The Match-Up Hypothesis: Physical Attractiveness, Expertise, and the Role of Fit on Brand Attitude, Purchase Intent and

Brand Beliefs.(Industry Overview)

Journal of Advertising | September 22, 2000 | Till, Brian D.; Busler, Michael | Copyright

Two studies are presented that examine the role of attractiveness and expertise in the "match-up hypothesis." Much "match-up hypothesis" research has focused on physical attractiveness. Study One examined physical attractiveness as a match-up factor and its impact on brand attitude, purchase intent and key brand beliefs. In a 2 X 2 experiment, endorser attractiveness and product type are manipulated. Results indicated a general "attractiveness effect" on brand attitude and purchase intent but not the match-up predicted in previous literature. Study Two considered expertise as the match-up dimension. The second experiment manipulated product and endorser type. A match-up effect was found as the athlete was most effective as an endorser for the energy bar in increasing brand attitude, but not purchase intent. The variable "fit" or "belongingness," was shown to play an important role in match-up effects. Examining the effectiveness of endorsers is important for practitioners and academics (Kaikati 1987). Over the years there have been a number of studies that have examined under what conditions endorsers, oftentimes celebrities, are appropriate for products (e.g., Agrawal and Kamakura 1995; Atkin and Block 1983; Freiden 1984; Kamins 1989; Kamins, Brand, Hoeke and Moe 1989; Ohanian 1991; Tripp, Jensen and Carlson 1994). A variety of mechanisms such as the Elaboration Likelihood Model (Petty, Cacioppo and Schumann 1983), associative learning (Till and Shimp 1998), social adaptation theory (Kahie and Homer 1985), schema theory (Lynch and Schuler 1994), attribution theory (Mowen and Brown 1981; Tripp, Jensen and Carison 1994), culturalmeaning transfer (McCracken 1989) and source-characteristics (i.e. attractiveness, expertise, trustworthiness, etc.) (Ohanian 1991) have been invoked to explain the effectiveness (or ineffectiveness) of celebrity endorsers. The "match-up hypothesis" (e.g., Kamins 1990) suggests that endorsers are more effective when there is a "fit" between the endorser and the endorsed product. Most of the empirical work on the match-up hypothesis has focused on the physical attractiveness of the endorser. The conceptual argument is that attractive celebrities are more effective endorsers for products which are used to enhance one's attractiveness (lead to higher brand attitude and purchase intentions). Empirical support for this attractiveness match-up, as will be discussed in the following sections, has been elusive.

The purpose of this research is to examine the effectiveness of physical attractiveness and expertise as relevant match-up factors and the role of endorser/product fit in these effects. Physical attractiveness has been traditionally offered as an appropriate match-up factor (Kahle and Homer 1985; Kamins 1990). Expertise has also been considered relevant for understanding celebrity endorser effects (Ohanian 1991) and is examined for its role in potentially driving match-up effects. We present two studies which are very similar in design and method. Study One investigates the effectiveness of physical attractiveness as a match-up factor, while Study Two focuses on the role of expertise as a match-up factor. In each of these studies, key dependent variables include brand attitude, purchase intent, endorser fit with the product and several cognitive belief measures. Physical Attractiveness and Communicator Effectiveness Casual observation suggests that marketing and advertising practitioners believe that using physically attractive spokespeople is effective. The behavior of advertisers is consistent with a review article (Eagly, et al. 1991) in which the researchers conducted a meta-analysis of 76 studies and found that, in general, physically attractive people are viewed more favorably on a variety of personality traits such as social competence, intellectual competence, concern for others and integrity. Given that attractive people are imbued with a host of other positive traits, it is natural that advertisers would wish to associate their products with attractive individuals. Research in the advertising and marketing literature has been equivocal, however, with respect to the impact that physically attractive spokespeople have on ad and product evaluations. Baker and Churchill (1977) found that using attractive models had a positive effect on affective evaluations of the ad (but not on purchase intentions). Both Petroshius and Crocker's (1989) and Patzer's (1983) studies found that physically attractive models used in advertising led to more favorable attitudes toward the ad and stronger purchase intentions. Two studies, though, were not able to detect an effect of model attractiveness. Caballero, Lumpkin and Madden (1989) showed grocery shoppers videotapes of less/moderate/highly attractive models in ads for a soft drink and cheese, but did not find that the attractiveness of the model affected purchase intent. Caballero and Solomon (1984) used less/moderate/highly attractive models for instore displays for beer and facial tissues. They found no effect of attractiveness on beer sales yet, strangely, found that the less attractive model actually increased sales for facial tissue. The Match-up Hypothesis The genesis of the match-up hypothesis lies in advertising research that examined the differential impact that different types of endorsers, often celebrities, have on the endorsed brand. Early research found that the effectiveness of celebrities varies by product (Friedman and Friedman 1979). A study by Kanungo and Pang (1973) paired male and female models (non-celebrities) with different types of products and found that the effect of the models varied depending on the product

with which the models were paired. Kanungo and Pang (1973) explained their findings in terms of the "fittingness" of the model for the product. The fit between the celebrity and the product was captured by the term "match-up hypothesis" (Kahle and Homer 1985; Kamins 1990; Lynch and Schuler 1994; Solomon, Ashmore and Longo 1992). In several significant match-up research studies, the focus has been on a match between a celebrity and a product based on physical attractiveness (Kahle and Homer 1985; Kamins 1990). Specifically, the match-up hypothesis predicts that attractive celebrities are more effective when endorsing products used to enhance one's attractiveness. Invoking "social adaptation theory," Kahle and Homer (1985) argue for a match-up explanation of their findings, showing more favorable brand attitude when the product (razor blades) was paired with an attractive (rather than an unattractive) celebrity endorser. Kamins, in a 1990 study, tested the match-up hypothesis by pairing either an attractive (Tom Selleck) or unattractive (Telly Savalas) celebrity with either a product used to enhance one's attractiveness (luxury car) or a product not used to enhance one's attractiveness (home computer). The conclusions supporting the match-up hypothesis drawn by Kahle and Homer (1985) and Kamins (1990) bear closer scrutiny. While Kahle and Homer (1985) did find that the use of an attractive celebrity led to a more favorable attitude toward the endorsed brand, their test of the match-up hypothesis was incomplete in that they did not demonstrate that an attractive celebrity is less effective when endorsing a product not used to enhance one's attractiveness. Since Kahle and Homer (1985) tested only one product, razor blades, which they argue are used to enhance one's attractiveness, they had no data that shows the effect of attractive or unattractive celebrities on products not used to enhance one's attractiveness. The Kamins (1990) study was a full test of the match-up hypothesis--attractive celebrity paired with a product used to enhance one's attractiveness (Tom Selleck/luxury car), an unattractive celebrity paired with a product used to enhance one's attractiveness (Telly Savalas/luxury car), an attractive celebrity paired with a product not used to enhance one's attractiveness (Tom Selleck/home computer), and an unattractive celebrity paired with a product not used to enhance one's attractiveness (Telly Savalas/home computer). Subjects viewed a print ad representing one of those four conditions. Kamins (1990) collected data on seven dependent measures (advertiser believability, advertiser credibility, spokesperson believability, spokesperson credibility, brand attitude, attitude toward the ad, and purchase intention). The match-up hypothesis predicts a celebrity attractiveness by product type interaction. In the Kamins (1990) study, for example, the prediction is that Tom Selleck would be a more effective endorser for a luxury car than for the personal computer. The predicted interaction was found for only 2 of the 7 dependent measures and, it is important to note, not found for brand attitude nor for purchase intentions.

While early research (e.g., Friedman and Friedman 1979; Kanungo and Pang 1973) recognized the situational effectiveness of endorsers, two previously mentioned studies (Kahle and Homer 1985; Kamins 1990) specifically examining the match-up hypothesis and the role of physical attractiveness still left several issues open. The design of the Kahle and Homer (1985) study did not fully address the role of endorser attractiveness as it relates to products not used to enhance one's attractiveness, and the Kamins (1990) study did not find the predicted "match-up" interaction for brand attitude and purchase intentions. The purpose of the following studies is to consider the appropriateness of attractiveness and expertise as match-up factors, and to examine the role of a construct from associative learning, "fit" or "belongingness," in match-up effects. Associative Learning and Match-up Effects Associative learning theory is a useful framework for understanding match-up effects. Associative learning is a mechanism by which links or relationships between concepts can be established (Klein 1991; Martindale 1991). An associative network structure of memory consists of patterns of nodes (concepts) linked together (Anderson 1976, 1983). In the context of endorsements, both the endorser and brand represent units connected to other units based on our experiences with the brand and the endorser. These connections represent our association set for the brand and the endorser. Even attitude, our evaluation of the brand and the celebrity, is considered an element of our association sets (Berger and Mitchell 1989; Fazio, Powell and Williams 1989; Fazio et al. 1986; Judd et al. 1991; Noffsinger, Pellegrini and Brunell 1983). Indeed, Fazio et al. (1989) define attitude as the "association" between an object and the evaluation of the object. By repeated pairings of the brand and endorser, the brand and endorser be come part of each other's association set. An important variable in driving how easily an associative link is built between two concepts (such as a brand and an endorser) is belongingness, relatedness, fit, or similarity. Generally, the more similar two concepts are, the more likely the two concepts will become integrated within an associative network (e.g., Garcia and Koelling 1966; Hamm, Vaitl and Lang 1989; Rozin and Kalat 1971). It is this associative link, a perceived connection between a brand and an endorser, that drives predicted endorser effects. This theoretical perspective is consistent with hypothesizing by other researchers interested in match-up effects who propose the importance of "congruence" (Kamins 1990; Lynch and Schuler 1994), "fittingness" (Kanungo and Pang 1973), "appropriateness" (Solomon, Ashmore and Longo 1992), and "consistency" (Walker, Langmeyer and Langmeyer 1992). It is our contention that other aspects/characteristics of the endorser (such as expertise) may be more potent (provide an opportunity for greater fit) than physical attractiveness. This perspective is consistent with more recent work on the match-up hypothesis. Lynch and Schuler (1994) demonstrated that the muscularity of an individual affects perceived expertise with respect to exercise equipment. Lynch and Schuler (1994) did not find, though, that use of a muscular (vs. non-

muscular) spokesperson had a greater effect on cognitions for a product designed to enhance one's build (weight-lifting equipment). Although not a test of the match-up hypothesis, Ohanian (1991) found that of celebrity endorser source characteristics (physical attractiveness, trustworthiness and expertise), expertise was most closely associated with intent to purchase the endorsed product. Based on the above discussion, we conducted two studies. The first study examines the extent to which physical attractiveness serves as a match-up factor and its impact on brand attitude and purchase intentions. Study One follows closely in the tradition of several previous studies (e.g., Kahle and Homer 1985; Kamins 1990). Study Two considers the effectiveness of expertise as a match-up factor. Study One Much of the rationale for our hypotheses has been developed in the previous sections. We conducted two studies. The first study examined the role of physical attractiveness in affecting attitude toward the endorsed brand. Based on previous work on physically attractive spokespersons (both celebrity and non-celebrity) such as that of Baker and Churchill (1977), Kahle and Homer (1985), Kamins (1990), Patzer (1983), Petroshius and Crocker (1989), we expect that an attractive celebrity endorser will have a positive effect on brand attitude and purchase intentions: H1a: Attitude toward the brand will be more favorable when endorsed by an attractive spokesperson versus an unattractive spokesperson. H1b: Purchase intentions for brand will be greater when endorsed by an attractive spokesperson versus an unattractive spokesperson. Previous research on the match-up hypothesis (e.g., Kahle and Homer 1985; Kamins 1990) has examined the role of attractiveness as a match-up factor. Although empirical work in this area has not provided clear, consistent support for a match-up based on physical attractiveness, we also examine the extent to which attractiveness serves as an effective match-up dimension. As elaborated earlier, to the extent that attractiveness belongs with, fits with, or, in some other way is congruent with the endorsed product, we expect that: H2a: There will be an interaction between the endorser's attractiveness and the endorsed product. Specifically, the use of an attractive endorser with a product used to enhance one's attractiveness will have a greater positive effect on brand attitude than will the use of an attractive endorser with a product not used to enhance one's attractiveness. The use of an unattractive endorser with a product used to enhance one's attractiveness will have a greater negative effect on brand attitude than will the use of an unattractive endorser with a product not used to enhance one's attractiveness. H2b: There will be an interaction between the endorser's attractiveness and the endorsed product. Specifically, the use of an attractive endorser with a product used to enhance one's attractiveness will have a greater positive effect on purchase intent than will the use of an attractive endorser with a

product not used to enhance one's attractiveness. The use of an unattractive endorser with a product used to enhance one's attractiveness will have a greater negative effect on brand attitude than will the use of an unattractive endorser with a product not used to enhance one's attractiveness. Study One was a 2 X 2 design matching an attractive (or unattractive) endorser with a product used to enhance (or not enhance) one's attractiveness. Subjects were randomly assigned to one of the four conditions. Subjects read some background information on the endorser and then viewed a print ad featuring the endorser and the product. After viewing the print ad, subjects evaluated both the product and the endorser on assorted measures. Choice of Brand and Endorser For Study One we chose two products: a pen (pre-testing showed that a pen is a product not used to enhance one's attractiveness) and men's cologne (pre-testing showed that cologne is a product used to enhance one's attractiveness). In order to minimize brand familiarity effects, a fictitious brand name was used. The name for both products was Aerius 2. For the endorser, we created Phil Johnson, described as a member of the U.S. Olympic Water Polo team. The use of a fictitious endorser offers a key advantage with respect to experimental control. With the use of familiar endorsers, such as well-known celebrities, there can be a significant amount of variation in subjects' knowledge and attitude toward that familiar individual. And, although these variations are randomly distributed across experimental conditions, high within-group variation reduces the power of the study. Additionally, the use of a fictitious endorser minimizes the opportunity for unintended spurious confounds. For example, if one were to choose two different well-known endorsers who vary with respect to physical attractiveness, it is virtually certain that the two endorsers would vary on other dimensions as well. Even demonstrating that the two familiar endorsers are equivalent on some small set of characteristics (such as expertise) does not rule out the likelihood that they do vary on som e unmeasured characteristic(s). In short, we chose to sacrifice some ecological validity (by using a fictitious rather than a real endorser), but gained construct validity by minimizing spurious confounds and statistical conclusion validity by minimizing within-group variation with our manipulation. For our study, a close-up computer-scanned photograph of an attractive, athletic man was used to represent Phil Johnson in the "attractive" endorser condition. The scanned image representing Phil Johnson was modified by a professional graphic artist to reduce his attractiveness (mustache added, eyebrows thickened, eyes dulled slightly, etc.) for use in the "unattractive" condition. Materials and Procedures Materials. The packet of materials for Study One consisted of an introductory page that informed participants that they were being asked to assist in a study sponsored by a manufacturer of pens (men's fragrances). Participants were told that they would be viewing an ad for a pen (men's

cologne) featuring U.S. Olympic Water Polo athlete Phil Johnson, and that the manufacturer would like to get some input from them regarding their feelings about the advertised pen (cologne). Following this introductory page, we provided background information on Phil Johnson consisting of three facts (graduate of UCLA, member of U.S. Olympic Water Polo team, teaches swimming to disadvantaged youths every summer). All conditions received the same information. The purpose of providing this information was to establish Phil Johnson as a legitimate spokesperson and to create a generally positive impression of Phil Johnson. Next the stimulus ad featuring either an attractive or unattractive Phil Johnson was paired with a picture of either the pen or the cologne. Each ad had a headline and brief line of copy. These simple executions were designed primarily to visually pair the endorser with the product. Several pages of questions for the participants regarding their impressions of the pen (cologne) and of endorser Phil Johnson followed the stimulus ad. Procedures. Study participants were undergraduate students at several east coast and midwestern universities. The study was conducted during class time. The experimenter handed out the materials to the participants, randomly distributing the four different versions of the materials (attractive Phil Johnson, pen; attractive Phil Johnson, cologne; unattractive Phil Johnson, pen; unattractive Phil Johnson, cologne). Subjects were adequately spaced to prevent them from becoming aware of the different versions of the print ad. Participants were given a brief introduction by the experimenter. Participants then read the introductory page and the background information on Phil Johnson. Viewing of the stimulus ad was timed at 15 seconds. Participants then filled out the accompanying questions/measures at their own pace. A total of 212 student subjects participated in Study One. Analysis and Results Measures. Subjects evaluated the brand (pen or cologne) on 3 nine-point semantic differential scales (positive anchors: strongly like, favorable, and positive). For analysis these three items were averaged (coefficient alpha=.92) to represent brand attitude. Purchase intent was measured using 3 nine-point semantic differential scales (unlikely/likely, definitely would not/definitely would, improbable/probable) in response to "How likely is it that you would consider purchasing Aerius 2?" These three items were averaged (coefficient alpha=.95). Evaluation of Phil Johnson was on 15 nine-point semantic/differential scales from Ohanian (1990) and reflect the dimensions attractiveness (coefficient alpha=.94), trustworthiness (coefficient alpha=.95), and expertise (coefficient alpha=.94). Subjects were also asked, "As an endorser for Aerius 2, I think Phil Johnson is:" with two nine-point semantic differential responses, "inappropriate/appropriate" and "ineffective/effective. These responses were averaged for analysis (r=.80). Finally, subjects were asked to respond to the question, "I think the combination of Aerius 2 and Phil Johnson:" with three

nine-point semantic differential responses, "does not belong with/belongs with," "does not go together/goes together," and "does not fit together/fits together." These re sponses were also averaged for analysis (coefficient alpha=.99). Pretesting indicated that important attributes for cologne are "smells good" and "long lasting." For pens, important attributes are "writes smoothly" and "comfortable to write with." Evaluation of each of these attributes was measured using a ninepoint "strongly disagree" to "strongly agree" question (i.e., "To what extent do you believe that Aerius 2 cologne smells good?"). Manipulation Checks. Although the visuals of Phil Johnson were pretested to ensure a strong attractiveness manipulation, subjects in the study also evaluated Phil Johnson's attractiveness. The attractiveness mean for the "attractive" Phil Johnson was 5.8 and for the "unattractive" Phil Johnson, 3.7 (t=8.9, p[less than or equal to].01). It should be noted that the attractiveness manipulation also had an effect on the "trustworthiness" dimension (See Table 1--attractive Phil Johnson was perceived as more trustworthy than unattractive Phil Johnson). Although this confounding of attractiveness with trustworthiness is not desirable, it is virtually unavoidable given the extensive body of literature showing that physically attractive individuals are ascribed a multitude of additional positive traits (Eagly et al. 1991). The attractiveness manipulation did not have a significant effect on subjects' perceptions of Phil Johnson's expertise. In addition to the manipulation check for attractiveness, subjects were asked to rate (on a nine-point scale) the extent to which cologne and pens enhance or do not enhance one's attractiveness. The average rating for cologne was 6.3. The average rating for pens was 3.3. This difference was significant (t=18.0; p[less than or equal to].01). Hypotheses Tests. Table 2 presents the means for subjects' attitude toward the brand and purchase intentions. Hypothesis 1 predicted a more favorable brand attitude in the attractive endorser condition. The mean brand attitude in the attractive endorser condition is 5.4. The mean brand attitude in the unattractive endorser condition is 4.4. Table 3 presents the ANOVA results and indicates a significant main effect of endorser attractiveness (F1,198=44.2, p[less than or equal to]01). Similar results were obtained for purchase intentions (attractive endorser PI mean of 4.7, unattractive PI mean of 3.8). The main effect of endorser attractiveness on purchase intent was significant (F1,189= 12.3, p[less than or equal to]01). The effect of the attractive endorser was also tested separately for each product. Considering cologne, both brand attitude and purchase intent were significantly higher in the "attractive endorser" than in the "unattractive endorser" condition (p[less than or equal to]01). For the pen, brand attitude was also significantly higher in the "attractive endorser" condition (p[less than or equal to].01). Purchase intent for the pen was higher as well in the "attractive endorser" condition (p[less than or equal to].05).

Hypotheses 2a and 2b predicted a "match up" effect based on physical attractiveness. Support for the match-up hypothesis would be indicated by a significant endorser by product interaction. Results in Table 3 show no endorser attractiveness by product type interaction and do not support the match-up hypothesis based on endorser physical attractiveness. The power of the test of the endorser by product interaction is estimated at approximately .70 ("medium" effect, p[less than or equal to].05, Rosenthal and Rosnow 1984). The subject pool was fairly evenly split between males (112) and females (99). Separate analyses by gender revealed no differences in the pattern of results (significant effect of model attractiveness, no model by product interaction) from the pooled analyses presented here. Given a concern over the inadvertent manipulation of trustworthiness along with physical attractiveness, we examined the results of Study One using trustworthiness as a covariate. The results of this analysis were, again, a main effect of endorser attractiveness on attitude toward the brand (F1,196=56.2, p[less than or equal to].01) and no attractiveness by product type interaction (F1,196=1.8, p=.28). Covariate analysis for purchase intentions revealed similar effects in that the attractiveness main effect was significant (F1,188=30.5, p[less than or equal to].01). The attractiveness by product type interaction was not significant (F1,188=0.7, p=.42). Regarding brand cognitive beliefs, Table 4 presents means and standard deviations for each of the two products by endorser attractiveness. For the cologne, there was a significant endorser attractiveness effect for both "smells good" (t=2.7, p[less than or equal to].01) and "long lasting" (t=2.1, p[less than or equal to].05). There was not an endorser attractiveness effect for the pen on either of the two attributes. These findings indicate that the change in attitude toward the pen was more likely to be primarily affect-driven while the change in attitude toward the cologne was more likely to be influenced by both affective and cognitive processes. The results of Study One clearly demonstrate an "attractiveness" effect for brand attitude consistent with previous research showing that the use of an attractive spokesperson has a positive effect on brand attitude and purchase intentions (and brand beliefs for the cologne). Study One did not find a "match-up" effect based on physical attractiveness. The putative reason for expecting a "match-up" effect is that an attractive endorser fits with or is more compatible with a product used to enhance one's attractiveness. Pretesting and our manipulation check indicated that we successfully crossed an attractive/unattractive endorser with a product used to enhance/not used to enhance one's attractiveness. To examine the issue of "fit" with product, we present the means for the appropriateness of Phil Johnson (attractive and unattractive) as an endorser for Aerius 2 (pen and cologne) and the degree to which subjects perceive a good fit between Phil Johnson (attractive and unattractive) and Aerius 2 (pen and cologn e) (Table 5). These results indicate an endorser attractiveness main effect for both endorser appropriateness (Fl,204=31.9, p[less than or equal

to].01) and endorser/product fit (F1,203=33.2, p[less than or equal to].01). There was not a significant product by endorser interaction for either appropriateness or fit. This data suggests that the attractive endorser was considered more appropriate and a better fit for both products. Study Two Study Two examined the role of expertise in enhancing the perceived fit between the product and the endorser. Although there are many different ways in which a brand and a celebrity may "fit," expertise was the focus of this study. There is generally a positive effect of an expert source on persuasion particularly under situations of low involvement and in combination with other positive cues (McGuire 1985). While the main effects of expertise are of practical significance, much research has been focused on the boundary conditions and reversals associated with the "expertise effect" such as issue involvement (Petty, Cacioppo and Goldman 1981), timing of source identification (Homer and Kahle 1990; Sternthal, Dholakia and Leavitt 1978), timing of behavioral requests (Dholakia and Sternthal 1977), role of message comprehension (Ratneshawar and Chaiken 1991), and individuals' self-monitoring behavior (DeBono and Harnish 1988). H3a: There will be an interaction between the endorser expertise and the endorsed product. Specifically, the use of a celebrity endorser with a product that is consistent with the endorser's expertise will have a greater positive effect on brand attitude versus the use of an endorser with a product not consistent with the endorser's expertise. This research considers the role of expertise in driving a celebrity "match-up" effect. Consistent with the match-up hypothesis, it is expected that a high fit, based on expertise, between the celebrity and the endorsed product will be of greater value to the endorsed brand. Specifically: H3b: There will be an interaction between the endorser expertise and the endorsed product. Specifically, the use of a celebrity endorser with a product that is consistent with the endorser's expertise will have a greater positive effect on purchase intention versus the use of an endorser with a product not consistent with the endorser's expertise. Study Two considered that there may be an alternative dimension on which endorsers and products "match-up" which may be more appropriate, or provide a better fit, for the endorsement. Study Two explored the use of expertise as a more useful dimension for driving the match-up effect (Ohanian 1991). Although endorser status can emerge from a variety of sources such as TV or film acting (e.g., Jerry Seinfeld, Mel Gibson), business success (e.g., Bill Gates, Lee Iaccoco), music (e.g., Bruce Springsteen, Whitney Houston), etc., one major source of fame is athletic prowess. Athletes have appeared as endorsers for a variety of brands including Coca-Cola (Mean Joe Greene, Michael Jordan), Nuprin (Jimmy Connors) and Pizza Hut (Deion Sanders). Several brands, such as Miller

Lite and Wheaties, as well as organizations such as the Milk Promotion Board, have utilized a number of athletes as part of their marketing programs. We expect that athletes "fit" better with, and, therefore, would be more effective as endorsers for, products related to athletic performance. We believe that this "fit" is closely related to expertise. Simply put, an athlete would be considered more of an expert on athletic products, and consumers would naturally see more of a fit between an athlete endorser and an athletic product than between an athlete and a non-athletic product. Choice of Brand and Endorser Two different product categories were used, candy bars and energy bars (energy bars, such as PowerBar, typically provide a slow release of energy and are consumed either in preparation for physical activity or upon completion of exercise). Both product categories are relatively similar in terms of product form and price. Both products were given the fictitious name Toba, designed to be equally appropriate (neutral in meaning) as a name for both a candy bar and an energy bar. The endorser used, as in Study One, was created specifically for this research. Background information presented Ted Franklin as either a "stage and screen actor" or a "U.S. Olympic Track and Field athlete." An attractive male was used to represent Ted Franklin (the same visual was used to represent Ted Franklin, actor, and Ted Franklin, athlete, to avoid confounding expertise with attractiveness). Materials and Procedures Materials. Four different packets of materials were created for this study, representing each of the four study conditions (athlete/energy bar, athlete/candy bar, actor/energy bar, actor/candy bar). Each packet consisted of a cover page informing participants that they are participating in a study sponsored by a manufacturer of energy (candy) bars and that the manufacturer plans on utilizing stage and screen actor (U.S. Olympic Track and Field athlete) Ted Franklin as an endorser. The second page presented background information on Ted Franklin consisting of three facts. Two of the facts (graduated from UCLA and supervises summer camp for disadvantaged youth) were the same across all conditions. The third fact varied depending on whether Ted Franklin was described as an actor (had leading role in several Broadway plays and two motion pictures) or as an athlete (member of U.S. Olympic Track and Field team). Procedures. Study Two procedures were identical to Study One. Study participants were undergraduate students at several east coast and midwestern universities. The study was run during class time. The experimenter handed out the materials to the participants, randomly distributing the four different versions of the materials (athlete/candy bar, athlete/ energy bar, actor/candy bar, actor/energy bar). Participants were given a brief introduction by the experimenter and read the introductory page and the background information on Ted Franklin. Viewing of the stimulus ad was

timed at 15 seconds. Participants then filled out the accompanying questions/ measures at their own pace. A total of 213 student subjects participated in Study Two. Analysis and Results Measures. As in Study One, subjects evaluated the brand (candy bar or energy bar) on the same 3 nine-point semantic differential scales (positive anchors: strongly like, favorable, and positive). For analysis these three items were summed (coefficient alpha=.96). The same three purchase intent items from Study One were also used (coefficient alpha=.96). Evaluation of Ted Franklin was on 15 nine-point semantic differential scales from Ohanian (1990) and reflected the three dimensions of attractiveness, trustworthiness, and expertise (all coefficient alphaz[less than or equal to].91). Brand cognitive belief measures were also included in Study Two. The same nine-point "strongly agree" to "strongly disagree" scales were used. The relevant attributes for the candy bar were "tastes good" and "chocolate flavor." For energy bars, the attributes were "tastes good" and "gives extra energy." Manipulation Check. It was expected that the actor/athlete manipulation would affect subjects' perceptions of the endorser's expertise. Specifically, an endorser (actor or athlete) by product (candy bar or energy bar) interaction was expected, such that the athlete/energy bar combination would result in the greatest rating on endorser expertise. Table 6 presents the expertise, attractiveness and trustworthiness means by condition. The expertise means are in the expected direction and the interaction is significant (F1,209=3.2, p[greater than or equal to].10). The manipulation of endorser type (expertise) had no effect on perceptions of endorser's trustworthiness or physical attractiveness. Hypothesis Test. Table 7 presents the means for subjects' attitude toward Toba and purchase intent by condition. Hypothesis 3 predicts a significant endorser by product type interaction. ANOVA results are given in Table 8. For attitude toward the brand, the pattern of Toba means are in the predicted direction, and the interaction is significant (F1,208=4.5, p[greater than or equal to].05). Although the purchase intent means are directionally consistent with match-up predictions, the interaction is not significant. For the candy bar, there was no significant difference in brand attitude based on the type of endorser (athlete or actor). However, for the energy bar, the brand attitude was significantly higher when the en dorser was an athlete than when the endorser was an actor (t=3.8, p[greater than or equal to].01). To examine the issue of "fit" between product and endorser as an explanation of the observed interaction, we present the means for the appropriateness of Ted Franklin as an endorser for Toba and the degree to which subjects perceive a good fit between Toba and Ted Franklin (Table 9). These results indicate a significant endorser by product type interaction for appropriateness

(F1,207=4.1, p[greater than or equal to].05) and fit (F1,207=3.5, p[greater than or equal to].10) such that athlete Ted Franklin was most appropriate and had the best fit for Toba energy bar. Table 10 presents means and standard deviations for cognitive belief measures by endorser condition and product type. Endorser type (actor or athlete) had no effect on the cognitive belief measures for the candy bar. For the energy bar, endorser type had no effect on the "tastes good" belief, but the athlete endorser led to a significantly greater belief that Toba energy bars "give extra energy" (t=2.5, p[greater than or equal to].05). Discussion These two studies point to expertise as possibly being more important than physical attractiveness for matching a brand with an appropriate endorser. While Study One found an attractiveness effect, it did not detect evidence of a product type by endorser attractiveness interaction as would be expected under the traditional view of the "match-up" hypothesis. Study Two found a match-up effect based on expertise, suggesting that expertise, rather than attractiveness, may be more appropriate for matching products with celebrities. Additionally, these studies point to the role of fit or belongingness in driving the match-up process. Further, we note that brand cognition may also play a role in this process. In this research, the use of an attractive endorser had an effect on brand beliefs for the cologne, but not for the pens. We also found that the use of an athlete endorser had a positive effect on the belief that the energy bar "gives extra energy." Despite the findings of these two studies, it may be premature to dismiss the role physical attractiveness might play in the match-up hypothesis. The difficulty in detecting a match-up effect based on physical attractiveness could be due to methodological idiosyncrasies of previous studies (including the studies presented here) that have inhibited detecting an effect. The subjects, manipulations, procedures, measures and/or stimuli of previous studies (again, including ours) were not sufficiently potent to detect an attractiveness match-up effect. The premise, in this case, is that attractiveness is a useful dimension on which to match celebrities and products, but studies to date have not been adequately designed to substantiate this. The value of associative learning as a framework for examining marketing and advertising phenomena has continued to gain currency (e.g., Grossman and Till 1998; Kim, Allen and Kardes 1996; Shimp 1991). While our studies were not focused on associative learning per se, the construct of "fit" or "belongingness" was borrowed from associative learning as a potentially useful mechanism in understanding match-up effects. Indeed, findings from these two studies do suggest that "fit" plays an important role in the match-up hypothesis. We speculate that "fit" (or lack thereof) may explain why physical attractiveness may not be, in many cases, the most powerful dimension for generating a match-up because the connection between "attractive celebrity" and "product used to enhance one's attractiveness" may not seem particularly

obvious. This interpretation is consistent with our failure to find a product type by endorser interaction in Study One to the extent to which the endorser and the product belonged together. In the Kamin's study (1990), for example, Tom Selleck (attractive celebrity) was paired with a luxury car. While a luxury car can be used to enhance one's attractiveness, it also serves many other purposes including basic transportation, pleasure, status, etc. Buying a luxury car to enhance one's attractiveness may not be exceptionally salient for consumers. Perhaps a physically attractive celebrity endorsing cosmetics provides a more obvious link for consumers, although one could also argue that the attractive celebrity is more "ex pert" in cosmetics (given their attractiveness) than an unattractive celebrity. More research in this area is needed. One concern with the endorser manipulation in Study One is the confounding of attractiveness with trustworthiness. Specifically, our intended manipulation of attractiveness, while successful, had the unintended effect of also influencing endorser trustworthiness. Previous research (e.g., Kamins 1990) attempted to avoid this problem by choosing different endorsers who varied in perceived attractiveness (e.g., Telly Savalas and Tom Selleck), yet were equivalently trustworthy. Although this approach would seem to solve the confounding problem, caution must be exercised in concluding that confounding has been eliminated. It is clear that Telly Savalas and Tom Selleck are not completely identical except for physical attractiveness. Therefore, while it appears that they are equivalently "trustworthy," it is almost certain that confounding occurred on some other, albeit unmeasured, dimension. Whether or not this unknown confounding affected brand evaluations would, at this juncture, be mere speculation. Including " trustworthiness" as a covariate in our analysis did not change the basic conclusions of Study One as there was still a main effect of physical attractiveness and no attractiveness by product type interaction. Future research could continue in several areas. Above, we alluded to the idea that physical attractiveness is not a strong match-up factor because the link from "attractive endorser" to "product used to enhance one's attractiveness" may not be readily apparent. Future research could more strongly operationalize the salience of "product used to enhance one's attractiveness." A study which manipulates attractiveness of endorser, product used to enhance one's attractiveness, and the salience of the "used to enhance one's attractiveness" dimension of the product could explore this issue. The salience of the "used to enhance one's attractiveness" dimension could be enhanced (or not) by having subjects read a carefully crafted article prior to viewing the stimulus ad or by selecting those products for which "enhancing one's attractiveness" is a highly salient feature of the product. There is also an opportunity to explore the role of involvement in match-up effects. Consider, for example, an attractive endorser paired with either a product used to enhance one's attractiveness or not used to enhance one's attractiveness. Under low involvement an attractive endorser may serve as a simple peripheral cue leading to a positive effect on attitude toward both product types (as was

found in our Study One). Under high involvement, subjects may more readily see the connection between "attractive endorser" and "product used to enhance one's attractiveness," leading to a match-up effect based on physical attractiveness that to date has been somewhat elusive. The focus of this research has been on the appropriateness of physical attractiveness and expertise as match-up factors. The findings suggest that expertise may be better match-up factor than physical attractiveness. There are, however, many examples of endorsers who seemingly have no particular expertise with the products they endorse. Such examples include Michael Jordan and his own cologne; Arnold Palmer and motor oil; Jack Nicklaus and turf products; Candice Bergen and long distance service, etc. Examples of such endorsements, even when considered successful, serve not to invalidate the perspective offered here. Jack Nicklaus may make a fine endorser for turf products. Expertise as a driver of the match-up hypothesis would suggest, however, that Jack Nicklaus would likely make a better endorser for golf clubs. Brian D. Till (Ph.D., University of South Carolina) is an Assistant Professor at Saint Louis University. Michael Busler (ABD, Drexel University) is an Assistant Professor at Cabrini College. References Agrawal, Jagdish and Wagner A. Kamakura (1995), "The Economic Worth of Celebrity Endorsers: An Event Study Analysis," Journal of Marketing, 59 (July), 56-62. Anderson, John R. (1976), Language, Memory, and Thought, Hillsdale, NJ: Lawrence Erlbaum Associates. ----- (1983), "A Spreading Activation Theory of Memory," Journal of Verbal Learning and Verbal Behavior, 22 (3), 261-295. Atkin, Charles and Martin Block (1983), "Effectiveness of Celebrity Endorsers," Journal of Advertising Research, 23 (1), 57-61. Baker, Michael J. and Gilbert A. Churchill, Jr. (1977), "The Impact of Physically Attractive Models on Advertising Evaluations," Journal of Marketing Research, Vol. X1V (November), 538-555. Berger, Ida E. and Andrew A. Mitchell (1989), "The Effect of Advertising on Attitude Accessibility, Attitude Confidence, and the Attitude-Behavior Relationship," Journal of Consumer Research, 16 (December), 269-279. Caballero, Marjorie J., James R. Lumpkin and Charles S. Madden (1989), "Using Physical Attractiveness as an Advertising Tool: An Empirical Test of the Attraction Phenomenon," Journal of Advertising Research, August/September, 16-22. ----- and Paul J. Solomon (1984), "Effects of Model Attractiveness on Sales Response," Journal of Advertising, 13 (1), 17-23, 33.

DeBono, Kenneth G. and Richard J. Harnish (1988), "Source Expertise, Source Attractiveness, and the Processing of Persuasive Information: A Functional Approach," Journal of Personality and Social Psychology, 55 (4), 541-546. Dholakia, Ruby Roy and Brian Sternthal (1977), "Highly Credible Sources: Persuasive Facilitators or Persuasive Liabilities?" Journal of Consumer Research, 3 (March), 223-232. Eagly, Alice H., Richard D. Ashmore, Mona G. Makhijani and Laura C. Longo (1991), "What is Beautiful is Good, But...: A Meta-Analytic Review of Research on the Physical Attractiveness Stereotype," Psychological Bulletin, 110 (1), 109-128. Fazio, Russell H., Martha C. Powell and Carol J. Williams (1989), "The Role of Attitude Accessibility in the Attitude-to-Behavior Process," Journal of Consumer Research, 16 (December), 280-288. -----, David M. Sanbonmatsu, Martha C. Powell and Frank R. Kardes (1986), "on the Automatic Activation of Attitudes," Journal of Personality and Social Psychology, 50 (2), 229-238. Freiden, Jon B. (1984), "Advertising Spokesperson Effects: An Examination of Endorser Type and Gender on Two Audiences," Journal of Advertising Research, 24 (5), 33-41. Friedman, Hershey H. and Linda Friedman (1979), "Endorser Effectiveness by Product Type," Journal of Advertising Research, 19 (5), 63-71. Garcia, John and Robert A. Koelling (1966), "Relation of Cue to Consequence in Avoidance Learning," Psychonomic Science, 4 (3), 123-124. Grossman, Randi Priluck and Brian D. Till (1998), "The Persistence of Classically Conditioned Brand Attitudes," Journal of Advertising, 27 (1), 23-31. Hamm, Alfons O., Dieter Vaitl and Peter J. Lang (1989), "Fear Conditioning, Meaning, and Belongingness: A Selective Association Analysis," Journal of Abnormal Psychology, 98(4), 395-406. Homer, Pamela M. and Lynn R. Kahle (1990), "Source Expertise, Timing of Source Identification, and Involvement in Persuasion: An Elaborative Processing Perspective," Journal of Advertising, 19 (1), 30-39. Judd, Charles M., Roger A. Drake, James W. Downing and Jon A. Krosnick (1991), "Some Dynamic Properties of Attitude Structures: Context-Induced Response Facilitation and Polarization," Journal of Personality and Social Psychology, 60(2), 193-202. Kahle, Lynn R. and Pamela Homer (1985), "Physical Attractiveness of the Celebrity Endorser: A Social Adaptation Perspective," Journal of Consumer Research, 11 (March), 954-961. Kaikati, Jack G. (1987), "Celebrity Advertising: A Review and Synthesis," International Journal of Advertising, 6, 93-105. Kamins, Michael A. (1989), "Celebrity and Noncelebrity Advertising in a Two-Sided Context," Journal of Advertising Research, June/July, 24-42.

-----, Meribeth J. Brand, Stuart A Hoeke and John C. Moe (1989), "Two-Sided Versus One-Sided Celebrity Endorsements: The Impact on Advertising Effectiveness and Credibility," Journal of Advertising, 18 (2), 4-10. ----- (1990), "An Investigation into the 'Match-Up' Hypothesis in Celebrity Advertising: When Beauty May Be Only Skin Deep," Journal of Advertising, 19 (1), 4-13. Kanungo, Rabindra N. and Sam Pang (1973), "Effects of Human Models on Perceived Product Quality," Journal of Applied Psychology, 57 (2), 172-178. Kim, John, Chris T. Allen and Frank Kardes (1996), "An Investigation of the Mediaional Mechanisms Underlying Attitudinal Conditioning," Journal of Marketing Research, 33 (August), 318-328. Klein, Stephen B. (1991), Learning: Principles and Applications, New York: McGraw-Hill. Lynch, James and Drue Schuler (1994), "The Matchup Effect of Spokesperson and Product Congruency: A Schema Theory Interpretation," Psychology and Marketing, 11 (5), 417-445. Martindale, Colin (1991), Cognitive Psychology: A Neural-Network Approach, Pacific Grove, CA: Brooks/Cole. McCracken, Grant (1989), "Who is the Celebrity Endorser? Cultural Foundations of the Endorsement Process," Journal of Consumer Research, 16 (December), 310-321. McGuire, William J. (1985), "Attitudes and Attitude Change," in Handbook of Social Psychology, Gardner Lindzey and Elliot Aronson, eds., New York: Random House. Mowen, John C. and Stephen W. Brown (1981), "On Explaining and Predicting the Effectiveness of Celebrity Endorsers," Advances in Consumer Research, 8, Kent M. Monroe, ed., 437-441. Noffsinger, Edward B., Robert J. Pellegrini and George M. Brunell (1983), "The Effect of Associated Persons Upon the Formation and Modifiability of First Impression," Journal of Social Psychology, 120 (2), 183-195. Ohanian, Roobina (1990), "Construction and Validation of a Scale to Measure Celebrity Endorsers' Perceived Expertise, Trustworthiness, and Attractiveness," Journal of Advertising, 19 (3), 39-52. ----- (1991), "The Impact of Celebrity Spokespersons' Perceived Image on Consumers' Intention to Purchase," Journal of Advertising Research, February/March, 46-54. Patzer, Gordon L. (1983), "An Experiment Investigating the Influence of Communicator Physical Attractiveness on Attitudes," in Proceedings of AMA Educators' Conference, Patrick E. Murphy et al., eds., 25-29. Petroshius, Susan M. and Kenneth E. Crocker (1989), "An Empirical Analysis of Spokesperson Characteristics on Advertisement and Product Evaluations," Journal of the Academy of Marketing Science, 17 (3), 217-225.

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 (November), 847-855. -----, ----- and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146. Ratneshwar, R. and Shelly Chaiken (1991), "Comprehension's Role in Persuasion: The Case of Its Moderating Effect on the Persuasive Impact of Source Cues," Journal of Consumer Research, 18 (June), 52-62. Rosenthal, Robert and Ralph L. Rosnow (1984), Essentials of Behavioral Research, New York: McGraw-Hill. Rozin, Paul and James W. Kalat (1971), "Specific Hunger and Poison Avoidance as Adaptive Specializations of Learning," Psychological Review, 78 (6), 459-486. Shimp, Terence A (1991), "Neo-Pavlovian Conditioning and Its Implications for Consumer Theory and Research," in Handbook of Consumer Behavior, Thomas S. Robertson and Harold H. Kassarjian, eds., New York: Prentice-Hall. Solomon, Michael H., Richard D. Ashmore and Laura C. Longo (1992), "The Beauty Match-up Hypothesis: Congruence Between Types of Beauty and Product Images in Advertising," Journal of Advertising, 21 (4), 23-34. Sternthal, Brian, Ruby Dholakia and Clark Leavitt (1978), "The Persuasive Effect of Source Credibility: Tests of Cognitive Response," Journal of Consumer Research, 4 (March), 252-260. Till, Brian D. and Terence A Shimp (1998), "Endorsers in Advertising: The Case of Negative Celebrity Information" Journal of Advertising, 27 (1), 67-82. Tripp, Carolyn, Thomas D. Jensen and Les Carlson (1994), "The Effects of Multiple Product Endorsements by Celebrities on Consumers' Attitudes and Intentions," Journal of Consumer Research, 20 (March), 535-547. Walker, Mary, Lynn Langmeyer and Daniel Langmeyer (1992), "Celebrity Endorsers: Do You Get What You Pay For?" Journal of Services Marketing, Vol. 6 (4), 35-42.

http://www.accessmylibrary.com/article-1G1-70037081/match-up-hypothesis-physical.html