacted more favorably to image-oriented advertisements, were willing to pay morefor products if they were advertised with an image orientation, and were morewilling to try a product if it was marketed with an image appeal. By contrast, lowself-monitoring individuals reacted more favorably to product-quality-oriented ads,were willing to pay more for products if they were advertised with a quality ori-entation, and were more willing to try a product if it was marketed with a qualityclaim. Theoretical and practical implications of these findings for advertising strat-egies, as well as theoretical implications of these findings for the nature of attitudes,are explored.

It has been called "the most potent influencein adapting and changing [our] habits andmodes of life, affecting what we eat, what wewear, and the work and play of the whole na-tion" (Coolidge, cited by Fox, 1984, p. 97). Ithas been compared with "such long-standinginstitutions as the school and the church in themagnitude of its social influence" (Potter,1954, p. 167). It "dominates the media, it hasvast power in the shaping of popular standards,and it is really one of the very limited groupsof institutions which exercise social control"(Potter, 1954, p. 167). "It" is advertising, whosemessages reach out in words and in pictures,from newspapers and magazines, from radiosand televisions.

For most people, advertising is all but im-possible to avoid or to ignore. On televisionalone, viewers are exposed to tens of thousandsof commercial messages every year (Hacker,1984). These messages may be viewed as at-tempts at persuasion and social influence. In-deed, one of the giants of the advertising in-

This research and the preparation of this manuscriptwere supported in part by National Science FoundationGrant BNS 82-07632 to Mark Snyder. We thank RussellNettle for his assistance in conducting the investigations,and Allen M. Omoto, Jeffry A. Simpson, and Dave Smithfor their helpful comments on the manuscript.

Requests for reprints should be sent to Mark Snyder,Department of Psychology, University of Minnesota, 75East River Road, Minneapolis, Minnesota 55455.

dustry, William Bernbach of the Doyle DayneBernbach agency once said that "Advertisingis persuasion, and persuasion is ... an art.Advertising is the art of persuasion" (cited byFox, 1984, p. 251). As persuasion artists, thecreators of advertising seem to belong to twoschools, each of which has flourished through-out the history of the advertising industry.Within the trade, these two schools are known,respectively, as the "soft-sell" approach and the"hard-sell" approach (cf. Fox, 1984).

Practitioners of the soft-sell approach typi-cally create ads that appeal to the images as-sociated with the use of the product, imagesthat one may gain and project by using theproduct. Practitioners of this image-based ap-proach to advertising believe that how a prod-uct is packaged by its advertising is as impor-tant as the product itself. Therefore, they tendto create advertisements that are very strikingin their visual appeal; they pay particular at-tention to the finer details of form and color.Typically, the copy associated with these adsemphasizes the image of the product or, morespecifically, the images associated with the useof the product. These image-oriented adsrarely, if ever, make any explicit mention ofthe quality of the product, and, indeed, littleemphasis is placed on the product itself.

Consider the case of the ad campaign forArrow shirt collars, which focused not on theproduct itself but on the image of the man

586

THE PSYCHOLOGY OF ADVERTISING 587

who used it. In this case, the advertising agency"created a campaign stressing the accessoriesand background of the man who wore theproduct. Instead of picturing the collar by it-self, [they] put it around the neck of a stylishyoung man, impossibly clear of eye, clean ofjowl and square of jaw and surrounded withopulent possessions and women" (Fox, 1984,p. 44). Or, consider the case of the DeSoto carads, which pictured young people, happily atplay, along with the simple message, "Expectto be stared at." Perhaps the best known ex-ample of image-oriented advertising is theMarlboro man, emphasizing the rugged, mas-culine image of the man who smokes Marl-boro cigarettes. In not one of these cases doesthe ad contain any explicit information aboutthe product itself, only allusions to the imagesto be gained by identifying oneself with theproduct.

By contrast, practitioners of the hard-sellapproach to advertising have created ratherdifferent ads, ones that focus on claims aboutthe intrinsic merit, inherent quality, and func-tional value of the product itself. Their ads tellthe consumer how good the product is, howwell it works, or, in the case of things to eatand drink, how good they taste. For these ad-vertisers, it's the "matter, not the manner" thatcounts (Fox, 1984, p. 324). The earliest effortsof this school were on behalf of patent medi-cines and cure-alls, one of the very earliest ofwhich was Lydia E. Pinkham's VegetableCompound, advertised as "a sure cure for allfemale weaknesses, . . . efficacious and im-mediate in its effects" (cited by Fox, 1984, p.141). Recent advertisements for Total cereal,emphasizing the nutritional benefits of the ce-real, clearly fall into this category. So too doPepsi Cola's now-famous series of Pepsi chal-lenge taste tests, designed to communicate thesupposedly superior taste quality of Pepsi and,in so doing, counter Coke's claim that "It's thereal thing."

There is no doubt about it. The history ofadvertising testifies to the faith that these twostrategies have engendered in their proponents(for a review, see Fox, 1984). Of course, someads feature elements of both strategies, pre-senting information both about image andabout quality. Nevertheless, the fundamentalquestion to be asked is that of the basis of the

effectiveness of either type of advertising ap-peal. What is it that makes image and qualityappeals succeed in engaging, motivating, andpersuading consumers? In other words, whatare the psychological mechanisms involvedthat render each of these strategies successful?

One way to address a question such as thisone is to identify categories of individuals whoare especially responsive to either type of ad-vertising: one category of individuals who areespecially responsive to image-based advertis-ing appeals and a contrasting category of in-dividuals who are especially responsive toproduct-quality-based advertising appeals.Once identified, these categories of individualscan provide a more thorough understandingof the reasons why these two markedly differenttypes of appeals are effective. Members of eachcategory may serve as the ideal candidates inone's investigation of the dynamics of eachstrategy in operation. Moreover, susceptibilityto one or other type of advertising strategy maybe but one specific feature of a larger and moreextensive syndrome of regular and consistentdifferences in how members of these categoriesthink, feel, and act. Then the differing gener-alized interpersonal orientations characteristicof these contrasting categories of individualsmay provide a framework within which to un-derstand the effectiveness of advertising. (Fora more elaborated discussion of this strategyof inquiry in personality and social psychology,see Snyder & Ickes, 1985.)

Are there, then, these two categories of in-dividuals who are differentially influenced bythese two types of advertising appeals? Specif-ically, are there individuals who typically areparticularly responsive to advertisements thatstress the image associated with the use of aproduct and, conversely, are there individualswho typically are especially influenced by ad-vertisements that emphasize the quality of theproduct? There are reasons to believe that thesetwo contrasting categories of individuals maybe identified with the psychological constructof self-monitoring (Snyder, 1974, 1979).

High self-monitoring individuals typicallystrive to be the type of person called for byeach situation in which they find themselves(Snyder & Monson, 1975). They are adept attailoring their behavior to fit social and inter-personal considerations of situational appro-

588 MARK SNYDER AND KENNETH G. DEBONO

priateness (Lippa, 1976) and, as a result, theirbehavior often displays marked situation-to-situation shifts in the images they convey toother people (Danheizer & Graziano, 1982;Shaffer, Smith, & Tomarelli, 1982). These highself-monitoring individuals are identified bytheir relatively high scores on the Self-Moni-toring Scale (Snyder, 1974).

Because of their concerns with being theright person in the right place at the right time,these high self-monitoring individuals oughtto be very sensitive to the images of self thatthey project in social situations (indeed, thereis some suggestion that they may be; Snyder,Berscheid, & Click, 1985), and, as such, theymay be especially attentive to and influencedby advertising messages that convey infor-mation about the images that they acquire andproject by virtue of using particular consumerproducts. In other words, to the extent that anadvertisement allows high self-monitoring in-dividuals to perceive that a given product hasthe potential to be used to create or enhancean image, they should react favorably to it.They should respond to the cigarette ad thatpromises sophistication (even if the ad saysnothing about the quality of the tobacco in thecigarette), to the car ad that features a sporty-looking car (even if the ad says nothing aboutthe performance and handling characteristicsof the car), and to the toothpaste ad that offerswhiter teeth and brighter smiles (even if the adsays nothing about the cavity-fighting capa-bility of the toothpaste).

By contrast, low self-monitoring individualstypically do not attempt to mold their behaviorto fit situational and interpersonal considera-tions (Snyder & Monson, 1975). Instead, theseindividuals tend to guide their behavioralchoices on the basis of information from rel-evant inner sources, such as attitudes, feelings,and dispositions (Snyder & Tanke, 1976). Asa result, low self-monitoring individuals typi-cally display substantial correspondence be-tween their private attitudes and their actualbehavior in social contexts (Snyder & Swann,1976; Zanna, Olson, & Fazio, 1980). Low self-monitoring individuals are identified by theirrelatively low scores on the Self-MonitoringScale (Snyder, 1974).

Unlike their high self-monitoring counter-parts, low self-monitoring individuals are less

concerned with the images they project to oth-ers in social situations; instead, they are moreconcerned that their behavior in social contextsbe an accurate reflection of their underlyingattitudes, values, and dispositions. As such,they may be particularly responsive to adver-tisements that feature appeals to a product'squality. Information about product qualitymay be readily interpreted by these individualsin terms of their underlying attitudes, values,and other evaluative reactions. Take, for ex-ample, a low self-monitoring individual whoenjoys the taste of Scotch whisky and who alsodrinks Scotch whisky. To maximize the con-sistency between this individual's favorable at-titude toward Scotch and the behavior ofdrinking a Scotch whose taste he or she actuallyenjoys, as well as to maximize the consistencybetween this individual's behavior of drinkingScotch and the accompanying evaluative re-action, "I am enjoying the taste of this Scotch,"this individual ought to drink only thosebrands of Scotch that taste the way good Scotchshould taste, brands whose taste this individualtherefore would enjoy. Hence this individualshould be particularly attentive and responsiveto advertisements that inform him or her aboutthe good taste of particular brands of Scotch.Such ads would provide information useful tothis individual in making consumer decisionsthat provide opportunities to act in a mannerconsistent with underlying attitudes (in thiscase, to drink a Scotch the taste of which heor she enjoys), as well as opportunities to max-imize the consistency between the behavior ofdrinking Scotch and the evaluative reaction ofenjoying the taste of the Scotch he or shedrinks.

Are high and low self-monitoring individ-uals differentially responsive to ads that prom-ise images and to appeals that feature productquality? In our research, designed to answerthis question, we have created advertisementsthat, in pictures and words, represent image-based and product-quality-based messages toconsumers. Reactions to our ads, collected ina series of empirical investigations, suggest thathigh self-monitoring individuals are particu-larly responsive to image-oriented advertisingappeals and that low self-monitoring individ-uals are particularly responsive to product-quality advertising appeals.


Study 1

In Study 1, we created three sets of magazineadvertisements, each set containing two ad-vertisements for a particular product. We ad-vertised whisky, cigarettes, and coffee, threefrequently advertised consumer products. Thetwo advertisements for each product wereidentical in all respects except for one: thewritten message or slogan associated with thepicture. One slogan was an appeal to the imageassociated with the use of the product; theother slogan was a claim about the product'squality.

Method

Participants

Fifty male and female undergraduates enrolled in intro-ductory psychology at the University of Minnesota partic-ipated in this investigation for course credit. On the basisof a median split of their Self-Monitoring Scale scores(Snyder, 1974), obtained in an earlier questionnaire survey,

we categorized half of the participants as high self-moni-toring individuals and half as low self-monitoring individ-uals.

Stimulus Materials

We created three sets of magazine type advertisements,each set containing two advertisements for a particularproduct. Specifically, we advertised Canadian Club whisky,Barclay cigarettes, and Irish Mocha Mint flavored coffee,with the following contents.

Canadian Club. The picture for this set of advertise-

ments prominently displayed a bottle of Canadian Clubresting on a set of house blueprints. The written copy forthe image-oriented advertisement stated, "You're not justmoving in, you're moving up," and the product-quality-oriented advertisement claimed that "When it comes togreat taste, everyone draws the same conclusion."

Barclay cigarettes. Here, the pictorial content depicteda handsome gentleman, about to light up a cigarette, look-

ing into a mirror at his female companion. The woman'shand is shown resting on the gentleman's shoulder. Theimage-oriented message read, "Barclay . . . you can seethe difference," and the product-quality-oriented messageread, "Barclay . . . you can taste the difference."

Irish Mocha Mint. For this set, the picture showed a

man and a woman relaxing in a candle-lit room, smilingat each other while drinking Irish Mocha Mint coffee. The

image-oriented ad promised to "Make a chilly night becomea cozy evening with Irish Mocha Mint," and the product-

quality ad informed the consumer about "Irish MochaMint: A delicious blend of three great flavors—coffee, cho-colate, and mint."1

Procedure

When participants arrived for their individual appoint-ments, the experimenter informed them that their task

would be to help in evaluating the relative merits of ad-vertisements currently being studied by researchers at theuniversity. The experimenter then presented, sequentially,the three sets of advertisements (with their order of pre-

sentation counterbalanced across participants). After eachset, participants filled out a 12-item questionnaire on whichthey were to make evaluative comparisons between thetwo advertisements in the set. For example, the question-naire asked "Overall, which ad do you think is better?""Which one appeals to you more?" "Which ad do you

think would be more successful?"

Results and Discussion

We predicted that high self-monitoring in-dividuals would react more favorably than lowself-monitoring individuals to the image-ori-ented advertisements, but that low self-moni-toring individuals would respond more favor-ably than high self-monitoring individuals tothe product-quality-oriented advertisements.To derive an index of favorability toward theadvertisements, we assigned participants ascore of 1 each time that, in making evaluativecomparisons between two ads, they favored theimage-oriented ad, and a 0 each time they fa-vored the product-quality-oriented ad. Foreach product, the 12-item index created by thisscoring procedure was internally consistent(for Canadian Club, KR-20 = .91; for Barclay,KR-20 = .95; for Irish Mocha Mint, KR-20 =.91). Therefore, to obtain for each product anindex of favorability toward each type of ad,we summed the participants' scores over the12 items for each product. This in essence cre-ates, for each product, a 0-12 index of favor-ability toward the two types of ads; higherscores indicated greater favorability toward theimage-oriented ad and lower scores indicatedgreater favorability toward the product qualityad. Means for this index of favorability towardeach type of ad are presented, for each product,in Table 1.

1 As a check on the effectiveness of the manipulation,

we presented 10 judges with the three sets of advertise-ments and, after every set, asked them, "Which ad do youthink is directed at concerns about product quality?" and"Which ad do you think is directed at concerns aboutproduct image?" All of the judges correctly classified allof the advertisements into the image-oriented and product-quality-oriented categories they had been designed to rep-resent.


Table 1Evaluations of Advertisements: Study I

Product

Self-monitoring Canadian Irish Mochacategory Club Barclay Mint

HighLow

6.84

4.885.003.68

7.406.08

Note. Range = 0-12. Higher scores indicate greater favor-ability toward image-oriented advertisements, and lowerscores indicate greater favorability toward product-quality-oriented advertisements.

We then entered the participants' scores on

the favorability index into an analysis of vari-

ance (ANOVA) with one between-subject factor

(self-monitoring) and one within-subject factor

(product). This analysis revealed a statistically

significant main effect for self-monitoring

propensity, F(l, 48) = 4.21, p < .05; that is,

as predicted, high self-monitoring individuals

reacted more favorably to the image-oriented

ads than did low self-monitoring individuals,

and low self-monitoring individuals reacted

more favorably than did high self-monitoring

individuals to the product-quality-oriented

ads. Moreover, the differential favorability of

high and low self-monitoring individuals to-

ward the image and quality advertisements was

evident, to some extent, for each of the three

products. Tests of simple main effects (cf.

Keppel, 1982) indicated that the cell means

on the favorability index were in the predicted

direction for the Barclay ad, the Canadian Club

ad, and the Irish Mocha Mint ad: r(48) = 1.36,

p< .10; r(48) = 1.84, p < .05; and /(48) =1.34, p < .10, respectively.

Clearly, individuals high and low in self-

monitoring have different evaluative reactions

to advertising that is based on appeals to image

and claims about product quality. Although

the effectiveness of an advertising campaign

can, in part, be measured by the evaluative

reactions to the advertisements, this is surely

not the whole picture. The goal of any adver-

tising strategy is not only to elicit a favorable

reaction from the consumer, but also to induce

the individual to purchase the particular

product. Therefore, to determine how gener-

alized these evaluative reactions to image-ori-

ented and product-quality-oriented advertising

are, and to examine the effects of these differ-

ential evaluations on actual consumer behav-

ior, we next examined the impact of these two

modes of advertising on one important featureof consumer behavior.

Study 2

In Study 2, we sought to determine whether

the way a product is advertised would have

any impact on how much consumers would

be willing to pay for that product. On the basis

of the results of the first study, we predicted

that high self-monitoring individuals would be

willing to pay more for products if they were

advertised with appeals to their images than

with claims about their quality. By contrast,

we predicted that low self-monitoring individ-

uals would be willing to pay more for identical

products if their advertising focused on quality

rather than images.

Method

Participants

Forty male and female undergraduates enrolled in in-troductory psychology at the University of Minnesota par-ticipated for course credit. On the basis of a median splitof their Self-Monitoring Scale scores (Snyder, 1974), ob-tained in an earlier questionnaire survey, we categorizedhalf of the participants as high self-monitoring individualsand half as low self-monitoring individuals.

Procedure

The experimenter informed participants, scheduled ingroups of 3 to 4 and assigned randomly to one of twoconditions, that as part of an ongoing research programon advertising, they would respond toanumberof adver-tisements that presently were being pretested. The exper-imenter then presented participants in one condition with

three image-oriented advertisements, and those in the othercondition with three product-quality-oriented advertise-ments (which in each case were the same three ads devel-oped for Study 1, presented sequentially in a counterbal-anced order).

After viewing each ad, participants filled out a ques-tionnaire, the critical item of which was "How much wouldyou be willing to pay for this product?" To avoid extremeresponses on this question, the experimenter gave partic-ipants ranges from which they could select their prices.The ranges provided for Canadian Club, Barclay, and IrishMocha Mint were $5.00-15.00, $0.50-1.50, and $2.00-5.00, respectively, ranges chosen so as to approximately


represent prices at which these products could be purchased

at the time of the investigation.


The mean prices participants were willingto pay for each product are presented inTable 2. To test our hypotheses about the in-fluence of advertising, we entered participants'"willingness to pay" prices into a 2 X 2 X 3(Self-Monitoring X Type of Ad X Product)ANOVA; self-monitoring and type of ad werebetween-subjects factors and product was awithin-subjects factor.

This overall analysis revealed a highly sig-nificant Self-Monitoring X Type of Ad inter-action, F(l, 36) = 23.87, p < .001. As pre-dicted, high self-monitoring individuals werewilling to pay more for the products if theywere advertised with an image orientation thanif they were advertised with a quality orien-tation, ?(36) = 1.93, p < .05, and low self-mon-itoring individuals were willing to pay morefor the products if they were advertised withclaims about their quality than if they wereadvertised with appeals to their images, ((36) =2.14, p < .05. Moreover, the absence of a sig-nificant three-way interaction, F(2, 12) < 1,suggests that there were no reversals of thetrend across the three products. However,analyses of the Self-Monitoring X Type of Adinteraction for each product considered sep-arately revealed that this effect was particularlystrong for the Canadian Club ads, F( 1, 106) =14.92, p < .01, less so for the Irish Mocha Mintads, f\l, 106) = 2.14, p < .15, and even lessso for the Barclay ad, F < 1, m, although inall three instances the means were in the pre-dicted direction.

These results indicate that the differentialfavorability of high and low self-monitoringindividuals to image and quality appeals is ac-companied by differences in at least one rel-evant consumer behavior, namely, how muchthey would be willing to pay for products. Forhigh self-monitoring individuals, favorableimages were worth money; by contrast, for lowself-monitoring individuals, quality was worthpaying for. We next sought to determinewhether these differing orientations towardadvertising extend so far as to influence actualdecisions to consume or not to consume prod-

Table 2

Price Willing to Pay: Study 2

Type ofadvertisement

Canadian Club"Image

QualityBarclay*

ImageQuality

Irish Mocha Mint"Image

Quality

Self-monitoring

High

9.758.24

0.890.89

3.433.28

category

Low

7.508.64

0.740.94

2.973.50

Note. Dollar signs omitted.1 Range = $5.00-$ 15.00. b Rang: =S2.00-S5.00.

$0.50-$ 1.50.° Range =

ucts on the basis of information gleaned fromtheir advertising.

Study 3

Are high self-monitoring individuals morewilling to actually consume a product if it isadvertised with an appeal to image than witha claim about product quality? And, con-versely, are low self-monitoring individualsmore willing to actually consume a product ifit is advertised with a quality orientation thanwith an image orientation? To answer thesequestions, we conducted Study 3, in whichconsumers were offered the opportunity to tryout a new shampoo.

Method

Participants

Forty introductory psychology undergraduates at theUniversity of Minnesota participated in this study forcourse credit. On the basis of a median split of their scores

on the Self-Monitoring Scale (Snyder, 1974), we categorizedhalf of the participants as high self-monitoring individualsand half as low self-monitoring individuals.

Procedure

The experimenter, blind to the participant's self-moni-toring score and posing as a market researcher, contactedparticipants by telephone and offered them the followingopportunity to participate in a test marketing study:

Hi, __ and 1 work for a mar-keting research firm here in the Twin Cities. Presently,we are surveying college students to see if, in the future,


they would be interested in trying out a new shampoothat our client plans to market. However, before you tellme if you would be interested in trying this shampoo,let me tell you a little about it.

At this point, the experimenter delivered one of the fol-lowing messages, determined by random assignment.

Image message: The results of recent laboratory testshave indicated that while compared to other shampoos,this brand usually rates about average in how it cleansyour hair, it consistently rates above average in how goodit makes your hair look.

Quality message: The results of recent laboratory testshave indicated that while compared to other shampoos,this brand usually rates about average in how good itmakes your hair look, it consistently rates above averagein how clean it gets your hair.

After delivering one of the two messages, the experi-menter asked air of the participants the following two ques-tions:

Now, , what I would like you to do is to indicatefor me which of the following verbal descriptions bestdescribes your willingness to use this shampoo: definitelynot, probably not, unsure, probably yes, or definitelyyes. Also, with 0% indicating not at all and 100% indi-cating definitely willing, what percentage do you thinkbest describes your willingness to try this shampoo?

The experimenter recorded the participants' responses,

answered any questions they had, and then thanked themfor their time.


To create an index of willingness to use theshampoo, we coded the participants' answersto the first dependent variable in the followingmanner: 1 point for answering definitely not,2 points for probably not, 3 points for unsure,4 points for probably yes, and 5 points for def-initely yes. Because the participants' responsesso scored were highly correlated with theirpercentage estimate scores on the second de-pendent variable (r= .865,p< .01), wedecidedto form a composite index of willingness touse the shampoo by summing the participants'responses over the two measures. To accom-plish this, we first standardized scores on eachmeasure and then used the sum of each par-ticipant's standardized scores as an indica-tion of willingness to try the shampoo prod-uct. Means for this composite standardized"willingness to use" index are presented inTable 3.

We then entered the participants' scores onthe composite index into a 2 X 2 (Self-Moni-

Table 3Willingness to Use the Product: Study 3

Type of message receivedSelf-monitoring

category Image Quality

HighLow

.7626

-.9106

-.7008

.9201

Note, Higher scores indicate greater willingness to use theproduct.

toring X Message) ANOVA. This analysis re-vealed a statistically reliable interaction, F(l,36) = 7.18, p < .05. Evidently, high self-mon-itoring individuals were more willing than lowself-monitoring individuals to try the shampooif they believed that it would leave their hairlooking good, ;(36) = 2.06, p < .025, and lowself-monitoring individuals were more willingthan high self-monitoring individuals to try theshampoo if they believed that it would leavetheir hair very clean, /(36) = 1.99, p < .05.

Clearly, then, when it came to actual choicesto use or not to use this consumer product,high self-monitoring individuals invoked con-siderations of the images associated with aproduct (choosing, in this case, to use theshampoo that would make their hair lookgood, even if it meant that their great-lookinghair would be less than perfectly clean) andlow self-monitoring individuals responded toattributes of the product's quality in perform-ing its defining function (choosing, in this case,to use the shampoo that would get their hairvery clean, even if their very clean hair wouldhave a less-than-beautiful look).

General Discussion

On the basis of the converging pattern ofresults of these three investigations, it seemsthat we have succeeded in identifying two typesof advertising strategies (image and quality)and two types of individuals (those high andthose low in self-monitoring) who react differ-entially to these two strategies. High self-mon-itoring individuals react favorably to image-oriented ads, they are willing to pay moremoney for products if they are advertised withan image orientation, and they will agree to


try a product if it is marketed with an imageappeal. By contrast, low self-monitoring in-dividuals react favorably to product-quality-oriented ads, they are willing to pay moremoney for products if their advertisementsstress product quality, and they will agree totry a product if an appeal is made to its quality.

How are we to understand the effectivenessof appeals to images and claims about productquality? One basis for such an understandingis provided by the generalized interpersonalorientations characteristic of high and low self-monitoring individuals. In other words, thepropensity to respond favorably, at the level ofevaluative reactions and at the level of behav-ioral choices, to image-oriented or quality-ori-ented advertising appeals may be one featureof the larger syndrome of differing cognitive,motivational, and behavioral processes thatguide the behavior of high and low self-mon-itoring individuals in social contexts (for a re-view of the defining and characteristic featuresof these generalized interpersonal orientations,see Snyder, 1979). Responding favorably toimage-based advertising may be yet anothermanifestation of a striving to be a pragmaticcreature of one's situations, to project imagesappropriate to one's circumstances, a strivingthat is particularly pronounced in high self-monitoring individuals; thus these individualsare ideal candidates for the study of the roleof appeals to images in the advertising process.Similarly, responding favorably to quality-ori-ented advertising may be yet another manifes-tation of a quest to be a principled being, onewith congruence between one's actions andunderlying attitudes, values, and dispositions,a quest that is especially evident in low self-monitoring individuals; thus these individualsare ideal candidates for the study of the roleof claims about quality in the advertising pro-cess.

Before considering further the implicationsof our findings for understanding advertisingas a process of persuasion and social influence,we examine more carefully the distinction be-tween image-oriented and quality-orientedadvertising strategies. As clearly denned as thedistinction may be at the level of the operationsused in these investigations, some further de-cision rules may be necessary in order to iden-tify the general features that define and sepa-

rate the two categories of advertisements. Forexample, is an advertisement whose messageconcerns "people of good taste" or one whosemessage concerns "quality-conscious individ-uals" to be classified as an appeal to image ora claim about quality? From our perspective,if the message appeals to the image of havinggood taste or to the image of quality con-sciousness that one acquires by virtue of usingthe product, but at the same time the messagedoes not claim that the product tastes good orthat it has high quality, then the advertisementis an image-oriented one. As such, it ought toappeal preferentially to high self-monitoringindividuals. By contrast, if the message actuallymakes claims about the good taste or the highquality of the product as expressions of one'sintrinsic desires for things that taste good orone's inherent valuing of things high in quality,then the advertisement is a quality-orientedone. As such, its claims ought to be particularlyeffective with low self-monitoring individuals.

Furthermore, it is necessary to specify theconditions under which the differential reac-tions of individuals high and low in self-mon-itoring to image-oriented and quality-orientedadvertising can be expected to be translatedinto actual consumer behaviors. Can high self-monitoring individuals be expected to buy ev-ery product that promises an image? And canlow self-monitoring individuals be expected tobuy every product that possesses high quality?Clearly not. How, then, can one specify whichproducts advertised with appeals to images willactually be used by high self-monitoring in-dividuals and which products advertised withclaims about quality will actually be used bylow self-monitoring individuals?

High self-monitoring individuals typicallychoose the images that they project on the basisof information about the appropriateness ofparticular images to specific social situationsand interpersonal contexts. Such situationalconsiderations also may be relevant in theirresponsiveness to information provided by ad-vertising. Thus the high self-monitoring malesmoker who seeks to project an image of rug-ged masculinity on his weekend trips with hishunting buddies to the hunting lodge may, inaddition to choosing words, deeds, and ex-pressive mannerisms appropriate to that im-age, also choose, at least while on hunting trips,

594 MARK. SNYDER AND KENNETH G. DEBONO

to smoke the cigarettes (and perhaps promi-nently display the package) whose advertisingpromises just such an image of rugged mas-culinity. As well, this individual may choosenot to smoke, at least in these same circum-stances, the brand whose ads appeal to an im-age of cultural sophistication (even though, ifasked, he might grant its ads highly favorableevaluative reactions). More generally, althoughhigh self-monitoring individuals may be ex-pected to be favorable toward advertising thatappeals to images, their actual choices amongproducts so advertised may reflect the imagesthat they actually choose to project in theirsocial circumstances.

In the case of low self-monitoring individ-uals, just as their behavior in specific socialcontexts is thought typically to reflect their owngeneral attitudes and dispositions, so too oughtsuch considerations of their own personal at-tributes be relevant in their responsiveness toinformation provided by advertising. For ex-ample, the low self-monitoring individual whoenjoys the taste of gin but who does not enjoythe taste of bourbon ought to be more likelyto purchase the brand of gin whose advertisingconveys information about its fine gin tastethan the brand of bourbon whose advertisingspeaks of its fine bourbon taste, even thoughthat individual ought to be expected to regardboth advertisements as particularly appealing,engaging, and effective ones. More generally,low self-monitoring individuals may be ex-pected to be generally favorable toward adver-tising that makes claims about product-qualityattributes, but to choose among products ad-vertised with this orientation on the basis ofthe fit between these products and their ownattitudes, preferences, and dispositions. Theywould make these choices in ways that maxi-mize the consistency between these personalattributes and their consumer behavior.

Not only does the self-monitoring concep-tual framework provide a theoretical under-standing of the effectiveness of advertising, butit provides some practical hints as well. Onehint is that the most potent image-orientedadvertisements are those that most effectivelyconvey the message that the images associatedwith using the advertised products are partic-ularly appropriate ones to project and are im-ages that, if conveyed, will increase the likeli-

hood that one will fit into important life sit-uations. Another hint is that the most effectivequality-oriented advertisements are the onesthat most successfully include the message thatone will, by using the advertised products, begaining opportunities to be true to one's ownpersonal attitudes and important values.

Our findings and our interpretation of themalso provide a new perspective on the twolongstanding traditions in the advertising in-dustry: the image-oriented soft-sell approachand the claim-oriented hard-sell approach. In-deed, it very well may be that these approacheshave succeeded, survived, and flourished be-cause each one has managed to appeal tomembers of one category of consumers. Image-oriented advertising campaigns that haveworked may have worked because they havesucceeded in engaging and motivating the im-age concerns of high self-monitoring consu-mers. Claim-oriented advertising campaignsthat have succeeded may have succeeded be-cause they have been effective in engaging andmotivating the quality concerns of low self-monitoring consumers.

Perhaps, too, advertisements that featureelements of both strategies (information aboutimage and information about quality) maysucceed because each type of information suc-ceeds in drawing different sets of consumers;that is, the image elements of such hybrid adsmay draw high self-monitoring consumers atthe same time as their quality elements drawlow self-monitoring consumers. However, aword of caution ought to be associated withthis generalization from our findings: It is notyet known whether the relative preferences ofhigh and low self-monitoring individuals forimage-oriented and quality-oriented advertis-ing are accompanied by an actual aversion tothe other type of advertising. For example, arelow self-monitoring individuals, who preferquality-oriented messages, simply indifferentto image-oriented messages (so that a hybridad would appeal to them as much as a quality-only ad) or are they perhaps actively opposedto image-oriented messages (so that a hybridad might appeal to them less than a quality-only ad)? Answering this question would re-quire comparisons of image and quality adswith "no-message" ads (ones that lack bothappeals to images and claims about quality),


although it is not clear, in the context of thedesign of our investigations, what operationswould constitute such a "no-message" com-parison condition.

Our concerns with appeals to image andclaims about quality are not limited to the do-main of advertising. Indeed, they ought to beconsidered in the context of several distinctionsthat have been made concerning the processesof persuasion. For one, Petty and Cacioppo(1981) introduced the distinction between thecentral route to persuasion, whereby persua-sion is due to extensive thinking about andelaboration of the persuasive message, and theperipheral route to persuasion, whereby per-suasion is mediated by nonmessage factorssuch as source attractiveness. In addition,Chaiken (1980) distinguished between the sys-tematic processing of a persuasive message,whereby, as Petty and Cacioppo's central route,persuasion is dependent on extensive thinkingabout the content of the persuasive message,and the heuristic processing of a persuasivemessage, whereby agreement with the messageis based on the use of simple decision rules(e.g., "Experts are usually correct").

Although the conceptual distinctions theseinvestigators make may appear to be similarto the conceptual distinction we are advancing(e.g., one may be tempted to equate image-based appeals with the peripheral route to per-suasion or to link them to heuristic processing),we believe that the research strategies used bythese investigators and our research strategyare sufficiently dissimilar to render the con-ceptual similarity more apparent than real. Inmost persuasion research in which either cen-tral versus peripheral routes or systematic ver-sus heuristic processing is examined, the con-tent of the persuasive message is held constantand another variable (e.g., a source factor orsubject's level of involvement) is manipulated(e.g., Petty, Cacioppo, & Goldman, 1981). Bycontrast, in our research, the only differencebetween the two advertisements for any givenproduct was the message content; no periph-eral cues were manipulated. Therefore, anydifferences that the image and product-qualityadvertisements engendered must have beendue to reactions to the message content itself.From this perspective, responsiveness to image-oriented and quality-oriented messages may

both represent instances of central routes topersuasion.

Whether the processing of the message con-tent itself was systematic or heuristic, we can-not say. One could engage in either systematicor heuristic processing of either type of per-suasive message. Thus, for example, in pro-cessing the quality-oriented messages of ourfirst two studies, one could invoke the heuristic,"If it promises taste, buy it," or one could sys-tematically assess the fit between claims aboutthe product's taste and one's abiding prefer-ences. And, in processing the image-orientedmessages of those studies, one could use theheuristic, "If it promises image, buy it," or onecould systematically evaluate the strategicutility of using the product for particular im-age-fashioning purposes in specific situations.We suspect, in line with our previous discus-sion of when evaluative reactions to advertisingare translated into consumer behavior, that ul-timate behavioral responsiveness to advertisingis the outcome of systematic processing, bothfor low self-monitoring individuals (who, wehave suggested, assess the fit between productsand their own attitudes, preferences, and dis-positions) and for high self-monitoring indi-viduals (who, we have suggested, evaluate thepotential of products to aid them in creatingthe images they choose to project in social cir-cumstances).

There is, in addition to the distinctions be-tween central and peripheral routes to persua-sion and between systematic and heuristicprocessing, another distinction that may berelevant to understanding responsiveness tomessages that make appeals to image and thosethat make claims about quality. Kelman (1961)posited three processes of agreement with acommunication: compliance, identification,and internalization. For our purposes, the mostmeaningful of this trio are the processes ofidentification and internalization. Identifica-tion is said to occur when an individual agreeswith a message because that agreement allowshim or her to project a desirable image and tofit into important situations and peer groups.This agreement is considered somewhatephemeral in that it can easily change whenthe image projected is no longer desirable. Assuch, identification processes may be involvedin the preferential responsiveness of high self-


monitoring individuals to image-based ap-peals. By contrast, internalization is said tooccur when an individual accepts a messagebecause the message is congruent with the in-dividual's personal value system. Accordingly,internalization processes may be implicated inthe preferential impact of quality-based ad-vertisements on low self-monitoring individ-uals.

Nevertheless, the applicability of Kelman's(1961) theory to our particular investigationsmay be limited by the fact that his theory isconcerned primarily with reactions to attri-butes of the source of a persuasive message(i.e., expert sources tend to produce agreementby means of internalization, attractive sourcestend to produce agreement by means of iden-tification). Yet in our investigations we focusednot on the source of the message but rather onthe message itself. Nevertheless, it may be thatmessages that are linked to sources of highcredibility are particularly effective with lowself-monitoring individuals and that the verysame messages, when linked to sources of highpersonal or social attractiveness, are particu-larly effective with high self-monitoring indi-viduals.

In addition to having implications for un-derstanding advertising strategies, our findingsalso may have implications for understandingthe nature of attitudes themselves. Taken to-gether, our studies provide evidence, albeit in-direct, that attitudes may be serving differentfunctions for high and low self-monitoring in-dividuals. In our studies, high self-monitoringindividuals seem to have formed more favor-able attitudes toward objects that they poten-tially could use for the purpose of creating anddisplaying social images. In keeping with theirpragmatic orientation toward their socialworlds, high self-monitoring individuals re-acted positively to objects that may have beenuseful to them as means for achieving the goalsof presenting images appropriate to their socialsituations. By contrast, and in keeping withtheir principled orientation toward their socialenvirons, low self-monitoring individuals inour studies reacted positively to objects andissues that allowed them the opportunity toexpress their underlying attitudes and values.

This possibility—that the reactions of in-dividuals high and low in self-monitoring to

advertising may reflect differing functionalbases of their attitudes toward consumer prod-ucts—is an intriguing one. The notions thatindividuals may hold similar attitudes for dif-ferent reasons, and that the same attitude mayserve different functions for different individ-uals, are the fundamental tenets of the func-tional theories of attitudes (e.g., Katz, 1960;Smith, Bruner, & White, 1956). Although itmay not be not possible, on the sole basis ofthe results of this series of investigations, tospecify conclusively which of the particularfunctions proposed by these theories underlieeither the specific domain of reactions to ad-vertising or the general domain of social atti-tudes, theoretical analyses of self-monitoring(e.g., Snyder, 1979) do provide some basis forconjecture.

To the extent that the characteristic inter-personal orientation of high self-monitoringindividuals is a pragmatic one of fitting them-selves to their social circumstances, this char-acteristic interpersonal orientation may alsoinclude social attitudes that are formed on thebasis of how well they serve the ends of be-having in ways appropriate to the various ref-erence groups that form one's social circum-stances. As such, the social attitudes of highself-monitoring individuals may be said, in thelanguage of the functional theorists, to beserving a social adjustive function. By contrast,to the extent that the characteristic interper-sonal orientation of low self-monitoring indi-viduals is a principled one of choosing behav-iors that accurately reflect and meaningfullycommunicate their own personal attributes,that interpersonal orientation may also includesocial attitudes formed on the basis of how wellthey reflect and communicate more funda-mental underlying values. As such, the socialattitudes of low self-monitoring individuals,once again in the language of the functionaltheorists, may be said to be serving a value-expressive function.

These conjectures about the links betweenthe psychological construct of self-monitoringand the functional bases of social attitudes mayhold the potential for dealing with a problemthat has long plagued the functional theorists.The Achilles' heel of the functional theorieshas been the lack of a way to systematicallytest these notions (cf. Kiesler, Collins, & Miller,


1969). To assess the validity xif a functionaltheory of attitudes, one must be able to identifya priori the functions being served by a givenattitude for a given individual. In the past, thishas proven very difficult to do and may, inpart, explain why the functional theories havebeen in a state of hibernation for some twodecades now.

However, the results of this series of studiessuggest one way of overcoming this hurdle. Justas we were able to identify categories of indi-viduals who were especially responsive to im-age-oriented and quality-oriented advertisingmessages, so too may it be possible to identifycategories of individuals who are especiallylikely to hold attitudes serving particular typesof functions. Thus, for example, if one wereseeking candidates for the study of the socialadjustive function of attitudes, one might seekout members of the category of high self-mon-itoring individuals. Similarly, if one wished toprobe the dynamics of the value-expressivefunction of attitudes, one might focus one'sefforts on members of the category of low self-monitoring individuals. In either case, theidentification and investigation of these cate-gories of individuals would serve not as endsunto themselves but as vehicles for under-standing the functional underpinnings of at-titudes. The application of this hybrid "per-sonality-social psychology" strategy to thestudy of attitudes very well may have the po-tential to awaken the functional theories of at-titudes, persuasion, and social influence fromtheir long winter of hibernation.

References

Chaiken, S. (1980). Heuristic versus systematic informationprocessing and the use of source versus message cues in

persuasion. Journal of Personality and Social Psychology,39, 752-766.

Danheiser, P. R., & Graziano, W. G. (1982). Self-moni-toring and cooperation as a self-presentational strategy.Journal of Personality and Social Psychology, 42, 497-505.

Fox, S. (1984). The mirror makers. New York: Morrow.Hacker, A. (1984, June 24). Poets of packaging, sculptors

of desire. New York Times Book Review, p. 1.

Katz, D. (1960). The functional approach to the study ofattitudes. Public Opinion Quarterly, 24, 163-204.

Kelman, H. C. (1961). Processes of opinion change. Public

Opinion Quarterly, 25, 57-78.Keppcl, G. (1982). Design and analysis: A researcher's

handbook (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall.Kiesler, C. A., Collins, B. E., & Miller, N. (1969). Altitude

change: A critical analvsis of theoretical approaches. NewYork: Wiley.

Lippa, R. (1976). Expressive control and the leakage ofdispositional introversion-extraversion during role-played teaching. Journal of Personality, 44, 541-559.

Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and per-

suasion: Classic and contemporary approaches. Du-

buque, IA: Wm. C. Brown.Petty, R. E., Cacioppo, J. X, & Goldman, R. (1981). Per-

sonal involvement as a determinant of argument-basedpersuasion. Journal of Personality and Social Psychology,

41, 847-855.Potter, D. M. (1954). People of plenty: Economic abundance

and the American character. Chicago: University of Chi-

cago Press.Shaffer, D. R., Smith, J. R., & Tomarelli, M. (1982). Self-

monitoring as a determinant of self-disclosure reciprocityduring the acquaintance process. Journal of Personality

and Social Psychology, 43, 163-175.Smith, M. B., Bruner, J. S., & White, R. W. (1956). Opin-

ions and personality. New York: Wiley.Snyder, M. (1974). The self-monitoring of expressive be-

havior. Journal of Personality and Social Psychology, 30,

526-537.Snyder, M. (1979). Self-monitoring processes. In L. Ber-

kowitz (Ed.), Advances in experimental social psychology

(Vol. 12, pp. 85-128). New York: Academic Press.Snyder, M., Beracheid, E., & Click, P. (1985). Focusing on

the exterior and the interior: Two investigations of theinitiation of personal relationships. Journal of Personality

and Social Psychology, 48, 1427-1439.Snyder, M., & Ickes, W. (1985). Personality and social be-

havior. In G. Lindzey & E. Aronson (Eds.), Handbook

of social psychology (3rd ed., pp. 883-948). New York:

Random House.Snyder, M., & Monson, T. C. (1975). Persons, situations,

and the control of social behavior. Journal of Personality

and Social Psychology, 32, 617-644.Snyder, M., & Swann, W. B., Jr. (1976). When actions reflect

attitudes: The politics of impression management. Jour-

nal of Personality and Social Psychology, 34, 1034-1042.Snyder, M., & Tanke, E. D. (1976). Behavior and attitude:

Some people are more consistent than others. Journal

of Personality, 44, 510-517.Zanna, M. P., Olson, J. M., & Fazio, R. H. (1980). Attitude-

behavior consistency: An individual difference approach.Journal of Personality and Social Psychology, 38, 432-

440.

Received September 25, 1984

Revision received March 25, 1985 •

Documents

Appeals to Image and Claims About Quality: Understanding the