Content-Type: text/html Understanding Celebrity Endorsers in Cross-Cultural Contexts An Exploratory Analysis of South Korean and US Advertising by Hye-Jin Paek, a doctoral student University of Wisconsin-Madison School of Journalism and Mass Communication 5167 Vilas Hall, 821 University Ave., Room 5115 Madison, WI 53706-1497 Phone) 608-265-2222 Fax) 608-262-1361 E-mail) [log in to unmask] Student competitive paper submitted to the Association for Education in Journalism and Mass Communication 2003 Annual Conference -Advertising Division July 30-Aug. 2 2003, Kansas City, MO Understanding Celebrity Endorsers in Cross-Cultural Contexts An Exploratory Analysis of South Korean and US Advertising Abstract This study links McCracken's "cultural meaning transfer" model (1989) with Hofstede's cultural typology, in order to understand the cultural meanings of celebrity endorsers in cross-cultural advertising. The content analysis of South Korean and US newspaper ads finds that ads in a high uncertainty avoidance and power distance culture employ a greater number of celebrity endorsers and that there is some possibility for international advertisers to use standardized celebrity endorser strategy. Implications and directions for future studies are discussed. INTRODUCTION Using celebrity endorsers is one of the most popular strategies of advertising in the US (Agrawal & Kamakura, 1995). According to industry sources, approximately a quarter of all television commercials feature a celebrity person (Erdogan, Baker & Tagg, 2001). US companies paid more than $1 billion to nearly 2,000 athletes for endorsement deals and licensing rights in 1996 (Belch & Belch, 1998). Basketball icon Michael Jordan, one of the most successful advertising endorsers, makes an estimated $40 million a year in endorsement fees from companies such as McDonald's, Nike, General Mills, Hanes, Quaker Oats (makers of Gatorade), and MCI COLLECT CALL (Horovitz, 1999). Ten out of the top 20 most effective and remembered TV commercials in the US, as reported by Advertising Age (2002), were celebrity-endorsed ads, including Pepsi Cola (Britney Spears), KFC (Jason Alexander), and Hanes (Michael Jordan). Recently, a rising TV idol "Joe Millionaire" (Evan Marriott) joined existing pitchman Jason Alexander in two upcoming TV commercials plugging KFC Boneless Wings. However, the cost is not always worth it. For instance, the #1 rental car company, Hertz, used O. J. Simpson as its spokesperson for 20 years. The company lost all of that equity when Simpson was accused of murdering his ex-wife (Belch & Belch, 1998). The global brand Pepsi Cola hurt their name value seriously when the tarnished personal lives of their three celebrity endorsers -- Mike Tyson, Madonna, and Michael Jackson -- became public (Till & Shimp, 1998). A plethora of studies have examined celebrity endorser effectiveness and consumer responses (e.g., Bower, 2001; Bower & Landreth, 2001; Frieden, 1984; Goldsmith, Lafferty, & Newell, 2000; Kahle & Homer, 1985; Kamins, 1989, 1990; Kamins, Brand, Hoeke, & Moe, 1989; Stafford, Stafford, & Day, 2002). However, little is known about the impact of cultural contexts on roles of celebrity endorsers in ads (see McCracken, 1989, for exception). McCracken (1989) explains that the success of celebrity-endorsed ads depends on whether the celebrity endorser is meaningful to consumer values and norms within a culture. Interpreting celebrity endorsers as cultural value representatives, this study attempts to go further: In this complex global market situation, how can celebrity endorsers work across cultures? Companies continue to expand their markets beyond national borders and have an increasing interest in global marketing strategy, but international marketers, advertising agencies and academicians have been debating the applicability of globalized or localized international advertising for almost four decades (see Agrawal, 1995; Onkvisit & Shaw, 1999, for detailed review). These debates involve disagreement over whether consumer needs become homogeneous (e.g., Levitt, 1983; Kanso, 1992), or whether consumer reaction differs across cultures, reflecting their indigenous cultural norms and values (e.g., Lin, 2001; Mueller, 1987, 1992; Zandpour, Campos, Catalano, & Chang, 1994). Although international advertising researchers have stressed the importance of cultural factors in making strategic decisions about globalization or localization and have assessed the values reflected in the ads (Lin, 1993, 2001; Mueller, 1987, 1992, 1996), practitioners lean toward globalization because of greater cost effectiveness, simplified strategic planning, and consistent brand image (see Agrawal, 1995). In relation to this globalization, international advertisers and brand managers increasingly consider the potential global popularity of their celebrity endorsers (Howard, 2003). For instance, Yao Ming, an NBA player, is contracted to endorse the sports drink Gatorade, which the advertiser attempts to push globally. Given that celebrity endorsement of certain global brands assists in building a consistent brand image worldwide, it is worthwhile to understand what kind of cultural meanings celebrity endorsers generate across cultures. In so doing, this study is a small step in exploring how celebrity endorsers in advertising are presented and interpreted in relation to cultural meanings and values. This study compares US and South Korean newspaper ads. South Korea (hereafter, Korea) was selected because of its economic importance, as the US' ninth largest import and export market in 1998 (Moon & Franke, 2000), and as the world's tenth largest advertising market in 1999 (Ad Age International, 1999). According to a recent Advertising Age report (2003), a Korean advertising agency, Cheil Communications, is ranked 19th among the world's top 50 advertising organizations (ranked by advertising revenue in 2002) Because of its close ties between the two countries and importance of Korea as a global market, a number of cross-cultural studies between the US and Korea have been conducted, including examinations of cultural values reflected in ads (Cho, Kwon, Gentry, Jun, & Kropp, 1999; Han & Shavitt, 1994; Tak, Kaid, & Lee, 1997), advertising execution styles and appeals (Miracle, Chang, & Taylor, 1992), consumer attitudes (Yoon, Muehling, & Cho, 1996), and work ethics (Moon & Franke, 2000). This study adds to the knowledge of advertising and values for these two national cultures by examining celebrity endorsers in ad contents, based on the Hofstede's (1991) cultural framework (uncertainty avoidance, power distance) linked with McCracken's (1989) "cultural meaning transfer" model. As such, this study will give some insights of a particular type of global marketing strategy to practitioners. Celebrity Endorsers in US and Korean Ads A celebrity is "known for being well-known" (Boorstin, 1961, p.57) and is defined as an individual (actor, sports figure, entertainer, politician, etc.) who is known to the public for his or her achievements (Blackwell, Miniard, & Engel, 2001). These achievements are often in areas unrelated to the product class being promoted (e.g., Britney Spears for Pepsi Cola, Tiger Woods for Buick cars, Kobe Bryant for McDonald's). Frieden (1984) defines celebrities as well-known individuals who are directly or indirectly associated with their endorsing product category (e.g., Michael Jordan for Nike athletic shoes, Kobe Bryant for Adidas basketball shoes, Tiger Woods for Nike golf equipment). In their endorsed advertising, celebrities have played a variety of promotional roles: giving a testimonial citing, recommending the products' benefits, lending their names to a product (e.g., Jordan cologne, Nike Air Jordan), playing as an actor in the advertisement, and representing a brand or company as a spokesperson while the contract is valid. Although US advertisers vigorously embrace using celebrity endorsers in their ads, Korean advertisers rely even more heavily on celebrity endorsements. Of the TV commercials screened, about 32% of the ads included celebrity endorsers, and 59% of prime time TV commercials used celebrities as their endorsers (Son, 2001). The popularity of endorsers depends on the advertising belief that a celebrity endorser is the only way to differentiate their advertised products from those of their competitors. Ever since the Korean advertising market opened for foreign countries in 1988 and the use of Western endorsers has been deregulated, it is also an increasing trend to use Western celebrities. For instance, Brooke Shields pushed Aloe drinks, and Kenny G represented Inkel audio. Shannon Doherty, from the popular TV show Beverly Hills 90210, appeared on LG cosmetics, while Kim Basinger pitched Vitamin Shampoo. Cindy Crawford played a role as an endorser in LG fashion; sexy star Sharon Stone of the sex thriller movie Basic Instinct made the Hanhwa energy ad hot; Claudia Schiffer endorsed GV jeans; Naomi Campbell pushed Mercoledi fashion clothing. Not only have these movie stars played role models in Korean advertising, but so have Western scholars, politicians, and news reporters. One of the most successful uses of Western celebrities was the Daewoo car advertising campaign using Karl Bernstein, who won the Pulitzer Prize for his coverage of Watergate and Nixon. Likewise, the futurist Alvin Tofler endorsed Hyundai, and British ex-prime minister Margaret Thatcher appeared on Samsung corporate image advertising (Lee, 2001). The use of various Western celebrities has not always been successful. It is noteworthy that many campaigns in which Western celebrities use their sexuality and physical attractiveness have had disappointing results, while ads using credible and powerful Western celebrities were successful (Lee, 2001). One exceptional example of a global brand using Western celebrities is Vidal Sassoon Shampoo (Kim, 2001). Kim (2001) explains that the Vidal Sassoon Shampoo campaign became successful by adopting a 'globalization' (or standardization) strategy using Vidal Sassoon himself as an endorser. This standardization strategy followed Vidal Sassoon's 'localization' strategy, which used Korean celebrity endorsers showing off their glamorous hair but was not effective. Ever since, more global product advertisers have adopted a standardization strategy (Kim, 2001). Nonetheless, Lee (2001) has cautioned that global brand managers and advertisers should be careful of using Western celebrities, and that differing cultural tastes, norms, and values must be taken into account. Given that, how can such various results of using Western celebrity endorsers in global/domestic products advertising be explained in cultural context? What makes celebrity endorsers appeal to consumers across cultures? Cultural differences between the US and Korea Culture is defined as "the interactive aggregate of common characteristics that influence a human group's response to its environment" (Hofstede, 1980, p.19) and as the behavioral norms, attitudinal tendencies, and beliefs shared among people from the same group (Gundykunst, 1995). Thus, working as a "lens," culture determines how the world is seen (McCracken, 1986, p.72). Understanding cultural differences is often considered a prerequisite for successful international advertising (de Mooij, 1998). Consumers grow up in a particular culture and become accustomed to that culture's value systems, beliefs, and perception processes. Consequently, they respond to advertising messages that are congruent with their culture, rewarding advertisers who understand that culture and tailor ads to reflect its values (Zhang & Gelb, 1996). In a cultural context, the celebrity endorser is a cultural hero (Hofstede, 1991). Cultural heroes are "persons, alive or dead, real or imaginary, who possess characteristics that are highly prized in a society, and thus serve as role models for behavior" (de Mooij, 1998, p.45). McCracken (1989), in a similar context, argues that celebrities are those who "draw these powerful meanings from the roles they assume in their television, movie, military, athletic, and other careers and each new dramatic role brings the celebrity into contact with a range of objects, persons, and contexts" (p. 315). In his "cultural meaning transfer" model, celebrity endorsers should posess shared cultural meanings, imbue cultural meanings into products, and deliver such meanings to the consumers throughout the subsequent processes. Britney Spears, for instance, transferred her distinctive cultural images based on her records, live performances, and video appearances into the Pepsi Cola ads, which successfully promoted its product concept of a "new generation" to its young target audiences (Peter & Olson, 1993). Consistent with this concept of meaning transfer, Cohen (1992) found that advertisements with Asian spokespersons produce more favorable attitudes toward high technology engineering products than do ads with white spokespersons, while the reverse is observed for products associated with status. Gwinner and Eaton (1999) have adopted the 'cultural meaning transfer' model to hypothesize that a sporting event's image is transferred to a brand through event sponsorship activity. Hirshman and Thompson (1997) agree that the relative success of a celebrity-endorsed ad depends on how well its images of cultural icons fit to its promotional purposes and consumer tastes. Taken together, "cultural meaning transfer" has some important implications for companies using celebrity endorsers. Marketers should first decide on the image or symbolic meaning important to their target audience for this particular product, service, or company, and then determine which celebrity best represents the meaning or image to be projected. Given that celebrities who have been typecast are more likely to have shared cultural meanings that can be associated with a product, how would the celebrity endorsers be different in other global markets? In applying McCracken's model -- that celebrity endorsers should have shared cultural meanings--, this study also adopts Hofstede's typology of culture (1980, 1983, 1991) as a framework for testing cross-cultural differences in order to understand the cultural meanings of celebrity endorsers. Hofstede's cultural typology consists of four dimensions on which societies differ, representing the cultural value system of the majority of the middle class in the workplace. These dimensions are: individualism/collectivism, power distance, uncertainty avoidance, and masculinity/femininity. This study adopts the two cultural dimensions of power distance and uncertainty avoidance in particular, in that these are distinctly different cultural dimensions between the US and Korea and fit well into discussing the different cultural meanings of celebrity endorsers. Uncertainty avoidance The uncertainty avoidance dimension involves "the extent to which the members of a culture feel threatened by uncertain or unknown situations" (Hofstede 1991, p. 113). Individuals in low uncertainty avoidance cultures, such as the US, are relatively comfortable with ambiguity and are tolerant of others' behaviors and opinions. Where uncertainty avoidance is high, such as Japan and Korea, consumers rely more on formal rules (Gudykunst et al., 1996), absolute truth (Hofstede, 1983), and the advice of those whom they consider to be experts (Hofstede, 1980). This dimension has been adopted in cross-cultural advertising studies that investigate the ethical perception of advertising practitioners between the US and Korea (Moon & Franke, 2000), the recall of and attitude toward comparative advertising (Donthu, 1998), and the difference of the form and content of political advertising messages between the US and Korea (Tak, Kaid, & Lee, 1997). For instance, Donthu (1998) hypothesized that low uncertainty avoidance cultures may be very receptive to comparative advertising, whereas high uncertainty avoidance cultures may not be comfortable with such advertising. This discomfort results from the fact that comparative advertising is an unfamiliar approach to advertising that is therefore considered ambiguous and risky. Further, he found that consumer attitudes toward comparative ads were not very positive in a high uncertainty avoidance culture. Likewise, Zandpour et al. (1994) analyzed the content of TV commercials collected from different cultures (the US, Mexico, the UK, France, Germany, Spain, Taiwan, and Korea) and found that cultures with high uncertainty avoidance relied more frequently on a trustworthy source to provide them with logical reasoning and visual information. Thus, the implication for celebrity endorsement advertising is that uncertainty reduction requires credible, trustworthy, and knowledgeable endorsers in high uncertainty avoidance cultures. Especially for a high involvement product that is newly introduced and very expensive (and thus a risk-taking purchase), people in a high uncertainty avoidance culture may rely more on experts or credible celebrities than people in a low uncertainty avoidance culture. Power distance Power distance in Hofstede's typology involves the extent to which people accept the unequal distribution of power in society and organizations (Hofstede, 1980, 1991), or the extent to which people see authority as a basic facet of the society (Gudykunst et al., 1996). Hofstede (1980) claims that the cultural dimension of power distance explains how different societies have addressed basic human inequalities in view of social status, prestige, wealth, and power. High power distance countries, such as Korea, are more tolerant of hierarchies and autocratic leadership and more likely to expect clear directions. In contrast, low power distance countries, such as the US, are more likely to seek factual evidence and reasoning in relation to a particular course of action (Hofstede, 1991). People in a high power distance culture, therefore, tend to obey the recommendations of public and authority figures such as celebrities and high-status figures, in comparison with low power distance cultures (Zandupor et al., 1994). Thus, in a high power distance culture, people with power are considered to be right, and function as referent groups (Albers-Miller & Gelb, 1996). To apply this to celebrity endorsement ads, celebrity endorsers -- who wield their salience and fame as a power to guide consumers -- might be more prevalent in a high power distance culture than for those in a low power distance culture, which has little tolerance for authority, and focuses more on self-interest, autonomy, and independence (Hofstede, 1980). Celebrities in a high power distance culture transfer their publicly salient and powerful image to consumers, and deliver these images via product ads. Therefore, the following hypotheses can be proposed: H1: Ads in a high power distance and high uncertainty avoidance culture (Korea) will have a greater number of celebrity endorsers than those in a low power distance and low uncertainty avoidance culture (the US). H2: Ads in a high power distance and high uncertainty avoidance culture (Korea) will have a greater number of expert (product-relevant) celebrity endorsers than those in a low power distance and low uncertainty avoidance culture (the US). Product involvement interacting with cultural characteristics Product involvement has been one of the most important and widely employed concepts in the advertising research domain, particularly when a psychological process of advertising is examined (e.g., Petty & Cacioppo, 1986; Petty, Cacioppo, & Schumann, 1983; Petty & Priester, 1994). Here, product involvement, according to Bowen and Chaffee (1974), is a relation between consumer and product that theoretically operates as a contingency or necessary condition, governing the relevance of an ad's appeal to the consumer. In this content-analytic study, the degree of product involvement is determined by factors including how expensive a product is, how frequently the product is purchased, how long it lasts, and what the consequences are of a product choice. How will this product involvement variable interact with cultural values of uncertainty avoidance and power distance to explain presence and meaning of celebrity endorsers? Following cultural characteristics, high involvement product ads in a high uncertainty avoidance culture should employ celebrity endorsers (and especially the endorsers' expertise). This is because high involvement product ads can be considered highly risky, and ads in a high uncertainty culture should meet consumers' expectations, since those consumers want to avoid high risk and uncertainty. However, such reasoning contradicts dual-processing persuasion literature (e.g., the Elaboration Likelihood Model and the Heuristic Systematic Model), which posits that product endorsers (i.e., experts or celebrities) exert greater impact under low involvement conditions (Petty, Cacioppo, & Schumann, 1983). However, in their experiment to investigate the interactive role of source expertise, time of source identification, and involvement, Homer and Kahle (1990) found that source expertise can serve as the central persuasion cue under the proper conditions, implying advertisers should not assume that all source information is peripherally processed. Given this inconsistent and contradictory account of the role of celebrity endorsers in advertising, it is interesting to see the role of product involvement in celebrity-endorsed ads across cultures. Therefore, the following research question is addressed: RQ1: With low and high product involvement considered, how differently are celebrity endorsers presented in ads across the two cultures? International products and celebrity endorsers in Korean ads As discussed earlier, international marketers and advertisers increasingly execute their product ads in Korean media, but little is known about their celebrity endorser strategy. The terms and definitions of domestic and international products have been various; Samli (1995) defined the former as products developed to satisfy very local and specific needs, and the latter as products designed to satisfy common needs around the world among different nations. For instance, in the same computer product category, such a brand as Hyunju (a Korean brand) is a domestic product, whereas IBM and Macintosh would be international/global products. Based on McCracken's "cultural meaning transfer" model (1989), either domestic or foreign celebrities, whichever are the most culturally acceptable and meaningful for the endorsed ads, and eventually for consumers, are likely to be shown. In addition, considering some contingent conditions such as standardization vs. specialization strategy and complex Korean media environments importing MTV and Hollywood movies, how many international product ads are executed in Korean newspaper? How many global brand advertisers adopt standardized or localized celebrity endorser strategies? Taking these considerations together, then, the following research question is addressed: RQ3: Among international product ads in Korea, how frequently are domestic and foreign celebrity endorsers presented? METHOD Content analysis was used to examine celebrity endorsers shown in US and Korean newspaper advertising. Newspaper advertisements were chosen because they are the most comparable and generalizable for all target segments[1], but are relatively neglected in cross-cultural advertising studies. Sampling. Advertisements in Chosun-ilbo (Korea) and the New York Times (U.S) were selected randomly. These newspapers are relatively well matched (Tak et al., 1997) in that each is arguably the most salient and influential newspaper in each country and has nationwide readership. Based on constructed week sampling, newspapers throughout the year 2000 were collected. Sunday-issued newspapers, duplicate ads, local ads, classifieds, and movie ads were excluded to avoid sampling bias and lack of matching. The unit of analysis is a considerable-size ad[2], which totals 1318 (694 for Chosun-ilbo; 624 for The New York Times) ads. Coding Procedure. Two bilingual graduate students coded the ads. Taking into account the potential for gender differences and cultural biases (Miracle, 2001), one female bilingual Korean American who has lived in the US for more than 15 years coded the entire data set. The other coder was a Korean male who had just arrived in the US. He coded 170 randomly selected sample ads -- more than ten percent (Wimmer & Dominick, 1994) -- to calculate intercoder reliability. The coders worked independently and were blind to the hypotheses. Before coding, six training sessions were conducted, in which the coders practiced coding and contributed to the coding manuals. The coders were instructed to enter a simple "yes" or "no" to indicate the presence or absence of human models, and then celebrities, for each ad. Human models should show more than the upper torso, so that it can be told whether they are Eastern or Western, male or female, and celebrities or non-celebrities. Celebrity endorsers are operationalized as "publicly well-known individuals such as TV stars, movie actors/actresses, sports figures, musicians, scholars, politicians, etc (Blackwell, Miniard, & Engel, 2001). Out of these celebrity endorsers, expert celebrities are operationalized as "product relevant celebrities," such as sports players endorsing sports products (e.g., Tiger Woods in golf equipment ads, Kobe Bryant endorsing basketball shoes). Non-expert celebrities indicate "product-not-relevant celebrities" (e.g., Jamie Lee Curtis endorsing Sprint PCS, Jason Alexander endorsing KFC). These different types of celebrities are again categorized and coded as three different age groups; 0 to 25 years old, 26 to 50 years old, 51-and higher years old. For intercoder reliability, this study adopted Krippendorff's alpha, which takes chance agreements into account (Krippendorff, 1980). All intercoder reliability values of the variables were at or above the conventionally accepted value of .75 (Wimmer & Dominick, 1994), ranging from .87 to .99 (see APPENDIX A). Measure. The product involvement scale was constructed with four item variables: product price [PRICE], period of use of product [DURAB], frequency of product purchase [FREQ], and consequences of a product choice [CONSEQ] (see APPENDIX A for more detailed operationalization). The four items were summed to construct the product involvement scale, and the value of the scale, from 0 to 7, is dichotomized into low and high product involvement. Cronbach's alpha reliability of the scale was .91, which indicates very high internal reliability. Exploratory factor analysis using Principle Component Analysis (PCA) also showed that the scale forms one factor with eigenvalue, 3.12 and 78.02 % of the total variance being explained. The communalities of each item are as follows: PRICE= .77, DURAB= .75, FREQ= .82, CONSEQ= .78. RESULTS For the comparison of frequency of celebrity endorsers between Korean and US ads, a series of Chi-square statistics were performed. Out of the total ads, ads endorsed by human models were 326 for Chosun-ilbo and 199 for The New York Times. Korean ads contained a significantly greater number of human models (47% out of the total ads) than US ads (31.9%, _2 (df=1) = 31.19, p < .001). The first hypothesis, positing that ads in a high power distance and high uncertainty avoidance culture (Korea) would have a greater number of celebrity endorsers than those in a low power distance and low uncertainty avoidance culture (the US), was supported. As shown in table 1, Korean ads had a greater number of celebrities (24.1%, n = 167) � whether the celebrities are product-related or not � than the US ads (9.9%, n = 62, _2 (df=1) = 45.68, p < .001). However, the second hypothesis, predicting that Korean ads would have a greater number of expert (product-relevant) celebrity endorsers than the US ads, was not supported. Rather, the result was in the opposite direction, in that the US ads employed a greater number of product- relevant celebrity endorsers (77.4%, n = 48) than Korean ads (38.3%, n = 64, _2 (df=1) = 27.66, p < .001). The first research question relates celebrity endorsers in different cultural contexts with product involvement. Ads for high involvement products in the Korean newspaper accounted for 69.8% of the total ads (n=395), while 67% of total ads in the US newspaper were for high involvement products. The difference of the frequency between the two country ads was not statistically significant (_2 (df=1) = .92, p = ns). The first research question asked how product involvement works with different cultural characteristics to explain presence of celebrity endorsers in ads in each culture. First of all, among high involvement products, Korean ads carried significantly more celebrity endorsers (21.5%, n=85) than their US counterparts (9.1%, n=29, _2 (df=1) = 20.32, p < .001). However, in terms of different types of celebrities (i.e., product-relevant or product-not-relevant), the US ads employed significantly more product-relevant celebrity endorsers than the Korean ads. Specifically, 55.2 percent of the celebrity-endorsed, high involvement product ads in the US newspaper (n = 16) employed product-relevant celebrities, while their Korean counterparts employed only 29.4 percent (n = 25, _2 (df=1) = 6.23, p < .05). This relatively high proportion of product-relevant celebrities in the US ads is seen among low involvement product ads as well. Among low involvement products, Korean ads had a significantly greater number of celebrity endorsers than the US ads, but within celebrity-endorsed ads, the US ads carried product-relevant celebrity spokespersons more than the Korean ads. 96.1 percent of the total celebrity endorsers presented in the low involvement product ads in the US newspaper were product-relevant celebrities, while about half of the Korean low involvement product ads employed product-relevant celebrities (see table 1). ------------------------------------------- TABLE 1 ABOUT HERE ------------------------------------------- The second research question asked about the relationship between international product ads and the nationality of celebrity endorsers within Korean ads. Out of a total of 694 ads, only 8.1 % were international product ads (n = 56). Of 56 international product ads in the Korean newspaper, as shown in table 2 below, only four ads had celebrity endorsers (7.1%), while 163 domestic product ads had celebrity endorsers (25.5%, _2 (df=1) = 9.54, p < .01). Foreign celebrity endorsers were present in three out of the four international product ads, while domestic endorsers were present in only one international product ad. It is noteworthy that 12 foreign celebrities (7.7%) endorsed domestic product ads within Korean newspaper. ------------------------------------------- TABLE 2 ABOUT HERE ---------------------------------------- DISCUSSION This is the first attempt to explore cultural meanings of celebrity endorsers in advertising across cultures. While numerous studies have focused on effectiveness and psychological processing of celebrity endorsers, the cultural meaning of celebrity endorsers has been relatively neglected. Expanding the scope of McCracken's "cultural meaning transfer" model (1989), this study investigates how celebrity endorsers are interpreted in a cross-cultural context and what kind of cultural meanings they could transfer across borders. The results show that Hofstede's cultural value framework was helpful in searching for cultural meanings of celebrity endorsers. Ads in the high uncertainty avoidance and high power distance culture of Korea employed more celebrity endorsers than ads in the low uncertainty avoidance and low power distance culture of the US. Such results may imply that celebrity endorsers can play a more significant role in guiding consumers, in less risky and uncertain ways, in Korean society than in the US. This also implies that advertising within a culture should use and portray celebrities congruently with ways that consumers can accept cultural meanings of the famous spokespersons. One caution concerns specific types of celebrity endorsers. Expert celebrities, operationalized as "product-relevant" celebrities in this study, were hypothesized to be more prevalent in Korea than in the US, because such source expertise (or product relevancy) may contribute to reducing the risk of unknown product attributes. However, the hypothesis was not supported; rather, the result was in the opposite direction. Although different operationalization and measure may have different results, perhaps cultural characteristics that Korean consumers hold may offer unique insights. Korean consumers are high trend pursuers. Once trends arrive, Korean consumers seek after such trends unanimously; otherwise, they will be regarded as "fashion-laggards" (Cheil Communication Monthly, 2000). However, such trends change at lightning speeds. Yesterday's fashion, therefore, will become history tomorrow. Also, such trends mostly come from TV stars and famous singers who are currently the most salient. Clothes that a pop star wore in last night's TV show come out in downtown stores in a few days, and form a fashion trend among Korean consumers. The pop star is likely endorsing several products in a couple of ads or more, and will probably become a cultural hero sooner or later. Such a short-breath circle of cultural trends is not uncommon in Korean society. In this unique cultural context, expertise in the product may not matter. What does matter is which celebrities catch the biggest popularity, and which advertisers preempt the most recent and hottest cultural idols. Taken together, the degree of salience of celebrity endorsers may be overwhelming source experts, such that the celebrities are performing well enough (or at least perceived to be doing so) to relieve consumers' uneasiness about the product attributes. To transfer their cultural meaning to Korean consumers via product ads, in other words, celebrity endorsers may not require specific knowledge pertinent to the endorsed product brands, but rather simply being the most salient cultural icon and symbol possible. Just as Hofstede's cultural framework of uncertainty avoidance and power distance gives only a partial explanation of celebrity endorsers, so does product involvement. According to dual-processing persuasion model, the literature predicts that low involvement product ads would be likely to employ more celebrity endorsers, specifically experts, to give consumers peripheral cues to recall and buy the products (e.g., Petty et al., 1983), this paper shows somewhat mixed results. The US ads followed the persuasion model prediction, in that, among low involvement and celebrity endorsed product ads, the proportion of product-relevant celebrities overwhelmed product-not-relevant celebrities (96.1% vs. 3.9%); however, the Korean ads did not follow this direction with same proportion between product-relevant and product-not-relevant celebrity endorsers. Among high involvement product ads, likewise, the results were complex: the US ads had a slightly more than half product-relevant celebrities, while less than 30 percent out of the total celebrity endorsers in the Korean ads were product-relevant celebrities. On the other hand, high involvement product ads were more prevalent in both countries' newspaper ads. One possible explanation concerns the unique characteristics of the newspaper medium, which has no limit about time and space to advertise, and consumers are more actively involved in looking each ad; accordingly, advertisers may want to advertise high involvement products that are expensive and are not frequently purchased (Rothschild, 1987). Finally, in terms of global advertising strategy, the results show that only four international product ads employed celebrity endorsers, but three out of four ads had foreign celebrity endorsers. Interestingly, all these three cases were international female fashion magazine ads. By showing their cover models -- Cameron Diaz in Cosmopolitan, Cindy Crawford in Elle, and Elizabeth Hurley in Marie Claire � these magazines promoted their endorsed products. With the rising tide of globalization, thin Western models with a blonde hair, blue eyes and long legs have become a standard of beauty. Along with the visibility of such global magazine brands, well-known Western fashion models and celebrities become salient by acquiring a cultural meaning of "global beauty standard." It is likely that other beauty related product ads covering cosmetics, clothes, and fashion accessories have the possibility of adopting a global celebrity endorser strategy. Such an argument is supported by de Mooij (1998), who claims that standardization advertising strategy can be more likely applied to such product ads as cosmetics, toiletries, clothing, and footwear. In addition, it should be noted that some domestic product ads employed foreign celebrities as well. One salient example is computer software emperor (as he is called in Korea) Bill Gates, who endorsed the Sambo Dream notebook computer. Interestingly, he was not shown in person, but on the cover picture of his book, "Speed of Thought." It seems that the book publisher and the computer company did co-opt marketing, so as to promote both products to evoke a synergy effect. The fact that intelligent and knowledgeable celebrities may be effective can be explained by unique cultural norms prevalent in Korean society (see Paek, Nelson & McLeod, 2002, for detailed discussion). Paek, Nelson and McLeod (2002) argue that, although Korea is known as a collectivistic culture, Korean society demonstrates vertical and individualistic characteristics such as 'competition' and 'achievement.' It is already well known that aspiration for higher education is widespread and deeply ingrained in Korean society. Such education-oriented cultural values and norms can be detected in Korean ads, in that about 17% of total ads in Korean newspaper in 2000 were for education-related product ads, mostly textbooks enhancing English or other foreign language skills (e.g., TOEFL, TOEIC, TESOL) and self-study guidebooks for college entrance exams. It is not surprising to see Bill Gates in Korean ads, because his image of intelligence may meet Korean consumers' norms and expectations in a Korean cultural context. Furthermore, several Korean celebrities who have a similar image to that of Bill Gates endorse Korean brands as well. For instance, Cheol Soo Ahn, who invented an anti-virus computer software program, is a famous cultural icon for his intelligence and brilliance. He has been endorsing several products pertinent to computers, computer program software, and education. Overall, although it is rarely generalizable, this study implies that it is possible to adopt a consistent and homogeneous celebrity endorser strategy, if the celebrity endorsers are well known and culturally acceptable. As discussed earlier, the success of such foreign celebrity endorsers as Karl Bernstein, Alvin Tofler, and Margaret Thatcher may be a few salient precursors predicting that powerful, credible and intelligent celebrities may work if well coordinated with a certain type of product. Need celebrity endorsers be culturally meaningful to perform their tasks successfully in the endorsed ads? This study seems to answer that, yes; celebrity endorsers are cultural symbols that represent contemporary cultural trends and can converse with consumers, but only when the celebrities are culturally meaningful. And in the cross-cultural setting, this study suggests that celebrity endorsers should be cautiously selected to transfer the meaning of the advertised products to consumers, by understanding the very cultures where international advertisers want to penetrate. LIMITATIONS AND FUTURE STUDY This study should be evaluated more as a theoretical exploration than an empirical achievement, because many curious questions remain unanswered with answers open to future studies. Firstly, the sampling of newspaper ads might be a considerable limitation. Much literature is devoted to how influential and prevalent celebrity endorsers are in TV commercials; however, few studies paid attention to role of celebrity endorsers in print ads carried via newspaper and magazine. In analyzing Korean and US newspaper ads, this study might add some insights to a dearth of celebrity advertising endorser literature in print media; at the same time, though, this study lacks an analysis of the mainstream role of celebrity endorsers that can be found in TV commercials. Secondly, this content analytic study examined presence and frequency of celebrity endorsers in advertising for different types of products. It will be interesting to see how consumers respond to different roles and types of celebrity endorsers in different conditions. Consumer surveys will suffice for such consumer response- relevant research questions. In addition, different operationalization and different measures of content analysis of advertising may be able to explain roles of celebrity endorsers in ads in more detailed and different ways. Finally, this study opens the possibility of linking other contingent theories and concepts in order to study celebrity endorsers and consumer responses in cross-cultural advertising domains. In so doing, this paper requests to see advertising in a bigger picture of the whole society and culture in more integrated and synthetic ways. Table 1. Chi-square test of celebrity endorsers in Korean and US ads Variables Frequencies % (n) Korea US Chi-sq P-value Endorsers 47.0 (326) 31.9 (199) 31.19 .000 Celebrity endorsers 24.1 (167) 9.9 (62) 45.68 .000 Product-relevant celebrities 38.3 (64) 77.4 (48) 27.66 .000 Among high involvement product ads Celebrity endorsers 21.5 (85) 9.1 (29) 20.32 .000 Product-relevant celebrities 29.4 (25) 55.2 (16) 6.23 .013 Among low involvement product ads Celebrity endorsers 32.7 (56) 15.9 (25) 12.46 .000 Product-relevant celebrities 50.0 (28) 96.1 (24) 15.91 .000 Note: Degrees of freedom =1 Table 2. Relationship between product nationality and celebrity endorsers within Korean ads Variables Frequencies % (n) Domestic International Chi-sq P-value Endorsers 47.6 (304) 39.3 (22) 1.45 ns Celebrity endorsers 25.5 (163) 7.1 (4) 9.54 .002 Among celebrity endorsers Domestic endorsers 92.3 (56) 25.0 (1) 20.65 .000 Foreign endorsers 7.7 (12) 75.0 (3) Note: Degrees of freedom =1 REFERENCE Ad Age International. (1999). Agency report, 70 (April 19), s37. Advertising Age. (2003). 2003 Ad Agency Report [online], retrieved from http://www.adage.com/page.cms?pageId=974 Advertising Age. (2002). December 30 [online], retrieved from http://www.adage.com/news.cms?newsId=36811 Agrawal, M. (1995). Review of a 40-year debate in internationial advertising: Practitioner and academician perspectives to the standardization/adaptation issue. International Marketing Review, 12(1), 26-48. Agrawal, J. & Kamakura, W. (1995). The economic worth of celebrity endorsers: An event study analysis, Journal of Marketing, 59(3), 56-. Albers-Miller, N. D. & Gelb, B. D. (1996). Business advertising appeals as a mirror of cultural dimensions: a study of eleven countries. Journal of Advertising, 15(4), 57-70. Belch, G. E. & Belch M. A. (1998). Advertising and promotion: An integrated marketing communications perspective (4th ed.), Irwin: McGraw-Hill. Blackwell, R. D., Miniard, P. W. & Engel, J. E. (2001). Consumer behavior and marketing strategy (9th ed). Fort Worth: Harcourt College. Boorstin, D. J. (1961). The Image. New York: Atheneum. Bowen, L. & Chaffee, S. (1974). Product involvement and pertinent advertising appeals, Journalism Quarterly, 51(Winter), 613-15. Bower, A. B. (2001). Highly attractive models in advertising and the women who loathe them: The implications of negative affect for spokesperson effectiveness. Journal of Advertising, 30(3), 51-. Bower, A. B. & Landreth, S. (2001). Is Beauty Best? Highly Versus Normally Attractive Models in Advertising. Journal of Advertising, 30(1), 1-12. Cheil Communications Monthly. (2000). Cheil Communications, March, Seoul, Korea. Cho, B. J., Kwon, U., Gentry, J.W., Jun, S. K. & Kropp, F. (1999). Cultural values reflected in theme and execution: a comparative study of U.S. and Korean television commercials. Journal of Advertising, 28 (4), 59-73. Cohen, J. (1992). White consumer response to Asian models in advertising. Journal of Consumer Marketing, 9, 17-27. de Mooij, M. K. (1998). Global marketing and advertising: understanding cultural paradoxes. Thousand Oaks, CA: Sage Publications. Donthu, N. (1998). A cross-country investigation of recall of and attitude toward comparative advertising. Journal of Advertising, 27(2), 111-122. Erdogan, Z. B., Baker, M. J. & Tagg, S. (2001). Selecting celebrity endorsers: The practitioner's perspective. Journal of Advertising Research, 41(3), 39-48. Frieden, J. B. (1984). Advertising spokesperson effects: An examination of endorser type and gender on two audiences. Journal of Advertising Research, 24(5), 33-41. Goldsmith, R. E., Lafferty, B. A. & Newell, S. J. (2000). The impact of corporate credibility and celebrity credibility on consumer reaction to advertisements and brands. Journal of Advertising, 29(3), 43-54. Gudykunst, W. B. (1995). Bridging differences: Effective intergroup communication. Thousand Oaks: Sage Publications. Gudykunst, W. B. & Kim, Y. Y. (1995). Communication with strangers: an approach to intercultural communication. New York: McGraw-Hill. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K. S. & Seyman, S. (1996). The influence of cultural individualism-collectivism, self construals, and individual values on communication styles across cultures. Human Communication Research, 22 (June), 510-543. Gwinner, K. P. & Eaton, J. (1999). Building brand image through event sponsorship: The role of image transfer. Journal of Advertising, 28(4), 47-57. Han, S. P. & Shavitt, S. (1994). Persuasion and culture: advertising appeals in individualistic and collectivistic societies. Journal of Experimental Social Psychology, 30, 326-350. Hirshman, E. C. & Thompson, C. J. (1997). Why media matter: Toward a richer understanding of consumers' relationships with advertising and mass media. Journal of Advertising, 26(1), 43-60. Howard, T. (2003). Gatorade gets Yao factor ; Chinese NBA star signs multiyear endorsement deal. USA Today, Feb 7, B3. Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hills, CA: Sage Publications. Hofstede, G. (1983). The cultural relativity of organizational practices and theories. Journal of International Business Studies, 14 (Fall), 75-89. Hofstede, G. (1991). Cultures and organizations: software of the mind. New York: McGraw-Hill Book Company. Homer, P. M. & Kahle, L. R. (1990). Source Expertise, Timing of Source Identification, and Involvement in Persuasion: An Elaborative Processing Perspective. Journal of Advertising, 19 (1), 30-39. Horovitz, B. (1999). Endorsement career far from done. USA Today. Jan., 13, 12C. Kahle, L. R. & Homer, P. M. (1985). Physical Attractiveness of the Celebrity Endorser: A Social Adaptation Perspective. Journal of Consumer Research, 11 (March), 954-961. Kamins, M. A. (1989). Celebrity and noncelebrity advertising in a two-sided context. Journal of Advertising Research, 29(2), 24-42 . Kamins, M. A. (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. Kamins, M. A., Brand, M. J., Hoeke, S. A., & Moe, J. C. (1989). Two-sided versus one-sided celebrity endorsements: The impact on advertising effectiveness and credibility. Journal of Advertising, 18 (2), 4-10. Kanso, A. (1992). International advertising strategies: global commitment to local vision. Journal of Advertising Research, 32 (1), 10-14. Kim, S. Y. (2001). Product and message decide the endorsers, Advertising Information, Nov., 39-43 (in Korean). Lee, W. Y. (2001). It passed when we admired 'blue eyes with blonde hairs,' Advertising Information, Nov., 62-68 (in Korean). Levitt, T. (1983). The globalization of markets. Harvard business Review, 61(3), 92-102. Lin, C. A. (1993). Cultural differences in message strategies: A comparison between American and Japanese TV commercials. Journal of Advertising Research, 33(4), 40- 48. Lin, C. A. (2001). Cultural values reflected in Chinese and American television Advertising. Journal of Advertising, 30 (4), 83-94. McCracken, G. (1986). Culture and consumption: A theoretical account of the structure and movement of the cultural meaning of consumer goods. Journal of Consumer Research, 13(June), 71-84. McCracken, G. (1989). Who is the celebrity endorser? Cultural foundations of the endorsement process. Journal of Consumer Research, 16 (Dec.), 310-321. Miracle, G. E., Chang, K. Y. & Taylor, C. R. (1992). Culture and advertising executions: a comparison of selected characteristics of Korean and U.S. television commercials. International Marketing Review, 9(4), 5-17. Miracle, G. E. (2001). Process globalization: the significance of reliable and valid information. In Roberts, M. S. & King, R. L. (Eds.), The Proceedings of the 2001 Special Asia-Pacific Conference of the American Academy of Advertising (pp. 1-5). Moon, Y. S. & Franke, G. R. (2000). Cultural influences on agency practitioners' ethical perceptions: a comparison of Korea and the U.S. Journal of Advertising, 26 (1), 51-65. Mueller, B. (1987). Reflections of culture: an analysis of Japanese and American advertising appeals. Journal of Advertising Research, 27 (3), 51-59. Mueller, B. (1992). Standardization vs. specialization: an examination of westernization in Japanese advertising. Journal of Advertising Research, 32 (1), 15-24. Mueller, B. (1996). International advertising. Wadsworth Publishing Company Onkvisit, S. & Shaw J. J. (1999). Standardized international advertising: some research issues and implications. Journal of Advertising Research, 39(6), 19-24. Paek, H-J, Nelson, M. R., & McLeod, D. M., (2002), Is culture going global? Comparison of South Korean and U.S. newspaper ads in the new millennium," paper presented at the annual conference of the Association for Education in Journalism and Mass Communication, August, Miami, Florida. Peter, P. J. & Olson, J. C. (1993). Consumer behavior and marketing strategy (3rd ed). Homewood: Irwin. Petty, R. E. & Cacioppo, J. T. (1986). Communication and persuasion: central and peripheral routes to attitude change. New York: Springer-Verlag. Petty, R. E., Cacioppo, J. T. & Schumann, D. (1983). Central and peripheral routes to advertising effectiveness: the moderating role of involvement. Journal of Consumer Research, 10(2), 135-146. Petty, R. E. & Priester, J. R. (1994). Mass media attitude change; implications of the elaboration likelihood model of persuasion. In J. Bryant & Zillmann, D. (eds), Media effects: advances in theory and research, (pp. 91-122). Hillsdale, NJ: Lawrence Erlbaum Associates. Rothschild, M. L. (1987). Advertising: from fundamentals to Strategies. Lexington, Mass: Heath Co. Samli, C. A. (1995). International Consumer Behavior. Westport, CN: Quorum Books. Son, T. W. (2001). Success of advertising depends on the appropriateness of celebrity use, Advertising Information, November, 440-50 (in Korean). Stafford, M. R., Stafford, T. F., & Day, E. (2002). A contingency approach: The effects of spokesperson type and service type on service advertising perceptions, Journal of Advertising, 31(2), 17-35. Tak, J. Y., Kaid, L. L.& Lee, S. B. (1997). A cross-cultural study of political advertising in the United States and Korea. Communication Research, 24(4), 413-430. Till, B. D. & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27(1), 67-82. Wimmer, R. D. & Dominick, J. R. (1994). Mass media research: an introduction. 4th ed. Belmont, CA: Wadsworth Publish Co. Yoon, K., Muehling, D. & Cho, J. H. (1996). A comparison of Korean and American attitudes toward advertising. Journal of Marketing Communications, 2, 231-245. Zandpour, F., Campos, V., Catalano, J. & Chang, C. (1994). Global reach and local touch: achieving cultural fitness in TV advertising. Journal of Advertising Research, 34 (5), 35-43. Zhang, Y. & Gelb, B. D. (1996). Matching advertising appeals to culture: the influences of products' use conditions. Journal of Advertising, 25 (3), 29-46. APPENDIX A. Operationalization of the Variables Expert celebrity endorsers: 1. Product-relevant celebrity endorser with age 0 to 25 (.97) 2. Product-relevant celebrity endorser with age 26 to 50 (.98) 3. Product-relevant celebrity endorser with age 51 + (.99) Non-expert celebrity endorsers: 1. Product-not-relevant celebrity endorser with age 0 to 25 (.95) 2. Product-not-relevant celebrity endorser with age 26 to 50 (.92) 3. Product-not-relevant celebrity endorser with age 51 + (.92) Non-celebrity endorsers: 4. With age 0 to 25 (.92) 5. With age 26 to 50 (.89) 6. With age 51 + (.93) Involvement Scale: (. 91)* 1. Price of product: low, medium, high (.98) Low: less than $20 e.g.) Stationeries, foods, hair care, battery, OTC medicine, toothbrush/paste, Medium: About $21- $200. e.g.) fashion apparel, watches, airline ticket, hotels, camera, telephone/cellular service High: more than $200 e.g.) car, jewelry, computer, electronics, business products such as real estates, insurance, consulting and finance, company networking/hosting service 2. Durability, period of use of product: short, medium, long (.98) Short: one-time use or used in a few times e.g.) Foods, hair care, battery, airline ticket, hotels, home/office goods Medium: used about in a year e.g.) Vitamins, diet pills, clothes/shoes, Long: durable for a long time, or never worn out e.g.) car, jewelry, computer, electronics, telecommunication service, business products, home/office furniture, publication 3. Frequency of product purchase: frequent, medium, less frequent (.91) Frequent: more than one time per month e.g.) Foods, hair care, battery, OTC medicine, home/office goods Medium: a couple of times per year e.g.) fashion apparel, watches, airline ticket, hotels, publication Less frequent: only one or a few times for lifetime e.g.) car, jewelry, computer, electronics, diet pills, business products, home/office furniture 4. Consequences of bad decisions of product purchase: low, high (.97) Low: Less effective. Not much damage. Okay just for trial e.g.) Foods, hair care, battery, airline ticket, hotels, home/office goods High: Very much effective. Financial and mental damage. When not satisfied with these products, you might want to tell your friends or families not to buy those products. e.g.) car, computer, electronics, business products, home/office furniture Product nationality (.96) Domestic product: essentially made for the purpose of serving domestic customers. Made, sold, and advertised domestically. Including joint companies, with foreign companies, M&A, or just export products International (global/transnational/multinational) product: well-known, worldwide brands. Made, sold, and advertised globally. Made for and used by customers worldwide. Eastern model (.99): Mostly easily picked as pacific islanders. Koreans, Japanese, Chinese, South Asians, and Mongolians Western model (.87): White Americans, Hispanics, Blacks, Europeans. Basically all races excluding Asians Note * indicates Cronbach's alpha reliability; otherwise, intercoder reliability [1] For instance, the magazine medium in Korea is not well developed due to the small market size and the low readership; TV commercials may not be comparable, either, because Korea has only 15-second TV commercials and no ads inserted during programming. [2] The measurement of Korean newspaper ads is called TAN. The size of Korean ads adopted in this study is 7 1/2 TAN (About A4 size, 36.5 cm diagonally). Because the US newspaper is narrower in width than the Korean newspaper (59cm: 62cm diagonally), the size of the US sample advertisements was calculated as 59: x = 62: 36.5.