PLACE ll RETURN BOX to remove thh chockout from your mood. TO AVOID FINES return on or More data duo. DATE DUE DATE DUE DATE DUE —_-J— :1 “TIP—51m MSU I: An Affirmdlvo Adm/Equal Opportunity Initiation emails-9.! COMPARISON AND EVALUATION OF INFORMATION CONTENT IN 0.5. AND SOUTH KOREAN TELEVISION ADVERTISING BY Kyu Yeol Chang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Communication Arts and Sciences - Mass Media 1991 654 - (a /:r ABSTRACT COMPARISON AND EVALUATION OF INFORMATION CONTENT IN U.S. AND SOUTH KOREAN TELEVISION ADVERTISING by Kyu Yeol Chang This study compares the information content of U.S. and Korean television advertising. It focuses on the number of information cues, information types, and modes of presenting information. Preliminary research indicated that the frequently used Resnik-Stern 14-cue method was inadequate. The study utilized a coding scheme comprised of 33 information cues. Visual and auditory modes of presenting information cues were coded separately. Five groups of hypotheses were derived from the literature. It was predicted that U.S. commercials contain more information cues than Korean commercials, that there are differences regarding the types of information delivered, and that Korean commercials use more visual presentations while U.S. commercials use more auditory means. Product life cycles and product categories were also used to predict the information content of commercials. The study analyzed the information content of 1,228 0.5. and 867 Korean television commercials for products and services Kyu Yeol Chang nationally advertised. Data coding instruments and codebooks were developed in the two languages. Sixteen U.S. and twelve Korean coders were recruited and trained. The study found that there are no significant differences regarding the number of information cues between Korean and U.S. commercials. But differences were found regarding information types and the modes of presenting information. The anticipated product life cycle effects regarding the number of information cues in commercials were not confirmed in either nation. However, the product category effects were found in both nations. Cross-national differences were found in the number of information cues in commercials for physical products and for services in the same stage of the product life cycle. Differences were also found across product categories. The results confirmed expectations that advertising information content is associated with cultural variables. It was found that the coding instrument utilized for the study satisfactorily identified advertising information. The instrument made it possible to show the differences in the modes of presenting information. Managerially, the localized approach to international advertising between the two nations seems to be appropriate regarding the information content of television advertising. Limitations of the study were identified and future directions for research were suggested. Copyright Kyu Yeol Chang 1991 Dedicated to my Parents, Dong Wook Chang and Sa Joong Shin, Whose understanding, sacrifices, and support inspired myself and their other children. ASKEQELEDQEHEHIé The research reported herein has been enriched by the contributions of many individuals with whom I have been associated throughout the course of my graduate studies at Michigan State University. The Chairperson of my Guidance Committee and Dissertation Director has been Dr. Gordon B. Miracle, Professor of Advertising. Dr. Miracle has led and guided my academic development in a collegial manner, which I shall remember as a superb example of the relationship between a mentor and a student. He has been exceptionally supportive and consistently been available for priceless advice. His support has meaning beyond what can be described in these acknowledgements and for that support I shall always be grateful. Other members of the Committee have been generous in their willingness to provide stimulation and suggestions, and to expand my horizons in regard to my graduate program and the writing of this dissertation. Dr. Gina M. Garramone, Professor of Advertising, Dr. S. Tamer Cavusgil, Professor of Marketing and Director of the International Business Development Center, and Dr. Joseph Straubhaar, Associate Professor of Telecommunication have each made unique and significant contributions which I deeply appreciate. vi The data base used in this research was developed as part of a broader research program by Dr. Miracle with the assistance of myself who served as his chief research assistant and other research assistants, especially Hae Kyong Bang, Jae Yung Cho, Hiromi Maenaka, Sachiko Takaoka, and Charles R. Taylor. The research group met regularly to design, test, revise and finalize the many facets of the research. Generous funding was obtained by Dr. Miracle from Michigan State University, the Hoso Bunka Foundation (Japan), the Korea Broadcast Advertising Corporation (Korea), ASI Market Research (Japan), the Yoshida Hideo Memorial Foundation (Japan), and Youngshin Academy (Korea). I must also express my deep indebtedness to Dr. Bruce Vanden Bergh, Professor of Advertising, for his support as the Chairperson of the Department of Advertising throughout my graduate studies. Personal acknowledgements are very much in order. My parents, Dong Wook Chang and Sa Joong Shin, have always encouraged my participation in the educational process and helped in many quiet ways to make this accomplishment possible. They have been a source of inspiration for myself throughout this endeavor. I cannot imagine how I could have passed through the gates of the academic world and arrived at this juncture without the unstinting, selfless, and constant support of my wife, Hye Myong Chang. vii Her support and willingness to encourage my graduate studies and research have been constant. I admire, respect, and am eternally grateful for her patience, involvement and understanding. My three-year-old daughter, Connie Juhee Chang, can fully claim credit for enlightening me with her never-ending happiness and for reminding me of a Daddy’s many responsibilities. viii LIST OF TABLES . . . . CHAPTER I. Background . . Purposes and Contributions Research Questions Summary . . . . . CHAPTER II. LITERATURE Standardization of Advocates for Advocates for TABLE OF CONTENTS INTRODUCTION . . . . . REVIEW . . International Advertising Standardization . . . . . Localization Contingency Approach . . Standardization and the Dissertation . Information Content in Advertising . . . . . Measurement of Advertising Information Studies on Advertising Information . . Advertising Information and the Disserta Product Life Cycle Product Categories Cultural Characteristics . . . Culture and Advertising . Individualism and Collectivism Low and High Context Cultures . Written versus Oral Communication Confucianism and Buddhism . . . . Cultural Characteristics and theD Methodological Concerns . . . Content Analysis . . . . Cross-Cultural Research . Research Hypotheses Summary . . . . . CHAPTER III. RESEARCH METHODOLOGY . Measurement Scale Development Definition of Advertising Problems of the Resnik-Stern Measure Number of Information Cues Separation of Presentation Modes Pilot Study . . Procedures of the Pilot Study . . Lessons from the Pilot Study . . ix ooooogooooooo Information ii in rt ooooouooooooooflooooooo oooooP-oooooooogooooo O a o a o o o o o o .1 .3 .7 .8 .9 1o 11 12 14 15 16 1_§_ 27 27 29 31 31 35 37 4O 43 43 43 44 48 51 52 52 52 54 58 61 61 64 Content Analysis for the Dissertation Research . Summary . . . . . . . . . . . . . . Taping Commercials and Developing Data Base . Category Development . . . . . . . . . . . . Data Coding Instrument and Codebook . . . . Coders, Coder Training and Coding . . . . . . Coding of Product Categories, Commercial Lengths, and PLC Stages . . . . . . Reliability . . . . . . . . Administration and Data Input . . CHAPTER IV. ANALYSIS, RESULTS AND DISCUSSION . . . . Number of Information Cues . . . . . . . . . . . . . . Types of Information . . . . . . . . . . . . . . . . . Presentation Modes . . . . . . . . . . . . . . . . Product Life Cycle . . . . . . . . . . . . . . . . . . Product Categories . . . . . . . . . . . . . . . . . . Summaries . . . . . . . . . . . . . . . . . . . . . . CHAPTER V. CONCLUSIONS . . . . . . . . . . . . . . . . . Summary of Findings . . . . . . . . . Theoretical Conclusions . . . . Methodological Conclusions . . . Managerial Conclusions . . . . . Limitations . . . . . . . . . . Directions for Future Research . APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX Culture and Advertising Information . . . Product Life Cycle and Product Categories . . A. COMPARISONS BETWEEN SOUTH KOREA AND THE U.S.A. B. DIMENSIONS OF INTERNATIONAL ADVERTISING TO BE CONSIDERED FOR STANDARDIZATION . . . . C. RESNIK AND STERN’S FOURTEEN CRITERIA FOR CLASSIFICATION . . . . . . . . . . D. INFORMATION CUES FOR THE PILOT STUDY . . . . . E. RESULT OF THE PILOT STUDY: DIFFERENCES OF ADVERTISING INFORMATION FREQUENCY . . . . . F. DATA CODING INSTRUMENT . . . . . . . . . . . 66 . 67 68 . 69 . 71 74 74 82 .102 .105 .105 .105 .106 .107 .114 .115 .117 .118 .120 .121 .122 .124 .125 .126 APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX G. H. N. O. VIDEOTAPING FOR U.S. COMMERCIALS . . . . . VIDEOTAPING FOR KOREAN COMMERCIALS . . . . CODEBOOK . . . . . . . . . . . . . . . . . DEFINING THE PRODUCT LIFE CYCLE . . . . . PRODUCT CLASSES REPRESENTED IN COMMERCIALS COMMERCIAL LENGTHS . . . . . . . . . . . . RELIABILITY . . . . . . . . . . . . . . . DATA CODING INSTRUMENT (IN KOREAN) . . . . CODEBOOK (IN KOREAN) . . . . . . . . LIST OF REFERENCES . . . . . . . . . . . . . . . . . . xi .128 .129 .136 .134 .135 .137 .138 .139 .141 .145 Table Table Table Table Table Table Table Table Table Table Table Table Table Table 10. 11. 12. 13. 14. LIST OF TABLES Page Mean Number of Information Cues in Different Length Commercials . . . . . . . . . 76 Mean Number of Information Cues in lS-second TV Commercials . . . . . . . . . . . 77 Mean Number of Information Cues in 30-second TV Commercials . . . . . . . . . . . 78 Mean Number of Information Cues in TV Commercials of All Lengths . . . . . . . . 79 Frequency (t) of Commercials for Each Information Cue . . . . . . . . . . . . . . . . . . . . . . . 84 Modes of Presenting Information (Mean Frequency in 15-second Commercials) . . . . 87 Modes of Presenting Information (Mean Frequency in 30-second Commercials) . . . . 87 Modes of Presenting Information (Mean Frequency in Commercials of All Lengths) . 87 Mean Number of Information Cues Across PLC Stages for 0.8. Television Commercials . . . . . . . . . 91 Mean Number of Information Cues Across PLC Stages for Korean Television Commercials . . . . . . . . 92 Mean Number of Information Cues in PLC Stages (ls-second Commercials) . . . . . . 93 Mean Number of Information Cues in PLC Stages (30-second Commercials) . . . . . . 94 Mean Number of Information Cues in PLC Stages (Commercials of All Lengths) . . . 94 Mean Number of Information Cues Across Product Classes (U.S. Television Commercials) . . . . . . 98 xii Table Table -able Table 15. 16. 17. 18. Mean Number of Information Cues Across Product Classes (Korean Television Commercials) . . . Mean Number of Information Cues by Product Categories (ls-second commercials) Mean Number of Information Cues by Product Categories (BO-second commercials) Mean Number of Information Cues by Product Categories (Commercials of All Lengths) xiii 99 .101 .101 .102 CHAPTER I INTRODUCTION 3.352152132111151 Advertising is expected to provide useful consumer information, which helps consumers to make better purchasing decisions (Backman 1968; Borden 1942; Dameron 1942; Kaldor 1950; Laband 1986; Nelson 1974). As a result of this expectation, there are studies dealing with information content of advertising in different media, in different types of messages, and in cross- cultural settings. Most of these studies use a scheme of measuring advertising information developed by Resnik and Stern (1977). These studies have broadened existing knowledge of the current status of advertising practices and improving information content in advertising. This dissertation research compares and evaluates the information content of U.S. and South Korean television advertising. While identifying methodological problems associated with the widely-utilized Resnik-Stern measure of advertising information, the dissertation research develops, introduces and utilizes an improved scheme for measuring information content in television commercials. The advertising industry of South Korea (henceforth, Korea) has a history of more than a hundred years. The nation's advertising expenditure was $2.3 Billion in 1989, surpassing 1.3% of the nation’s Gross National Product (GNP). The average annual growth rate of the advertising industry has been over 25% during the past decade in its expenditure level (Shin 1989). There have been, however, only a limited number of studies about Korean advertising (Moon and Franke 1987; Shin 1982, 1989). U.S. advertising expenditures were $124 Billion in 1989, approximately 2.5% of the nation's GNP (Advertising Annual 1989; Coen 1991). Appendix A shows comparisons between the two countries regarding selected basic economic statistics and advertising expenditures. The Korean advertising market is opening its doors to foreign investment in 1991. Thus, it is especially pertinent to study Korean advertising in order to provide international advertisers with information about the country's advertising (Lee 1990; Shin 1989). This research will investigate whether or not Korean is different from U.S. advertising, by comparing the content of the two nations' television advertising. An improved measurement scale of advertising information will be used for the research procedure and analysis. The research also provides information about the standardization and localization issue in advertising information between the U.S.A. and Korea. To derive the hypotheses to be tested in this research, the literature on international marketing, advertising information, and cultural characteristics was reviewed. Literature searches on cross-cultural research methods and content analysis method were also performed to identify and solve methodological problems for the study. Wises The purposes of this research are: (1) to investigate cross- cultural differences and similarities between the U.S.A. and South Korea regarding the information content of television advertising: especially, the number of information cues, information types, and presentation modes of information, (2) to identify specific cultural differences between the two nations that may relate to or are associated with differences and similarities in the information content of television advertising, and (3) to identify relevant marketing theories and determine if they are useful to predict or explain the information content of television advertising. The scope of this research is limited to finding cross- cultural differences and similarities in information content between U.S. and Korean television advertising. The identification of problems with the established and widely-used method of measuring advertising (Resnik and Stern 1977), and the development of a new scheme of measurement is also within the scope of the research. The scope of the dissertation research does not, however, include measuring and evaluating effects of advertising information on consumer attitudes or behavior. This is beyond the scope of a content analysis. The study's emphases instead were placed on developing well-defined content categories and research materials, and on establishing cross-cultural equivalence in the content analysis of the research. The dissertation research will only deal with the two nations; the U.S.A. and Korea, limiting the scope of the study in a geographic sense. Affective aspects, such as feeling, mood, etc., of advertising messages also are not within the scope of the research. The research will only deal with an objective measure of information cues contained in television advertising messages. These limitations suggests possible extensions of the study's scope for the future research. This research contributes a better understanding of cross- cultural differences and similarities in communication patterns that are reflected in the information content of television advertising. Concepts in the cross-cultural studies such as individualism-collectivism (Gudykunst, Yoon and Nishida 1987; Hofstede 1984a, 1984b; Hui and Triandis 1986), high and low context languages and cultures (Hall 1983), written versus oral cultures (Choe 1980), and influences of religious philosophy (Yum 1987) are reviewed to investigate whether these cultural concepts are related to, or associated with, differences and similarities of advertising communication patterns in the U.S.A. and Korea. The study also contributes improvements in advertising information measurement. Problems are identified within the widely-utilized Resnik-Stern (1977) l4-cue measure of advertising information, and an improved measure is developed to overcome these problems. Detailed discussion of the methodological improvement is included in the methodology section. The improved measure is utilized for the dissertation research. Another methodological contribution is made by applying cross-cultural research methods which produce a more reliable and valid cross-cultural comparisons. In order to establish conceptual, functional, and linguistic equivalence of measurement material, the translation and back-translation techniques (Brislin 1970; Triandis 1976, 1980) are carefully and extensively used in developing research materials such as data coding instruments and codebooks in both languages. To maintain internal consistency among coders, intercoder reliability is checked for each information cue in both countries. Whether it is better to standardize or to localize marketing and advertising efforts has been a theoretical as well as a practical issue for a long time. This study contributes a practical understanding of this issue by presenting differences and similarities between U.S. and Korean advertising in information content. The dissertation research contributes a better theoretical understanding regarding a marketing theory, the Product Life Cycle (PLC) theory. The study tests the explanatory value of the theory with respect to the number of information cues in Korean and U.S. advertising. The theory suggests that in the earlier stage on the product life cycle, the more information is needed to increase the awareness and knowledge level of consumers about the product class and brand (Day 1981; Midgley 1981; Rice and Lu 1988). The theory has been initiated and developed mostly in the U.S. and Western setting. Thus, this study contributes to a better understanding of the way the product life cycle model works in another cultural setting. This study contributes to the broadening of the theory's application to a cross-cultural level. Different product categories are expected to deliver varying numbers and types of information in their advertising messages (Hong, Muderrisoglu and Zinkhan 1987; Murphy and Enis 1986). The research also attempts to confirm this idea, by comparing information content among different product categories in each country. This effort contributes a better understanding of the ‘product category’ effect on marketing strategies such as advertising message development. Another contribution of the study is in providing an empirical finding regarding differences and similarities of the information content of U.S. and Korean television advertising. Whether to confirm or reject the assumption about the content of Korean advertising relative to the U.S. counterpart discussed above makes a significant contribution to the advertising research (Shin 1984). Finally, this study will help international marketing and advertising managers. This research provides them with information about the information content of U.S. and Korean television commercials, standardization and localization issue, and product category and product life cycle effects on advertising information. 8W With the above contributions in mind, the major research questions investigated by the dissertation research are; 1) What are the cross-national differences and similarities between U.S. and Korean television advertising regarding information content: the total number, types, and presentation modes of information? 2) Are the differences and similarities related to or associated with differences in cultural characteristics: individualism-collectivism, high-low context languages and cultures, written-oral cultures, and influences of religious background? 3) Are the differences in information content between commercials in each country associated with 'product life cycles' and 'product categories’? Summarx The Chapter has discussed the background, the scope and purposes, the expected contributions of the dissertation research. It also proposed research questions that the study purports to deal with. CHAPTER II LITERATURE REVIEW Given the increase in global markets and international trade, important questions concerning marketing and advertising have arisen. Is the standardization or localization of advertising desirable; and to what extent, if any? Numerous failures of applying successful domestic marketing strategies to international markets have been reported (Ricks, Arpan and Fu 1974; Liesse 1991; Tanzer 1986). Due to the high cost of these mistakes, questions and suspicions about whether a company should standardize its international marketing and advertising programs abound (Bartels 1968; Britt 1974; Buzzell 1968; Sorenson and Wiechmann 1975). Some advocate standardization on the basis of the universal needs of global consumers (Fatt 1967), and the scale advantage of magnitude in management (Levitt 1983). Additionally, Sheth (1986) and others suggest a contingency framework that identifies situations under which standardization will be successful and other situations under which localization is necessary. With respect to this issue, three schools of thought are identified; (1) advocates for standardization, (2) advocates for localization, and (3) an in-between position advocating the 10 contingency approach. WM Blinder (1961) has been considered to be the researcher who initiated the modern controversy over the standardization issue of international marketing. Due to a trend toward convergence in consumer behavior and the mobility of consumers among European countries, he argues that standardization of advertising for the European market is possible and necessary. Other researchers have also advocated standardization of international marketing and advertising. Some, because of universal appeals common to average consumers in any nation and because of advancement of transportation and communication means (Fatt 1967). Others, in order to have the economy of scale in operation and to cater to consumers efficiently with consistency and improved centralized planning and control (Buzzell 1968). In a classic article, Levitt (1983) argues that the globalization of markets due to advanced common systems of technology, communications, and transportation would lead international marketers to adopt standardization strategies more frequently. He predicted, however, that standardization in areas of branding and manufacturing may be easier to accomplish than in other areas of marketing such as advertising and promotion. Pointing out a marketers' need to build consistent images 11 in the global market, Peebles (1989) also maintains that managers should strive to find standardized advertising approaches when it is feasible. El ! E I 1' !° Weissman (1967) maintains that there are more important deep-rooted dissimilarities rather than universality or convergence among those living in different cultural societies. He argues that marketers would be better off by taking localized approaches in advertising, especially for consumer goods. Citing several failure stories of standardization approaches in international marketing, Ricks, et al. (1974) emphasizes the importance of knowledge about cultures that advertising operates in. Keegan (1989, p.256) reported the experience of the Parker Pen Company to show how the globalization approach can fail if it is planned without a good understanding of cross-cultural differences. Hornik (1980) found that localized advertising themes were more preferred by consumers, in his empirical comparative study of different advertising themes of U.S. advertising campaigns between the U.S.A. and Israel. Frith and Frith (1990a), Kaynak and Mitchell (1981) and Sugiura (1990) also argue that international marketers should recognize the advantages of localization in marketing communications strategies, stressing the importance of intercultural differences. Robinson (1986) maintains that international 12 markets are more likely to become increasingly fractured by culturally and environmentally based preference, though not necessarily organized on a national basis. Thus, he is strongly in favor of the adaptation approach. Harris (1984) admits that there are a number of product categories that could adopt the standardization approach in advertising, but he argues that the number is limited. He also argues strongly in favor of the local adaptation of advertising and marketing strategies. Martensen (19 9) suggests that there should be differences between high- and low-context cultures: high-context culture perceives nonverbal communication as being more informative than does a low-context culture. She thus implies that marketers need to pay more attention to cultural differences than to similarities to effectively execute marketing and advertising strategies. W11 Several researchers argue that a variety of contingencies and environmental variables have to be considered in order to determine the extent of standardization which should exist in international advertising (Jain 1989; Kreutzer 1988; Peebles, Ryans and Vernon 1977; Roostal 1963). They suggest that attempts be made to identify contingencies under which standardization is feasible. Walters (1986), in his review article on the standardization 13 issue, points out that the issue of process standardization has been taken up frequently in recent studies. Walters (1986) cites Miracle (1968) as one of the first researchers to adopt the process standardization position suggesting that standardization of approach in making advertising decisions be required in developing international marketing strategies. The work of Sorenson and Wiechmann (1975) is also cited as arguing that the process standardization is advisable based on their survey of managers in subsidiaries of multinational corporations. Dunn (1976), Peebles, Ryans and Vernon (1978), and Shuptrine and Toyne (1981) are also cited as adopting and recommending the process standardization or the programmed management approach. Sheth (1978) reviews the two opposite schools of thought on the standardization issue and summarizes eight different levels of standardization and adaptation, from the complete extension to the complete adjustment, depending upon the contingencies. He suggests three major variables which determines the degree of standardization: consumer expectations, the encoding-decoding process, and silent languages between countries. Boddewyn, Soehl and Picard (1986) found, from their survey of B.B.C. firms, that standardization was in fact practiced by those firms, albeit with some substantial barriers. Four major variables were found to be related to the success of the standardization approach: the nature of the product, the extent of national differences, the level of competition, and economic 14 conditions. In a replicated study of the Boddewyn, Soehl and Picard's study (1986) with U.S. multinationals, Hite and Fraser (1988) found that the standardization strategy was practiced by only a very small number of firms, and that some combination of standardization and local adaptation was more widely accepted and practiced. Blackwell, Ajami and Stephen (1991) also suggest a combined approach of globalization and localization for sharing market information, sharing creative ideas and producing marketing materials. E: l' . E! i ii !i l !l Ii ! !' Appendix B was prepared by Professor Gordon B. Miracle of Michigan State University during the early stages of this research. It shows a broad range for consideration when making decisions regarding standardization of international advertising. Many studies, however, neither cover all of these areas nor specify even which of these areas were involved in their studies. This dissertation research limits its scope to the total number, types, and presentation modes of advertising information in message strategies (see Appendix B). Although theoretical arguments and researches based on survey results with managers may be useful in determining the extent of advertising standardization, some empirical investigations and comparisons of advertising content in cross- national settings are also useful in indicating and determining 15 necessary degrees of standardization or local adaptation. There is only a limited amount of literature of this nature. This study attempts to provide empirical data on information content of U.S. and Korean television advertising with the standardization issue in mind. Important implications for marketing and advertising managers will be provided by this research. These implications, concerning the standardization/ localization issue, focus primarily on the number of information cues in ads, information types, and modes of presenting advertising information. WWW One of the major roles of advertising is to provide useful consumer information. Along with persuasive strategies, an advertising message is expected to contain some forms of information (Hunt 1976). Although there has been a controversy over the question whether advertising is information or persuasion, researchers seem to agree that advertising is an information medium for consumers as well as a persuasion means for marketers in order to increase the consumer demand (Backman 1968; Borden 1942; Dameron 1942; Hunt 1976; Kaldor 1950; Laband 1986; Marquez 1977; Nelson 1974). In addition, many critics of advertising argue that advertising can be a useful means of communication only when it contains information (Tom, et al. 1984; Kaldor 1950). 16 Hunt (1976) points out the difficulty of measuring the amount of information objectively, because the quantity of information may not be independent of the message's quality or usefulness of information. In addition, the quantity of information may be individual-specific, meaning that the same message can be interpreted by different people as containing different amounts of information. The latter problem of individual-specific interpretation can be solved in a content analysis with multiple coders and with careful reliability checking. The former problem of information quality or usefulness can also be solved if a researcher can develop a comprehensive list of information cues that can be delivered in advertising messages. Resnik and Stern developed an objective measurement scheme of 14 "information cues" to measure the level of informativeness of advertising messages (Resnik and Stern 1977, see Appendix C). The authors define advertising information as that which permits a typical consumer to make a more intelligent buying decision after being exposed to an advertising message than before the exposure. Coders are instructed to record any of the 14 cues- present in an advertisement. The authors define that an advertising message to be informative when it contains at least one information cue. Soley and Reid (1983a) compared quantity and type of 17 advertising information to see which of the two determined the consumers' perception of informativeness. The authors found, in a sample of advertisements for industrial products, that perceived informativeness was determined more by the type of information rather than by the quantity of information. Aaker (1984) maintains that measuring the informativeness of television advertising through content analysis does not capture the entire information content of the ad. The content analysis method also falls short in that it depends on the subjective judgement of the researchers to categorize the information they observe. In addition, it is difficult to create a list of cues which are of equal relevance to consumers, valid for all products and reliable for all modes of television advertising. Finally, Aaker (1984) points out that television commercials are so difficult to codify due to their gestalt-like nature; the whole is more than the sum of their parts. As a partial solution, Aaker suggests adding a measure of the consumer’s perception of informativeness to the objective content measures. The problem of coding television commercials pointed out by Aaker (1984) may be significantly reduced by separately evaluating the (visuals on screen) and auditory (verbal with sound) presentation of information. Although some limitations with the objective measure of advertising information have been identified, it still has merits. The objective measure allows cross-cultural and 18 longitudinal comparisons of information levels in advertising messages. It provides policy makers with an objective baseline from which to work (Resnik and Stern 1977). 5! 1' E: !i i I E !' Studies about information content of advertising have increased in number since Resnik and Stern (1977) introduced their scheme. Many studies have investigated the information content of advertising messages of different media and message types (Harmon, Razzouk and Stern 1983; Healy and Kassarjian 1986; James and Vanden Bergh 1990; Laczniak 1979; Moon and Franke 1987; Ng and Reid and Rotfeld 1981; Supaporn.1990; Resnik and Stern 1977; Stern, Krugman and Resnik 1981; Stern, Resnik and Grubb 1977; Stewart and Furse 1986; Stout and Moon 1990; Tom, Calvert, Goolkatsian and Zumsteg 1984). They also analyzed content and compared information levels in advertisements cross-nationally (Dowling 1980; Hong, Muderrisoglu and Zinkhan 1987; Huang and Hon 1987; Johnstone, Kaynak and Sparkman 1987; Sepstrup 1985; Madden, Caballero and Matsukubo 1986; Martensen 1987; Renforth and Raveed 1983; Rice and Lu 1988; Weinberger and Spotts 1989). With their own 14-cue measure, Resnik and Stern analyzed information content of 378 U.S. television commercials and found that 49.2% of commercials were informative according to their definition of including at least one information cue (Resnik and 19 Stern 1977). They suggest, with the finding, that the non- informative advertising policy, adopted by over 50% of the sample, may be self-destructive by virtue of the fact that the better-educated and more-sophisticated consumers are seeking relevant product information, and these consumers would purchase products which could show real benefits over existing alternatives. Stern, Resnik and Grubb (1977) report a further analysis of the same 378 television commercials used in the Resnik and Stern study (1977). They indicate that television commercials have room for improvement in providing more useful information to consumers. The authors also suggest that there might be other cues that may be considered relevant information by consumers, in addition to the original 14 provided by Resnik and Stern. Marquez (1977) analyzed the content of 600 newspaper and magazine advertisements on a 5-point measurement scale of messages types. These points are: (1) basic persuasion, (2) basic information, (3) high in persuasion but low in information, (4) high in information but low in persuasion, (5) mainly intimidation. He found that the question whether the advertising content is basic persuasion or basic information depends on the type of product being advertised. The findings indicate that products which are inexpensive and non-technical in nature and operation tend to be advertised with basic persuasion. By the same token, products which require advocacy of a point of view, 20 mechanical operations, terms of purchase, or if the product itself is information, these products are likely to be advertised with basic information. Laczniak (1979) did a content analysis of 380 magazine advertisements, using the Resnik-Stern 14-cue measure. He found that 92% of the sample had at least one information cue. Thus the majority of print advertising was found to be informative and significantly different from television commercials in the amount of information it presents. In this study, he compares the Resnik-Stern 14-cue measure (1977) and the Marquez's S-point judgmental measurement scale (1977) to find that the Resnik-Stern measure results in a more objective and reliable outcome. Dowling (1980) uses the 14-cue method in studying advertising information cross-nationally. He analyzes the content of 163 Australian television commercials to report that 74% of these commercials contains one or more of the 14 cues, and compares the finding with the U.S. findings of having only 49.2% informative messages from the Resnik and Stern study (1977). He attributes the difference to the strict Australian government- and self-regulation on advertising content, implying that more regulation may lead to higher informativeness in advertising. He also found that in both countries advertising informativeness is dependent upon television dayparts and product classes advertised. Pollay, Zaichkowsky and Fryer (1980) content analyze 884 21 U.S. and Canadian television commercials from two separate periods; 1971-1973 and 1977 with an extended list of 19 information cues. They found 84.4% of all commercials (85.2% of Canadian and 83.2% of American ads) were informative, containing at least one cue. They also found that the level of informativeness has not been changed between the two periods, regardless of a resurgence of regulatory vigilance and some redrafting of legislative or regulatory codes on advertising practices during the 1970s. Stern, Krugman and Resnik (1981) found that 86% of 1,149 U.S. magazine advertisements had at least one information cue, reconfirming the Laczniak's finding (1979) that print advertisements were generally more informative than broadcast commercials. The study also confirms that informativeness varies by product categories being advertised. Reid and Rotfeld (1981) replicated the Resnik and Stern study with a sample of 324 Saturday morning television commercials, which were primarily directed toward children, finding that 42.3% were informative. They argue that commercial messages in children's programs are not significantly less informative than commercials broadcast at other dayparts and during other program formats. Using the 14-cue method, Renforth and Raveed (1983) reported a cross-country comparison between American, Australian, and Ecuadorian television commercials concerning the amount of 22 information presented in advertising. The authors found that 82.4% of 108 Ecuadorian commercials were informative as compared with the U.S. (49.2%) and Australian (74.0%) results. They attributes this cross-national differences to the different stages in the product life cycle of products being advertised and to the different levels of economic development among the three countries. Harmon, Razzouk and Stern (1983) add one more cue, energy, to the 14-cue list. They examine comparative and non-comparative advertisements. They found that comparative advertisements contained more information cues. They evaluate that the Resnik and Stern classification system provides coders with a reliable and objective means of evaluating each advertisement against a uniform set of criteria. The effects of government regulation on the informativeness of magazine advertising was studied by Healy and Kassarjian (1983). They looked specifically at the regulation of the Federal Trade Commission (FTC)'s substantiation program. Healy and Kassarjian found that mere regulation does not necessarily contribute to the increase of information in advertising messages over time. Tom, Calvert, Goolkatsian and Zumsteg (1984) replicated the 1977 Resnik and Stern study, analyzing the content of 348 U.S. television commercials on the 14-cue list. They found 52.7% of commercials satisfied the one-cue criteria to be considered 23 informative. They concluded that information content of television commercials had not been improved in the five years since the Resnik and Stern (1977) study. Additional analyses by broadcasting dayparts and by product categories reveals similar trends to the Resnik and Stern's initial findings (1977). Sepstrup (1985) developed another extended version of information list of 24 cues, and content analyzed and compared commercials on Sky Channel, West German Channel 2, ZDF and print advertisements from U.S. magazines and Danish newspapers. He found that television advertising has a limited value since information relevant for most purchasing decisions is not communicated to a degree that seem to have any practical significance for consumers. Furthermore, the information content of print advertising was found to be clearly better than that of television commercials. Madden, Caballero and Matsukubo (1986) compared 832 U.S. and 608 Japanese magazine advertisements with the 14-cue method, and found that 85% of Japanese and 75% of U.S. advertisements were informative. These differences between two cultures regarding information amount were attributed to different cultural characteristics, different communication styles, different education levels of average consumers, and different basic approaches of advertising practitioners. Stewart and Furse (1986) extended the list of information cues to 26 information cues and used it in their study of 1,059 24 television commercials. They found U.S. television commercials contained higher levels of information than previous studies did. This higher level of informativeness was apparently due to the additional information cues that were not included by these using the 14-cue method. In order to investigate cross-cultural and cross-national differences in marketing communications, Johnstone, Kaynak and Sparkman (1987) compared U.S., French Canadian and English Canadian television commercials regarding information content. They found that 69.2% of U.S. (139 out of 201), 75.0% of French Canadian (57 out of 76), and 60.4% of English Canadian (64 out of 103) commercials were informative with at least one of the 14 information cues. Statistical differences were found only between U.S. English and French Canadian commercials, showing that such differences appeared to be due to an interaction between nation and culture, although other possibilities might exist. Hong, Muderrisoglu and Zinkhan (1987) compared Japanese and U.S. magazine advertisements on three dimensions; mood rating, informativeness, and comparativeness. In dealing with informativeness, they used the 14-cue method and found that 95% of Japanese and 90% of U.S. magazine advertisements were informative, containing at least one information cue. Contrary to their expectation, U.S. ads were not found to be more informative than Japanese ads. They explained that this finding 25 might be due to the characteristics of the medium, magazines, which usually contain large amounts of information in its advertising, and due to the higher involvement level of magazine readers compared to listeners or viewers of broadcast media. Martensen (1987) compared Swedish and U.S. television commercials and found that 45% of Swedish (out of 102) and 36% of U.S. (out of 22) commercials were informative with the Resnik and Stern method. Moon and Franke (1987) content analyzed 573 Korean magazine advertisements with the 14-cue method. They found that 80% of the messages were informative. They also found that advertisements for international brands carry a little lower level of information than those for Korean domestic brands. They also report varying information amount by product categories. 472 Chinese magazine advertisements were content analyzed by Rice and Lu (1988) with the 14-cue method, and all commercials (100%) in the sample were found to be informative containing at least one information cue. They attributed the higher degree of informativeness in Chinese advertising to different stages of economic development, different stages of the product life cycle, different consumer involvement levels in consumers’ decision making, and different regulation patterns. Weinberger and Spotts (1989) compared 566 U.S. and 301 U.K. television commercials regarding information content with the 14- cue method. They found that 65.3% of the U.S. sample and 55.8% 26 of the U.K. sample were informative. This difference was found to be significant, and they explained the finding with different cultural characteristics and regulation systems of the two countries. Direct response advertisements, product/store image advertising copies and institutional advertising messages in magazines were compared by James and Vanden Bergh (1990). They found that 99.7% of direct response, 92.6% of product/store, and 94.4% of institutional advertising messages were informative. They conclude that direct response advertising seems to include more and diverse information cues than product/store or institutional advertising messages, because the medium is supposed to ask for more immediate response than others. Ng and Supaporn (1990) used the 14-cue method to compare information content of U.S. television advertising longitudinally, covering a period of 1975 to 1990. They found that the overall trend indicates that information amount has increased over time. Stout and Moon (1990) content analyzed 655 U.S. magazine advertisements with the 14-cue method, and found that advertisements with endorsers generally contained less information than advertisements without endorsers. They suggest a need to improve the Resnik-Stern measure to include additional information cues that individuals may use in making purchase decisions. 27 Adzsrtisins_Inf2rmafign_and_the_nisssrtatign With a small number of exceptions (Harmon, et al. 1983; Pollay, et al. 1980; Sepstrup 1985; Stewart, et al. 1986), most of the studies on advertising information utilized the Resnik- Stern (1977) 14-cue measure. This dissertation research, however, does not use the 14-cue method. Instead, the study identifies perceived problems with the 14-cue method, and develops an alternative measurement scheme to be used in measuring advertising information. Details on the methodological adjustment will be discussed in the following methodology chapter. The research compares U.S. and Korean television commercials regarding information content; t9ta1_nunbers_ef_inf9rmation_sues. types and ngsgn§3319n_mggg§. By so doing, the study contributes to the tradition of advertising information research. The study also analyzes advertising information by pgggggt_lifig_gyglg stages. by nrsdust.satssgriss. and by seamersial_lsngth§. BIQQHQS_Lifi§_Q!§l§ The major thrust of the product life cycle (PLC) theory is that a product’s sales potential and profitability change over time. The product life cycle theory is an attempt to recognize distinct stages in the sales and profit history of a product, and it suggests four stages; introduction, growth, maturity and decline (Kotler 1984, and see Appendix J). 28 The product life cycle theory indicates that more information may be required in the earlier stages because manufacturers need to build consumer awareness about the existence, attributes, and performance of the product, with more informative messages (Midgley 1981). The relationship between product life cycle stages and the number of information cues has been found to be consistent in previous studies on advertising information. They found that product classes in the earlier stages of the PLC did contain more information (Stern, Resnik and Grubb 1977; Renforth and Raveed 1983; Rice and Lu 1988). This research attempts to confirm the relationship between product life cycle and informativeness levels in Korean as well as in U.S. television advertising. Another important aspect of comparing the two countries’ advertising content by the product life cycle stages is related to the study’s cross-cultural nature. Triandis (1980) pointed that almost all theories in social sciences are figments of Euro- American imagination, and questioned if they provide a truly universal or general account of phenomena that they deal with. Lee (1991) also suggests that theory building in consumer behavior is not mature enough to modify consumer behavior theories and models "made in U.S.A." to improve their explanatory power and generalizability for the local consumers of different cultural backgrounds. The product life cycle (PLC) theory was also initially 29 developed in the U.S. setting and tested mostly in the U.S. and Western settings. This research contributes to confirm the generalizability of the product life cycle (PLC) theory in another cultural setting, by investigating whether the advertising informativeness can be associated with the product life cycle stages in Korea with the same pattern as in the U.S.A. Renforth and Raveed (1983) and Rice and Lu (1988) report that cross-national differences regarding informativeness of advertising are affected by the overall differences of product life cycle stages of products in each nation. They found that advertising messages of lesser developed nations contained relatively more informative content because products of these countries were at the earlier stages of the product life cycle. The dissertation research will also investigate cross-national differences of product life cycle stages of product categories and their association with the informativeness level of advertising messages. EIQQBQS_§§§§99212§ Previous studies on advertising information found different levels of advertising informativeness between different product categories (Aaker 1984; Dameron 1942; Miracle 1965; Healy and Kassarjian 1983; Stout and Moon 1990; Resnik and Stern 1977; Stern, Krugman and Resnik 1981; Harmon, Razzouk and Stern 1983; Madden, Caballero and Matsukubo 1986; Hong, Muderrisoglu and 30 Zinkhan 1987). Murphy and Enis (1986) review literature on product classification. They propose that an integrated product classification scheme which adds ’preference’ products to the conventional classification: convenience, shopping, and specialty products. They suggest that different product characteristics require different marketing and promotional strategies. The study attempts to examine whether the different product characteristics lead to differences in informativeness of television advertising content in the U.S.A. and Korea. The research attempts to confirm whether the p;g§gg§_gg;ggg;y_gfifgg§ matters in determining information levels of television advertising in the U.S.A. and Korea. The dissertation research divides products into three categories based on durability; nondurable goods, durable goods, and services. A nondurable good is consumed in one or a very limited number of uses. A durable good lasts for many uses. Unlike nondurables and durables which are both tangible, services are intangible activities, benefits, or satisfactions that are offered for sale (Scheoll and Guiltinan 1990, p.295). 31 e d v The relationship between advertising content and the cultural values of a society has been recognized by researchers. Advertising content is believed to reflect cultural values of a society (Frith 1990; Frith and Frith 1990a, 1990b; Holbrook 1987; Hong, Muderrisoglu and Zinkhan 1987; Madden, Caballero and Matsukubo 1986; McLuhan 1964; Miracle, Chang and Taylor 1989; Mueller 1987; Unwin 1974). McLuhan noted that the advertisements are the richest and most faithful daily reflections that any society ever made of its entire range of activities (McLuhan 1964, p.232). Frith and Frith (1990a) reviews cultural differences between the Western and the Eastern cultures that are reflected in advertising contents. These cultural characteristics were direct- and indirect discourse, individualism and collectivism, horizontal and vertical identification of personal relationships, determinism and fatalism, and logical and intuitive problem- solving. Advertising’s influence on society has also been recognized. Pollay (1983, 1986a, 1986b, 1987) notes that advertising has inescapable and profound effects, mostly negative, on a society. He summarizes a review of the work of humanities and social science scholars, indicating that they saw advertising as reinforcing materialism, cynicism, irrationality, selfishness, 32 anxiety, social competitiveness, sexual preoccupation, powerlessness, and a loss of self-respect (Pollay 1986a). Major interest of the research is in the former type of association between culture and advertising. The study investigates if cultural characteristics of two nations are reflected in or associated with information contents of U.S. and Korean television commercials. This section reviews literature on cultural characteristics of Korean and U.S. cultures to derive research hypotheses concerning information content of U.S. and Korean television commercials. Four cultural characteristics are reviewed as related and relevant cultural characteristics; individualism- collectivism (Becker 1986; Gudykunst, Yoon and Nishida 1987; Hofstede 1984a, 1984b; Hui and Triandis 1986), high- and low- context cultures and languages (Hall and Hall 1983; Martensen 1989), cultural preferences for written than oral communication (Choe 1980; Park, Cambra and Klopf 1979; Unwin 1974), and influences of confucianism and buddhism (Lee 1991; Kim 1986; Yum 1987). Although there certainly are additional cultural characteristics that may be relevant to advertising content, these four characteristics seem to be useful to explain some of the different patterns of advertising expression regarding the number of information cues between U.S. and Korean television commercials. 33 W The individualism-collectivism dichotomy has been identified by several cross-cultural researchers (Hui and Triandis 1986; Hofstede 1980; Hofstede and Bond 1984; Lee 1991; Waterman 1981). Individualism is defined as "a situation in which people are supposed to look after themselves and their immediate family only," whereas collectivism is defined as "a situation in which people belong to in-groups or collectivities which are supposed to look after them in exchange for loyalty (Hofstede and Bond 1984)". Studies have reported that the U.S. culture is highly individualistic, and Chinese and Japanese cultures are highly collectivistic (Hsu 1981; Hofstede 1980). Korean culture is found to be collectivistic (Kelly, Whatley, Worthley and Lie 1986; Klopf 1981; Lee 1991), even more collectivistic than Japanese culture (Gudykunst, Yoon and Nishida 1987; Hofstede 1934). Hofstede (1984) found, in his cross-cultural study on work- related concepts, that the Korean culture is highly collectivistic, showing a score of 18 on an individualism scale, compared to 91 of the U.S. and 46 of Japanese cultures. He points out that the fundamental issue addressed by the concept is the degree of ineergenengenee an individual maintains in a society. Klopf (1981) reported that in Korea the family unit is more important than the individual, and decisions are made in 34 favor of the entire family, rather than for the sole benefit of a single individual of the family. In a collectivistic society, an individual is hardly supposed to be heard strongly on his or her own, without authority, social status or group consensus, and people tend to move toward uniformity, harmony and/or standardization rather than toward individuality. Because of this characteristic, Koreans may less appreciate arguments of strength in messages when communicating with each other, instead putting higher priority on conformity with authority and other people’s opinions (Becker 1986). Lee (1991) indicates that people in a collectivistic society such as Korea can sacrifice themselves for the benefit of a group. These benefits largely accrue to a particular social unit, or even to society as a whole: Koreans are more motivated toward conforming to the norms of the group that they belong to than are people in an individualistic society. This cultural characteristic of collectivism, together with its manifestations such as standardization and conformity inclination, may lead Koreans to put less emphasis on individual differences or strength in argument, and instead put more emphasis on shared feeling and consonance among persons when they process persuasive messages such as advertising. In the context of advertising message development, this cultural characteristic of avoiding individual confrontation and 35 of seeking harmony and conformity may influence content of Korean advertising to contain relatively less direct information about product, brand and company than that of an individualistic society such as the U.S.A. WW Language and culture are so closely tied that it may not be possible to understand the full meaning of verbal utterances without a complete understanding of cultural nuances (Kim 1985). Hall (1976) indicates that cultures of the world can be arranged on a sliding scale from high to low context. He explained the difference as follows; A high context (HC) communication or message is one in which most of the information is already in the person, while very little is in the coded, explicit, transmitted part of the message. A low context (LC) communication is just the opposite; i.e., the mass of the information is vested in the explicit code. People in high context cultures have a general tendency to fill blanks or gaps based upon the nature of the context or circumstantial information when they encounter incomplete and indirect messages. They are apt to become impatient or irritated when people from a low—context culture give them more information than they need. Conversely, people from low context cultures are at a loss when people from high context cultures do not provide enough direct information (Hall 1976). In a low-context culture, things have to be described clearly and accurately for 36 effective and proper communication to take place (Unwin 1974). People in high context cultures may not need the direct utterance of every bit of information that builds the structure and meaning of a message. Instead, they rely more upon contextual information, such as situational characteristics and interpersonal relationships. Providing detailed information may sometimes even harass receivers of a message or decrease the level of effectiveness and efficiency of communication, even though the information is relevant to the topic. Hall and Hall (1983) suggest that Japan and other Oriental cultures can be classified as a high context culture. Gudykunst and Nishida (1986) indicates that most Asian cultures, e.g., Japanese, Chinese, and Korean, fall at the high end of the context continuum. Yum (1987) indicated that in Korea Nun-chi was regarded as an important communicative quality. The Nun-chi is roughly translated as sensitivity with eyes, or an ability to perceive what takes place around a conference table or in other persons’ minds based upon looking around. It is through Nun-chi that one understands what happens without being told in detail. Catching a correct idea about what other persons have in their minds with nun-chi provides a solid basis to achieve a high level of effectiveness in interpersonal communication. The important point here is that practicing Nun-chi means that there is no or very little verbal communication among 37 communicators. Instead, they fill in any communicative gaps, which in most cases are related to affective or affect-laden cognitive aspects. Other related evidence to show the Korean culture’s high- context nature includes verbal phrases frequently used in Korea, which can be translated as ”there is a bone within what he is saying", ”truth within a joke", and "he hears one and understands ten" (Yum 1987). Due to the high-context characteristic of Korean culture and language, the advertising communication style may place less emphasis on a list of articulated strengths and informative cues about products advertised in television commercials. Instead, situational cues and circumstantial expression may be more important in developing and interpreting advertising messages in Korea. Therefore, the content of Korean advertising may be expected to contain relatively less direct information about brand, product, or company than U.S. advertising. WM Comparing communication styles of the Western and Oriental cultures, Choe (1980) stated that the Oriental view of communication was oriented relatively more to written words, while the Western view was directed relatively more to spoken words. He explained that this difference stemmed from the religious background of each culture, where the theological 38 meaning depended heavily on spoken words at the beginning of Christianity and written words had a relatively greater significance at the beginning of Confucianism. The clergymen of Christianity are expected to speak to the public, while Confucian scholars are required to read and record with written words (Choe 1980). Spoken words have been valued less in the traditional Korean and most Oriental societies, whereas participation with spoken words has been encouraged throughout the Western history. Choe (1980) maintains that the Korean social norm does not place great values on eloquence, refutation, disputation of right or wrong, argumentation, and explanation with spoken words. To read and write are supposed to be the profession of scholars and to speak is supposed to be the act of menials. Frequent usage of bulletin board posting for demonstrating political arguments compared to face-to-face debates in election processes, the greater emphasis given to reading and writing than to verbal skills in the language teaching, and the preferred usage of written examinations compared to the formal face-to-face interview when evaluating candidates may all exemplify the cultural tendency toward written evidences over verbal articulation (Yoo and Lee 1987). Another important aspect related to this cultural characteristic in Korean traditional society is with the means of writing. Although an efficient phonetic system, Hangul, was 39 invented as early as in the 15th century, the usage of the Hangul has been officially discouraged and limited to backrooms until the end of the 19th century. The major means of writing among the ruling classes instead was Hanja, Chinese characters, ideographs that were difficult to learn for middle and lower class Koreans. Furthermore, instruction to learn Hanja was limited to the ruling class of society. Therefore, the general public in traditional Korea did not have any effective means and channels of delivering their arguments and messages through the social structure, because carrying their messages with spoken words has been despised by cultural norms and they did not possess an effective and officially approved means of delivering their messages through social channels. With this cultural and historic background, Koreans may need to avoid direct utterance of their arguments, and instead may need to crystalize their points of communication into a written format. Koreans may expect communicative messages to carry fewer points of argument in accordance with their cultural tendency of discouraging verbal articulation of ideas. Korean television advertising may deliver more information with the visual means than with the auditory means, in the culture of discouraging verbal articulation of ideas. Thus, the visual means, including putting informative cues on screen, may be more frequently used in Korean television advertising than the auditory means of information delivery. A cross-cultural 40 comparison may be made with the U.S. advertising, where relatively more auditory means are expected than the visual means of delivering information. Traditional respect given to the Hanja system (Chinese characters) and discouragement of using Hangul (phonetic system) in Korea may have influenced the Korean public and culture on the visualizer-verbalizer scale (Richardson 1977; Rossiter and Percy 1980). The system of Hanja is ideographic with its origin in ancient hieroglyphics, conveying meanings with visual images with a single character. Having used the system for centuries, Korean people may have developed a different way of processing information relying more on visual aspects of communication than on verbal aspects. From this observation, television advertising of Korea is expected to deliver information through more visual (picture on screen) means than auditory (sound) means, compared to U.S. television advertising. ; E . . l E 1:]. As a philosophy of humanity, Confucianism stresses the importance of maintaining proper human relationships and providing proper ways to handle rituals which function to maintain the social order. Yum (1987) argues that this tradition of putting the highest value on human relationship results in the implicitness and indirectedness of Korean language. She explains that indirect communication helps avoid the 41 embarrassment of rejection by the other person, leaving the relationship intact. In many cases, the human relationship is perceived to be more important than the business at hand, even . though the business initiated the interpersonal interaction (Lee 1991). The Confucianism also admonishes people to suppress their emotions and feelings. It maintains that human passions are the major impediments to the pursuit of profound knowledge and truth (Yum 1987). Indirectedness and elusiveness in the Korean communication style are illustrated by such Confucian lessons, as one should curb one’s anger, hide one’s sorrow, not be obvious in joy, and so on. The cultural tradition of Confucianism is also related to the above-mentioned ‘preferred written communication style of the society’. There may be a tendency not to have too much verbal expression of emotion where such direct articulation has been curbed by the Confucian standard, which had been an official social norm imposed by the Yi dynasty (1392-1910). Confucianism emphasizes written communication and de-emphasizes oral communication, while the Christian tradition of the Western world encourages oral communication. While the clergymen of Christianity were asked to speak, Confucian scholars were asked, first of all, to read and record in written words. While the former were orators, the latter were literati (Choe 1980). Kim (1986) also points out that communication styles used in 42 Korean advertising have been significantly influenced by the Confucian tradition of the society, with many indirect and roundabout messages. Kim’s (1986) argument is based mainly upon his industry experience and observation, and this view about Korean advertising’s expression is shared by other advertising professionals (Shin 1984). Even though Buddhism has been oppressed in Korea by the central Confucian government throughout the Yi dynasty, more recently Buddhism has been widely accepted due to its egalitarian nature, especially among the lower class public and women who were deprived of social status under Confucian ethics. In Buddhism, however, there is a general distrust of communication, written or spoken (Yum 1987). Buddhists believe that words are only approximations of any ideas or events and cannot convey what we really mean. Thus, implicit communication and the ability to discern hidden meaning are highly valued in the Buddhist tradition. Influences of Confucian and Buddhist traditions on Korean communication styles may lead Koreans to develop their messages in more indirect and subtle forms and include less direct argumentation even when they need to persuade others with their messages. Thus, advertising messages with many informative cues supporting the message may not be as effective in Korea as in the U.S.A. 43 c ' t' ° t The major thrust of the dissertation research is to find and to check if there is some association between cultural characteristics and informativeness level of television advertising. The association between cultural characteristics and advertising content may provide an important theoretical as well as practical clue to international advertising scholars and practitioners with respect to the globalization and localization issue. Information about this association also contributes to the theoretical reasoning process regarding the relationship between culture and advertising content. WW QQBSQE§_An§l¥§1§ Content analysis is the study of the communication message itself, and it is a research technique for enjeeeige, eyseenaeic, and geeneieQELEe description of the manifest content of communication (Berelson 1952). It, thus, does not deal with the communicator or the audience. Kassarjian (1977) notes that content analysis is necessary when the subject’s own language and mode of expression is crucial to the investigation. Both points are relevant for this dissertation research. The linguistic aspect of culture is extremely important in cross-cultural research, and modes of expression are one of the major investigation areas of the 44 research. Kassarjian (1977) further indicates that content analysis can be particularly helpful when the objective and systematic evaluation of material is needed that is much too voluminous for a single investigation. In this dissertation research, the large number of commercials required the use of such a systematic technique so the above advantage could be realized. In a content analysis, areas of critical decision include unit of analysis, categories of analysis, coder selection and training, and reliability checking (Kassarjian 1977). This dissertation research requires not only a rigorous systematic analysis scheme for the content analysis but also a mechanism to maintain a certain level of equivalence between the different cultural settings. The next section deals with general concerns in cross-cultural research. WW General concerns in performing cross—cultural research include the complexity of research design, recruiting competent and comparable research participants, developing equivalent research materials, maintaining equivalent research procedures, and analyzing results (Douglas and Craig 1983). Ennigelenee_1eenee Campbell (1964) argues that differences in behavior cannot possibly be interpreted and compared unless a common underlying process exists. Gordon and Kikuchi (1966) 45 points out the danger of using the same research material without checking and solving possible equivalence problems between cultures. Bhalla and Lin (1987) specify four major types of equivalence that must be considered in planning cross-cultural testing of a theory; construct equivalence, operational equivalence, scalar equivalence, and linguistic equivalence. genee;eee_egnige1enee deals with the equality of meaning, the meaning of stimuli, and the concepts or overt behavior under study. Construct equivalence has three distinct components; conceptual equivalence, functional equivalence, and category equivalence (Douglas and Craig 1983). Conceptual equivalence is concerned with the interpretation that people in different cultures place on stimuli, product, and behavior. For example, because of different socio—cultural norms in Far Eastern societies, comparative advertising may not be considered as culturally acceptable or desirable as in the U.S.A. Functional equivalence deals with the similarity of goals of behavior across different cultures. Meanings attached to automobile ownership in different cultures may vary according to each culture’s stage of economic development and societal environment. Category equivalence refers to the way that objects, stimuli, and behaviors are grouped. Socioeconomic and demographic variables used to classify individuals and their responses may not be equivalent across cultures. For example, 46 the category ’household’ is subject to variation across cultures, depending upon typical household sizes, whether there are joint families or extended families, and so forth. Qne:e§iene1_egnigelenee deals with questions about whether there are shared operational definitions of a construct. While ‘leisure’ may be ope: -onalized with hours spent in watching television, it may not be a good operational definition in societies that do not have 24-hour television transmission (Bhalla and Lin 1987). §calar egeivelenee deals with ways measurement scales are structured to measure a construct. Not all countries express equal familiarity with all scales and scaling procedures. In the U.S.A., a five- or seven-point scale may suffice, while in other countries a 10-, 12-, or 20-point scale may be more appropriate (Douglas and Craig 1983). Lingnieeie_egeiye1enee refers to equivalence problems in both spoken and written languages applied to developing questionnaires, interviewing, training research personnel, probing open-ended responses, etc. The goal of linguistic equivalence is to maintain commonality in understanding and interpreting concepts between different languages (Berry 1969; Brislin 1983). In developing research materials for cross-cultural research, it is required to maintain linguistic equivalence. Respondents in each culture should interpret and respond to research materials with the same understanding, if valid 47 comparisons are to be made. Effective translation can help overcome some problems of equivalence. In all fairness, though, a very accurate translation is not always foolproof or enough (Werner and Campbell 1970). If similar research materials are to be used in two or more cultures, translation and back-translation should be done to check whether there may be any discrepancies in meanings and understandings between different languages (Brislin 1970; Miracle 1988). The technique is to have written material translated from one language into another, and then back-translated independently into the original language. The original materials can then be compared and evaluated with the back-translated versions to identify problems of equivalence and to determine if these problems are etic or merely difficult emic problems (Pike 1966; Jones 1979; Miracle 1988). ear ' e A comparison between more than one culture regarding a theory calls for a research design involving multiple cultures. Designing research for such cross- cultural settings may be considerably more complex than when a single country is concerned (Douglas and Craig 1983). eenjeeee Relevant and comparable subject groups may differ across cultures by their demographic identification. Although college students are intensively used as research subjects and participants in many U.S. studies because of availability and convenience, some other group or combination of groups may have 48 to be used to acquire a relevant, comparable, and representative response patterns in other cultures. t ' en 'ons Some adjustments may have to be made in a cross-cultural setting regarding administering experimental sessions, such as in sessions for coder training and reliability checking. Efforts should be made to ascertain equivalence between coordinators in charge of each nation’s research tasks regarding their understanding of all research procedures. 8W According to the two cultural differences between the U.S.A. and Korea, low- and high-context culture and Confucian and Buddhist influences in Korea, a directional hypothesis regarding the number of information cues can be stated. The hypothesis controls for product life cycle stages and product categories, because there may be extraneous effects from these variables on the total number of information cues in advertising content. E1: For the same product classes in the same stages of the product life cycle, U.S. television commercials will contain more information cues than Korean television commercials. Individualism-collectivism, and low- and high-context cultural differences between the two nations indicate that the types fo information cues contained in television commercials may also be different between the two countries. In this 49 hypothesis, there are 33 information cues included in the analysis, and a specific direction of each cue is not specified at the outset. 32: 0.8. television commercials are different from Korean television commercials, in terms of information types delivered in advertising messages. Also, the differences between the two countries; in written and oral cultures, especially influence of the Hanja writing system, and influences of Confucianism and Buddhism, lead to the following directional hypotheses regarding modes of presenting advertising information in U.S. and Korean commercials; 83a: Korean television commercials deliver information by more visual means on screen than 0.8. television commercials. 33b: U.S. television commercials deliver information by more auditory means than Korean television commercials. Product life cycle (PLC) theory indicates that there are relationships between PLC stages and the number of information cues in advertising messages. The first two hypotheses deal with intra-cultural comparison between different PLC stages regarding the number of information cues in television commercials. The third hypothesis is structured to make a cross-national comparison regarding differences in the number of information cues in different stages of the product life cycle. The hypothesized direction is based upon the review of the cultural 50 characteristics discussed with H1. 84a: Television advertising for products in the introductory and growth stages of the product life cycle contains more information cues than television advertising for products in the maturity and decline stages of the product life cycle in the U.S.A. 34b: Television advertising for products in the introductory and growth stages of the product life cycle contains more information cues than television advertising for products in the maturity and decline stages of the product life cycle in the Korea. 34c: U.S. television advertising contains more information cues than Korean television advertising for products in the same stages of the product life cycle. Product category effects indicate that there are relationships between product categories and the number of information cues in advertising messages. The first two hypotheses deal with intra-country comparisons regarding the relationships. The third hypothesis is to compare between the two countries the patterns of differences in the total number of information cues by product categories. The hypothesized direction is also based upon the cultural characteristics discussed with H1. 35a: U.S. television advertising messages for different product categories--durable goods, non-durable goods, and services-- contain different numbers of information cues. 51 HSb: Korean television advertising messages for different product categories--durable goods, non-durable goods, and services-- contain different numbers of information cues. 85c: U.S. television advertising contains more information cues than Korean television advertising for the same product categories: non-durables, durables and services. Summary Previous studies about related issues, both academic and practical, were reviewed. These areas include standardization of international advertising, advertising information, product life cycle and product categories, and cultural characteristics. In addition, the literature on methodological concerns were also reviewed. Five groups of hypotheses were derived from the literature review. CHAPTER III RESEARCH METHODOLOGY This chapter provides an explanation of methodologies utilized in this dissertation research and detailed information on research procedures. Me eme t Sca ve o t De ' ' ' ve Information has been defined in various ways. Littlejohn (1983) summarized major information theories: (1) the mathematical model of communication, (2) the semantic information theory, and (3) the effective information theory. The mathematical model of communication (Shannon and Weaver 1949) views information in three ways: a measure of uncertainty (entropy) in a situation, the number of messages required to reduce the uncertainty, and the number of alternatives available in predicting the outcome. The semantic information theory (Bar-Hillel and Carnap 1953) defines information as a message that not only reduces the uncertainty but also affects the receiver’s choice of alternatives. The effective information theory (Ackoff 1958) emphasizes the impact of the message on the receiver’s goal. Communication 52 53 may affect the purposeful state of a receiver by conveying information. Information alters the probabilities of choice, changes the efficiencies of the course of action, or changes the values of the outcome. Most previous studies of advertising information used a measurement scheme developed by Resnik and Stern (1977). They defined an "advertising information cue" as that which permits a typical buyer to make an intelligent purchase decision among alternatives after being exposed to an advertising message. The term "intelligent", however, is ambiguous and subject to subjective interpretation. For this research, a definition developed by Maenaka, Miracle and Chang (1991) is used. Advertising information is defined as a message presented by sound or picture, by which the advertiser intends: (1) to reduce the listener’s and/or viewer’s uncertainty about the product advertised, (2) to provide the alternatives of choice, or (3) to effect a change in the mental state or behavior of the target audience about purchasing or using the product. The advertiser puts information in a commercial anticipating that the target audience will use it: (1) to decide whether or not, when, where or from whom to buy a product or service, (2) to discriminate a certain brand from others, or (3) to establish and maintain loyalty. A "message" can be oral or pictorial, and it can be verbal or nonverbal. It can include consumer benefits as well as 54 product attributes, and a reminder of what the consumer may already know, as well as new information. With this definition, it seems to be appropriate to consider simple mentioning of brand name, company name, and showing product or service as information. These cues can serve as reminder that evoke top-of-the-mind awareness, which is directly related to the cognitive process. "Amount of information" refers to the number of different types of information, and "presentation mode" refers to means of delivering information cues on television, picture (visuals on screen) or sound (auditory). o ' -S 8 Using the same method in different studies under different settings is desirable and important where a comparison between different settings is concerned and a replication of the same research method is necessary. The replication of a study in other situations and at other points of time is desirable to reinforce the validity of the original study and to observe if there have been any changes over time and at different situations. One of the major requirements to apply the same method, however, is that the method should supply a valid measure and provide a reliable result. Studies on advertising information amount have been successfully replicated by using the Resnik and Stern (1977) 14-cue method. Although the measure has been 55 effective in identifying and evaluating the informativeness level of advertising messages with an accepted objectivity and reliability (Laczniak 1979; Stern, Krugman and Resnik 1981), the 14-cue method is perceived to have some problems in its format with respect to its validity. There have been related studies on advertising information which suggested some guidelines that can help improve the method. These studies are discussed in the following paragraphs. The first problem that was recognized was regarding the measure’s inadequate definition of ’advertising information’, as discussed in the above section. The second problem with the 14-cue method is with the number of information cues. Sepstrup (1985), Stewart and Furse (1986), and Pollay, Zaichkowsky and Fryer (1980), for example, have indicated that the Resnik-Stern measure may underestimate the informativeness of advertising messages. They suggest that advertising may deal with more than 14 cues. There can be more cues that advertising messages can contain to help consumers to make purchasing decisions. They recognize a need to significantly extend the list of information cues to include more information cues. Thirdly, it is difficult, with the Resnik-Stern measure, for coders to handle ambiguities caused by the cross-modal effects of communication between visual (pictorial) and verbal (aural) modes of presenting information (Jacoby, Hoyer and Zimmer 1983; Liu and 56 Stout 1987). This problem is perceived to be especially serious in television messages that use bi-modal communications, particularly in missing visual message components. The definition of advertising information has already been addressed. The next section deals with two remaining issues: the number of information cues, and the separation of presentation modes. e o nfo a io ues With the l4-cue measure, it may be possible that important information cues cannot be counted because of the limited number of information cues. The development of a new measure for this research required first a literature review on advertising information, and then informal interviews and brainstorming sessions by advertising instructors, students, and research participants, both Korean and U.S. natives, at Michigan State University. The focus of these activities was to find useful information cues in television commercial messages that can be identified and used by consumers. Dameron (1942) lists 46 different information elements that may serve consumers, based upon his observation of advertising messages. He maintains that advertising messages contained much more information than was generally assumed in 1940s. After the Resnik and Stern study was reported, some researchers attempted to extend or adapt the 14-cue list to have 57 more information cues, so as to measure information in advertising messages more properly (Sepstrup 1985). Although Harmon, Razzouk and Stern (1983) utilize the basic structure of the Resnik-Stern measure, they added one more cue, energy, to have a 15-cue list to analyze information content of comparative magazine advertisements. It is notable that one of the authors of the Resnik and Stern study (1977) participated in this study. Stern, Resnik and Grubb (1977) also suggested that there may be more than 14 cues in advertising messages that are important and relevant information to consumers. Pollay, Zaichkowsky and Freyer (1980) analyzed the content of U.S. and Canadian television commercials, with an adapted list of 19 information cues. They explain that the added cues may result in a higher percentage of informative commercials in the sample, and also indicate that the extended list may more adequately measure the informativeness of advertising messages. Sepstrup (1985) analyzed the content of German and Danish television commercials with an extended list of 24 information cues. Stewart and Furse (1985, 1986), in their studies of television commercials, used a list of 26 information cues. Moon and Franke (1987), in a study to measure information content of Korean domestic and international magazine advertisements, extended the 14-cue list to 20 information cues. The literature review and group discussions to decide the 58 proper number of information cues for the data base used for this dissertation indicated that an extension of the 14-cue list was necessary, and a new list of 33 information cues (Appendix F) was developed. The decision was made in part by a pretest of U.S. and Korean television commercials. Detailed discussion on the pretest will be given in a following section. 5 !. E E ! !' H 1 By its nature, television has a strength that other media cannot duplicate, delivering messages with sound and picture simultaneously. Liu (1986) points out that the basic information presented in television advertising is not at all unique to the medium; the same basic information can always be presented in print, still pictures, or oral presentation. He emphasizes that what is unique about television advertising is the fern in which the information is presented; the audiovisual mode. Harris, Sturm, Klassen and Bechtold (1986) also note that the impact of the information in the advertisement is partially dependent on the medium in which it is presented, and that linguistic factors may be differentially influenced by modes of presentation. In this research, separation of the presentation modes was considered necessary for the following two reasons: First, there are cultural implications of the presentation modes in the 59 communication styles, as discussed earlier in the section on cultural variables. Information about cultural differences regarding presentation nodes may be important for practitioners when they try to develop messages for different target audiences and different cultural groups, domestically and internationally. For example, sometimes, a written message may be preferred over an oral message in Korea. Secondly, separation of the presentation modes is important to help coders do an accurate content analysis of television commercials. In the rest of this section, the literature supporting the need for the separation of presentation modes into visual and auditory is reviewed. Chaiken and Eagly (1976) specify that messages conveyed through video, audio, or written modes differ in the impact they have on learning and attitude change. Richardson (1977) categorizes individual differences regarding cognitive styles, visualizer and verbalizer, and indicates that this distinction may be extended to cultures. Visualizers are relatively more dependent upon the visual cues of communications in interpreting message contents, while verbalizers are more relying on the verbal articulation of messages. Rossiter and Percy (1978), in a study of individual differences in visual imaging ability, distinguish between verbal comprehension response and visual imagery response from advertising messages, and compare the effectiveness of visually oriented advertising and verbally oriented advertising. They also suggest that both 60 the visual and verbal components of advertising have effects that lead to visual imagery response and verbal comprehension response to messages. The effects of the verbal and visual components of advertising on brand attitudes and attitudes toward advertising messages are studied by Mitchell (1986), and the two different components are found to have different levels of effects on dependent measures. Jacoby, Hoyer and Zimmer (1983), in a cross- media comparison between radio, television, and print advertising regarding miscomprehension levels, found that there are significant differences between media types and miscomprehension. Especially with regard to advertising messages, overall miscomprehension level for television was one of the highest among different media types. They suggest that television may not be the best medium for conveying information if an objective is to have it understood. Haley, Richardson and Baldwin (1984) and Stewart, Becker and Graham (1987) stress the importance of nonverbal components of marketing, and especially the effects of television advertising on consumers’ persuasion, recall, awareness, and comprehension of messages. In addition, a general review on the role of imagery in information processing was done by MacInnis and Price (1987). This line of research on the nonverbal aspects of advertising, and imagery, or imaging ability, might be better served if the video (picture) and auditory (sound) components of 61 advertising content are analyzed separately. In this dissertation research, auditory (sound) and visual (picture) presentation modes are handled separately on the data. coding instrument. Also they are handled separately in coding the content analysis procedure (see Appendix F). Detailed information about the data coding instrument and coding tasks will be discussed in a following section. W A list of 36 information cues was developed from a literature review of Dameron (1942), Sepstrup (1985) and Stewart and Furse (1986), and from research group discussion sessions (see Appendix D). The pilot study was performed to examine whether or not these cues were valid for measuring advertising information in each country. In addition, the pilot study was conducted to detect whether or not equivalence problems existed in other research materials, methods and procedures, so that improvements could be made for the dissertation research. WW Television commercials from two countries were videotaped off-the-air in May 1987 in East Lansing, Michigan and Seoul, Korea, respectively. Only national, not local or regional, commercials for products and services were included in the data base after editing. The finalized data base for the pilot study 62 included 331 U.S. and 282 Korean commercials. The information list had four choices to be checked by coders for each information cue; Xee;eenne, gee;nie§n;e, Xee-boen, and Ne. In preparing coding instruments and codebooks in two languages, a procedure of translation and back-translation was utilized between Korean and English to minimize any linguistic equivalence problems. Six native speakers for coding each nation’s commercials were recruited, in East Lansing, Michigan, as coders for the pilot study. They were students in advertising programs, undergraduate or graduate, at Michigan State University. Coders were instructed only to hear commercials without looking at television monitor to code the Xee;eeeng column throughout all 36 cues, and they were instructed to only watch the same commercial to code the gee;n1e§nze column with sound turned off. This was done to minimize possible cross-mode confusions between the two presentation modes, auditory (sound) and visual (picture). After finishing the first two lee columns, coders were instructed to process the commercial with both presentation modes, hearing and viewing to see if there were any informatiOn cues not recognized with either of two modes alone. In some cases, it was found that certain information cues that could only be identified with both the visual and auditory components of the commercials. Although it happened rarely, coders were asked to 63 check on the third column, lee;ne§n when it happened. The third column, Xee-noen was also used when both of the first two leg columns were checked. If coders found both (sound and picture) columns were checked after coding on the first two Xee columns, they were instructed to erase both checks and to check only on the third column, Isszbeth- If none of these three columns were checked, coders were asked to check the Ne column. To check intercoder reliability, the first 30 commercials in the data base were used. The procedure involved assigning 30 commercials to pairs of coders for analysis, and asking each coder to code independently under identical conditions controlled by a coordinator. For each information cue, a chi-square test was performed to see if there were cross-national differences regarding the frequency of presence. The total number of information cues in each commercials was also examined to see if there were any differences in information amount between the two nations. To check differences in presentation modes, column checks on Xee;eenng and BOLD were added together to get the sum of Sound, and column checks of Xee;nieenze and 3253 were also added together to get the sum of Picture. A t-test was performed on these aggregated measures to check whether there were cross- national differences regarding presentation modes. Areas to be reviewed and checked with the pilot study were 64 the development of data coding instruments and codebooks, the coder training and instruction, the intercoder reliability and the statistical analysis. 0 ' ' u (1) Native Koreans were recruited as coders for the pilot study. They seemed, however, to have some difficulties in coding in Korean because they had stayed in the United States to study at school with English as foreign students. During the coding procedures, they sometimes apparently used English words and U.S. or Western perspectives in interpreting and understanding advertising messages, even though they were initially instructed to think, process and code only in Korean in processing Korean commercials. From this finding, a strong need was recognized to recruit native Koreans living in Korea as coders for the dissertation study to minimize the ’acculturation’ problem and to maintain the maximum equivalence in coders’ thinking and interpreting. (2) Some linguistic equivalence problems were identified with respect to development of data coding instruments and codebooks. It was found that there was a need to use somewhat longer codebook definitions and to use more examples. (3) It was found to be important to have a large enough sample of commercials for each country’s data base in order to maintain large enough cell sizes to be analyzed whenever 65 subsectional and crosssectional analyses are needed. (4) The level of intercoder reliability was found to be satisfactory, showing the average reliability over 85% of coder agreement in both countries (Kassarjian 1977). A few cues, however, showed relatively lower reliability levels, indicating that additional coder training is needed for the main study. (5) In terms of the number of information cues, it was found that some adjustments were needed. Information regarding company was found exist in a very small number of commercials in both countries (see Appendix E). Thus, seven information cues were combined into one cue, company information. Four additional cues were found to be necessary to be included in the list; size. W 1191119121911. taste. Wasting: Taste and availability or location were information cues initially included in the Resnik-Stern (1977) measure. Thus, it was necessary to put them back in the list. The quantity and supply available or limitation and size were cues that were not included initially and frequently asked by coders on how to deal with. Therefore, it was decided to include them. The final list of information cues include 33 cues (see Appendix F). (6) The separation of presentation modes of advertising information (see Appendix F) and the data analysis statistics and techniques seemed to be properly selected and performed for the 66 content analysis purposes. No specific coding problems and statistical problems were found. Commercials to be analyzed were videotaped off-the-air in E.Lansing, Michigan in the U.S.A. and Seoul in Korea during the Fall of 1988 and the Winter of 1989 (see Appendices G and H for taping schedules). Initial videotaping included program contents as well as commercials on air in order to avoid missing any advertising messages by chance. The taping for the U.S. data was done by the author of this dissertation and the taping for the Korean data was done by staff members of the Korea Broadcast Advertising Corporation (KOBACO), a supporting and funding institution for the study in Korea.» It should be noted that taping a sufficient number of commercials took only four weeks for the U.S. sample, while the same task required twelve weeks in Korea. This may indicate that there are more identical television commercials on the air in Korea than in the U.S.A. It may also reflect a government regulation in Korea that only 8% of total broadcast time can be used for advertising purposes (Advertising Annual 1989). Specific taping instructions were developed and used in taping commercials to acquire representative and proportionate samples of programs and commercials, covering all major dayparts; 67 primetime, morning, and weekend afternoon (see Appendices G and H). Three major networks, CBS, ABC, and NBC, and a cable channel, WTBS for the U.S. sample and three major national networks, KBS1, KBSZ, and MBC for the Korean sample were used. An extensive two-step editing process was done. The first step deleted program contents and the second editing excluded any duplicate and disqualified advertising messages such as local, political advocacy, government, and generic demand advertisements. Thus, the samples include only qualified product or service advertising materials aired nationally. The final samples include 1,228 U.S. and 867 Korean commercials. The samples are comprised of a comprehensive range of product categories normally sold and advertised on national television in each country (see Appendix K). te v c As reviewed at the above section, a list of 33 information cues was prepared for the measurement of information content (see Appendices F and I). Categories for the product life cycle coding were developed based upon a literature review on the theory (Day 1981, Midgley 1981, Dodge and Fullerton 1984, Kotler 1986). They include four categories; introduction, growth, maturity, and decline (see Appendix J). Products and services were grouped into 43 different 68 categories (see Appendix K). nag; Ceding Insteument and Cedebook The data coding instrument and codebook were developed first in English, then they were translated and back-translated to resolve any possible equivalence problems between the two cultures and languages (see Appendices F, I, N, and O). The final data coding instrument with 33 information cues contains all 14 cues included in the Resnik-Stern (1977) measure. The procedures to code sound and picture were the same as described earlier for the pilot study. Another problem perceived with the Resnik-Stern 14-cue measure was the simplicity of codebook definitions on information cues. In developing codebooks, thus, a more detailed elaboration on each cue was attempted and examples were used whenever they were available in either country (see Appendix I). ; 1 E l I i . i : !' Sixteen U.S. coders and twelve Korean coders were recruited for the study. The 0.8 coders were advertising students at Michigan State University, and the Korean Coders were advertising or communication students in Seoul, Korea. The KOBACO helped in recruiting the Korean coders. Detailed coder instructions on how to deal with information cues were prepared in written form, based on lessons learned in 69 the pilot study. Coder training sessions were held both in E.Lansing and Seoul, presided by each country’s native research personnel. Important points in coder training were to train coders to be familiar with coding cues and codebook definitions and to maintain equivalent meaning and interpretation among coders and between coder groups of both countries. Thus, the Korean trainer, the author of this dissertation, first participated in all training sessions held for U.S. coders, and then went to Korea to train the Korean coders. The Korean trainer’s participation in all sessions of U.S. coder training prior to the coder training sessions in Korean was particularly important, because it contributed to the maintaining of equivalence between U.S. and Korean coders regarding interpretation of information cues and codebook definitions. .dj,.. . - .. p .t-.. '-\ .uu- '. -,.. ,; ,.-5 Product categories were coded (see Appendix K) into 43 product and service categories. The product categories were regrouped into three categories for the analysis purpose; non- durable goods, durable goods, and services. Appendix K shows how each of the 43 product categories was reclassified into one of these three categories. Commercial lengths were measured by a stopwatch and coded to the nearest second. And, for purposed of the analysis, coding 70 results were regrouped into eleven categories; 7-11 seconds, 12- 16 seconds, 17-21 seconds, 22-26 seconds, 27-31 seconds, 32-36 seconds, 37-41 seconds, 42-46 seconds, 47-51 seconds, 52-56 seconds, and 57-61 seconds. These categories were then recoded for purposes of analysis as 10 seconds, 15 seconds, 20 seconds, 25 seconds, 30 seconds, 35 seconds, 40 seconds, 45 seconds, 50 seconds, 55 seconds, and 60 seconds, respectively. Special panel coding was used to code product life cycle (PLC) stages of products and services. The Korean panel was composed of three persons: a Ph.D. in communication, and two M.A.s in advertising. They were all native Koreans and one of the M.A.s was the author of this dissertation. The U.S. panel was composed of a Ph.D. in marketing, and an M.B.A. in marketing. Panels were given definition guidelines (see Appendix J) prepared by Professor Gordon E. Miracle at Michigan State University for helping panel members to judge stages of the product life cycle. Initial PLC coding used four categories: introduction, growth, maturity, and decline. Decisions were made by panel discussion following each member’s trial independent coding based on the guidelines. However, because there were very few products in the introductory and decline stages, it was necessary for purposes of analysis to recode products and services into two categories: (1) the earlier PLC stages (introduction and growth) and (2) the latter PLC stages (maturity and decline). 71 1911M At the end of each nation’s coder training, all coders were instructed to code the first 30 commercials in each nation’s data base to check intercoder reliability (Kassarjian 1977, see Appendix M). To calculate levels of reliability, a matrix was developed for each information cue, and the level of agreement among coders was checked. The matrix was 16 * 30 for the U.S. sample and 12 * 30 for the Korean sample (16 coders for the U.S., 12 coders for the Korean samples * 30 commercials). Results of the reliability check are shown in Appendix N. The overall mean reliability levels were above the .85 level recommended by Kassarjian (1977). A few cues, however, showed reliabilities a little less than .85, so that additional training for these cues was done after the reliability check (see Appendix N). Aeminisezatien end Data Input After the reliability checking sessions, the rest of commercials were divided among coders and each coder coded commercials independently thereafter. A typical U.S. coder coded about 75 commercials and a typical Korean coder coded about 70 commercials, in addition to the 30 commercials coded in the reliability checking session. Completed data coding instruments were collected by a coordinator in each country, and the Korean data coding instruments were sent to E.Lansing for data input and analysis 72 purposes. The functions of the coordinators were important in keeping the research procedure orderly and in maintaining equivalence between nations. They not only collected data coding instruments from coders, but also provided guidelines and answers regarding coders’ questions on coding, functioning as a problem solver for coders. They met coders regularly during the individual codings to collect the data coding instruments and to have group meetings to solve whatever problems that coders had. The coordinators were also instructed to review the work of coders and to clean the data coding instruments if there were any problems on the instrument. The U.S. coordinator was an M.B.A. in Marketing, and the Korean coordinator was an M.A. in Advertising (both studying for the Ph.D). Data input was done with the SPSS/PC+ Data Entry (version 2.0) package, by which a direct on-screen input of data was possible. Developing the finalized data base included completing the data input and cleaning the data. This process required approximately six months. In order to perform these tasks, the personnel had to be trained to be thoroughly familiar with codebook definitions and with the cross-cultural scope of the research project. 73 Summazy This Chapter discussed the methodology that the current study used for the investigation. The discussion dealt with the development of the improved measure, the pilot study, and a detailed information about research methods and procedures that the main study utilized. CHAPTER IV ANALYSIS, RESULTS AND DISCUSSION A series of statistical tests was performed, utilizing SPSS/PC+ (version 3.0) package, to compare the number of information cues, types of information delivered, and delivery modes. The ability of the product life cycle and the product categories to predict information content was also tested in each country and compared between the two nations. W The literature review on cultural characteristics suggested the following directional hypothesis regarding the number of information cues contained in U.S. and Korean television advertising: Hi: For the same product classes in the same stages of the product life cycle, 0.8. television commercials will contain more information cues than Korean television commercials. To test this hypothesis, the average total numbers of information cues were compared between the two countries, for 15- second and 30-second commercials. Although the majority (37.1%) of Korean commercials are 20 seconds, the 20-second commercials were not compared between the nations because there were only eleven (11) U.S. 20-second 74 75 commercials in the data base. The small U.S. cell size did not allow meaningful statistical testing (see Appendix L). The two major categories of commercial lengths (15 and 30 seconds) covered 91.5% of U.S. but only 57.8% of Korean commercials in the data base. To avoid losing information in the Korean data base, the average total numbers of information cues for commercials of all lengths were also compared. As an alternative way to avoid losing information from the Korean data, a method to normalize the total number of information cues was considered. In this method, the total number of information cues of each commercial would be divided by each commercial’s length, so as to get a normalized total. Then the mean of normalized totals for all commercials would be compared between the two nations. This method, however, assumes that television advertising messages of different lengths contain information cues roughly proportionate to the commercial length, that is, 30-second commercials contain roughly as twice information cues as 15- second commercials. However, this assumption proved not to be true (see Table 1). Therefore, this normalization method was not utilized. 76 Table 1 Mean Number of Information Cues in Different Length Commercials 11.5.15... Mean (N1 M2191 15 seconds 5.2186 (366) --(divide by 15)--> .3479 20 seconds 4.7273 ( 11) --(divide by 20)--> .2364 30 seconds 5.9550 (756) --(divide by 30)--> .1985 BOISE 15 seconds 6.0218 (275) --(divide by 15)--> .4015 20 seconds 6.0217 (322) --(divide by 20)--> .3011 30 seconds 6.4115 (226) --(divide by 30)--> .2137 To test H1, one-tailed t-tests were performed on the mean total number of information cues in Korean and U.S. television commercials for the same product classes and in the same stage of the product life cycle. The classification of product categories included durable goods, non-durable goods, and services. The stages of the product life cycle were reduced to two categories: (1) the earlier stages (introduction and growth) and (2) the latter stages (maturity and decline). The mean total number of information cues was compared between the U.S.A. and Korea for the different combinations of the three product categories and the two stages of the product life cycle, separately for 15- second, 30-second and all commercials. The table 2 shows the mean number of information cues contained in 15-second U.S. and Korean commercials. A significant difference between the nations was found in only one of the six combinations at.the .05 significance level; in non- 77 durable goods in the latter stages of the product life cycle. It is notable that Korean commercials have more information cues than U.S. commercials, contrary to the hypothesis. Table 2 Mean Number of Information Cues in 15-second TV Commercials 1-tailed - Disses __§__ __2_ Non-durable goods/ Early PLC 5.9231 6.7037 -1.45 .078 ( 13) ( 27) Non-durable goods/ Latter PLC 5.2552 6.2229 -6.73 .000* (286) (166) Durable goods/ Early PLC 4.0000 6.0000 n/a n/a ( 1) (15) Durable goods/ Latter PLC 4.8182 5.2593 -1.37 .088 ( 33) ( 54) Services/ Early PLC 5.0000 4.1667 1.14 .150 ( 2) ( 6) Services/ Latter PLC 5.1304 5.5000 -.29 .386 (23) ( 2) * significant at .05 level. The analysis of 30-second commercials shows significant differences in three combinations (see Table 3): non-durable goods in the earlier PLC stages, non-durable goods in the latter PLC stages, and services in the latter PLC stages. It is also notable that Korean commercials for non-durable physical products have more information cues than U.S. commercials for the same products. Service categories, however, 78 show support for the hypothesis; U.S. commercials contain more information cues than Korean commercials. Table 3 Mean Number of Information Cues in 30-second TV Commercials 1-tailed 915181 .Kezes_ __£__ __£_ Non-durable goods/ Early PLC 6.4737 7.3478 -1.82 .038* ( 19) ( 23) Non-durable goods/ Latter PLC 5.7967 6.6696 -4.88 .000* (418) (115) Durable goods/ Early PLC 6.0625 6.5714 -1.00 .165 ( 16) ( 7) Durable goods/ Latter PLC 5.8739 5.7556 .39 .348 (119) ( 45) Services] Early PLC 6.3846 6.3636 .03 .488 ( 26) ( 11) Services/ Latter PLC 6.3926 5.3684 1.93 .028* (140) ( 19) * significant at .05 level. The analysis using all commercials shows significant cross- national differences in two combinations (see Table 4): non- durable goods in the latter PLC stages and services in the latter PLC stages. The former group shows that Korean commercials contain more information cues than 0.8 commercials, while in the latter group U.S. commercials have more information cues than Korean commercials. 79 Table 4 Mean Number of Information Cues in TV Commercials of All Lengths 1-tailed 111.5181 .8213; t _p_ Non-durable goods/ Early PLC 6.3333 6.8519 -1.55 .063 ( 36) ( 81) Non-durable goods/ Latter PLC 5.6693 6.4172 -7.51 .000* (765) (477) Durable goods/ Early PLC 6.1667 5.7234 1.11 .135 » ( 18) ( 47) Durable goods/ Latter PLC 5.6667 5.5372 .70 .241 (165) (188) Services/ Early PLC 6.1333 5.4091 1.37 .088 ( 30) ( 22) Services] Latter PLC 6.2707 5.1333 2.66 .004* (181) ( 30) * significant at .05 level. Hypothesis H1 was based upon the literature review on the differences in the two cultures discussed in the Chapter II. The hypothesis suggested not only that there would be cross- cultural differences but also the direction of the differences, namely that U.S. advertising contains more information cues than Korean advertising. Results, however, were mixed and on balance show no support for the hypothesis. The hypothesized cross-national difference was found only with commercials for the service categories at the 80 latter PLC stages. Commercials for physical products were found to be either not significantly different or different in the opposite direction of the hypothesized prediction. These findings provide important implications for future research. The hypothesis was based only upon the review of the four cultural characteristics discussed earlier. The findings of no significant difference and differences in the opposite direction for physical products suggest that there may be additional cultural variables that the current study did not identify. Additional cultural variables may also help to understand cross-cultural similarities as well as differences. These findings also suggest that there may be non-cultural variables that can also explain advertising informativeness. Possible variables include: (1) stages of economic development, (2) rules and regulations governing advertising content and controls on the advertising industries, (3) structures of the advertising industries, and (4) demographic characteristics of advertising personnel. These possibilities suggest directions for future studies. The non-significant differences also suggest that a standardized approach to international advertising in terms of the number of information cues does not appear to be unsuitable between the two countries for durable product categories. However, this suggestion may be limited only to the number of information cues. Additional investigation is needed to reach at 81 a more comprehensive conclusion about the advertising standardization issue. Finally, support for hypothesis (H1) in cross-national comparisons for the service categories suggests an interesting managerial implication. This finding and other findings discussed above suggest that there are significant cross-cultural differences. Therefore it seems as if a localized approach rather than a standardized approach to international advertising may be more appropriate between the two countries for developing advertising messages for services. Again, a more comprehensive conclusion about the standardization issue must be reserved at this point pending additional research. 82 W The literature discussed in Chapter II suggested the following second hypothesis: Hz: 0.8. television commercials are different from Korean television commercials, in terms of information types delivered in advertising messages. Chi-square tests were performed on frequencies of information cues for all commercials from the two countries, to see whether there were any differences between countries. Among 33 information cues, 26 cues were found to differ significantly at the .05 level (see Table 5). There were 16 information cues that occurred more frequently in U.S. than in Korean commercials: yelee, glee, eeeneeyL_eeyinee, eeeelyi guantitx_axailable. methgd_9f_nsxment. QBIBhili§!L_QSP§EQeDili£¥. MW. milabiliuLlssatien. W warranties, indenenden; research, eomeeny research, enidentified geseagcn, Beefernenee, eaeiefecgienl loyalty, seeeriogity cleim, and W. Ten information cues occurred more frequently in Korean commercials: erend heme, egeeuce name, on n nam , erice, veriety e: eneeuee, eeeliey, neg nzeeuegz new or ineneved features. 952.929.99.193. W. and W There were seven cues for which no significant differences were found: eeeee, eenee;y_infernee;en, ceneenentsl centeneez ingzeeienee, eeckaeingz snaee, safety, new ideesl new uses, 83 co v n' nc n se. These findings generally support the hypothesis that information types differ between Korean and U.S. television commercials. U.S. commercials contain more information cues regarding individual attributes and performance of products and services while Korean commercials contain more information cues related with general information, consumer characteristics, and company information. This finding is consonant with the cultural characteristics of the two countries. All four cultural characteristics discussed in Chapter II suggested that the content of Korean advertising should contain less direct and less specific information about product, brand and company than the content of U.S. advertising. Based on current practices in the two countries, these differences regarding information types used in advertising messages suggest that the localization approach is likely to be more suitable than the standardization approach for developing international advertising messages between the two countries. This suggestion, however, does not necessarily imply that effective standardization approaches cannot be found. Therefore, future research should include effectiveness studies, particularly with regard to the types of information found in this content analysis. 84 Table 5 Frequency (%) of Commercials for Each Information Cue 2-tailed U.SLA. ___Kgres__ shizsgi __2_ Brand name 1,182 (96.3) 862 (99.4) 20.18 .000* Product 1,185 (96.5) 851 (98.2) 4.51 .034* Company name 683 (55.6) 823 (94.9) 386.55 .000* Price 129 ( 9.8) 218 (25.1) 87.62 .000* Variety of product 264 (21.5) 366 (42.2) 102.73 .000* Value 54 ( 4.4) 7 ( .8) 21.92 .000* Quality 111 ( 9.0) 107 (12.3) 5.60 .018* Size 76 ( 6.2) 14 ( 1.6) 24.76 .000* Economy] savings 123 (10.0) 25 ( 2.9) 38.30 .000* Supply] quant. avail. 32 ( 2.6) 5 ( .6) 10.92 .000* Method of payment 34 ( 2.8) 1 ( .1) 20.20 .000* Dependability 79 ( 6.4) 16 ( 1.8) 23.70 .000* Nutrition] health 154 (12.6) 80 ( 9.2) 5.32 .021* Taste 282 (23.0) 189 (21.8) .34 .558 Sensory information 234 (19.1) 183 (21.1) 1.20 .274 Components] contents 425 (34.6) 291 (33.6) .20 .653 Availability] location 125 (10.2) 57 ( 6.6) 7.88 .005* Packaging] shape 63 ( 5.1) 52 ( 6.0) .58 .447 Guarantees] warranties 47 ( 3.8) 6 ( .7) 19.01 .000* Safety 25 ( 2.0) 18 ( 2.1) .00 1.000 Independent research 23 ( 1.9) 1 ( .1) 12.35 .000* Company research 23 ( 1.9) 1 ( .1) 12.35 .000* Unidentified research 64 ( 5.2) 2 ( .2) 39.71 .000* New ideas] new uses 21 ( 1.7) 11 ( 1.3) .40 .528 Performance 555 (45.2) 183 (21.1) 128.18 .000* Satisfaction] loyalty 133 (10.8) 48 ( 5.5) 17.38 .000* Superiority claim 330 (26.9) 21 ( 2.4) 216.09 .000* Convenience in use 110 ( 9.0) 85 ( 9.8) .34 .562 Special offer, event 119 ( 9.7) 47 ( 5.4) 12.12 .001* New product] new improved features 189 (15.4) 163 (18.8) 3.99 .046* Use occasion 47 ( 3.8) 182 (21.0) 152.03 .000* Image of users 133 (10.8) 279 (32.2) 145.27 .000* Company information 78 ( 6.4) 126 (14.5) 37.78 .000* Sample sizes (N) 1,128 867 * significant at .05 level. 85 e ntat'o es According to cultural differences between the two nations suggested by the literature reviewed in the ear 9r chapter, directional hypotheses were proposed as follows; 83a: Korean television commercials deliver information by more visual means on screen than 0.8. television commercials. H3b: 0.8. television commercials deliver information by more auditory means than Korean television commercials. To test these hypotheses regarding modes of presenting advertising information, the frequencies of sound presentation and the frequencies of nieenne presentation were calculated. These results were labeled as "Information by sound (IS)", and ”Information by picture (IP)", respectively. IS and IP were aggregated for all 33 cues to produce a total of each. Analyses for 15-second commercials and 30-second commercials were done separately with one-tailed t-tests on the mean frequencies of IS and IP between the two countries. In addition, the same analysis for commercials of all lengths was also done. As shown in Tables 6, 7, and 8, hypotheses were confirmed that Korean commercials use more visual presentations of information on screen and that U.S. commercials use more presentations by auditory means to deliver information cues. Hypothesis H3a concerning information by picture (IP) is supported in 15-second, 30-second, and commercials of all lengths. Significant differences regarding information by sound 86 (IS) were also found in 15-second, 30-second, and commercials of all lengths, confirming the hypothesized direction in H3b. These findings may have significant theoretical, methodological and managerial implications. First, the literature review suggested different verbal and nonverbal components of communication messages. The current study suggested that cross-national differences in terms of these two presentation modes seem to be especially associated with the following cultural characteristics: written versus oral communication, especially the Hanja writing system, and Confucianism and Buddhism. These findings suggest the need for further research on the effects of different presentation modes to convey messages in cross-cultural settings. Secondly, the method used to separate presentation modes seems to perform satisfactorily in distinguishing distinct modes of presenting advertising information and in providing results showing cross-cultural differences. 87 Table 6 Modes of Presenting Information (Mean Frequency in 15-second Commercials) _Q1§181_ _KQISQ_ Info. by Sound (IS) 4.5355 4.2182 Info. by Picture (IP) 4.0656 5.3527 Sample sizes (N) 366 275 1-tailed t ._JL__ 2.75 .003* -11.23 .000* * significant at .05 level. Table 7 Modes of Presenting Information (Mean Frequency in 30-second Commercials) _QL§LA&_ .52ISQ. Info. by Sound (IS) 5.2285 4.8805 Info. by Picture (IP) 4.7371 5.6062 Sample sizes (N) 757 226 l-tailed __11_. __2__ 2.41 .008* -6.65 .000* * significant at .05 level. Table 8 Modes of Presenting Information (Mean Frequency in Commercials of All Lengths) l-tailed _n1§181_ _Kerea_ t ._2__ Info. by Sound (IS) 5.0847 4.5260 6.90 .000* Info. by Picture (IP) 4.6336 5.4014 -10.14 .000* Sample sizes (N) 1,228 867 * significant at .05 level. 88 Thirdly, the significant differences between U.S. and Korean commercials suggests a managerial implication: For advertising messages to Korean consumers, pictorial presentations on screen may be more frequently required than in commercials for U.S. consumers. Conversely, commercials for U.S. consumers may need more verbal presentations of information cues than those for Korean consumers. These results suggest the local adaptation of messages between the U.S.A. and Korea. This suggestion, however, does not necessarily imply that effective standardized approaches cannot be found. 89 W The literature review of the product life cycle theory suggested the following hypothesized relationships between stages of the product life cycle and the number of information cues contained in advertising messages in each country: H4a: Television advertising for products in the introductory and growth stages of the product life cycle contains more information cues than television advertising for products in the maturity and decline stages of the product life cycle in the U.S.A. 84b: Television advertising for products in the introductory and growth stages of the product life cycle contains more information cues than television advertising for products in the maturity and decline stages of the product life cycle in the Korea. To test hypotheses H4a and H4b, t-tests were performed on the differences in the mean number of information cues in each country, with two groups of the product life cycle stages: the earlier stages (introduction and growth) and the latter stages (maturity and decline). Separate analyses of 15-second and 30- second commercials were done. In addition, 20-second commercials were also analyzed for the Korean comparisons. Also an analysis of commercials of all lengths was done. H4a about U.S. commercials is not confirmed for 15- and 30- second commercials. There were small differences in the numbers, in the hypothesized direction, but the differences were not statistically significant. H4a is, however, mildly supported when U.S. commercials of all lengths were analyzed, showing that 90 products at the earlier stages of the product life cycle contain more information cues than products at the latter stages. However, the absolute difference is still small (see Table 9). The results provide only limited support for proposition that the product life influences the number of information cues. This finding may indicate that U.S. advertisers do not use a substantially different number of information cues for products at different stages of the product life cycle. They may be well advised to include more information cues in commercials for products in the earlier stages to gain higher levels of awareness and knowledge for the products and services (Midgley 1981). This finding may also be due to the small cell sizes of the products at the earlier stages (17 15-second and 63 30-second commercials; see Table 9) of the product life cycle. This possibility may suggest a need for a further investigation with larger samples. 91 Table 9 Mean Number of Information Cues Across PLC Stages for U.S. Television Commercials 1-tailed Earlier_§tages Latter_§tasss __t__ __n__ 15 seconds 5.6471 5.1977 1.17 .123 ( 17) (349) 30 seconds 6.2222 5.9375 1.18 .115 ( 63) (688) All 6.1685 5.7749 1.87 .030* ( 89) (1,133) * significant at .05 level. The results of the analysis of Korean commercials also only partially support hypothesis H4b. No significant differences were found in comparisons of 15-second, 20-second, and commercials of all lengths. A significant difference was found only in the 30-second comparison, indicating that commercials for products in the earlier stages of the product life cycle contain more information cues than commercials for products in the latter stages. Korean commercials appear to be similar to U.S. commercials regarding the product life cycle effects. The results indicate that Korean advertisers also do not use a substantially different number of information cues for products at different stages at the product life cycle. In addition, it is also possible that 92 the small sample sizes may contribute to this inconsistent finding as in the U.S. case. Table 10 Mean Number of Information Cues Across PLC Stages for Korean Television Commercials 1-tailed W __s__n_ 15 seconds 6.1176 5.9820 .61 .272 ( 51) (222) 20 seconds 5.8333 6.0752 -.94 .175 ( 54) (266) 30 seconds 6.9286 6.2951 2.18 .015* ( 42) (183) All 6.2756 6.1062 1.17 .120 (156) (706) * significant at .05 level. Comparisons of the number of information cues in commercials in the two countries by stages of the product life cycles were done to test the following hypothesis: 84c: U.S. television advertising contains more information cues than Korean television advertising for products in the same stages of the product life cycle. In 15-second commercials, a cross-national significant difference was found for products in the latter stages of the product life cycle (see Table 11). The finding indicates that commercials for Korean products in the latter stages of the 93 product life cycle contain more information cues than those of U.S. products at the same PLC stages. However, between products of the two countries at the earlier stages, no significant difference was found. 30-second Korean commercials contain significantly more information cues in both the earlier and the latter stages of the product life cycle than U.S. commercials (see Table 12). The analysis of commercials of all lengths shows that Korean products in the latter stages contain significantly more information cues than U.S. products at the same stages (see Table 13). Table 11 Mean Number of Information Cues in PLC Stages (15-second Commercials) 1-tailed .915181 .39121 __I__. __n__ Earlier Stages 5.6471 6.1176 -1.09 .140 ( 17) ( 51) Latter Stages 5.1977 5.9820 -6.10 .000* (349) (222) * significant at .05 level. 94 Table 12 Mean Number of Information Cues in PLC Stages (BO-second Commercials) 1-tailed _H1§1A1 .3912! __§__ __P__ Earlier Stages 6.2222 6.9286 -2.14 .018* ( 63) ( 42) Latter Stages 5.9375 6.2951 -2.36 .010* (688) (183) * significant at .05 level. Table 13 Mean Number of Information Cues in PLC Stages (Commercials of All Lengths) 1-tailed _HL§&AL. .KQIBA __§__. ._Jl__ Earlier Stages 6.6185 6.2756 -.46 .649 ( 89) (156) Latter Stages 5.7749 6.1062 -3.83 .000* (1,133) (706) * significant at .05 level. These results indicate that there are significant differences in the opposite direction of the hypothesis between the two countries’ commercials for products and services in the latter stages of the product life cycle in terms of the number of information cues; Korean commercials contain more information cues. In addition, for products and services at earlier stages, Korean commercials also seem to contain more information cues 95 although differences were not statistically significant. This pattern was consistent in both 15-second and 30-second commercials. These findings also show that Korean products and services are generally in earlier stages than U.S ones. The data base contains 156 Korean commercials (18.0% of Korean commercials) for products and services in the earlier stages of the product life cycle. Only 89 U.S. commercials (7.2% of U.S. commercials) were for products and services at the same stages. This fact can partly explain the higher-than-expected level of informativeness in Korean commercials. Previous studies also found that commercials in lesser developed countries tend to contain more information cues because their products and services are generally in the earlier stages, requiring advertisers to provide consumers with more information (Renforth and Raveed 1983; Rice and Lu 1988). These findings may also suggest that the localization approach is more pertinent for international advertisers than the standardization approach. This suggestion does not necessarily imply that effective standardized approaches cannot be found. 96 W Hypotheses H5a and H5b were presented as follows, as suggested by the literature review: 35a: U.S. television advertising messages for different product categories--durable goods, non-durable goods, and services-- contain different numbers of information cues. ESb: Korean television advertising messages for different product categories--durable goods, non-durable goods, and services-- contain different numbers of information cues. Analysis of variance (ANOVA) tests were performed for three product categories from each country: durable goods, non-durable goods and services. Separate analyses of 15-second and 30- second commercials and an analysis using commercials of all lengths were done. In addition, the 20-second commercials were also analyzed for the Korean data. Across product categories, U.S. 30-second commercials were found to be significantly different, as were commercials of all lengths. 15-second commercials also show differences, but not statistically significant. This result may be due to the small cell sizes for durable goods and services (see Exhibit 14). These findings generally support H5a, showing differences in the number of information cues between different product categories. The product category effects on the number of information cues are confirmed and it was found that U.S. television commercials for different product categories contain 97 different numbers of information cues. This finding confirms that different message strategies are needed for commercials for different product categories, regarding the total number of information cues. The pattern of information cues is also clear from results. Commercials for service categories seem to contain more information cues than other categories, with those for non- durable goods containing the fewest cues. Those for durable goods are in the middle. However, this overall pattern is not clear in 15-second commercials, possibly due in part to the small sample segments for durable goods and services. 98 Table 14 Mean Number of Information Cues Across Product Classes (U.S. Television Commercials) 2-tailed Nen:dnrables Durables Ssrxisss __E__ __n__ 15 seconds 5.2843 4.7941 5.1200 1.567 .210 (299) '( 34) ( 25) 30 seconds 5.8250 5.8897 6.3916 6.015 .002* (440) (136) (166) All 5.6990 5.7120 6.2512 7.418 .001* (804) (184) (211) * significant at .05 level. Note: The following numbers in each pair in this table differ significantly at the .05 level, as indicated by the t-test. 1) 30-second commercials: 5.8897 and 6.3916 6.3916 and 5.8250 2) all commercials: 5.6990 and 5.7120 5.7120 and 6.2512 6.2512 and 5.6990 Significant differences were found for Korean commercials of all lengths: 15 seconds, 20 seconds, 30 seconds. In addition, significant differences were also found when all commercials were analyzed (see Table 15). Thus, the product category effects on the number of information cues are confirmed for Korean television commercials. The most informative Korean commercials are for non-durable products, the second are those for durable products, and the least informative are those in the service category. These findings suggest that different message 99 strategies regarding the number of information cues are appropriate for different product categories in developing television commercials in Korea. 15 seconds 20 seconds 30 seconds All Table Mean Number of Information Cues Across Product Classes 15 (Korean Television Commercials) Nenzdurables 6.3077 (195) 6.4619 (197) 6.7770 (139) 6.4848 (560) 5.4203 ( 69) 5.4587 (109) 5.8654 ( 52) 5.5612 (237) 4.5000 ( 8) 4.4444 ( 9) 5.7333 ( 30) 5.2500 ( 52) Durables serxiees __£;_ __2__ 2-tailed 15.782 .000* 17.544 .000* 8.664 .000* 37.768 .000* * significant at .05 level. Note: The following numbers in each pair significantly at the .05 level, as 1) 15-second commercials: 2) 20-second commercials: 3) 30-second commercials: 4) all commercials: 6.3077 4.5000 6.4619 4.4444 6.7770 5.7333 in this table differ indicated by the t-test. and and and and and and 6.4848 and 5.5612 5.2500 and 6.4848 5.4203 6.3077 5.4587 6.4619 5.8654 6.7770 100 Comparisons of the number of information cues in commercials in the two countries by product categories changes were done to test the following hypothesis: KSc: 0.8. television advertising contains more information cues than Korean television advertising for the same product categories: non-durables, durables and services. In the analysis of 15-second commercials, it was found that Korean commercials contain more information cues for non-durable and durable goods than U.S. commercials for those products. The analysis of 30-second commercials shows that Korean commercials for non-durable goods contain more information cues than U.S. commercials (see Table 17). The analysis of commercials of all lengths found that Korean commercials for non-durable good contain more information cues than U.S. commercials for the same products. But U.S. commercials for services have more information cues than Korean commercials for services. The testing of HSc shows that there are cross-national differences in the Opposite direction to the hypothesis regarding the pattern of containing information cues in television commercials for different product categories. For non-durable products, Korean commercials appear to contain more information cues than U.S. commercials, while U.S. commercials contain more information cues for services than Korean commercials. These differences suggest that the local adaptation 101 regarding the number of information cues, by product categories, is more suitable for international advertisers between the U.S.A. and Korea. This suggestion does not necessarily imply that ' effective standardized approaches cannot be found. Table 16 . Mean Number of Information Cues by Product Categories (15-second commercials) 2-tailed .lhjhlu. _Keree t ._JL__ Non-durable goods 5.2843 6.3077 -7.46 .000* (299) (195) Durable goods 4.7941 5.4203 -2.10 .019* ( 34) ( 69) Services 5.1200 ' 4.5000 .98 .167 ( 25) ( 4) * significant at .05 level. Table 17 Mean Number of Information Cues by Product Categories (BO-second commercials) 2-tailed Q,S.A. Korea t e Non-durable goods 5.8250 6.7770 -5.79 .000* (440) (139) Durable geods 5.8897 5.8654 .09 .464 (136) ( 52) Services 6.3916 5.7333 1.56 .060 (166) ( 30) * significant at .05 level. 102 Table 18 Mean Number of Information Cues by Product Categories (Commercials of All Lengths) 2-tailed U.§.A, Korea t Non-durable goods 5.6990 6.4848 -8.37 .000* (804) (561) Durable goods 5.7120 5.5612 .91 .363 (184) (237) Services 6.2512 5.2500 3.06 .002* (211) ( 52) * significant at .05 level. This Chapter presented results of statistical analyses to test the hypotheses given in Chapter II, and provided discussions of their meaning and implications. No support was acquired for H1 which suggested that Korean television advertising contains more information cues than U.S. commercials. Further research on additional cultural and other related variables with respect to advertising informativeness was suggested. H2 predicted cross-national differences regarding types of information cues delivered, and was supported by the results. Twenty six (26) out of thirty three (33) cues were found to differ significantly between Korean and U.S. commercials. Support for the localized approach to international advertising 103 was suggested from the findings, and a need to further investigate these findings was indicated. The results provided support for H3a and H3b. Korean commercials use more visual presentations than U.S. commercials while U.S. commercials use more auditory means to communicate advertising information to consumers. The findings were in harmony with what the cultural variables had suggested and suggested the localization approach over the standardization approach in international advertising between the two countries. H4a and H4b, which predicted the relationships between stages of the product life cycle and the number of information cues in each nation, did not receive support. However, the results showed cross-national differences in the opposite direction of the hypothesis H4c, indicating that Korean commercials contain more information cues for products in the latter stages of the product life cycle. Managerial recommendations to use more information cues for products and services at the earlier PLC stages in each country and to take the localization approach in developing international advertising between the two nations were suggested. H5a and H5b were supported by the results, indicating that the numbers of information cues contained in television commercials are different depending on product categories. H5c regarding cross-national differences in the number of information cues was not supported. This was because there were cross- 104 national differences in the opposite direction of the hypothesis especially between non-durable goods and between services of the two countries. From these findings, it was suggested that different message strategies are required for different product categories in each country, and that the localization approach is likely to serve better than the standardization strategy between the two nations in developing advertising messages. CHAPTER V CONCLUSIONS W The hypothesis predicting cross-cultural differences in terms of the number of information cues contained in television commercials (H1) did not acquire support from the results. Furthermore, the results report contradicting findings to the hypothesized direction for physical products, showing that Korean commercials contain more information cues than U.S. commercials. The hypotheses dealing with types of information did, however, have support from the results. Significant cross- cultural differences were found in 26 cues on the 33-cue list, regarding frequencies of uses of those cues (H2). These differences suggest that cultural characteristics are associated with the types of information cues used frequently in television commercial of each nation. It was further found that Korean commercials use more visuals on screen than U.S. commercials, while U.S. commercials use more auditory means than Korean commercials, to deliver information cues. This finding suggests that the previously presented cross-cultural differences are associated with these two modes of presenting advertising information. It also 105 106 suggests the need for further investigation of advertising. These findings showing cross-national differences suggest that the localized approach may be more suitable than the standardization approach between the two countries in developing international advertising messages. However, this does not necessarily mean that effective standardized approaches cannot be found. P c c 'e Although most previous studies confirmed the relationships between the stages of the product life cycle and the amount of advertising information, the results from the current study did not find the relationship as predicted. Only partial support could be given to the hypothesis in either country. Cross- national comparisons show that Korean commercials generally contain more information cues than U.S. commercials in all stages of the product life cycle. The number of advertising information cues was found to differ significantly in different product categories in both countries. Cross-national comparisons also show that Korean commercials contain more information cues than U.S. commercials for commercials for physical products. But U.S. commercials for services contain more information cues than Korean commercials. 107 Theorericel gonelueions The major purpose of the current study was to investigate whether cultural characteristics are associated with the information content of U.S and Korean television commercials. The hypotheses were based on the following cultural characteristics: (1) individualism and collectivism, (2) high and low context culture, (3) written and oral communication, and (4) influences of Confucianism and Buddhism in Korea. Theory based on these cultural characteristics suggested the nature of the information content in television commercials from the two countries, as expressed in the hypotheses. To the extent that the hypotheses are accepted or rejected, the underlying theories based on the indicated cultural characteristics are supported or not supported. Hall (1983) indicated that the level of context, high or low, has implications for communication content and style. People in high context cultures such as Korea may feel others are talking down to them when they present too much information. Conversely, too little information may mystify people in low context cultures such as those of the U.S.A. or Germany. Yum (1987) pointed out that the Korean communication style can be characterized as implicit and indirect as a consequence of the religious and philosophical legacy from both Confucian and Buddhist tradition. In contrast, the U.S. communication style tends to be relatively explicit and direct. Therefore, it was 108 inferred and hypothesized that the content of Korean advertising would contain less information than U.S. advertising. The results, however, showed that these characteristics did not predict the comparative number of information cues in television commercials of the two countries. In contrast to theory, the results showed that Korean commercials generally contain more information cues than U.S. commercials. These results cast doubt on the ability of these two cultural variables to predict or explain the total number of information cues in advertising. This study, however, found cross-national differences between the two nations regarding the types of information cues and modes of presenting information. It was inferred, from the collectivism literature (Hui and Triandis 1986; Hofstede 1980 and 1984; Hofstede and Bond 1984) that collectivism in Korea, together with its manifestation such as conformity and inclination to seek harmony, may influence the country’s advertising content to contain relatively less direct and confronting information about specific brands and products than that of an individualistic society such as that of the U.S.A. The high and low context culture dichotomy (Hall 1976, Hall and Hall 1983, Gudykunst and Nishida 1986) was also used to derive the hypothesis regarding information types. It was inferred that, due to the high-context characteristics of the Korean culture and language, the advertising communication style 109 may place relatively little emphasis on the articulated strengths of products and brands advertised on television. Instead, situational cues and circumstantial expression may be more important in Korean commercials than in U.S. commercials. This expectation was generally confirmed by the results of the study. It was found that U.S. commercials contain more information cues regarding individual attributes and performance of products and services while Korean commercials contain more information cues related to general information, consumer characteristics, and company information. Also, the more frequent uses of company name and company information in Korean commercials than in U.S. commercials reflect the cultural inclination toward collectivism in Korea. Korean advertisers may use these types of information cues more frequently for consumers who rely relatively more on the good name and reputation of manufacturers rather than relying on attributes of the product or the brand. These results suggest that cultural characteristics can indeed be used to predict or explain types of information content and the modes of presenting information. Consonant with the expectation based upon the high and low context cultural dichotomy (Hall 1976, Hall 1983), Korean commercials were found to less frequently use information cues directly related to the specific product or brand than U.S. commercials. Korean advertisers may not include many information cues directly related to the brand in their 110 commercials, because they know that Korean consumers may not need to hear or see those types of cues. General information such as mere identification of product, brand, or company name, and situational information such as use occasion and image of users are more frequently used in Korean commercials. These reflect the high context culture that does not need as much specific direct information about the product or the brand. U.S. commercials contain more direct information cues related to performance and attributes of a specific product or brand, such as durability, nutrition or health, research evidence, performance, superiority claims, special offers or events. These results regarding high and low context cultures also suggest an adjustment in the content of the Hall’s (1976, 1983) definition. In its present form, the definition seems to deal only with the amount of information that a message contains, indicating that messages of high context cultures contain less information, and those of low context cultures contain more information in the explicit code. But, the results of this study suggest that the definition should also deal with the types of information in addition to the amount of information. The following revised definition may serve more properly: A low context (LC) communication or message is one in which relatively much direct or issue-relevant information is contained in the explicit code of the message. A high context (HC) communication or message is just the opposite; i.e., relatively little direct or issue-relevant information is contained in the explicit code of the message. The high context message instead contains relatively more information 111 providing contextual and circumstantial ideas, and information suggesting that the message source can be trusted and depended upon. The dichotomy of low and high context communication, thus, is based on the types as well as the amount of information contained in the message. It should be understood in the above paragraph that "direct or issue-relevant" information is used in a sense that is appropriate for the English language and the U.S. culture. In the U.S.A., such information may be considered to include primarily the advantages of purchasing and using a particular brand or product. In Korea, however, direct or issue-relevant information may be considered to include primarily information for a consumer to select a trusted and dependable company or source instead of brand or product features or benefits. Cross-national differences regarding the modes of presenting information cues also provided support for the inference that cultural variables are related to written and oral communication, especially the use of the Hanja writing system in Korea, and the Confucian and Buddhist written tradition manifested in the Korean culture. Choe (1980) pointed out that the Oriental view of communication was oriented relatively more to written words, while the Western view was directed relatively more to spoken words. Yum (1987) also reported that the Confucian tradition in Korea may encourage the written form of communication while suppressing oral communication. Indeed the religious and philosophical background in Korea seems to discourage the use of 112 auditory means of presenting information. From these arguments and observations, it was inferred that Korean commercials are expected to use more visual means on screen and U.S. commercials are expected to use more auditory means to deliver information cues. The results confirmed the expectation. It was indeed found that Korean commercials use more visual means and U.S. commercials use more auditory means in delivering information cues. Therefore, the cultural differences based on the written and oral communications seem to be useful in explaining the differences in presentation modes. Theory building regarding influences of different religions and philosophies on the content of advertising seems to be needed. Also, further efforts to develop theoretical explanations about the relationship between the presentation modes and consumer reactions seem to be worthwhile for future research. The current study also attempted to investigate the relationships between marketing theories and the number of information cues contained in television advertising and marketing theories, namely, product life cycles and product categories, within each nation and between the two nations. The results did not support the product life cycle theory. Products and services at different stages of the product life cycle did not contain different numbers of information cues, as suggested by the theory. Also, contrary to the theory, the 113 results also showed that Korean commercials contain more information cues than U.S. commercials for products in each stage of the product life cycle. The direction is contrary to the hypothesized expectation. However, the results showed that commercials for different product categories contain different numbers of information cues in each country. Commercials for Korean non-durable goods contain the most information cues, those for durables the second most, and those for services the least. Among U.S. commercials, those for services contain the most information cues, those for durable goods are second, and those for non-durables are third. These findings seem to be consistent across different commercial lengths. This finding indicates that ’product category’ is useful to predict the information content of advertising. However, a comprehensive explanation of these patterns is not within the scope of the current study. Further efforts are needed to build a theory regarding this issue. In sum, the usefulness of cultural variables in explaining some aspects of the information content of advertising was demonstrated by the current study. Individualism and collectivism, high and low context cultures, written and oral communication, and influences of Confucianism and Buddhism were found to be associated with the information content of U.S. and Korean television advertising, especially with respect to types of information cues and presentation modes. 114 Methodologieal Conclusions The results of this study show that the new methods for the current study was useful to identify advertising information cues properly. On average, the U.S. and Korean television commercials were found to contain five to six information cues. Those who used the Resnik-Stern 14-cue measure usually found one to two information cues per advertisement (Resnik and Stern 1977; Hong, et al. 1987; Madden, et a1. 1986; Rice and Lu 1988). This difference suggests that advertising may be much more informative than indicated by those who used the Resnik-Stern limited list of information cues. Another notable aspect of the improved measure is the separation of visual and verbal modes of presenting advertising information. This separation significantly reinforces the measure of advertising information by providing an additional dimension of the measurement of advertising information. The study was able to find that advertising messages are significantly different cross-nationally along the dimension of presentation modes, in accordance with the hypotheses which were based upon cultural characteristics. It is recommended that those who study advertising information should investigate both visual and verbal cues, especially if the study deals with television advertising or if it involves cross-cultural comparisons. Cultural variables appeared to be associated with the modes of presenting of information content. 115 Due to the cross-cultural scope of the study, special concerns were taken inevitably throughout the research procedures. It was important and critical to maintain equivalence of research methods and procedures between the two nations. Those areas include; the development of data coding instruments and codebooks, the recruitment of research personnel and coders, the coder training and coder instructions, control of the coding process, the administration and the data input, and the analysis and the interpretation of results. The methods developed for cross-cultural studies such as translation and back-translation techniques were utilized for the study. MenagerieLgenelusiebs No significant differences were found in the overall numbers of information cues between U.S. and Korean commercials. Korean and U.S. television commercials, however, were found to be different in modes of presenting information used and types of information cues delivered. Cross-national differences regarding advertising informativeness were also found to be associated with different stages of the product life cycle and different product categories. Discussions of the standardization of international advertising were based on the results of this study in Chapter IV. The findings from the current study suggested that the local adaptation approach was recommended for international 116 advertisers between Korea and the U.S.A. Differences in cultural characteristics appear to be associated with the information content of television commercials from the two countries. It seems clear that cultural characteristics are associated with information types and modes of presenting advertising information. Although the current study deals only with the content of domestic advertising messages of the two countries, the differences found from the results may guide international marketers and advertisers. The content of domestic advertising reflects important cultural characteristics. Thus the certain culturally based content can be accepted readily in each country, but sometimes not in the others. Therefore, the results of this study can be used to prepare guidelines for international advertisers interested in doing business between Korea and the U.S.A. Those cross-national differences include: (1) U.S. commercials contain more specific information regarding attributes and strengths of products than Korean commercials, (2) Korean commercials use more visual presentations while U.S. commercials use more auditory means to present advertising information, and (3) U.S. commercials for service categories contain more information cues than Korean commercials, while Korean commercials for non-durable products contain more information cues than U.S. commercials. 117 Findings regarding the product life cycle suggest that marketing and advertising managers should pay more attention to the managerial implications of the theory. The theory suggests that products in the early stages of the cycle need to contain more information cues in their advertising messages. But the results show that there were no clear differences in the number of information cues in commercials for products and services in different stages of the product life cycle. A key limitation of this dissertation research is that the study is descriptive in nature. It analyzes the content of television commercials and does not involve any investigations of the effects of advertising information on consumers. However, as Kassarjian (1977), Holbrook (1977) and Holsti (1968) suggested, a content analysis is especially appropriate when objective systematic evaluation of material is too voluminous for a single investigator, and when the subject’s own language and mode of expression is crucial to the investigation. The problem of subjective coding of information cues was a methodological problem. Although efforts were made to develop clearly defined cues and codebook definitions of the cues, and to assure the cross-cultural equivalence of research materials and procedures, a certain level of subjectivity seemed to be inevitable in deciding whether each information cue on the list 118 was present in each commercial. The addition of a measure of the consumers’ perception of informativeness to the objective content measure might provide a partial solution to this problem of subjectivity, as suggested by Aaker (1984). The country-specific findings of the research may be another limitation. The findings and conclusions of the study might be appropriate only for comparisons of U.S. and Korean television advertising messages. To overcome this limitation, future studies dealing with other countries will also be needed. Directions for Eerure Beeeerch To extend the scope of the current study, a further investigation of the effects of information cues on consumer reactions may be suggested. According to Aaker (1984), a consumer perception measure of advertising informativeness, as well as to the measure provided by content analysis, is an important guide to future research. A content analysis can illustrate the characteristics of messages. It cannot, however, show what impacts the messages may have on consumers, regarding effectiveness measures such as recall, retention, comprehension, persuasion, attitudes toward advertisements, attitudes toward brands, intentions to purchase, and actual purchase. Content investigation of television commercials from other countries will help overcome the country-specific limitation of the current study. This extension should attempt to provide a 119 more comprehensive understanding of the relationships between cultures and effective advertising content. Investigation of additional cultural characteristics and their association with advertising informativeness is needed to have a more comprehensive understanding about cross-cultural and cross-national similarities as well as differences. Other cultural variables may include power distance, masculinity, uncertainty avoidance, etc. (Hofstede 1984b). Investigation on non-cultural variables related to economic and societal conditions was also suggested earlier. It is speculated that one of the most fruitful topics for further research is to identify and test the relationships between these additional variables and the advertising informativeness. Finally, there seems to be a need to review the entire body of literature dealing with characteristics of different cultures as they relate to marketing communication. Such a review will provide a base for research to develop a linkage between cultural characteristics and social phenomena such as advertising. APPENDIX A COMPARISONS BETWEEN SOUTH KOREA AND THE U.S.A. 120 Appendix A betw e ou 0 nd U. . . I. General Statistics U.S.A. Koren, Population 248,777,000 (’89) 42,380,000 (’89) Density per sq. mi. 68 (’89) 1,106 (’89) Gross National Product $4,880,600M (’88) $112,947M (’87) Advertising EXpenditure $123,930M (’89) $2,303M (’89) Per Capita Income $19,860 (’88) $2,690 (’87) Literacy 95.5% (’80) .92.7% (’81) Average household size 2.6 (’88) 4.0 (’87) Average household income $32,144 (’87) $8,510 (’87) Newspaper circ./ 1,000 pop. 255 (’88) 265 (’86) Television ownership 1 / 1.3 persons 1 / 4.9 persons Teieenene oenerehin 1 L 1.9 bereene, 1 l 4.1 pereene Source: Encyclopedia Britannica 1990. II. Advertising Expenditure (1989)' (in $,000s) U.S.A. Koree Newspapers 32,368,000 (26.1%) 903,972 (39.2%) Magazines 6,928,000 ( 5.6%) 121,631 ( 5.3%) Television 26,891,000 (21.7%) 773,972 (33.6%) Radio 8,323,000 ( 6.8%) 92,117 ( 4.0%) Direct Mail 21,945,000 (17.7%) Outdoors 1,111,000 ( 0.9%) Misceiieneees 2 364 00 2 .3 41 404 7.9 * TOTAL 123,930,000 2,303,096 Sources: Ha 1990 and Coen 1991. Applied exchange rate for Korea (US $:Korean Won) = 1:679.60 * Miscellaneous in Korea includes outdoors, direct mail, overseas, and other miscellaneous advertisements. APPENDIX B DIMENSIONS OF INTERNATIONAL ADVERTISING TO BE CONSIDERED FOR STANDARDIZATION 121 Appendix B Dimensions of International Advertising t b ’ a 'zat' oca izat'o ° I. Advertising Objectives - Communication tasks - Defined target audience - Specified periods of time II. Advertising Message Strategies - Theme - Appeals - Positive vs. negative approach - Benefits or product/service attributes - Information amount and types - Image vs. product attributes - Comparisons vs. no comparisons - Spokesperson vs. voice-over - Color mix, etc. III. Advertising Media Strategies - Choice of media categories and vehicles - Size and length of advertising materials - Scheduling patterns; continuity, flighting, pulsing IV. Advertising Budgets - Heavy, medium, or light weight - By product - By market area - By customer classification - By selected media - For selected creative or message approaches - Match or not match competitors - Contracyclical or not V. Advertising Decision Processes ' Prepared by Professor Gordon E. Miracle. APPENDIX C RESNIK AND STERN’S FOURTEEN CRITERIA FOR CLASSIFICATION 122 Appendix C Resnik and Stern’s Fourteen (14) Criteria' f as ' 'cat' as a 've o No - o at've Price-Value - What does the product cost? - What is the need satisfaction capability] dollars? - What is its value retention capability? Quality - What are the product’s characteristics that distinguish a particular product from competing products based upon an objective evaluation of workmanship, engineering, durability, excellence of materials, structural superiority, superiority of personnel, attention to detail, or special services? Performance - What does the product do and how well does it do what it is designed to do in comparison to alternative purchases? Components or Contents - What is the product composed of? - What ingredients does it contain? - What ancillary items are included with the product? Availability - Where can the product be purchased? - When will the product be available for purchase? Special Offers - What limited-time non-price deals are available with a particular purchase? Taste - Is evidence presented that the taste of a particular product is perceived as superior in taste by a sample of potential customers? (The opinion of the advertiser is inadequate.) Nutrition - Are specific data given concerning the nutritional content of a particular product or is a direct specific comparison made with other products? 9. 10. 11. 12. 13. 14. 123 Packaging or Shape - What package is the product available in which makes it more desirable than alternatives? - What special shapes is the product available in? Guarantees and Warranties - What post-purchase assurances accompany the product? Safety - What safety features are available on a particular product compared to alternative choices? Independent Research - Are results of research gathered by an "independent" research firm presented? Company Research - Are data gathered by a company to compare their product with a competitor’s presented? New Ideas - Is a totally new concept introduced during the commercial? - Are the advantages of the idea presented? ’ Source: Stern, Resnik and Grubb (1977). APPENDIX D INFORMATION CUES FOR THE PILOT STUDY 1. 3. 5. 7. 9. 11. 13. 15. 17. 19. 21. 23. 25. 27. 29. 31. 33. 35. 124 Appendix D io es 0 Brand name Company name Variety of the product Quality Durability Aesthetic appeals Methods of purchasing Guaranty/warranty Nutrition/health Research done by company New uses/application of the product Consumer satisfaction Usage convenience New product/improved features Users’ characteristics Company’s experience and size Location of company Policies toward government t 2. 4. 6. 8. 10. 12. 14. 16. 18. 20. 22. 24. 26. 28. 30. 32. 34. 36. t est Product name Price Value Economic value/savings Sensory information Contents/ingredients Packaging Safety Research from outside company Research from unknown sources Results of using Product superiority claim Special event or special offer Usage occasion Image, reputation of company Variety of business Policies toward public Policies toward employees APPENDIX E RESULT OF THE PILOT STUDY: DIFFERENCES OF ADVERTISING INFORMATION FREQUENCY 125 Appendix E Results of the Pilot Study: HifisAs KQBEA Snizégs 1. Brand name 313 282 13.95 .0002 2. Product 326 280 .30 .5828 3. Company name 195 272 116.21 .0000 4. Price 27 84 46.60 .0000 5. Variety of product 66 126 42.19 .0000 6. Value 21 13 .58 .4484 7. Quality 47 37 .07 .7877 8. Economy/Savings 29 14 2.81 .0938 9. Dependability 33 9 9.93 .0016 10. Sensory Information 110 149 23.19 .0000 11. Components/Contents 138 77 13.22 .0003 12. Availability 24 13 1.44 .2308 13. Packaging 73 174 97.85 .0000 14. Guarantees/Warranty 16 12 .02 .8825 15. Safety 5 11 2.55 .1105 16. Nutrition/Health 48 71 10.42 .0012 17. Independent research 2 0 .36 .5505 18. Company research 4 0 1.82 .1774 19. Research from unid. source 23 4 9.79 .0018 20. New uses 9 8 .00 1.0000 21. Results of using 132 119 .25 .6173 22. User’s satisf.]Loyalty 97 107 4.74 .0295 23. Superiority claim 121 17 79.61 .0000 24. Convenience in use 58 18 16.39 .0001 25. Special offer or event 26 14 1.64 .2005 26. New product] 78 33 13.66 .0002 New or improved features 27. Use occasion 130 138 5.39 .0202 28. Characteristics or image 133 66 18.79 .0000 of users 29. Company image or 19 44 15.01 .0001 reputation 30. Company age or size 7 8 .09 .7532 31. Variety of company 2 0 .36 .5505 business 32. Location of business 4 3 .00 1.0000 33. Policies toward public 1 3 .44 .5066 34. Policies toward government 0 0 n.a. n.a. 35. Policies toward employees 0 0 n.a. n.a. 36. Other information 0 50 61.56 .0000 SAMPLE SIZES (n) 331 282 - Reported are number of commercials having each information cues in each country, chi-square, p-value, and * is marked where significant difference is at .05 alpha level. APPENDIX F DATA CODING INSTRUMENT 9. 10. 11. 12. 13. 14. 15. 16. 17. Appendix F WW YES Picture Both 2 3 Brand name Product name Company name Price Variety of Product Value Quality Size Economy or savings, quantity/price Supply, quantity available or limitation Method of payment, cash, credit, etc. Dependability, reliability or durability Nutrition or health Taste Sensory information (other than taste) Components, contents or ingredients Availability, location NO 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. Packaging or shape Guarantees or warranties, return, privilege, return service Safety Independent research Company sponsored research Research from unidentified source New ideas, new uses Performance, results of using User’s satisfaction or Loyalty Superiority claim Convenience in use Special offer or event New product or new or improved features Use occasion Characteristics or image of users Company information 127 APPENDIX G VIDEOTAPING FOR U.S. COMMERCIALS Fall 1988 Winter 1989 Monday Tuesday Wednesday Thursday Friday Saturday Sunday 128 Appendix G Teeing Eeriods for U.S. Commercials October 10, 1988 - October 23, 1988 January 16, 1989 - January 29, 1989 (4 weeks) c . omm ' s (63 hours per week) 10 a.m. - 2 p.m. 6 p.m. - 11 p.m. ABC (10-12 noon) CBS (6-9 p.m.) NBC (12-2) WTBS (9-11) CBS (10-12) ABC (6-9) WTBS (12-2) NBC (9-11) NBC (10-12) NBC (6-7) ABC (12-2) WTBS (7-8) CBS (8-11) NBC (10-12) CBS (6-8) CBS (12-2) ABC (8-11) ABC (10-11) ABC (6-8) NBC (11-1) CBS (8-9) WTBS (1-2) NBC (9-11) 1 p.m. - 5 p.m. 6 p.m. - 11 p.m. ABC (1-3) NBC (6-9) CBS (3-5) WTBS (9-11) NBC (1-3) ABC (6-9) CBS (3-4) CBS (9-11) WTBS (4-5) APPENDIX H VIDEOTAPING FOR KOREAN COMMERCIALS 129 Appendix H Inning Eerieds for Kereen Commercials Fall 1988 October 24, 1988 - December 4, 1988 Winter 1989 January 16, 1989 - February 26, 1989 (12 weeks) V o 'n c o ea Co cia s (63 hours per week) 6 a.m. - 10 aeme 6 poms - 11 p.m. Monday KBS-1 (6-8 a.m.) KBS-1 (6-7 p.m.) KBS-2 (8-9) KBS-2 (7-9) MBC (9-10) MBC (9-11) Tuesday MBC (6-8) MBC (6-7) KBS-1 (8-9) KBS-1 (7-9) KBS-2 (9-10) KBS-2 (9-11) Wednesday KBS-2 (6-8) KBS-2 (6-7) MBC (8-9) MBC (7-9) KBS-1 (9-10) KBS-1 (9-11) Thursday KBS-1 (6-8) KBS-1 (6-7) KBS-2 (8-9) KBS-2 (7-9) MBC (9-10) MBC (9-11) Friday KBS-2 (6-8) KBS-2 (6-7) MBC (8-9) MBC (7-9) KBS-1 (9-10) KBS-1 (9-11) 1 p.m. - 5 p.m. 6 p.m. - 11 p.m. Saturday MBC (1-2) MBC (6-8) KBS-1 (2-3) KBS-1 (8-10) KBS-2 (3-5) KBS-2 (10-11) Sunday KBS-1 (1-2) KBS-1 (6-8) KBS-2 (2-3) KBS-2 (8-10) MBC (3-5) MBC (10-11) * KBS is Korea Broadcasting Systems, state-owned stations, and MBC is Moonhwa Broadcasting Company, state-controlled station, where higher than 51% of its financial rights are owned by the state. APPENDIX I CODEBOOK 10. 11. 130 Appendix I 9999.59.95. Brand name: Refers to the brand name(s) of the product(s) or service(s) advertised. Product: Refers to describing or showing of the product advertised. Company name: Refers to the name of the manufacturer or the service entity. Price: Refers to the amount the consumer must pay for the product or service; this may be in absolute terms, like a suggested retail price, or relative terms, like a 10 percent off sale. Variety of the product: Refers to claiming for or featuring more than one kind of product. Value: Refers to some combination of price and quality or quantity, as in more for the money, better quality at a low price, the best value for the dollar. Quality: Refers to how good the product or service is; may refer to craftsmanship and/or attention during manufacture, use of quality (better, best) ingredients or components, length of time to produce or create the product. May refer to or use symbol of quality standard, such as Good Housekeeping Seal of Approval in the USA or KS or Q in Korea. Size: Refers to the physical size or capacity of the product, how long, tall, wide, heavy, capacity to do particular size tasks. Economy/Savings: Refers to saving money or time either in the original purchase or in the use of the product relative to other products in the category. Supply, quantity available or limitation: Refers to how much or how many items are available and directly or indirectly the need to act before the supply is exhausted. Method of payment: Information on preferred method to pay, for example by credit card over the telephone. 12. 13. 14. 15. 16. 17. 18. 19. 131 Dependability/Reliability/Durability: Information concerning how long the product will last without repair, service records, and so on. Nutrition/Health: Information concerning the nutritional or health-related characteristics of the product--for example, "fortified with vitamin D," " the formula doctors recommend,” "relieves iron-poor blood. Taste: Primarily for food, drink or personal care product. Sensory information (other than taste) (fragrance, touch, comfort, styling, color, sound): Information concerning a sensory experience, appearance, classic beauty, beautiful sound, and so on of the product either when purchased or when prepared in final form; "smells April fresh," "feels silky smooth," "luxurious comfort." Components/Contents]Ingredients: What went into the making or manufacture of the product--for example, "contains lanolin," "made with pudding." These contents should be in the products purchased, not ingredients added to the product by the consumer in preparation for use. Can also refer to several items served in a restaurant meal, or components of other services such features in a life insurance policy. Availability: Any information concerning the place(s) the consumer may purchase or otherwise obtain the product--for example, "available in supermarkets," "look for it in the dairy section." May also refer to places where the product is not available-~for example, "not available in all areas." Packaging or shape: Information about the packaging of the product--for example, "look for the package with the red spoon,” "look for our special two in one package," "the package is reusable," "in the convenient one serving package." If no package, information about the shape, ease of handling, functions related to shape such as ease of storage, protected from damage. Guarantees/Warranty: Refers to any information concerning the presence of a guarantee or warranty, including but not restricted to money back offers, offers to repair or service the product in the event of problems, or offers to replace the product if the consumer is dissatisfied or has a problem. 20. 21. 22. 23. 24. 25. 26. 27. 132 Safety: Information concerning the safety of the product-- for example, "has a built-in cut-off switch," "won’t harm delicate hair," "nontoxic." Independent research results: Information offered about tests of the product or of product users that were carried out by an identified individual or organization other than the company manufacturing or distributing the product, such as Underwriter’s Laboratory, a leading university, or the U.S. government. Such test may concern objective product characteristics ("lasts twice as long") or may be related to user preferences ("preferred by two-thirds of the people surveyed”). Company research results: Information about tests of the product or users of the product that were carried out by the company manufacturing or distributing the product--for example, the Pepsi challenge. Research from unidentified source: Information about tests of the product or users of the product when the source of the test results is not identified. New ideas, new uses: Refers to any information about a new way to use an established product--for example, ”use x brand paper cups for sorting and storing nuts and bolts," "use Y baking soda to deodorize refrigerator." Performance, results of using (either tangible or intangible): Any information concerning the outcomes associated with the use of the product. These outcomes may be in a positive form-- "gives hair bounce," "gives a smoother ride," "makes you feel healthier,"--or a negative form--"won’t yellow floors.” Performance deals with whether the product accomplishes a consumer purpose, e.g. clean dishes. Includes side benefits such as softer hands after using a detergent. User’s satisfaction/loyalty: Refers to any information concerning users’ satisfaction, dedication, preference for the brand, or length of time consumer has used the advertised product--for example, "I’d never give up my Tide,” "I’ve always used ...." Superiority claim: Information that claims the advertised product is better than competitive products or better than an older version of the advertised product in some particular ways. 28. 29. 30. 31. 32. 33. 133 Convenience in use: Information concerning the ease with which the product may be obtained, prepared, used, or disposed of. Special offer or event: Information concerning special events such as sales, contests, two-for-one deals, premi- ums, or rebates to occur for a specified limited time. New product or new or improved features: Refers to any information concerning a new product introduction, new components, ingredients, features, or characteristics of an existing product or an improvement (qualitative or quantitative) in any feature, component, ingredient, or characteristic of an existing product--for example, "new and improved," ”now with 50 percent less sugar," ”new, milder.....” "new, stronger....,” "now with built-in flash." Use occasion: Information that clearly suggests an appropriate use occasion or situation for the product-- for example, "buy films for the Christmas season," "enjoy Jello at a birthday party," ”the beer for special occasions.” Characteristics or image of users: Refers to any information concerning the type(s) of individual(s) who might use the advertised product--for example, "for the young at heart," "for the busy career woman." Company information. May refer to any information about the image or reputation of the company that manufac- tures or distributes the product--for example, "we’ve been in business longer than anyone else," "we try harder," "babies are our business.” Indicates the company is competent, reputable or trustworthy in certain ways. May refer to company age or size such as claiming accumulated business experience of the company throughout its history, or bigness in terms of sales or employees. May refer to claims relating to varieties of business that the company is involved in, for example a conglomerate. May refer to location of business such as claiming advantages because of the geographical location of the company’s business, or its raw materials. APPENDIX J DIFINING THE PRODUCT LIFE CYCLE 134 Appendix J For purposes of this project, the life cycle is defined broadly by product categories (rather than by product type, form or brand). Thus we are concerned with categories such as word processors, Videocassettes, FAX machines, air conditioners, beer, fruit juice, passenger automobiles, cookies, cereals, disposable diapers, life insurance, shoes, and telephone answering machines. There are four stages in the product life cycle: 1. 2. 3. 4. Introduction or Development. When the product is first brought to market and sales are low. Growth. When demand begins to accelerate causing a rapid expansion of sales. Maturity. When demand levels off and grows only at the replacement rate or at the rate of growth in population or new families. Decline. When product sales begin to decline. There are several indicators by which to judge the stage of life cycle for each product: 1. Comparative sales volume over time; comparison with previous time periods; the rate of sales change; whether sales are increasing or decreasing. Sales are low, and increase only slowly when a product is in the introduction stage. Sales increase at an increasing rate in the growth stage. Sales increase only slowly and level off during the mature stage. Sales decline, first slowly, then more rapidly during the decline stage. Level of sales compared to potential sales. The ratio is low in the introduction stage, and increases up to the maturity stage, becoming very high in the decline stage. Variety of products. Only one or a few varieties initially; increasing variety in the growth stage; decreased variety in the maturity stage; and further reduced variety in the decline stage. Comparative advantage of products; competitive substitutes; product and brand differentiation. Great comparative advantage, few competitive substitutes and 135 great differentiation in the introduction and growth stages. In the mature stage products become more alike; parity products; small comparative advantages; close competitive substitutes; differences created largely by advertising and brand differentiation. This is carried to the extreme in the decline stage. 5. Manufacturer price variability or stability. In the introduction and growth stages there are substantial price differences among sellers. There is a tendency toward price parity in the maturity stage. Very little price variability among sellers in the decline stage; prices are essentially the same, with competitive pressure tending to force all prices downward, generally stable prices unless there is a price war. 6. Retailer price variability or stability. In the introduction stage there are high retail prices with little variability. In the growth stage there is an increase in variability in prices at which retailers sell the product. In the mature stage there is strong pressure on retailers to compete on a price basis and therefore relatively less retail price variability/greater price stability. This is continued to the extreme in the decline stage. 7. Number and types of competitors. There are few competitors in the introduction stage, gradually and then rapidly increasing into many competitors through the growth stage, and then a reduced number of competitors in the maturity stage. There are relatively few competitors in the decline stage. 8. Number of dealers, retailers. There are few dealers/retailers in the introduction stage, and an increasing number during the growth stage, and a leveling off and then a reduced number during the maturity stage. There are few during the decline stage. It should be noted that the above characteristics are interrelated and interdependent. Moreover, they are relevant and can be judged only with respect to each particular product category, and not on an absolute basis. Therefore it must be remembered that the final decision rests upon the coder’ 3 judgement, based on an understanding of the several complex characteristics relevant to each particular product category. ' Prepared by Professor Gordon E. Miracle and research staff for the project. APPENDIX K PRODUCT CLASSES REPRESENTED IN COMMERCIALS 136 H g... P‘H F‘H UNOBOF‘WIAP‘OIAC)N