An executive summary for managers and executive readers can be found at the end of this article

An examination on multiple celebrity endorsers in advertising

Daniella McDonald

Graduate Student, School of Business, Montclair State University, Upper Montclair, New Jersey, USA Keywords Celebrities, Advertising, Segmentation Abstract Celebrity endorsement advertising is a prevailing advertising technique. Some marketers choose to utilize multiple celebrities to promote their products or brands. Nevertheless, it is surprising that so little research has focused on this phenomenon. This research discussed advantages and potential concerns of multi-celebrity endorsement advertising and documented the actual use of multiple celebrity endorsers in the milk mustache campaign in the USA. We analyzed the content of the 50 milk mustache ads appearing on the http://www.whymilk.com Web site on a list of celebrity- or productrelated dimensions. Overall, we found that these milk mustache ads have matched their celebrities' gender, age and type of milk attributes in appealing to their female/male, teen/adult consumers. The results support that fit between the endorsed product and various celebrities is a key factor for using multiple celebrity endorsers in advertising.

Introduction Celebrity endorsement advertising has been recognized as ``a ubiquitous feature of modern marketing'' (McCracken, 1989). In the USA, the use of celebrity endorsers increased from a little over 15 percent to approximately 25 percent of all ads between 1979 and 1997 (Stephens and Rice, 1998). The celebrity heat is even more evident in Japan roughly 70 percent of Japanese commercials feature a celebrity, among which the majority are Japanese with a highly visible minority from Hollywood (Kilburn, 1998). These figures evidence the prevalence of the celebrity appeal as a method of persuasive communication. Marketers spend enormous amounts of money annually on celebrity endorsement contracts based on the belief that celebrities are effective spokespeople for their products or brands. Overall, research has found supporting evidence for such a belief in light of celebrity endorsers' impact on the audience's attention, recall, evaluations and purchase intentions. Marketers may rely on multiple celebrity endorsements In the advertising practice, it is very common that a product or brand be associated with one celebrity endorser over a considerably long period of time. Meanwhile, marketers may rely on multiple celebrity endorsements, in other words, the use of two or more celebrities in an advertising campaign. To name a few, Pepsi, Coke, and Nike have employed this technique in their advertising campaigns. One of the most recent, prominent examples is the milk mustache campaign in the USA. It has recruited more than 100 celebrities to promote milk since 1995. Indeed, when marketers decide to use celebrity endorsement advertising, they may face practical considerations about whether or not to use multiple endorsers and whether or not there are
The research register for this journal is available at http://www.emeraldinsight.com/researchregisters The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1061-0421.htm

JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002, pp. 19-29, # MCB UP LIMITED, 1061-0421, DOI 10.1108/10610420210419522

19

advantages in having a number of different celebrities endorse a product to offset the cost (Mowen and Brown, 1981). Furthermore, previous studies have identified match-up of the characteristics between the product and the endorser as an important factor to determine communication effectiveness (e.g. Kahle and Homer, 1985; Kalra and Goodstein, 1998; Kamins and Gupta, 1994; Misra and Beatty, 1990; Sengupta et al., 1997; Till and Busler, 1998). However, these studies tend to focus on the fit aspect in the context of a single endorser for a product. There is a distinct lack of research Despite the widespread use of multiple celebrity endorsements in advertising practice and abundance of celebrity endorsement research in marketing literature, there is a distinct lack of research on the use of multiple celebrity endorsers in advertising. This study seeks to further our understanding about the issue of multi-celebrities advertising. We discuss advertisers' motives and concerns about using multiple celebrity endorsers and highlight the importance of match-up between multiple celebrities and endorsed products. Furthermore, this study conducts a content analysis to provide descriptive data about the actual employment of multi-celebrity endorsement advertising. As an exploratory study, we focus on the milk mustache campaign in the USA, and our sample consists of 50 milk mustache ads appearing on the http://www.whymilk.com Web site. Specifically, we analyze these ads on who these celebrities are in terms of their type, gender, and age and their congruency with the campaign's target audiences, as well as milk attributes with which these celebrities are associated. Research background Multiple celebrity endorsers Multiple celebrity endorsements refer to use of two or more celebrities in an advertising campaign. This study adopts the definition of a celebrity endorser as any individual who uses his or her public recognition on behalf of a consumer good by appearing with it in an advertisement (McCracken, 1989). Traditionally, celebrity endorsers encompass entertainers (including actors, actresses, singers, and models), athletes, politicians, and business people. Following the notion that celebrity spokes-characters, like their human counterparts, have served as product endorsers (Callcott and Lee, 1995, 1994), anthropomorphic characters can be considered another type of celebrity endorser. When marketers decide to employ multiple celebrities, they can choose from the aforementioned categories of celebrity endorsers to appeal to their target audiences. Using multiple celebrities may help advertisers to affect consumer perception According to the attribution theory (Kelley, 1967), people assign causality to events on the basis of either their own behavior or the behavior of others. In the context of celebrity endorsement advertising, consumers might ask whether an endorser recommends a product because he/she actually believes the positive characteristics of the product (an internal attribution) or because he/she is paid for endorsing it (an external attribution). In particular, consensus is one of the attribution cues, which in an advertising context refers to the consumer's perception of whether other individuals, including other endorsers, view the product similarly to the endorser. Thus, using multiple celebrities to endorse a product may create a consensus and help advertisers to positively affect consumer perception (Mowen and Brown, 1981). Marketers generally assume that a spokesperson, as a reference group, should ``be like'' the target audience (e.g. same gender, same age) to enable the target audience to relate to the ad (Wells et al., 1989). Hence, use of a
20 JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002

mix of celebrity endorsers for one product can be beneficial for appealing to various audiences to which the product is aimed. Erdogan and Baker (n.d.) interviewed several advertising agency managers and reported these practitioners' perspective on this topic. It is indicated that ``a brand has a wide range of consumers and sometimes the use of multiple celebrities is needed to cover the whole target audience, though it must be made sure that each celebrity's values reflects core brand values'' (Erdogan and Baker, n.d., p. 13). Additionally, multiple celebrities may help marketers reduce audience boredom which may be caused by a single celebrity. However, the authors also cautioned that multiple celebrities can confuse consumers about the brand's identity and that it should be assured that ``each and every celebrity possesses compatible meanings that are sought for brands'' (Erdogan and Baker, n.d., p. 13). The above managerial perspective points to the importance of congruity, or matching, of the images of celebrity spokespersons and the advertised product. The fit factor is vital for the effectiveness of endorsements Previous research suggests that for a celebrity endorsement to be effective, the characteristics of the celebrity must match up with the attributes of the product. The perceived fit between the celebrity-product combination can be related to physical attractiveness, expertise or other highly relevant characteristics. Empirical studies have shown that endorser-product congruity positively affects consumers' perceptions of spokesperson credibility, attitudes, recall, recognition, purchase intention, and willingness to pay higher prices (e.g. Kahle and Homer, 1985; Kalra and Goodstein, 1998; Kamins and Gupta, 1994; Misra and Beatty, 1990; Sengupta et al., 1997; Till and Busler, 1998). Although these studies have examined the congruency issue in the context of a single endorser, we believe that the fit factor is also vital for the effectiveness of multiple-celebrity endorsements. Therefore, when employing a number of celebrities to be endorsers, marketers will need to carefully evaluate the match-up between each of the celebrities and their product. The milk mustache campaign The ``milk mustache'' campaign in the USA, recruiting more than 100 celebrities to wear a milk mustache in ads since 1995, can be considered one of the most noticeable multiple celebrities endorsements. Actually, Video Storyboard Tests' survey of 20,000 consumers rated the milk mustache campaign the ``Top print campaign of the year'' for 1995 (Rubel, 1996). The milk processors and New York agency Bozell launched the print campaign in January 1995. According to Jay Schulberg, Bozell's chief creative officer, the milk mustache campaign aimed to change consumer attitudes that caused decline in per capita consumption of milk and increase public awareness of nutritional facts (Rubel, 1996; Wallenstein, 1995). Also, milk was not perceived as an adult beverage and could not compete with the ``massive marketing investment by soft drinks and bottled waters'' (http:// www.bozell.com). The magazine ads were Bozell's solution to improve the image and awareness of milk. ``Image continues to be a huge deal for milk . . . which is why you see the celebrities'' (Talereco, 1999). In this campaign, celebrities' photos with milk mustaches are featured along with ad copy where the celebrities praise different kinds of milk for various nutritional benefits. The milk mustache ads initially targeted female adults aged 25 to 44, with a series of ads featuring such celebrities as Christie Brinkley, Kate Moss, Vanna White and actresses from Friends. In 1996, the campaign expanded to 12-49-year-old females and 18-34-year-old males, and widened its appeal
JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002 21

One of the most noticeable endorsements

with a more diverse mix of celebrities, including Spike Lee, Party of Five's Matthew Fox and Cal Ripken Jr (Gleason, 1996; Kinsella, 1998; Rubel, 1996). One of their ads featuring Britney Spears clearly appeals to teen consumers, with the popular teen singer claiming: ``Grow up. 15 percent of your height is added during your teen years and milk can help make the most of it'' (http://www.whymilk.com). Exploratory content analysis Research methodology In spite of the widespread use of multiple celebrity endorsements in advertising, no research has documented the nature of multiple celebrity endorsement ads. Content analysis is particularly useful in identifying frequently and infrequently used depictions in ads. Hence, to further our understanding about how multi-celebrity endorsements are used in advertising, we conducted an exploratory content analysis on the milk mustache campaign. This campaign was chosen due to its use of a large number of celebrity endorsers for a single product. Our sample consisted of all 50 milk mustache ads contained in the http://www.whymilk.com Web site, as of December 2000. These ads on the Web site are identical with the campaign's print ads placed in magazines. The Web site also includes the featured celebrities' brief bio. These ads were coded according to the following dimensions: celebrity's name, type (athlete, model, actor/actress, singer, politician, fictional character, other), gender (male, female, mixed group, and fictional character), age (child, teen, adult, fictional character), and key attributes of milk specified in ad copy (calcium, strong or growing bones, nutrients/vitamins/minerals/protein/potassium, osteoporosis, taste, no/ low fat). The two authors independently coded each of the 50 ads using the above coding scheme. After both authors had finished coding the ads, one of the authors identified any differences between their coding. They then reconciled differences by reviewing the ads and discussing the coding until a consensus was reached on how the item was best coded. The level of intercoder reliability, based on the coding before they were reconciled, was calculated by the Perreault and Leigh (1989) method and ranged from 94.5100 percent. Results Results of the content analysis are presented in the following paragraphs. It should be noted that since many cells in this data set contained fewer than five data points, we were not able to conduct further statistical analyses to determine relationships between variables. Forty-four percent of ads feature male celebrities Celebrity characteristics: type, gender, and age Table I presents the distribution of celebrity endorsers' characteristics in terms of type, gender, and age. Of the 50 ads, 16 ads feature actors/actresses (32 percent), 15 ads athletes (30 percent), 11 ads singers (22 percent), four ads fictional characters (8 percent), one ad politician (2 percent), and one ad an actor and a fictional character together (2 percent). Hence, the entertainer appears most frequently in this sample, followed by the athlete, and the politician is featured only once. Data also show that 22 ads (44 percent) feature male celebrities (17 ads with one male celebrity and five with a group of male celebrities in the ad), another 22 ads (44 percent) have female celebrities (18 with a single female celebrity and four with a group of female celebrities), four ads (8 percent) use fictional characters, one ad (2 percent) features a group of celebrities with both genders, and one ad (2 percent) has a
JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002

22

Celebrity characteristics Type Athlete Model Actor/actress Singer Politician Fictional character Fictional character and actor Gender Male Female Mixed group Fictional character Fictional character and a male Age Teen Adult Fictional character Fictional character and adult Total observations

Frequency 15 2 16 11 1 4 1 22 22 1 4 1 6 39 4 1 50

Percentage 30 4 32 22 2 8 2 44 44 2 8 2 12 78 8 2 100

Table I. Distribution of celebrity type, gender and age

male celebrity and a fictional character together. Overall, an equal use of male and female celebrities is observed. In terms of age, 39 of the 50 ads (78 percent) employ adult celebrities, six ads (12 percent) teen celebrities, four ads (8 percent) fictional characters, and one ad (2 percent) an adult celebrity plus a fictional character. Therefore, a majority of the milk mustache ads use adult celebrities. Six categories of milk attributes Milk attribute and celebrity type Copy of the 50 milk mustache ads in this sample highlights the following six categories of milk attributes (benefits): calcium (34 ads, 68 percent), strong or growing bones (20 ads, 40 percent), nutrients/vitamins/minerals/protein/ potassium (18 ads, 36 percent), osteoporosis (nine ads, 18 percent), no/low fat (three ads, 6 percent), taste (one ad, 2 percent). Some ads feature more than one attribute. Table II summarizes how different types of celebrities are associated with these milk attributes. Eight (53.3 percent) of the 15 ads with athlete celebrities, 23 (76.7 percent) of the 30 ads with entertainer celebrities, and three of the five ads with ``other-type'' celebrities stress the attribute of calcium. Fifteen (50 percent) of the 30 ads with entertainer celebrities, four of the five ads with ``other-type'' celebrities, but only one (6.7 percent) of the 15 ads with athlete celebrities tout milk's benefit for growing or strong bones. A majority of ads featuring athlete celebrities (12 ads, 80 percent)
Athlete n (%) 8 1 12 0 1 0 23.53 5 66.7 0 100 0 Celebrity type Entertainer Other n (%) n (%) 23 15 6 8 0 3 67.6 75 33.3 88.9 0 100 3 4 0 1 0 0 8.8 20 0 11.1 0 0 Total (%) 100 100 100 100 100 100

Milk attribute Calcium Growing/strong bone Nutrient/vitamin/etc. Osteoporosis Taste No/low fat

n 34 20 18 9 1 3

Table II. Associations of milk attribute and celebrity type

promote milk's value of containing nutrients, vitamins, minerals, protein, or potassium whereas only six (20 percent) of the 30 ads with entertainer celebrities and not a single ad with ``other-type'' celebrities advertise this aspect. Eight (26.7 percent) of the 30 entertainer celebrity ads and one (20 percent) of the five ``other-type'' celebrity ads stress the fact that milk prevents osteoporosis; none of the 15 athlete celebrity ads relates itself to this benefit of milk. The three ads promoting milk's low or no fat value feature entertainer celebrities exclusively. Finally, the single ad praising taste of milk uses an athlete celebrity endorser. Milk attribute and celebrity gender Table III indicates associations of celebrities' gender and milk attributes. Ads with either male or female celebrities are equally and frequently associated with the attribute of calcium (68.2 percent of each). Eight (36.4 percent) of the 22 ads with male celebrities and seven (31.8 percent) of the 22 ads with female celebrities contain the benefit of strong or growing bones. Nine (40.9 percent) of the 22 ads with male celebrities and eight (36.4 percent) of the 22 ads with female celebrities praise milk's value of offering nutrients, vitamins, minerals, protein, or potassium. Three (13.6 percent) of the 22 male celebrity ads and five (22.7 percent) of the 22 female celebrity ads highlight prevention of osteoporosis. The only ad that stresses milk's taste features a female celebrity. Milk's no/low fat value is advertised in a limited number of ads (one or 4.5 percent of male celebrity ads; five or 22.7 percent of female celebrity ads). Celebrities of different age groups Milk attribute and celebrity age Table IV presents how celebrities of different age groups are associated with the six categories of milk attributes. Calcium is stressed in five (83.3 percent) out of the six teen celebrity ads, 26 (66.7 percent) of the 39 adult celebrity ads, plus three (75 percent) of the four fictional character ads. A majority of
Celebrity gender Mixed Fictional Group Fictional and male n (%) n (%) n (%) 1 1 0 1 0 0 2.9 5 0 11.1 0 0 3 3 1 0 0 0 8.8 15 5.6 0 0 0 0 1 0 0 0 0 0 5 0 0 0 0

Milk attribute Calcium Growing/strong bone Nutrient/vitamin/etc. Osteoporosis Taste No/low fat

Male n (%) 15 8 9 3 0 1

Female n (%)

Total n (%) 34 20 18 9 1 1 100 100 100 100 100 100

44.1 15 44.1 40 7 35 50 8 44.4 33.3 5 55.6 0 1 100 33.3 2 66.7

Table III. Associations of milk attribute and celebrity gender

Fictional and adult n (%) 0 1 0 0 0 0 0 5 0 0 0 0

Total n (%) 34 20 18 9 1 3 100 100 100 100 100 100

Table IV. Associations of milk attribute and celebrity age

teen celebrity (five ads; 83.3 percent) and fictional character ads (three ads; 75 percent) promote strong or growing bones by drinking milk whereas this attribute appears in only 28.2 percent of ads with adult celebrities (11 cases). Attributes related to nutrients/ vitamins/ minerals/ protein/ potassium are featured in 17 (43.6 percent) adult celebrity ads and one (25 percent) fictional character ad, and not in any teen celebrity ads. Also, eight (20.5 percent) of the 30 adult celebrity ads and one (16.7 percent) teen celebrity ad highlight how milk can help prevent osteoporosis. Three (7.7 percent) and one (2.6 percent) of the 39 adult celebrity ads emphasize milk's no/low fat value and its taste, respectively. Celebrity endorsers represent a diverse mix of type, gender and age Summary of key findings Our sample shows that celebrity endorsers in the milk mustache campaign represent a diverse mix of type, gender and age. Entertainers and athletes have much higher appearances than fictional characters and political figures. Moreover, the findings reveal that target audience's and celebrity endorser's gender and age are matched. Given that milk's targets are 12-49-year-old females and 18-34-year-old males, a balanced use of male and female celebrities is arranged. Plus, adult celebrities (including younger celebrities in their 20s) and teens are featured in the campaign to appeal to the intended age segments. There are also some interesting findings regarding the associations of milk attributes and celebrity qualities. We observed an equal use of male and female celebrities to promote milk attributes that are important to both men and women (calcium, strong bones, nutrients, prevention of osteoporosis). Athlete celebrities are found to be most frequently associated with milk's nutritional value but have a lower association with other attributes of milk whereas entertainer celebrities tout more frequently milk attributes such as calcium, strong bones, and prevention of osteoporosis. Conclusion Given that an advertiser can afford to employ a number of celebrities, multicelebrity endorsement advertising may help the advertiser build a sense of consensus, avoid audience boredom, and appeal to multiple audiences. Once the marketer decides to use multiple celebrities rather than a single celebrity, the issue of which celebrities to be chosen naturally follows. Marketers will have to manage carefully the fit between the endorsed product and each of the endorsers in order to effectively appeal to the intended segments without confusing audiences about the product or brand image. This content analysis represents an exploratory effort in examining the multicelebrity endorsement advertising. Overall, our findings reveal that the milk mustache campaign has used a great variety of celebrity endorsers towards its multiple target segments via matching celebrity qualities (type, age, and gender) and their intended audiences as well as milk attributes. The authors recognize that this study is limited to a small number of ads from a single advertising campaign, and our findings are not able to be generalized across all multiple celebrity endorsement advertising. Preliminary findings generated from this study may best be applied to primary demand advertising, where a trade organization uses advertising to promote an entire product category (e.g. milk, beef, orange, credit card) instead of a specific brand. Surely, multiple celebrity endorsements can be used in selective demand advertising, seeking to promote a single brand. It is likely that since brands tend to possess well-defined images or personalities, they require a
JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002 25

Multi-celebrity endorsement advertising may appeal to multiple audiences

stricter set of criteria to pick a mix of celebrities who are compatible with such images or personalities. Celebrities possess different types of social powers Tom et al. (1992) suggest that celebrities possess different types of social powers expert power, referent power, legitimate power, coercive power and reward power that enable them to have an effect on consumers. In the case of the milk mustache ads, average consumers are likely to be familiar with milk, and the celebrities are not considered as having more knowledge (expert power) about milk. Referent power seems to be the most prevalent social power involved when we consider the celebrities who are a part of the milk mustache campaign. Referent power is defined as ``when an audience identifies with and aspires to be like the celebrity.'' Nonetheless, for other types of products, expert power or other types of social powers may be a more important consideration in determining the fit between the multiple celebrities and the advertised products. For example, sneaker marketers will most likely choose a number of athlete celebrities who are perceived to be experts in this product category to endorse their brands. Also, this study examined celebrity characteristics only in terms of their type, age and gender. Depending on characteristics of a product or brand, celebrities' personality and lifestyle meanings can be important qualities in determining the celebrities' congruity with the product or brand. Finally, we recognize that our use of content analysis has an inherent limitation of not permitting measurement of consumer responses. Therefore, we recommend experimental studies that further investigate issues in this area. For instance, what is the relative effectiveness of single versus multiple celebrity endorsers toward single versus multiple target audiences? Is there an optimal number of multiple celebrities given a specific time period? Does the multi-celebrity, multi-segment approach create audience confusion about the endorsed product's or brand's image, especially when one audience segment views ads intended for another audience segment? Research along these directions will provide more guidance for advertisers in employing multiple celebrity endorsers for their products.
References Callcott, M. and Lee, W. (1994), ``A content analysis of animation and animated spokescharacters in television commercials'', Journal of Advertising, Vol. 23 No. 4, pp. 1-12. Callcott, M. and Lee, W. (1995), ``Establishing the spokes-character in academic inquiry: historical overview and framework for definition'', in Kardes, F. and Sujan, M. (Eds), Advances in Consumer Research, Vol. 22, Chicago, IL, pp. 144-51. Erdogan, B.Z. and Baker, M.J. (n.d.), ``Celebrity endorsement: advertising agency managers' perspective'', http://www.execusports.com/br4.htm Gleason, M. (1996), ``Men are newest target for `milk mustache' ads'', Advertising Age, Vol. 67 No. 27, 1 July, p. 10. Kahle, L.R. and Homer, P.M. (1985), ``Physical attractiveness of the celebrity endorser: a social adaptation perspective'', Journal of Consumer Research, Vol. 11 No. 4, pp. 954-61. Kalra, A. and Goodstein, R.C. (1998), ``The impact of advertising positioning strategies on consumer price sensitivity'', Journal of Marketing Research, Vol. 35 No. 2, pp. 210-24. Kamins, M.A. and Gupta, K. (1994), ``Congruence between spokesperson and product type: a match-up hypothesis perspective'', Psychology and Marketing, Vol. 11 No. 6, pp. 569-86. Kelley, H.H. (1967), ``Attribution theory in social psychology'', in Levine, D. (Ed.), Nebraska Symposium on Motivation, University of Nebraska Press, Lincoln, NE, pp. 192-238. Kilburn, D. (1998), ``Star power'', Adweek (Eastern edition), Vol. 39 No. 2, pp. 20-1. Kinsella, B. (1998), ``Got milk? Booksellers do their own milk mustaches'', Publishers Weekly, Vol. 245 No. 43, 26 October. McCracken, G. (1989), ``Who is celebrity endorser? Cultural foundations of the celebrity endorsement process'', Journal of Consumer Research, Vol. 16 No. 3, pp. 310-21.
26 JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002

Misra, S. and Beatty, S.E. (1990), ``Celebrity spokesperson and brand congruity'', Journal of Business Research, Vol. 21 No. 2, pp. 159-73. Mowen, J.C. and Brown, S.W. (1981), ``On explaining and predicting the effectiveness of celebrity endorsers'', in Monroe, K.B. (Ed.), Advances in Consumer Research, Vol. 8, Association for Consumer Research, Chicago, IL. pp. 437-41. Perreault, W.D. Jr and Leigh, L.E. (1989), ``Reliability of nominal data based on qualitative judgements'', Journal of Marketing Research, Vol. 26 No. 2, pp. 135-48. Rubel, C. (1996), ``Mustache ads change attitude toward milk'', Marketing News, Vol. 30 No. 18, 26 August, p. 10. Sengupta, J., Goodstein, R.C. and Boninger, D.S. (1997), ``All cues are not created equal: obtaining attitude persistence under low-involvement conditions'', Journal of Consumer Research, Vol. 23 No. 4, pp. 351-61. Stephens, A. and Rice, A. (1998), ``Spicing up the message'', Finance Week, Vol. 76 No. 26, pp. 46-7. Talereco, S. (1999), ``Want milk?'', Dairy Field, Vol. 182 No. 11, p. 8. Till, B.D. and Busler, M. (1998), ``Matching products with endorsers: attractiveness versus expertise'', Journal of Consumer Marketing, Vol. 15 No. 6, pp. 576-86. Tom, G., Clark, R., Elmer, L., Grech, E., Masetti, J. Jr and Sandhar, H. (1992), ``The use of created versus celebrity spokespersons in advertisements'', Journal of Consumer Marketing, Vol. 9 No. 4, pp. 45-51. Wallenstein, A. (1995), ``Milk producers fight fat fatalism'', Advertising Age, 22 May, pp. 1, 12. Wells, W., Burnett, J. and Moriarty, S. (1989), Advertising Principles and Practice, PrenticeHall, Englewood Cliffs, NJ.

&

JOURNAL OF PRODUCT & BRAND MANAGEMENT, VOL. 11 NO. 1 2002

27