Citation
Standardization of International Advertising in Latin America: A Comparison of Advertisements from the United States and Costa Rica

Material Information

Title:
Standardization of International Advertising in Latin America: A Comparison of Advertisements from the United States and Costa Rica
Creator:
MILLWEE, CASEY HALE ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Advertising campaigns ( jstor )
Advertising rates ( jstor )
Advertising research ( jstor )
Brands ( jstor )
Comparative advertising ( jstor )
Cost accounting ( jstor )
Logos ( jstor )
Product standards ( jstor )
Slogans ( jstor )
Standardization ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Casey Hale Millwee. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
6/30/2007
Resource Identifier:
434595883 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

1 STANDARDIZATION OF INTERNATIONAL ADVERTISING IN LATIN AMERICA: A COMPARISON OF ADVERTISEMENTS FR OM THE UNITED STATES AND COSTA RICA By CASEY HALE MILLWEE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING UNIVERSITY OF FLORIDA 2006

PAGE 2

2 Copyright 2006 by Casey Hale Millwee

PAGE 3

3 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Jo rge Villegas, for his knowledge, support, and unending willingness to answer my questions. Thank you also to the other members on my supervisory committee, Dr. Chang-Hoan Cho and Dr. Marilyn Roberts, for their assistance and ideas. I could not have accomplishe d this study without their support. I would also like to thank my parents and friends for thei r continued understanding and encouragement. It is such a blessing to have friends and family to lean on. I would like to say tha nk you to the people of Costa Rica, for their generous hospitality. I have never met people that are more patient, helpfu l, and thankful for the life that they live. They truly take pride in th eir country and their culture. Finally, I owe my greatest thanks to my fianc, James, for hi s nonstop motivation and love. Without him, I would never have taken the opportun ity to pursue this stud y and travel to Costa Rica.

PAGE 4

4 TABLE OF CONTENTS Page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES................................................................................................................ .........7 ABSTRACT....................................................................................................................... ..............8 CHAPTER 1 INTRODUCTION................................................................................................................... .9 Costa Rica..................................................................................................................... ..........10 Purpose........................................................................................................................ ...........11 2 LITERATURE REVIEW.......................................................................................................12 Standardization, Adaptation, and Convergence......................................................................12 Brief History.................................................................................................................. .........14 Interpretations of Standardization...........................................................................................15 Organization and Deci sion Making Centers....................................................................16 Case Study: Goodyear in Latin America.........................................................................17 Models of Standardization......................................................................................................18 Harvey’s (1993) Model...................................................................................................18 Whitelock and Chung’s (1989) Model............................................................................19 Mueller’s (1989) Model..................................................................................................19 Backhaus, Mhlfeld, and van Doorn (2001)...................................................................20 Culture........................................................................................................................ ............20 Dimensions of Culture.....................................................................................................21 United States versus Costa Rica......................................................................................22 Past Suggestions Leading to Present Research.......................................................................24 Research Question and Hypotheses........................................................................................25 3 METHODOLOGY.................................................................................................................27 Sampling Design................................................................................................................ .....27 Variables...................................................................................................................... ...........28 Coding Procedure............................................................................................................... ....30 Intercoder Reliability......................................................................................................... .....30 Data Analysis.................................................................................................................. ........30 4 RESULTS........................................................................................................................ .......31 Standardization versus Adaptation.........................................................................................31

PAGE 5

5 Standardization versus Comparison Type..............................................................................33 Standardization versus Product Category Type......................................................................34 Standardized Elements vers us Type of Comparison..............................................................36 Standardized Elements versus Product Category Type..........................................................40 5 DISSCUSSION AND CONCLUSION..................................................................................43 Discussion..................................................................................................................... ..........43 Standardization versus Adaptation..................................................................................43 Standardized Elements vers us Type of Comparison.......................................................44 Standardized Elements versus Product Category Type...................................................46 Conclusion..................................................................................................................... .........47 Limitations.................................................................................................................... ..........48 Suggestions for Future Research............................................................................................49 APPENDIX A CODING SHEET...................................................................................................................51 B EXAMPLES OF COMPLETELY STANDARDIZED ADVERTISEMENTS.....................55 LIST OF REFERENCES............................................................................................................. ..59 BIOGRAPHICAL SKETCH.........................................................................................................61

PAGE 6

6 LIST OF TABLES Table Page 3-1 Magazines from which samples were taken......................................................................27 3-1 Continued.................................................................................................................. .........28 3-2 Operational definitions of coded variables........................................................................29 3-3 Product category types..................................................................................................... ..29 4-1 T-test results for standardization versus adaptation...........................................................33 4-2 Descriptive statistics for sta ndardization by type of comparison......................................33 4-3 ANOVA results for standardiz ation by type of comparison..............................................33 4-4 Product category within the overall sample.......................................................................34 4-5 Descriptive statistics for sta ndardization by product category type..................................34 4-5 Continued.................................................................................................................. .........35 4-6 ANOVA results for standardiz ation by product category type..........................................35 4-7 Post-Hoc Tukey mean difference results for standardization by product category type...36 4-8 Standardized variables within the overall sample..............................................................36 4-9 Standardized variable with in the brand comparison sample..............................................37 4-10 Standardized variables within th e product category comparison sample..........................38 4-11 Cross tabulation of type of co mparison by standardized variables....................................39 4-12 Cross tabulation of product ca tegory by standardized variables.......................................41 5-1 Standardized elements per product category.....................................................................47

PAGE 7

7 LIST OF FIGURES Figure Page 2-1 Hofstede’s Dimensions of Culture sc ore for United States and Costa Rica......................23 4-1 Histogram of standardization among overall sample.........................................................32 4-2 Histogram of standardization among brand comparisons..................................................32 4-3 Histogram of standardizati on among product category comparisons................................32 4-4 Means plot of standardi zation by type of comparison.......................................................33 4-5 Histogram of standardiza tion by product category type....................................................35 B-1 Armani Code from Costa Rica...........................................................................................55 B-2 Armani Code from the United States.................................................................................56 B-3 Damiani from Costa Rica...................................................................................................57 B-4 Damiani from the United States.........................................................................................58

PAGE 8

8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising STANDARDIZATION OF INTERNATIONAL ADVERTISING IN LATIN AMERICA: A COMPARISON OF ADVERTISEMENTS FR OM THE UNITED STATES AND COSTA RICA By Casey Hale Millwee December 2006 Chair: Jorge Villegas Major Department: Advertising This study explores the amount of standard ization in Latin American advertising, specifically Costa Rica, and the executional elements that are the most likely to be standardized. Using a content analysis of 101 pairs of ad s from the United States and Costa Rica, advertisements were compared across 21 executi onal elements, as well as by product category type of each advertisement. Advertisements we re compared as an overall sample as well as by brand comparison and by product category compar ison. Executional elements include general theme, general layout, size, as well as language , positioning, and content of picture, headline, explanatory copy, slogan, logo, and attributes highlighted. Results showed the advertisements to be more standardized than adapted overall, as well as by brand comparison and by product category compar ison. In addition, jewelry and accessories, as well as clothing and shoes, were the most sta ndardized product category types. In terms of standardized elements, language was found, in general, to be the most standardized. Specifically, attributes highlighted, slogan, and lo go language were the most often standardized. Picture was found to be the least of ten standardized element all around.

PAGE 9

9 CHAPTER 1 INTRODUCTION Standardization in internati onal advertising has been a l ong standing issue of debate (Cutler & Javalgi, 1992 ). Standardization in in ternational advertising i nvolves elements of the advertising campaign for a particul ar brand’s product or service be ing the same in each market that the brand’s product or service is sold. In other words, th e same brand’s product is advertised “in the same way in all markets of the worl d” (Onkvisit & Shaw, 1987, p. 43). Localization, or adaptation, refers to adapting an advertisi ng campaign to the local market. Localized advertisements are unique to their market, wher eas standardized advertisements are intended to be the same in every market. There are three basics schools of thought on the subject of intern ational advertising (Agrawal, 1995), as well as a wide variety of in terpretations for standard ization itself (Backhaus, Mhlfeld, & van Doorn, 2001; Ramaprasad & Ha segawa, 1992). Some believe that the only element of an advertisement that can be adap ted for an advertisement to be considered standardized is the language (Bac khaus et al., 2001; Cutler & Java lgi, 1992), which, because it is the most widely accepted definition, is what we will assume for the purposes of this thesis. Others believe that for an advertisement to be c onsidered standardized, at least one of three key parts of a campaign must be the same, with thes e parts being the languag e, the creative selling proposition, and the elements of th e advertisement itself (Backhaus et al., 2001). Elements of an advertisement include executional variab les such as picture, copy, and layout. Though Taylor (2002) suggests that there are too many studies d ealing with the content of international advertising while there is a need for research on why certain techniques work in certain markets, I believe that it needs to be know n what advertising content exists before further research can be conducted. There have been st udies conducted in North America, Europe and

PAGE 10

10 Asia dealing with standardization (Backhaus et al., 2001; Mueller, 1989; Ramaprasad & Hasegawa, 1992; Solberg, 2002; Whitelock and Chung, 1989), but there is very little research on international advertising in Latin America. Befo re any research can be done to determine which techniques work in Latin Ameri ca, or if any techniques would be more effective, something needs to be known about the curre nt techniques used in that area of the world. A step to understanding the current techniques will be knowledge of the extent of standardization used in Latin America. Costa Rica Costa Rica, known for its bananas, coffee, beach es, and rainforests, is Central America’s richest country (Ueltschy & Castillo, 2005). With a constitution in place since 1949, and a stable democracy since 1948, Costa Rica has enjoyed st eady economic growth over the past several decades (Country Profile, 2006; Risk Summary, 2006). In term s of economy, Costa Rica has high GNP, GNP per head, and exports compared to other Latin American countries, as well a low external debt. Tourism is one of the most vital parts of Costa Rica’s economy, with both tourist arrivals and currency fl ow from tourism steadily increa sing every year (Country Profile, 2006). This small country, which is about the si ze of West Virginia (Vines, 2005), has no army due a clause in its constitution a nd has not seen any civil unrest or violence, as well as has lower crime rates than other Latin American countri es. Costa Rica spends money that would go towards an army on its sophisticated education system. The literacy rate is upwards of 96%, which is one of the highest in Latin America. In addition to advanced education, Costa Rica has advanced telecommunication, with a higher density of telephone lines and high-speed internet than most other Latin American countries (Country Profile, 2006). As of now, te lephone and internet are run by the government, but they are set to open up competition for th ese services by the end of 2006, and for mobile

PAGE 11

11 phone service by the end of 2007 (Risk Summary, 2006) . Mass media are also highly developed, with high newspaper circulation a nd a variety of local radio and te levision stations. There are also a variety of cable stati ons, one which offers over 150 inte rnational channels (Country Profile, 2006). Advertising, especi ally television, print, radio, and billboard advertising, are developed and available (U eltschy & Castillo, 2005). For this study, Costa Rica was chosen to re present Latin America not only because it is a prime example of a stable Latin America econom y, but also because of the country’s love of United States citizens and United States products. Again, tourism is an important part of the Costa Rican economy, and also an im portant part of thei r lifestyle. There is a profusion of United States and international products, as well as retail distribution similar to that of the United States. Consumers in Costa Rica are brand c onscientious and view products from the United States as high quality, and are ge nerally willing to pay a little more for that quality (Ueltschy & Castillo, 2005). Purpose The purpose of this study is to determine th e amount of standardiza tion in international advertising in Costa Rica. In addition, th is study aims to determine which executional advertising elements are most often standardi zed in Costa Rica. Following is a review of literature that has helped to develop the th eories behind the research questions and the hypotheses. The literature review provided a basis for the ideas behind the research questions and hypotheses. The method used to answer the re search questions and test the hypotheses was a content analysis where 101 pairs of advertisements from the United States and Costa Rica were compared based on executional elemen ts suggested by reviewed literature

PAGE 12

12 CHAPTER 2 LITERATURE REVIEW This review will first discuss the schools of thought on international advertising. It will then move to a more in-depth discussion on th e interpretations of st andardization, including a case study on the convergence process and several m odels that have been proposed to determine standardization. Finally, the revi ew will lead to past suggestions that inspired the research questions and hypotheses of this study. Standardization, Adaptation, and Convergence According to Agrawal (1995), there are th ree schools of thought on the subject of international advertising: standardizati on, adaptation, and the convergence of the two. Those in favor of standardization feel that differences between count ries only vary in the degree to which the consumers differ, and theref ore advertisers should focus on the similarities instead of the differences (Agraw al, 1995). Here, the idea is that needs and wants are universal (Harris, 1996; Harvey, 1993; Onkvisit & Shaw, 1987) and “the same advertisement will work in any national market” (Harris, 1996, p. 551; Ryans, 1969). This school believes that the benefits of standardization lie within cost reductions (Agrawal, 1995; Backhaus et al., 2001), company image, and building an interna tional brand (Agrawal, 1995). Those in favor of the traditional approach of adaptation feel that differences such as product life cycle, media availability, legal rest rictions, culture, and economics in each country advertised in must be consid ered (Agrawal, 1995; Onkvisit & Shaw, 1987). Local advertising programs should be developed in each national ma rket to better meet the local needs and wants (Harris, 1996). De Mooij (2003 ) suggests that considering cu ltural values is essential for successful marketing. This school generally ha nds the blame of international advertising mistakes to not adapting advertising campaigns (Agrawal, 1995; Mueller, 1989).

PAGE 13

13 The final school of thought discussed by Ag rawal (1995) is a moderate convergence between standardization and adaptation. Those in favor of convergence feel that the most successful advertising campaigns will come from combining the two strategies after carefully evaluating the foreign situations, with the possibi lity of each situation differing greatly (Agrawal, 1995). Some elements are set to be standardized by the company, while other elements are left to the discretion of those responsible fo r each national market (Harris, 1996). According to Backhaus et al . (2001) and Harvey (1993), st andardization can result in reduced costs and efficiency gains within a firm . Image confusion and media spillover can be prevented in increasingly geographically mob ile consumers (Backhaus et al., 2001; Harvey, 1993). Harvey (1993) also adds that standardization allows fo r preserving a consistent image and sustaining planning and control across market s. Domzal and Kernan (1994) say that, in order for a brand to be global, it should have messages and advertisements that travel well. However, standardization decreases a firm’s ab ility to tailor the campaign to each country’s target consumers and their specific behaviors (Backhaus et al., 2001). In addition to this negative aspect of standa rdization, Harvey (1993) brings to li ght that though the standardization of advertisements may bring abou t the appearance of a consolid ated market, the illusion of a single market may harm the communications strate gy as well as the domesti c branches of larger firms. Solberg (2002) suggests that the more a firm knows about each local market that its products or services are a part of, the more they can look at the similaritie s and therefore standardize the advertising among them. This suggestion takes th e idea of localization and puts it into the perspective of standardization. If a firm looks at the individual mark ets’ cultures and legal restrictions, among other variables, it can find the similarities between them and design

PAGE 14

14 campaigns as to fit the markets as a whole. In 2000, Coca-Cola’ s CEO said that Coke’s biggest successes are in markets where they adapt thei r marketing and advertising (De Mooij, 2003). Brief History In the 1950s, adaptation of inte rnational advertisements seem ed to be the most popular train of thought. Both advertising practitioners and academicians thought that adaptation would make the industry more successful, due to the me ssage being meaningful for local people, lesser chance of offensiveness, and to appeal to the desires and behaviors of the local market. The 1960s took on a more confused atmosphere, with practitioners increasin g standardization and academicians beginning to follow a contingency approach, tailoring their decisions to the situation of the product, markets, and industry (Agrawal, 1995). Europe, during this time, saw high standardization (Ramaprasad & Hasegawa, 1992) . In the 1970s, practitioners began to turn back to adaptation and the academicians went back and forth between adaptation and contingency (Agrawal, 1995). In this decade, there was a movement towards realizing the cultural differences between countries and ma rkets (Ramaprasad & Hasegawa, 1992). The 1980s have the practitioners again changing th eir ideas and turning towards a trend in standardization, while the academicians begin to veer away from contingency and towards adaptation (Agrawal, 1995). This resurgence of the debate in the 1980s brought about a new desire for response and research (Ramaprasad & Hasegawa, 1992). Currently, there seems to be a trend towards convergence, which is standardization with the chance for some elements to be localized, though there is much debate and study as to what should be standardized, and to wh at degree. It should be kept in mind that today, compared to the late 1970s, it is much more feasible to crea te advertising at one f acility and disperse the resulting products to internati onal markets (Agrawal, 1995). Firm affiliates and regional offices are now able to understand the globa l nature of their parent firm’s business and are more likely

PAGE 15

15 to understand the importance of a global marketi ng image for their business, while still keeping locality in mind (Hanni, Ryans, & Vernon, 1995). Interpretations of Standardization It seems that research on the st andardization of advertising is heading in the direction of deciding which elements of advertisements can be standardized and/ or localized for the campaign to be considered standardized (Backha us et al., 2001; Harvey, 1993). Harvey (1993) points out that any promotional effort will probably ha ve to be adapted to some point, even if this means merely translating the language, and the issue has become not if to standardize other elements, but how much needs to be standardized. Goodyear developed a programmed manage ment approach in the 1970s before implementing a strategically converged advertis ing campaign in Europe, following the idea that the question was not if, but how much to standardi ze. This approach followed seven principles, which are still applicable to st andardization practices today (Ha nni et al., 1995). According to Hanni et al. (1995), the first principle involves kn owing the market, because local representatives are more likely to understand local markets than those working at headquarters. The second principle entails personnel between countries and markets being familiar with one another. The third principle states that the more personnel trav el to the local markets, the better they will understand the markets and the personnel represen ting those markets. The fourth principle encourages companies to use a networked advertis ing agency, because an agency with offices in the same countries that a company has products in is likely to have an understanding of those markets as well. An extension of the previ ous, the fifth principle emphasizes communication with foreign advertising account executives, as communication is vital to a successful campaign. In order to set the strategy, the sixth principle says that campaigns should have long lean planning times, and should be plan ned at least twelve months in advance. The final principle

PAGE 16

16 deals with budget, stating that the home office should mainta in budget approval authority in order to track spending in one place. Organization and Decision Making Centers Solberg (2002) proposes that a determinant fact or of standardization is the organization of the firm and where its decision making takes pla ce. According to Harri s (1996), the decision of determining international advertising policy fo r multinational companies is challenging and complicated, while Hanni et al. (1995) stress th e importance of standardization efforts being made in such a way that reduces concerns in local markets. Solberg (2002) describes four positions through which firms govern their decision making and marketing processes. The first of these is the local ba rony, where firms leave most of the de cisions up to local representatives rather than up to headquarters, and little about the local market conditions are known by headquarters. The next position is civil war, where firms’ headquarters have little knowledge about local market conditions, yet make decision s through headquarters. Following civil war is confederation, where there is local market knowledge at the firms’ headquarters and local representatives make a majority of the marke ting decisions. The fina l position is federation, where the firms’ headquarters have both local market knowledge and make a majority of the decisions. As predicted by Solberg (2002), local baronies adapt more then any other firm position. Surprisingly, Solberg found that firms in this position standardize their advertising more than only civil war groups, meaning they adapt their a dvertising more than the other two positions. Solberg feels that this is due to a lack of local market knowledg e at the headquarter level and a high level of decisive freedom for local representatives. Confederation and federation positions follow local baronies and civil wa r in level of adaptation, respec tively. In both of these cases, headquarters have a significant amount of local market knowledge, and may have a better

PAGE 17

17 understanding of what global techni ques will be most effective in the range of markets in which they are doing business. Case Study: Goodyear in Latin America In the 1970s, Goodyear implemented an adve rtising campaign in Europe that was a convergence of standardization and adaptation. Goodyear decided to implement a similar convergence strategy to a campaign in Latin Am erica in the 1990s, and the following is the process that Goodyear took in order to deve lop an the most eff ective combination of standardization and adapta tion (Hanni et al., 1995). To begin with, Goodyear believed that their produ cts were similar in each market and that their consumers’ needs were similar in each mark et as well. Based on these beliefs, Goodyear did not wish to abandon the idea of a global them e, yet they still wanted to incorporate views from the individual markets. They began with a Preliminary Orientation in September of 1992, where outlines of the expectations of a regional approach, includi ng cost-benefits, were laid out. In October 1992, A Regional Communications St rategy Definition Meeting was held with personnel from both the company and the advertisi ng agency (Leo Burnett) from each country, from which communications objectives were set and a creative assignment was given to the advertising agency. An Adve rtising Creative Review Meetin g in November 1992 produced a series of questions that needed to be answ ered about consumer attitudes and end-consumer reactions, which were researched throughout N ovember and December 1992 to find consistent results across all countries i nvolved. A Research Review Mee ting in January 1993 ended with an output of a clearly defined consumer propos ition, or the reason why consumers will change their beliefs about the products. The Final Cr eative Review in Ma rch 1993 consisted of presentations of several team concepts for the campaign, from which one was unanimously accepted (Hanni et al., 1995).

PAGE 18

18 From this point, the general campaign idea was developed into television commercials recorded with only visuals, leaving the firms in each country to add their own voice-overs and visual overlays when needed. This allowed for the local market concerns to be accounted for in terms of slight language differen ces. At the time the article was written, the campaign had not run long enough to evaluate its success in terms of sales effectiveness, nor have any follow-up articles been found, but Goodyear was pleased w ith the organization of their standardization strategy, and there have been visible positive eff ects from the commercials (Hanni et al., 1995). This case is important to the current research due its step by step analysis of the strategy put in place by Goodyear for a convergence of internationa l advertising, especially in Latin American countries, for which there is not a suffici ent amount of advertising research. Models of Standardization Harvey’s (1993) Model Harvey (1993) found that there was absence of a model to operationalize the analysis for the decision of whether to use st andardization or localiz ation. According to Harvey, advertisers seem to assume that the decision of standardizat ion versus localization is only relevant to the final advertisement, rather than to the whole ad vertising process. The model proposed a system that would allow advertisers to make a decision about which elements of the advertising process should be standardized, rather than just the elem ents that make up the final product. The model is based on six components: product variables, competitive variables, organizational experience and control variables, in frastructure variables, governmental variables, and cultural and societal variables. These components are measured in terms of elements of th e advertising process: research and development, creative, media, production, and post-advertising research. In Harvey’s model, the degree to which an adver tisement should be standardized depends on the points acquired through the cross measurement of the aforementioned co mponents and elements,

PAGE 19

19 with the higher the score meaning the higher the likelihood that sta ndardization would be relevant for the advertisement and/or product, as well as which elements should be standardized. Whitelock and Chung’s (1989) Model Whitelock and Chung’s (1989) model was developed to make comparisons of advertisements used in different countries. Though the study used advertisements from the United Kingdom and France, the model can and shoul d be applied to advertisements from other countries as well. Elements of an advertisemen t are taken into account: picture, general layout, caption, size, color, and explanat ory text, which are a ll assumed to influence the perception of standardization, and scores are acquired according to each criteria. It was found that the picture was the most significant element and general layo ut the least. While conducting this study, the authors found that there were three categories of standardizat ion, which happen to follow the three schools of thought discusse d earlier: complete standard ization, no standa rdization, or partial standardization. Mueller’s (1989) Model Mueller’s (1989) model also looks at the elemen ts that are likely to be standardized in international advertising. This model takes Am erican products that are advertised in Germany, Japan, and the United States and analyzes the advert isements based on criteria that is believed to influence the perception of similarity: packag ing, product name, version of product, theme, headline, slogan, body copy, subhead, spokespers on, visual/background scenes, and attributes highlighted, with the non-United St ates advertisements being compared against the United States advertisements. As with the Whitelock and C hung (1989) model, this model is meant to be applicable for use with other countries than thos e used in the study. Points were given for each criterion, with the total amongst all criteria being the score fo r each advertisement. Though no fully localized advertisements were used am ong 46 total campaigns, it was found that only 2

PAGE 20

20 were fully standardized, with the remaining 44 being a convergence between the two strategies. In addition, it was found that product name and packaging were the elements most often standardized, while attr ibutes highlighted and theme were less likely to be standardized. Backhaus, Mhlfeld, and van Doorn (2001) The study conducted by Backhaus et al. (2001) evaluates the elements considered to be important in consumer perceptions of sta ndardization using conjoint analysis and multidimensional scaling. The interest in this study came with the realization that no studies or models had been formed up to this date that involved consumer participation. Therefore, a sample was drawn for this study and participants were asked to judge their impressions of the similarities between objects in advertisements without a predetermined criteria set for ranking. Four advertising elements were included in the study: picture, general la yout, advertising topic, and language. Each advertising topic focused on a different attribute of the same brand. Though this study was conducted in only three countries, a nd therefore its results cannot be generalized for every country in the world, it was found that picture “plays by fa r the most significant role in determining perceptions of sim ilarity” (p. 58), with the other advertising elements not being nearly as important. In other words, when resp ondents were asked to determine which elements were similar among a group of advertisements, pi cture was the most significant element, which implies that the visual aspect of a campaign is the most important element to be standardized. Culture Culture is an important part of every count ry, and can be important in terms of the advertising produced in a country. Culture is important in term s of determining the wants and needs of consumers in each country, along with which messages and methods are effective in each country. Culture can be, and often is, defined in many ways. For the purposes of this review, culture is defined as the learned behaviors, beliefs, laws, customs, symbols, and habits

PAGE 21

21 shared by the members of a society (Mueller, 2004). Two key terms of this definition are learned and shared by the members of a society, or in th e case of this study, the members of a country. Dimensions of Culture Geert Hofstede (1983) framework of cultural di mensions has proved to be very important in terms of not only understanding culture, but also in terms of a pplying culture to international studies in many different fields. Hofstede’s original studies incl uded four dimensions of culture, with a fifth added later, with which countries were given a score from 0-100 for each. These scores were meant to show the difference betw een each country and still are often used to compare the relative cultural differences betw een countries (Mueller, 2004). The five dimensions are as follows (Hofstede, 1983; Mueller, 2004). Power distance (PDI) is based on the amount of equali ty between members of society in a country. High power distance implies acceptance of autocratic leadership and authority. Low power distance implies a negative view of au thority, as well as equality among members. Individualism (IDV), and its opposite, collectivism, are compared based on how much a society places importance on the group rather th an the individual. High individualism implies that the individual is important, and that laws an d rights are intended to pr otect the individual. On the other hand, low individualism implies that strong group ties, such as families and social ties, are important. Masculinity (MAS), and its opposite, femininity, ar e compared based on differences in gender roles and the values associated with each gender. High masculinity implies assertiveness, and a bigger difference between male and female . Low masculinity shows less of a difference between sexes, and often shows more nurturing and caring among both genders.

PAGE 22

22 Uncertainty avoidance (UAI) refers to how a country feels about uncertainty and ambiguity. High uncertainty avoidance implies a need for rules, expert s, and avoidance of conflict and competition. Low uncertainty avoidance implies tolerance of different views, as well as little threat coming from competition and conflict. Long-term orientation (LTO), and its opposite, short-term orientation, is based on the idea of tradition and commitment. High long-term or ientation implies a high importance of tradition and respect for commitment, where low long-term orientation implies a more rapidly changing culture with less impor tance put on tradition. United States versus Costa Rica The following comparison is based on Hofstede ’s (1983) dimensions of culture. The scores are based on a scale from 0-100, with 50 be ing the average score. There is no available long-term orientation score for Costa Rica, and therefore the c ountries will only be compared based on the first four dimensions (Figure 2-1). In terms of power distance, Costa Rica has a score of 35 and the United States a score of 40. This shows that both countries have a lower than average PDI, which suggests that both countries have a more negative view of authority and strive to have equality among citizens. Power distance is the dimension for which these two countries are the most similar. For individualism, Costa Rica has a score of 15, and the United States a score of 91. This shows that Costa Rica has low IDV and places em phasis on the group, especially the family. On the other hand, the United States has high IDV, wh ich suggests the importan ce of the individual, even if that means alienating the group. This could suggest that advert ising in Costa Rica may have a greater focus on groups and social activity, while the United States may be more likely to show individuals in advertisements.

PAGE 23

23 In terms of masculinity, Costa Rica has a scor e of 21, while the United States has a score of 62. Costa Rica has relatively low MAS, which suggests that there is not a great difference between gender roles. Women are likely to make almost the same amount of money as men, and both genders are likely to be nurturing and carin g. It is possible that there is a correlation between countries with low IDV and low MAS. The United States, however, has an above average MAS, which suggests that there is a re lative difference between gender roles and that men are more likely than women to work and make more money. In add ition, this suggests that even women in the United States are more asse rtive than women in Costa Rica. The comparison of these scores could suggest that advertising in the United States is more likely to show each sex in traditional roles, i.e. the woman clean ing, cooking, or taking care of children. Finally, for uncertainty avoidance, Costa Ri ca has a score of 86 and the United States a score of 46. Costa Rica has high UAI, which suggests that competition and conflict among members of the society are avoided if possible. In addition, it is more acceptable for people to show emotion, and to tension in ways other than confrontation, such as using their hands when talking or driving aggressively. The United St ates has just below average UAI, which suggests that the society is more likely to accept others’ views and are not bothered as much by conflict. Figure 2-1. Hofstede’s Dimens ions of Culture scores for United States and Costa Rica

PAGE 24

24 Past Suggestions Leading to Present Research According to Zou (2005), there has been much research in the field of international advertising, especially related to standardization, in the last fo rty-five years, but there is no integrative framework encompassing and synthe sizing the literature produced that would advance this knowledge to the next level. The models discussed above can be applied to more countries (Backhaus et al., 2001, Harvey, 1993, Muelle r, 1989), especially countries outside of Europe (Backhaus et al., 2001). Also, a study an d/or model encompassing the elements in the previously discussed model can be explored (B ackhaus et al., 2001). In addition, more case studies of companies that have implemented a standardization strategy could be useful to researchers and professional firm s alike (Hanni et al., 1995). In order to see where research in standardization of intern ational advertis ing is going, it must be seen where it has been, both in terms of where the research has taken place and in terms of the type of research. Taylo r’s (2005) content anal ysis of papers dealin g with international advertising published in the Journal of Advertising from 1965-2004 is a base for what has happened in international a dvertising research. The Journal of Advertising is one of the top journals relating to advertisi ng and advertising research. In the first 29 years of the Journal of Advertising (from 1965-1994), 29 papers on international advertising were pub lished, a vast majority of which were conceptual rather than empirical. Eleven of these papers dealt with the general subject of global advertising, seven dealt with advertising in Europe , five in Asia, and four in developing countries. From 1994 to 2004, 32 papers had been published on in ternational advertising in the Journal of Advertising , a majority of which were empirical rather than conceptual, with 34% using content analysis, followed closely by experiments and surveys. A majo rity of the papers dealt with advertising in Asia, followed by Europe, developing countries, and the general topic of global advertising

PAGE 25

25 (Taylor 2005). The focus changed from largely global to more specialized by continent, and shifted towards Asia, which may be viewed as de veloping more technologically than other areas of the world. Also shifting is the focus of th e research from conceptual to empirical, which implies that, as the theories involving internatio nal advertising have evol ved, researchers are now looking more into what exists, and why. Taylor (2005) suggests that there is a need for research involving the measurement of cultural dimensions, which explores how advertising impacts countri es with different cultures. Taylor (2005) also raises the need for in creased interaction am ong scholars and between practitioners and academicians, wh ich is supported by Agrawal’s (1995) article that lays out the differences between these groups over the past half century. In Taylor (2 005), it is shown that a great majority of papers published on the topic of standardization deal with Europe and Asia. Developing countries as a whole are mentioned, but nothing of great note (Taylor, 2005). From this, it seems that international advertising rese arch can be expanded to include Latin America. Hanni et al. (1995), in their discussion of the Goodyear campaign in Latin America, are a step in the right direction in terms of showing convergence of st andardization and adaptation in practice, especially in a little -researched global region. Further documentation of cases will not only act as an aid to companies in similar situat ions, but will also aid in the ongoing research of standardization in international advertising. Research Question and Hypotheses Studies in the United States, Europe, and Asia suggest that international advertisements are generally a convergence between standardization and adaptation, with more advertisements being completely standardized than co mpletely adapted (Backhaus et al., 2001; Mueller, 1989; Ramaprasad & Hasegawa, 1992; Solb erg, 2002; Whitelock and Chung, 1989).

PAGE 26

26 Though there seems to be little research on th e relationship between product category and standardization in international advertising, Ry ans (1969) points out that product category must be controlled when performing advertising resear ch, because the character of the product has a lot to do with how it is advertised. Ryans (1969) suggested that differe nt product categories will have different amounts of standardization. However, Cutler and Javalgi implicate that standardization is not “dependent upon product category” (p. 76). RQ1: Will print advertisements for brands and product categories advertised both in America and Costa Rica be more standardized or more adapted? RQ2: Will advertisements compared by brands or by product categorie s advertised in both America and Costa Rica be more standardized? RQ3: Which product category will be the most standardized for pr int advertisements advertised both in America and Costa Rica? Whitelock and Chung (1989) found picture to be the most significant standardized element in a study using advertisements from the United Kingdom and France. Bac khaus et al. (2001), in a consumer participation study, found picture to be the most significant element in determining perceptions of similarity. H1: Picture will be the most often standardi zed element in the overall sample of print advertisements advertised both in the United States and in Costa Rica. H2: Picture will be the most often standardized element in print advertisements for brands advertised both in The United States and in Costa Rica. H3: Picture will be the most of ten standardized element in print advertisements for product categories advertised both in Th e United States and Costa Rica. RQ4: Which element will be the most standardized for each individual product category for print advertisements advertised both in The United States and Costa Rica?

PAGE 27

27 CHAPTER 3 METHODOLOGY Onkvisit and Shaw (1987) suggest that the Unit ed States should be the point from where standardization is based, and therefore the standa rdization of advertisements in other countries should be measured according to th e United States market. This research aims to determine the extent of standardization of a dvertising in Latin America, name ly Costa Rica, and will do so by comparing magazine advertisement from Costa Rica to magazine advertisements from the United States using content analysis. Print advertisements in magazines were the chosen medium due to ease of comparison, as well as prin t being one of the most effective mediums in Costa Rica (Ueltschy & Castillo, 2005). Sampling Design 19 magazines of varying themes from AprilJune 2006 were collected from Costa Rica based on the top 50 magazines in the United Stat es in terms of circulation (MRIPocketPiece, 2006). 20 corresponding magazines were collected from the United States ranging from JuneOctober 2006. Magazines were chosen based on availability. Table 3-1 shows the magazines chosen. Table 3-1. Magazines from which samples were taken United States Costa Rica Men's Health Men's Health Shape Linea Bon Appetit Health & Fitness Family Circle Sabores Fit Pregnancy Padres e Hijos Mom & Baby Vanidades Vanity Fair Marie Claire Marie Claire Bazaar Bazaar Cosmopolitan Cosmopolitan Buen Hogar Good Housekeeping Vogue Vogue Glamour Lucky Surfos Glamour Maxim

PAGE 28

28 Table 3-1. Continued. United States Costa Rica Surfer PC World Maxim National Geographic PC World Newsweek National Geographic Casa Vida Newsweek El Mueble Veranda A total of 1,578 advertisements were cons idered for comparison, with 1,091 from the United States, and 487 from Costa Rica. The original sample of 1,578 advertisements was chosen on the requirements that the advertisements were at least a full page, were for international products, and were not classified advertisements. Of these advertisements, 51 from each country were compared based on brand and product category, and 50 from each country compared based on product category only. The se t of 51 advertisements was chosen because they were the only advertisemen ts that matched brand for brand as well as product category for product category. The set of 50 advertisements was chosen randomly from the remaining advertisements for international brands. For the first set of comparisons, two advertisements of the same brand and same product category we re compared, one advertisement from each country. For the second set of a dvertisements, two advertisements of the same product category, but different brand, were compared, one from each country. Variables Advertisements were compared across the tw o countries based on the following elements: theme, layout, size, picture, background, head line, explanatory copy, slogan, attributes highlighted, and logo or brand symbol. With the exception of theme, layout, and size, all of the variables were looked at in terms of language , content, and placement, making a total of 21 coded variables. Each set of advertisements was given an index score from zero to 21, with a score of zero meaning complete adaptation and a sc ore of 21 meaning comp lete standardization.

PAGE 29

29 The elements were chosen based on past re search (Backhaus, et al. 2001; Mueller 1989; Whitelock and Chung 1989). The following definiti ons were given to coders as a codebook to use for reference (Table 3-2). The coding sheet used by the coders can be seen in Appendix A. Table 3-2. Operational definitions of coded variables Variable Definition Language Refers to the language of the words in the advertisement. Content Refers to the actual subject matter of the element. For example, for headline content to be the same, the ac tual words must be the same in each compared advertisement. Placement Refers to the positioning of the element in the advertisement’s layout. Theme The overall look and subject of an advertisement. Layout The overall placement of the content in an advertisement. Size Number of pages covered by the advertisement. Picture The central and featured picture in an advertisement. Background The background images or colors in an advertisement Headline The text that is set apart from the rest of the text in an advertisement. Explanatory Copy The text in an advertisement that explains the product, brand, or advertisement. Slogan A phrase or statement that is meant to be memorable and often makes a claim about the product or brand. Attributes highlighted The product or brand features that are mentioned in the ad. Logo/Brand symbol The graphic element or icon representing a brand. (Backhaus et al., 2001; Mueller, 1989; Whitelock & Chung, 1989) Advertisements were also categorized into pr oduct category types (Tab le 3-3) in order to determine standardization of pr oduct categories and which of the above elements are most often standardized per product category. Table 3-3. Product category types Product Category Toiletries & Beauty Aids Jewelry & Accessories Clothing & Shoes Technology & Communication Automobiles Food & Beverages Miscellaneous

PAGE 30

30 Coding Procedure Advertisements were coded by two undergraduat e advertising students. Advertisements were distributed evenly among the coders, with a 10% overlap for reliability purposes. Each coder participated in a training session where they reviewed the code boo k and instructions and coded an example advertisement with the main researcher. Intercoder Reliability The following formula, from Holsti (1969) was us ed to establish inte rcoder reliability: 2M Reliability = _________ N1 + N2 M is the total number of responses that the coders an swered the same, and the Ns represent the number of responses of each code r, respectively. The coders analyzed the same 20 sets of advertisements representing 10% of the total samp le. Intercoder reliability was found to be 94.2%. Data Analysis Data was analyzed using SPSS 14.0 for Windows, wh ich is statistical analysis software. SPSS was used to calculate frequencies, cross tabul ations, and analyses of variance. Chi-square tests were also conducted in addition to the cro ss tabulations in order to determine significant differences between variables.

PAGE 31

31 CHAPTER 4 RESULTS This study used content analysis to determine if print advertisements in Costa Rica were standardized towards the United States or adap ted to Costa Rican culture, and to determine which elements were the most of ten standardized. The content analysis included a total of 101 pairs of advertisements, 51 being compared based on brand and product category, and the remaining 50 being compared on product category al one. Each set of advertisements was given an index score from zero to 21, based on the number of standardized elements in each pair, with the mid-point of the scale being 10.5. Therefore, advertisement pairs with an index score over 10.5 will be considered more sta ndardized, and those with an i ndex score less than 10.5 will be considered more adapted. Standardization versus Adaptation The first research question explores if the advertisements in the sample are more standardized or adapted. In or der to determine if the advertis ements were standardized or adapted, the mean index score for all 101 pairs wa s calculated, as well as the mean index score for each comparison type. Each group’s mean i ndex score was higher than 10.5, with the mean index score for the overall sample being 12.96 (F igure 4-1), the mean for brand comparison being 14.73 (Figure 4-2), and the mean for produc t comparison being 11.16 (Figure 4-3). To determine the significance of these findings, a one-sample t-test was performed for each group, where the mean index scores of the overall sa mple and brand comparisons were found to be significant, but the mean index score of product category comparisons was not. (Table 4-1). The histograms of standardization among th e overall sample, brand comparison, and product category comparison illustrate the tendencie s of the index scores (Figures 4-1-3).

PAGE 32

32 Figure 4-1. Histogram of standa rdization among overall sample Figure 4-2. Histogram of standa rdization among brand comparisons Figure 4-3. Histogram of standardiz ation among product category comparisons

PAGE 33

33 Table 4-1. T-test results for st andardization versus adaptation Comparison Type t d.f. p MeanMean Difference Overall Sample 5.37 100.0012.962.46 Brand Comparisons 7.29 50.0014.734.23 Product Category Comparisons 1.07 49.2911.16.66 Standardization versus Comparison Type The second research question explores whic h comparison type, brand comparisons or product category comparisons, are more standardi zed. In order to determine which comparison type was more likely to be standardized, br and comparison or product category comparison, and to determine if there was a significant difference, an analysis of varian ce (ANOVA) test was run. The standardization of brand comparison (M = 14.73) was significantly more than the standardization of product category comparison (M = 11.16, p < 0.01) (Table 4-2 and 4-3). The means plot of standardization by type of co mparison illustrates the significant difference of standardization between brand comparisons and product comparisons (Figure 4-4). Table 4-2. Descriptive statistics for standardization by type of comparison Type of Comparison Mean Std. DeviationN MinimumMaximum Brand Comparison 14.73 4.1451721 Product Category Comparison 11.16 4.3850318 Total 12.96 4.60101321 Table 4-3. ANOVA results for standa rdization by type of comparison Sum of Squares d.f. Mean SquareF p Between Groups 320.97 1320.9717.68.00 Within Groups 1796.88 9918.15 Total 2117.84 100 Figure 4-4. Means plot of standardization by type of comparison

PAGE 34

34 Standardization versus Product Category Type The third research question e xplores the product category with the most standardization. Of the overall sample of 101 ads, 32.7% were consid ered to be for toiletries and beauty aids, 20.8% for jewelry and accessories, 16.8% for cl othing and shoes, 10.9% for miscellaneous goods or services, 7.9% for technology and communica tions, 5.9% for food and beverages, and 5.0% for automobiles (Table 4-4). Table 4-4. Product category within the overall sample Product Category (n=101) FrequencyPercent Toiletries and Beauty Aids 3332.7 Jewelry and Accessories 2120.8 Clothing and Shoes 1716.8 Miscellaneous 1110.9 Technology and Communications 87.9 Food and Beverages 65.9 Automobiles 55.0 In order to determine which product category wa s the most standardized, and to determine if there was a significant difference, an anal ysis of variance (ANOVA) test was run. Product categories shown to be more standardized, from most standardized to l east, were jewelry and accessories (M =17.43), clothing and shoes (M=16.18) , toiletries and beauty aids (M=11.76), and food and beverages (M = 11.0, p < 0.01) (Table 4-5 and 4-6). Product categories shown to be more adapted, from most adapted to leas t, were automobiles (M=8.0), technology and communications (M = 8.5), and miscellaneous (M = 9.64, p < 0.01). The histogram of standardization by product category type illust rates the amount of standardization of each product category (Figure 4-5). Table 4-5. Descriptive statistics for standardization by product category type Product Category Mean Std. DeviationN MinimumMaximum Jewelry & Accessories 17.43 2.50211121 Clothing & Shoes 16.18 2.30171018 Toiletries & Beauty Aids 11.76 3.9433421 Food & Beverages 11.00 2.616916

PAGE 35

35 Table 4-5. Continued. Product Category MeanStd. DeviationN MinimumMaximum Miscellaneous 9.644.8811321 Technology & Communications 8.502.518513 Automobiles 8.01.145411 Total 12.964.60101321 Table 4-6. ANOVA results for standa rdization by product category type Sum of Squares d.f. Mean SquareF p Between Groups 1069.62 6178.2715.99.00 Within Groups 1048.22 9411.15 Total 2117.84 100 Figure 4-5. Histogram of standard ization by product category type In order to determine the significant differe nce between product categories, a post-hoc Tukey test was run. There was no significant difference between jewelry and accessories and clothing and shoes, which verifies that they were the most standa rdized categories. There was, however a significant difference between those tw o categories and every ot her product category. In addition, the five least standardized categor ies were not significantly different from one another (Table 4-7).

PAGE 36

36 Table 4-7. Post-Hoc Tukey mean difference resu lts for standardization by product category type Toiletries Food JewelryTech.ClothingAutos Misc. Toiletries & Beauty .76 *-5.67 3.26 *-4.42 3.76 2.12 Food & Beverages -.76 *-6.43 2.50 *-5.18 3.00 1.36 Jewelry & Accessories *5.67*6.42 *8.93 1.25 *9.43 *7.80 Technology & Comm. -3.26-2.50 *-8.93 *-7.68 .50 -1.14 Clothing & Shoes *4.42*5.18 -1.25 *7.68*8.18 *6.54 Automobiles -3.76-3.00 *-9.43 -.50 *-8.18 -1.64 Miscellaneous -2.12-1.36 *-7.79 1.14 *-6.54 1.64 d.f. = (6,94), *p < .05 Standardized Elements versus Type of Comparison The three hypotheses expect that picture wi ll be the most often standardized element among the overall sample, brand comparisons, and product category comparisons. The first hypothesis tested standardization elements among the overall sample. In terms of frequency, it can be seen that the overall element most often standardized was logo la nguage (96.0%), slogan placement and attributes highlighted placemen t following (82.2% each). Background and main picture content are the least ofte n standardized (33.7% & 27.7%, respectively) (Table 4-8). The results suggest that picture is not the most often standardized element. Table 4-8. Standardized variables within the overall sample Variable (n=101) FrequencyPercentx2 Logo Language 9796.0 *41.83 Slogan Placement 8382.2 *41.83 Attributes Highlighted Placement 8382.2 *41.83 Size 8180.2 *36.84 Slogan Language 8079.2 *34.47 Main Picture Placement 7170.3 *16.64 Attributes Highlighted Language 7069.3 *15.06 Attributes Highlighted Content 6766.3 *10.78 Layout 6766.3 *10.78 Explanatory Copy Placement 6766.3 *10.78 Logo Content 6564.4 *8.33 Slogan Content 6160.4 *4.37 Headline Positioning 6059.43.57 Theme 6059.43.57 Logo Placement 5655.41.20 Headline Language 5251.5.09 Explanatory Copy Language 5251.5.09 Explanatory Copy Content 3837.6 *6.19

PAGE 37

37 Table 4-8. Continued. Variable (n=101) Frequency Percentx2 Headline Copy 37 36.6 *7.22 Background 34 33.7 *10.78 Main Picture Content 28 27.7 *20.05 d.f. = 1, * p < .05 The second hypothesis tested standardized el ements among brand comparisons. In terms of frequency, it can be seen th at logo language (98.0%) was the el ement most often standardized among brand comparisons, followed by attributes highlighted placement and slogan placement (88.2% & 86.3%, respectively). The least often standardized elements among brand comparisons were explanatory copy content and main picture content (41.2% & 39.2%, respectively) (Table 4-9). The results suggest th at picture is not the most often standardized element. Table 4-9. Standardized variable within the brand comparison sample Variable (n=51) FrequencyPercentx2 Logo Language 5098.0 *47.08 Attributes Highlighted Placement 4588.2 *41.83 Slogan Placement 4486.3 *26.84 Slogan Language 4180.4 *18.84 Logo Content 4180.4 *18.84 Explanatory Copy Placement 4078.4 *16.49 Size 4078.4 *16.49 Attributes Highlighted Content 4078.4 *16.49 Attributes Highlighted Language 3874.5 *29.82 Layout 3874.5 *12.26 Main Picture Placement 3874.5 *12.26 Theme 3772.5 *10.37 Slogan Content 3670.6 *8.65 Logo Placement 3568.6 *7.08 Headline Positioning 3466.7 *5.67 Headline Language 3364.7 *4.41 Explanatory Copy Language 3160.82.37 Background 2549.0.02 Headline Copy 2447.1.18 Explanatory Copy Content 2141.21.59 Main Picture Content 2039.22.37 d.f. = 1, * p < .05

PAGE 38

38 The third hypothesis tested st andardized elements among product comparisons. In terms of frequency, it can be seen th at logo language (94.0%) was the mo st often standardized element among product category comparisons, followed by size, slogan language, and slogan placement (82.0%, 78.0%, & 78.0%, respectively). The leas t often standardized elements among product category comparisons were headline language and explanatory copy content (38.0% & 34.0%, respectively) (Table 4-10). The results suggest that picture is not the most often standardized element. Table 4-10. Standardized va riables within th e product category comparison sample Variable (n=50) FrequencyPercentx2 Logo Language 4794.0 *38.72 Size 4182.0 *20.48 Slogan Language 3978.0 *15.68 Slogan Placement 3978.0 *15.68 Attribute Highlighted Placement 3876.0 *13.52 Main Picture Placement 3366.0 *5.12 Attributes Highlighted Language 3264.0 *3.92 Layout 2958.01.28 Attributes Highlighted Content 2754.0.32 Explanatory Copy Placement 2754.0.32 Headline Positioning 2652.0.08 Slogan Content 2550.0.00 Logo Content 2448.0.08 Theme 2346.03.20 Explanatory Copy Language 2142.01.28 Logo Placement 2142.01.28 Headline Language 1938.02.88 Explanatory Copy Content 1734.0 *5.12 Headline Copy 1326.0 *11.52 Background 918.0 *20.48 Main Picture Content 816.0 *23.12 d.f. = 1, * p < .05 In order to explore the significant differe nces between the overall sample, brand comparisons, and product category comparisons in te rms of the standardized elements, the data was cross tabulated by type of co mparison and the variable elements. Chi-square tests were then performed to determine if there was a significant difference in the frequency of standardized elements with relation to the comparison types. Out of the 21 variables, ten were significantly

PAGE 39

39 different. While explanatory copy placemen t and attributes highlighted placement (x2 = 6.75 each, d.f. = 1, p < .05) are the most significant standardi zed variables for the overall sample and product category comparisons, the same is not so for brand comparisons. For brand comparisons, explanatory copy placement and attr ibute highlighted placement fall just behind logo content (x2 = 11.55, d.f. = 1, p < .05). The least standardi zed element for the overall sample, product category comparisons, and brand comparisons was main picture content (x2 = 6.79, d.f. = 1, p < .05) (Table 4-11). Table 4-11. Cross tabulati on of type of comparison by standardized variables Variable Type of Comparison x2 Brand ComparisonProduct Category Comparison Total Logo Content 41 80.4% 24 48.0% 65 64.4% *11.55 Background 25 49.0% 9 18.0% 34 33.7% *10.88 Theme 37 72.5% 23 46.0% 60 59.4% *7.38 Logo Placement 35 68.6% 21 42.0% 56 55.4% *7.25 Headline Language 33 64.7% 19 38.0% 52 51.5% *7.21 Main Picture Content 20 39.2% 8 16.0% 28 27.7% *6.79 Explanatory Copy Placement 40 78.4% 27 54.0% 67 66.3% *6.75 Attributes Highlighted Content 40 78.4% 27 54.0% 67 66.3% *6.75 Headline Copy 42 47.1% 13 26.0% 37 36.6% *4.82 Slogan Content 36 70.6% 25 50.0% 61 60.4% *4.47 Explanatory Copy Language 31 60.8% 21 42.0% 52 51.5% 3.57 Layout 38 74.5% 29 58.0% 67 66.3% 3.08 Attributes Highlighted Placement 45 88.2% 38 76.0% 83 82.2% 2.58 Headline Positioning 34 66.7% 26 52.0% 60 59.4% 2.25 Slogan Placement 44 86.3%39 78.0% 83 82.2% 1.18

PAGE 40

40 Table 4-11. Continued. Variable Type of Comparison x2 Brand Comparison Product Category Comparison Total Attributes Highlighted Language 38 74.5% 32 64.0% 70 69.3% 1.13 Logo Language 50 98.0% 47 94.0% 97 96.0% 1.08 Main Picture Placement 38 74.5% 33 66.0% 71 70.3% .88 Explanatory Copy Content 21 41.2% 17 34.0% 38 37.6% .55 Size 40 78.4% 41 82.0% 81 80.2% .20 Slogan Language 41 80.4% 39 78.0% 80 79.2% .09 d.f. = 1, * p < .05 Standardized Elements versus Product Category Type The fourth research question explores which element is most often standardized for each product category. In order to explore the sign ificant differences betw een the product category types in terms of standardized elements, the data was cross tabulated by product category type and the variable elements. Chi-square tests we re then performed to determine if there was a significant difference in the frequency of standa rdized elements with relation to the product category type. Due to small sample sizes , the product categories of technology and communications, food and beverages, and autom obiles were recoded into the miscellaneous category. Out of 21 variables, 13 were signifi cantly different. For the product category of toiletries and beauty aids, the most significant standardized element was slogan language (x2 = 12.44, d.f. = 3, p < .05) and the least significant standard ized element was explanatory copy content (x2 = 17.77, d.f. = 3, p < .05). For the category of jewelry and accessories, the most significant standardized element was attributes highlighted language (x2 = 27.00, d.f. = 3, p < .05) and the least sign ificant standardized element was theme (x2 = 10.01, d.f. = 3, p < .05). For the category of clothing and shoes, the most signi ficant standardized elements were explanatory

PAGE 41

41 copy language, attributes highlight ed language, and slogan language (x2 = 51.77, 27.00, & 12.44, respectively, d.f. = 3, p < .05) and the least significant standardized element was background (x2 = 11.43, d.f. = 3, p < .05). For the miscellaneous advertisements, the most significant standardized element was slogan language (x2 = 12.44, d.f. = 3, p < .05) and the least significant standardized element was headline copy (x2 = 39.32, d.f. = 3, p < .05) (Table 4-12). Table 4-12. Cross tabulation of produc t category by standardized variables Variable Product Category x2 Toiletries & Beauty Jewelry & Accessories Clothing & Shoes Miscellaneous Explanatory Copy Language 9 27.3% 20 95.2% 17 100% 6 20.0% *51.77 Headline Language 11 33.3% 19 90.5% 16 94.1% 6 20.0% *41.41 Headline Copy 9 27.3% 19 85.7% 9 52.9% 1 3.3% *39.32 Attributes Highlight Language 17 51.5% 21 100.0% 17 100.0% 15 50.0% *27.00 Slogan Content 20 60.6% 18 85.7% 14 82.4% 9 30.0% *20.64 Attributes Highlighted Content 17 51.5% 19 90.5% 16 94.1% 15 50.0% *18.19 Explanatory Copy Content 8 24.2% 12 57.1% 12 70.6% 6 20.0% *17.77 Headline Positioning 17 51.5% 19 90.5% 13 76.5% 11 36.7% *17.74 Layout 20 60.6% 20 95.2% 14 82.4% 14 43.3% *17.40 Explanatory Copy Placement 20 60.6% 18 85.7% 16 94.1% 13 43.3% *17.00 Slogan Language 22 66.7% 20 95.2% 17 100.0% 21 70.0% *12.44 Background 15 45.5% 10 47.6% 6 35.3% 3 10.0% *11.43 Theme 21 63.6% 16 76.2% 12 70.6% 11 36.7% *10.01 Logo Content 22 66.7% 17 81.0% 12 70.6% 14 46.7% 6.98 Attributes Highlight Language 25 75.8% 20 95.2% 16 94.1% 22 73.3% 6.63 Main Picture Content 10 30.3% 9 42.9% 1 5.9% 8 26.7% 6.57 Main Picture Placement 20 60.6% 19 90.5% 12 70.6% 20 66.7% 5.77

PAGE 42

42 Table 4-12. Continued. Variable Product Category x2 Toiletries & Beauty Jewelry & Accessories Clothing & Shoes Miscellaneous Slogan Placement 27 81.8% 19 90.5% 16 94.1% 21 70.0% 5.68 Size 27 81.8% 19 90.5% 12 70.6% 23 76.7% 2.68 Logo Language 31 93.9% 21 100.0% 17 100.0% 28 93.3% 2.53 Logo Placement 20 60.0% 12 57.1% 10 58.8% 14 46.7% 1.40 d.f. = 3, * p < .05

PAGE 43

43 CHAPTER 5 DISSCUSSION AND CONCLUSION Discussion The purpose of this study was to determine th e amount of standardization of international advertising in Latin America, specifically Cost a Rica, as well as which executional elements were most likely to be standardized. Following are some insights that can be taken from the results of the data analysis. Standardization versus Adaptation The first three research que stions posed in this study dealt with the amount of standardization in the sample of advertisem ents. The first question asked whether print advertisements for brands and product categories a dvertised both in The United States and Costa Rica be more standardized or more adapted. Th e mean of the index score for the overall sample, as well as for brand comparisons alone and prod uct category comparisons alone, suggests that the advertisements in Costa Rica are generally more standardized to advertisements from the United States rather than adapted to the Costa Rican market. However, the results for product category comparisons were not significant, and th erefore those results cannot be generalized. The second research question asked whether pr int advertisements co mpared by brands or by product categories advertised in both The Un ited States and Costa Rica would be more standardized. The mean index score for each comparison suggests that brand comparisons were more standardized than product category comparis ons, and the significance of the analysis of variance test ve rifies this. The third research question aske d which product category will be the most standardized for print advertisements advertised both in The Un ited States and Costa Rica. The analysis of variance test showed that advertisements for jewe lry and accessories were the most standardized,

PAGE 44

44 followed by clothing and shoes. This could be du e to Costa Rican’s love for products from the United States, which are more likely to be lifesty le products such as jewe lry and clothing, rather than household products such as food (Ueltschy & Castillo, 2005). Auto mobiles are the product category with the least amount of standardiza tion, followed by technology and communications. In fact, these categories were more adapted to the local market. Because the advertisements in the sample were for international products, as Solberg (2002) suggests, standardization could be a resu lt of the advertising decisions made at the company’s headquarters rather than in Latin Amer ica. Or, the decisions could have been made by someone that is not highly knowledgeable about the Latin American mark et. Conversely, it is possible that the reason for sta ndardization is that similariti es have been found between the United States and Latin America, which justifies standardizing advertising (Solberg, 2002). Also to be kept in mind is that ther e is a current trend towards conv ergence of standardization with some adaptation where needed (Agrawal, 1995). The mean of the overall sample was not completely standardized, but was not adapted either. This suggests a convergence between the two. Standardized Elements versus Type of Comparison The three hypotheses tested in this study dealt with which el ements of an advertisement were the most standardized in Costa Rica. The first hypothesis, that pict ure would be the most often standardized element within the overall samp le of advertisements advertised both in The United States and in Costa Rica, was not supported. In terms of frequency alone, logo language was the most often standardized and main picture content the least. Afte r being subjected to a chi-square test to determine significance, it was shown that explanatory copy placement and attributes highlighted content were the most significantly standardized elements, and main picture content the least.

PAGE 45

45 The second hypothesis, that picture will be the most often standardized element in print advertisements for brands advertised both in Th e United States and in Co sta Rica, was also not supported. Again, in terms of frequency, logo la nguage was the most often standardized and main picture content the least. After being subjected to a chi-square test to determine significance, it was shown that logo content was the most significantly standardized element, and, again, main picture content the least. The third hypothesis, that picture will be th e most often standardized element in print advertisements for product cate gories advertised both in The United States and Costa Rica, was also not supported. Consistent with the overa ll sample and brand comparisons, in terms of frequency, logo language was the most often standa rdized element and main picture content the least. After being subjected to a chi-square test to determin e significance, it was shown that explanatory copy placement and attributes high lighted content were the most significantly standardized elements, and, again, main picture content the least. While picture has proved to be the most significant element in studies done in Europe and Asia (B ackhaus et al., 2001; Whitelock & Chung, 1989), it is clear that picture is not the most standardized element in Costa Rica. These findings are interesting, not only b ecause picture was found to be not often standardized, but also because logo language, in terms of frequency, was the most often standardized. This could relate to Costa Ri cans liking products from the United States, and therefore maybe the logo being in English, or at least in the same language as it is in the United States (some brand names or logos may be in French, for example), suggests to Costa Rican consumers a higher perceived quality of the product (Ueltschy & Castillo, 2005).

PAGE 46

46 Whitelock and Chung’s (1989) study found picture to be the most significant standardized variable among adve rtisements from the United Kingdom and France. This could be attributed to both countries being in Europe, and being fairly close in proximity. Mueller’s (1989) study found product name to be the most of ten standardized element in advertisements from the United States, Germany, and Japan. This is similar to logo being standardized, which may again be attributed to the products being in ternational. In addition, in the Mueller (1989) study, attributes high lighted was found to be th e least often standardi zed element, along with theme. In the current study, attributes high lighted was among the most often standardized element in the overall sample and in product ca tegory comparisons. It is possible that the countries used in the Mueller (1989) study were much more di fferent culturally than the countries used in this current study. This difference in findings could also be attributed to a difference in proximity of the countries involve d. In the Backhaus et al. (2001) study, picture was found to be the most important element in term s of perceptions of standardization. This may suggest that though picture is the most important as far as percep tions of standardization, it may not necessarily be the mo st often standardized. Standardized Elements versus Product Category Type The final research question asked which elemen t will be the most standardized for each individual product category for print advertisem ents advertised both in the United States and Costa Rica. Due to a small sample size for se veral categories, food a nd beverages, technology and communications, and automobiles were all r ecoded into the miscella neous category. Table 5-1 illustrates the most signifi cant standardized elements and least for each product category:

PAGE 47

47 Table 5-1. Standardized el ements per product category Product Category Most Standardized Least Standardized Toiletries & Beauty Aids Slogan language Explanatory copy content Jewelry & Accessories Attributes highlighted language Theme Clothing & Shoes Explanatory copy language Attributes highlighted language Slogan language Background Miscellaneous Slogan language Headline copy Though each product category had a different element that was the most adapted, each product category’s most standardized element had to do with language. Because the advertisements dealt with were either in Englis h or Spanish, this means that the advertisements with standardized language were in English. It is possible that, again, Costa Rican consumers’ love for products from the United States (Ue ltschy & Castillo, 2005) is an explanation for language being standardized. Perh aps, if an advertisement for a product is in English, it will have a higher perceived quality. Often, jewelr y, clothing, and beauty aids are very personal products, and products that give co nsumers a sense of identity. Consumers may be more likely to buy a product from these categorie s if they perceive the quality to be higher, in order to give them a higher sense of identity and pride. An example of a pair of advertisements for Armani Code, a cologne advertisement in the toiletries and beauty aids product category, can be seen in Appendix A. This pair of advertisements is c onsidered completely standardized, and therefore has the same slogan language, which was found to be the most standardized element for this product category. The slogan on this advertisement is “The ultimate code of seduction for men.” Perhaps the language of the slogan is standardized to show the American quality of the product, or perhaps to promote the iden tity of “ultimate seduction.” Conclusion Using content analysis, this study compared 101 pairs of advertisements from Costa Rica and the United States to determine the standardiz ation of international advertisements in Costa

PAGE 48

48 Rica. The results from this study show that a dvertisements in Costa Rica are overall more standardized than adapted, and that advertisemen ts compared by brands are more standardized than those compared by product cate gory. It can also be said that advertisements for jewelry and accessories, as well as for clothing and shoes, are more standardized than other product categories. As mentioned earlier , this could have something to do with the positive Costa Rican attitude towards products from the United States (Ueltschy & Castillo, 20 05). There were not any advertisements that were completely adapted, but there were several that were completely standardized. Examples of two of these advertisements can be seen in Appendix B. The results also show that the most often st andardized elements have to do with language, placement, and content of copy and logos rather than with picture content or placement. This is not what was expected, and does not follow rese arch of international advertising on other continents (Backhaus et al ., 2001; Whitelock & Chung, 1989). Limitations There are several limitations that should be accounted for in future studies. First, the advertisements in the sample only covered a six month span. Though standardization is a practice that seems to continua lly change (Agrawal, 1995), a longe r sample-frame time may have provided different results. Though the magazines in the sample were as representative as possible, they may not have been completely representative of the magazi nes read by most Costa Ricans. The magazines collected were based on the top 50 magazines in terms of circulation in the United States (MRIPocketPiece, 2006), but the top magazines in te rms of circulation and readership in Costa Rica, and in Latin America in general, may be different. The overall sample size was only 101 pairs of advertisements. A larger sample size may have been more representative of the advertisements in both countries.

PAGE 49

49 Suggestions for Future Research Now that there is a study whic h looks at the amount of standa rdization in Costa Rica, there are a few suggestions for future a nd subsequent research. First of all, this study only looks at one Latin American country. A study that takes sample s from several or a majority of other Latin American countries would not only show the am ount of standardization in comparison to the United States, but also in comparison to Costa Ri ca. This type of study could be furthered to compare the standardization of advertisements be tween Latin American count ries. If these types of studies were to replicate th e current research, more variable s with more general definitions would be suggested. Taylor (2005) suggested that re search in international advert ising be expanded past content analysis. This study could be used as a stepping stone. Now that the amount of standardization has been researched, it would be interesting to look further into why advertisements are standardized towards advertisements in the Un ited States. In additi on, because picture was significant in Europe and Asia (Backhaus et al., 2001; Whitelock & Chung, 1989), it would be interesting to look at why the same is not so for Costa Rica. In response to Solberg’s (2002) study on decisi on making centers, it may be worthwhile to look at who makes the decisions for advertisemen ts in Costa Rica, and in Latin America in general. Research in this area could be us eful to the professiona l advertising industry. Another interesting suggestion would be to incorporate cultu re into the explanation of standardization. Based on Hofstede’s (1983) di mensions of culture, countries are rated on several different dimensions in an attempt to provide a framework for cross-cultural studies. This cultural framework could be used to comp are the advertising betw een countries, and could possibly be used to answer the question of why cer tain elements are or are not standardized, as well as the advertisements overall.

PAGE 50

50 Finally, and quite possibly most importantly, future studies need to examine not only why advertisements are standardized in Costa Rica, but also whether or not sta ndardization is actually effective there. Advertising is meant to be effective, but there are those that argue that standardized advertising is just an easy way to save time and money. Are advertisements in Costa Rica standardized for ease or do they really work?

PAGE 51

51 APPENDIX A CODING SHEET CASE ID#______ BRAND COMPARISON______ PRODUCT CATEOGORY COMPARISON______ General Theme & Layout (Theme) V1. General Theme 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V2. General Layout 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V3. Size 1) One Page 2) Two Pages 3) Three Pages Picture V4. Main Picture Content 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V5. Main Picture Placement 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________

PAGE 52

52 V6. Background 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ Headline V7. Headline Language 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V8. Headline Copy 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V9. Headline Positioning 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ Explanatory Copy (Copy) V10. Explanatory Copy Language 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________

PAGE 53

53 V11. Explanatory Copy Content 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V12. Explanatory Copy Placement 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ Slogan V13. Slogan Language 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V14. Slogan Content 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V15. Slogan Placement 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ Attributes Highlighted (Attributes) V16. Attributes Hi ghlighted Language 1) Same 2) Different 3) Can’t Code/Not Applicable

PAGE 54

54 If Different, how is it different? ____________________________________________________ V17. Attributes Hi ghlighted Content 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V18. Attributes Hi ghlighted Placement 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ Logo or Brand Symbol (Logo) V19. Logo or Brand Symbol Language 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V20. Logo or Brand Symbol Content 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________ V21. Logo or Brand Symbol Placement 1) Same 2) Different 3) Can’t Code/Not Applicable If Different, how is it different? ____________________________________________________

PAGE 55

55 APPENDIX B EXAMPLES OF COMPLETELY STANDARDIZED ADVERTISEMENTS Figure B-1. Armani Code from Costa Rica

PAGE 56

56 Figure B-2. Armani Code from the United States

PAGE 57

57 Figure B-3. Damiani from Costa Rica

PAGE 58

58 Figure B-4. Damiani from the United States

PAGE 59

59 LIST OF REFERENCES Agrawal, M. (1995). Review of a 40-year debate in international advertising. International Marketing Review, 12 (1), 26-48. Backhaus , K., Mhlfeld, K. , & van Doorn, J. (2001). Consumer perspectives on standardization in international advertising: A student sample. Journal of Advertising Research, 41 (5), 53-61. Country Profile (2006). Count ry profile Costa Rica. Economist Intelligence Unit , Retrieved September 27, 2006, from http://portal.eiu.com.lp.hscl.ufl.edu/ report_dl.asp?issue_id=1200843905&mode=pdf Cutler, R.D., & Javalgi, R.G. (1992). A cross-cult ural analysis of the visual components of print advertising: The United States and the European Community. Journal of Advertising Research 32 (1) , 71-80. De Mooij, M. (2003). Convergence and diverg ence in consmer behaviour: Implications for global advertising. International Journal of Advertising, 22 , 183-202. Domzal, T.J., & Kernan, J.B. (1994). Creative f eatures of globally-understood advertisements. Journal of Current Issues and Research in Advertising, 16 (1) , 29-47. Hanni, D.A., Ryans, Jr., J.K., & Vernon, I.R. (1995). Executive insights: Coordinating international advertising – the Goodyear case revisited for Latin America. Journal of International Ma rketing, 3 (2), 83-98. Harris, G. (1996). Internati onal advertising: Developmental and implemental issues. Journal of Marketing Management, 12, 551-560. Harvey, M.G. (1993). Point of view: A model to determine standardizatio n of the advertising process in international markets. Journal of Advertising Research , 33 (4), 57-64. Hofstede, G. (1983). National cu ltures in four dimensions. International Studies of Management & Organization, 13 (1/2) , 36-74. Hofstede website Holsti, O.R. (1969). Content analysis for the social sciences and humanities . Reading, MA: Addison-Wesley Publishing. Mueller, B. (1989). Degrees of gl obalization: An analysis of the standardization of message elements in multinational advertising. Current Issues and Research in Advertising, 12 (1), 119-133.

PAGE 60

60 Mueller, B. (2004). Dynamics of international adverti sing: Theoretical and practical perspectives. New York, NY: Peter Lang. Onkvisit, S. & Shaw, J.J. (1987). Standard inte rnational advertising: A review and critical evaluation of the theoreti cal and empirical evidence. Columbia Journal of World Business , 22 (3), 43-54. Ramaprasad, J., & Hasegawa, K. (1992). Creativ e strategies in American and Japanese TV commercials. Journal of Advertisi ng Research, 32 (1) , 59-67. Risk Summary (2006). Ce ntral America October 2006. Latin American Monitor, 23 (10) , 4. Ryans, J.K. (1969). Is it too soon to put a tiger in every tank? Columbia Journal of World Business, 4 (2) , 69-75. Solberg, C.A. (2002). The perennial issue of ad aptation or standardization of international marketing communication: Organizati onal contingencies and performance. Journal of International Marketing , 10 (3), 1-21. Taylor, C.R. (2002). What is wrong with international advertising research? Journal of Advertising Research , 42 (6), 48-54. Taylor, C.R. (2005). Moving intern ational advertising research fo rward: A new research agenda. Journal of Advertising , 34 (1), 7-16. Ueltschy, L.C., & Castillo, A. (2005). Marketing opportunities at the intersection of formal and informal economies. The Marketing Management Journal, 15 (1) , 69-80. Vines, E. (2005). South of the border. Shoot, 46 (25) , 26-28. Whitelock, J., & Chung, D. (1989). Cross-cu ltural advertising: An empirical study. International Journal of Advertising, 8 (3) , 291-310. Zou, S. (2005). Contributions to international advertising research: An assessment of the literature between 1990 and 2002. Journal of Advertising , 34 (1), 99-110.

PAGE 61

61 BIOGRAPHICAL SKETCH Casey Hale Millwee will graduate with her Ma ster of Advertising in December 2006. She received her B.A. in English from the University of Florida in May 2005. She plans to work in the advertising industry after graduating, and th en may return to school to pursue doctoral studies. Her research interests include international advertising, Latin American studies, and mass media effects on children.