"Graduate and Postdoctoral Studies"@en . "DSpace"@en . "UBCV"@en . "Dewhirst, Timothy"@en . "2009-11-14T15:18:54Z"@en . "2003"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Using an interdisciplinary and multi-method approach, this thesis examines Canadian tobacco industry practices that were responses to federal advertising regulations, with a specific focus on the role that sponsorship has played. By analyzing internal industry documents publicly accessible as a result of Canadian court proceedings, it is apparent that there were two defining periods for when sports and cultural sponsorship became an increasingly important component of the promotional mix for Canadian tobacco manufacturers. First, tobacco sponsorship became more prominent during the late 1960s and early 1970s when cigarette advertising was voluntarily withdrawn from television and radio. Second, expenditures substantially increased during the late 1980s and early 1990s with the implementation of the Tobacco Products Control Act (TPCA), which severely limited conventional forms of advertising. Case studies are provided for the marketing strategies of Player's, Export 'A' and Rothmans, thus a dominant trademark from each of Canada's three principal tobacco manufacturers is represented. Each case study employs a different framework to understand the persuasive elements of cigarette promotion and the importance of message repetition, continuity, and consistency. The effectiveness of Player's promotions are made clear by integrated marketing communication efforts, while Export 'A' and Rothmans promotional activities are understood in terms of rhetoric and intertextuality, respectively. The review of industry documents discloses that the primary objectives for sponsoring events are to increase brand awareness (through continued brand exposure) and to enhance or reinforce brand image. Tobacco brands continue to gain widespread exposure on television through event sponsorship, and in effect circumvent supposed bans on broadcast advertising. In an attempt to enhance or reinforce brand imagery, tobacco firms identify sports and cultural events that possess complementary symbolic properties, seeking a transfer of the imagery associated with the event, participants or sponsorship partners to the sponsoring cigarette trademark. Content analysis is a second research methodology employed in order to assess whether the character of the content of Canadian tobacco promotions has changed as a result of the TPCA's implementation. This thesis examines the consistent or changing nature of advertising content as the tobacco industry's marketing strategy shifted from traditional product advertising to sponsorship. This is done by comparing the content of print media promotions from the pre-TPCA era (1973-1988) with those from the post- TPCA era (1989-2002). It was found that Player's and Export 'A' represent two Canadian trademarks that have been effectively positioned toward the \"starters\" segment. Player's is positioned as a symbol of masculinity, independence, freedom, self-reliance, tradition, and modernity. The trademark attributes and personality of Export 'A' are masculinity, ruggedness, independence, self-determinedness, adventurousness, and escapism. Rothmans has been promoted as an expression of internationalism, premium quality, upward status and tradition, yet youth widely perceive the trademark as unpopular, 'old', and lacking contemporaneous. It is demonstrated that the Player's trademark is particularly effective in its communication strategies, abiding by key promotional principles relating to message repetition, continuity, consistency, and relevance."@en . "https://circle.library.ubc.ca/rest/handle/2429/14984?expand=metadata"@en . "37828074 bytes"@en . "application/pdf"@en . "T O B A C C O M A R K E T I N G A N D P U B L I C P O L I C Y IN C A N A D A , T H E R O L E OF SPONSORSHIP by T I M O T H Y DEWHIRST B.P.H.E. , University of Toronto, 1993 M . A . , Queen's University, 1995 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E STUDIES (Interdisciplinary Studies) We acceptJbis-Hpesis a>co1Tl^rmjrijj*t6 the required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A July 2003 \u00C2\u00A9 Timothy Dewhirst, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of l^Wii'Al^lM\j HyJlO The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A B S T R A C T Using an interdisciplinary and multi-method approach, this thesis examines Canadian tobacco industry practices that were responses to federal advertising regulations, with a specific focus on the role that sponsorship has played. B y analyzing internal industry documents publicly accessible as a result of Canadian court proceedings, it is apparent that there were two defining periods for when sports and cultural sponsorship became an increasingly important component of the promotional mix for Canadian tobacco manufacturers. First, tobacco sponsorship became more prominent during the late 1960s and early 1970s when cigarette advertising was voluntarily withdrawn from television and radio. Second, expenditures substantially increased during the late 1980s and early 1990s with the implementation o f the Tobacco Products Control Act ( T P C A ) , which severely limited conventional forms of advertising. Case studies are provided for the marketing strategies o f Player's, Export ' A ' and Rothmans, thus a dominant trademark from each of Canada's three principal tobacco manufacturers is represented. Each case study employs a different framework to understand the persuasive elements of cigarette promotion and the importance o f message repetition, continuity, and consistency. The effectiveness of Player's promotions are made clear by integrated marketing communication efforts, while Export ' A ' and Rothmans promotional activities are understood in terms of rhetoric and intertextuality, respectively. The review of industry documents discloses that the primary objectives for sponsoring events are to increase brand awareness (through continued brand exposure) and to enhance or reinforce brand image. Tobacco brands continue to gain widespread exposure on television through event sponsorship, and in effect circumvent supposed i i i bans on broadcast advertising. In an attempt to enhance or reinforce brand imagery, tobacco firms identify sports and cultural events that possess complementary symbolic properties, seeking a transfer of the imagery associated with the event, participants or sponsorship partners to the sponsoring cigarette trademark. Content analysis is a second research methodology employed in order to assess whether the character of the content of Canadian tobacco promotions has changed as a result of the T P C A ' s implementation. This thesis examines the consistent or changing nature of advertising content as the tobacco industry's marketing strategy shifted from traditional product advertising to sponsorship. This is done by comparing the content of print media promotions from the pre-TPCA era (1973-1988) with those from the post-T P C A era (1989-2002). It was found that Player's and Export ' A ' represent two Canadian trademarks that have been effectively positioned toward the \"starters\" segment. Player's is positioned as a symbol o f masculinity, independence, freedom, self-reliance, tradition, and modernity. The trademark attributes and personality of Export ' A ' are masculinity, ruggedness, independence, self-determinedness, adventurousness, and escapism. Rothmans has been promoted as an expression of internationalism, premium quality, upward status and tradition, yet youth widely perceive the trademark as unpopular, ' o ld ' , and lacking contemporaneous. It is demonstrated that the Player's trademark is particularly effective in its communication strategies, abiding by key promotional principles relating to message repetition, continuity, consistency, and relevance. T A B L E OF C O N T E N T S Abstract i i Table of Contents ; iv List o f Tables v i i i Acknowledgements ix Chapter One Introduction 1 K e y Principles of Advertising and Promotion 4 Research Methodology 8 A Thesis Overview 10 P A R T I: R E V I E W O F T O B A C C O I N D U S T R Y D O C U M E N T S Chapter Two Tobacco Sponsorship and Regulatory Issues in Canada . . . .16 A Shift to Sponsorship Once Broadcast Advertising Was Withdrawn (1972) 17 The Tobacco Products Control Act Takes Effect (1989) 20 The Tobacco Act : Another Policy Facing a Constitutional Challenge 28 Conclusion 42 Chapter Three Player 's\u00E2\u0084\u00A2 Cigarette Brand Marketing: Sponsorship as Part of an Integrated Marketing Commumcations Plan 46 Introduction 46 The Target Market o f Player's: Segmentation as a Marketing Tool 47 Sport Sponsorship: A n Increasingly Important Communications Tool 60 Brand Image Enhancement: A K e y Sponsorship Objective of I T L 61 Discussion 67 Conclusion 75 Chapter Four Export ' A ' \u00E2\u0084\u00A2 Cigarette Brand Marketing: Male Youth, Extreme Sports, and the Gendering o f Smoking . . . . 79 Advertising and Rhetoric 80 Export ' A ' : Segmentation as a Marketing Tool 83 Export ' A ' Brand Marketing (1972-2002) 86 Sponsorship: Becoming an Important Part of the Tobacco Marketing Plan 89 Discussion 93 Conclusion 100 Chapter Five Rothmans\u00E2\u0084\u00A2 Cigarette Brand Marketing: Intertextuality and the Decline of a Flagship Trademark 103 Intertextuality and the Nearly Infinite Promotional Chain 104 A n Overview of Ro thmans\u00E2\u0084\u00A2 Promotional Strategies 107 Rothmans Sponsorship Properties 113 Static in the Communication Process 118 Discussion and Conclusions 128 P A R T II: C O N T E N T A N A L Y S I S OF T O B A C C O P R O M O T I O N S Chapter Six Content Analyses of Cigarette Advertising: A Critical Review of the Literature 134 Literature Review 137 Discussion 166 Chapter Seven Content Analysis Study Comparing Traditional Product Advertising and Sponsorship Promotions: Persistent Lifestyle Messages in the Face of Regulation . . . 174 Hypotheses 175 Methodology 181 Results 189 Discussion and Conclusions 197 Chapter Eight Conclusion 205 Future Directions for Research 212 References 217 Appendix 1 Player's Filter ad: Couple Fishing (1977) 248 Appendix 2 Player's ad: Four Skiers (1980) 249 Appendix 3 Player's Light ad: Male Hiker (1988) 250 Appendix 4 Team Players Member: Mountain Biker 251 Appendix 5 Team Players Member: Rock Climber 252 Appendix 6 Molson Take Care ad: Greg Moore 253 Appendix 7 Export ' A ' ad: Good Companions 254 Appendix 8 Export ' A ' ad: Export, eh? 255 Appendix 9 Export ' A ' ad: Feeling Satisfied! 256 Appendix 10 Export ' A ' ad: A Taste for Adventure 257 Appendix 11 Export ' A ' Sponsorship ad: Motocross 258 Appendix 12 Export ' A ' Sponsorship ad: Loud, Fast, Hard 259 Appendix 13 Rothmans Packaging 260 Appendix 14 Rothmans ad: Horse and Coach 261 Appendix 15 Rothmans ad: Tobacco Leaves as Backdrop 262 Appendix 16 Rothmans Special M i l d ad: Sailing 263 Appendix 17 Rothmans Extra Light ad: Eating Food on Yacht 264 Appendix 18 Rothmans Plus ad: 1:15 A M 265 Appendix 19 Rothmans ad: Porsche Race Car 266 Appendix 20 Rothmans/Castrol ad 267 Appendix 21 Rothmans Racing ad: Jacques Villeneuve 268 Appendix 22 Rothmans International F i l m ad 269 Appendix 23 Rothmans World F i l m ad: Special Preview 270 Appendix 24 Reviewed Content Analysis Studies 271 Appendix 25 Content Analysis Coding Categories and Definitions 272 Appendix 26 Data Set Coding Form 283 Appendix 27 Intercoder Reliability (Percentage Agreement) by Advertisement 284 Appendix 28 Intercoder Reliability (Percentage Agreement) B y Instrument Item 285 Appendix 29 SPSS Data Analyses Pertaining to A l l Data-coding Instrument Items 286 Appendix 30 Export ' A ' Sponsorship A d and Packaging 345 v i i i LIST OF T A B L E S Table 1 Longitudinal Data Set (1973-2002) Sample Sizes 183 Table 2 Confirmation of 'Var ia t ion Over Time\" Hypotheses 193 A C K N O W L E D G E M E N T S First and foremost, I would like to thank Rick Pollay for being the ultimate supervisor and mentor over the past five years. I am grateful for your warmth, generosity, and encyclopaedic knowledge of tobacco marketing. Thanks for the continuously inspiring conversations (pertaining to tobacco and everything else). I w i l l always be on the lookout for additions to the A C M E collection. I would also like to express special gratitude to the additional members o f my dissertation committee: Bob Sparks, Patricia Vertinsky, and J im Frankish. Your expertise, support, and attention to detail were much appreciated. I am grateful to those at the Sauder School of Business. A s a student enrolled in interdisciplinary studies, it was possible to experience feelings of dislocation. Yet, professors and students in the marketing division (such as Chuck Weinberg, Peter Darke, Darren Dahl, Laurence Ashworth, and Robin Ritchie) were very supportive and welcoming. Margaret MacNe i l l deserves thanks for first advising me to pursue this field of study, and encouraging me to seek out Rick Pollay as a supervisor. I have now conducted research on tobacco marketing and public policy issues for roughly ten years, and I am grateful for insight provided by Bruce K i d d and Hart Cantelon during the initial stages o f my research endeavours. During my first year at the University of British Columbia, Chris Lovato provided guidance about research grant writing, which is gratefully acknowledged. X I would like to express special thanks to those who remind me about having fun: N ick Rintoul, Kate Ha l l , Laurence Ashworth (second mention), Matt Farish, Aaron Hunter, Maged Senbel, Anna Stokke,' Paul Janzen, Clare Mochrie, David Capie, the guys at Osborne Gym, Anna Tremp, Jessica Vexler, T i m Woodman, Brian Cook, Emma Cornish, Dave Cherry, Steve McCarthy, Michael Blecher, and L l o y d Grinham. Green College and St. John's College were my places of residence while in'Vancouver, which provided great settings for inspiration. I am grateful to Madeleine Thien for helping me make the move. Thanks to Jameson for the calming influence. Thanks to Emi ly L a i for your support and affection during the early stages of the research process. Thanks to Wonkyong Lee for giving me excitement about what the future holds. Thanks to my M o m , Dad, and sister (Tobi)\u00E2\u0080\u0094each o f you have been incredibly supportive and patient. Finally, I am grateful for funding received from Health Canada/Canadian Institutes of Health Research. Financial support through a National Health Ph.D. Fellowship permitted my efforts to be focused on the task at hand. 1 C H A P T E R O N E Introduction The regulation of tobacco advertising involves debates between two powerful players in the marketplace, with the tobacco industry a representative o f the private sector and the federal government a representative o f the public sector. Regulation, according to C. L loyd Brown-John, is defined as \"any constraint imposed upon the normal freedom of individuals by the legitimate activity of government\" (1981, p.7). Since regulation involves government activities that limit the choices available to individuals within society, it is typically controversial (Meier, 1985). The regulation of tobacco advertising is no exception. It is a debate grounded in the structure of the modern welfare state. To what extent can and should the government intervene in the lives o f citizens? Moreover, which government activities should be regarded as legitimate? The federal government justifies its regulatory role in the creation of tobacco control policies since cigarette smoking represents the single most important cause of preventable illness and premature death in Canada. Smoking has been linked to a number o f health problems, including chronic bronchitis, emphysema, strokes, heart disease, and cancer of the lung, lip, oral cavity, pharynx, larynx, esophagus, pancreas, bladder and kidney (U.S. Department of Health and Human Services, 1992). It is estimated that 45,000 Canadians die prematurely each year as a result of smoking. Tobacco product use is attributable to a greater number of deaths among Canadians than 2 the total caused by car accidents, suicides, murders, A I D S , and i l l ici t drug use combined (Cunningham, 1996). In legal proceedings assessing the constitutionality of Canadian federal tobacco control policies, the Crown has successfully argued that the \"peace, order, and good government\" clause of the Constitutional Ac t of 1867 gives the federal government the right to intervene in a public health issue for the public good. 1 Youth are considered to be a group particularly worthy of protection from tobacco inducements since smokers usually begin during their pre-teen or teenage years, and are known to highly underestimate the addictiveness of cigarettes. Industry-commissioned research indicates that adolescents who start smoking do not disbelieve the overall health consequences of smoking, but they almost universally assume these risks are non-applicable since they do not anticipate becoming addicted (Kwechansky Marketing Research, 1982). The health effects of smoking go beyond looking at the smoker per se. Second-hand smoke, also known as environmental tobacco smoke, is the combination of smoke produced by the burning o f tobacco and the exhaled smoke from a smoker. Second-hand smoke is comprised of gases and particles that contain greater than 4,000 chemicals and at least 43 of these chemicals are cancer-causing agents (Cunningham, 1996). It has been estimated that within Canada more than 1,000 deaths per year among non-smokers can be attributed to second-hand smoke. Smoking can no longer be considered an individual right or personal choice under all circumstances. The health consequences of smoking act as an important factor in government deliberations about the role that health care costs can and w i l l play in the reduction of the federal deficit. A reduction in overall tobacco consumption levels is regarded as a 3 valuable objective toward health care reform efforts and offsetting ever-increasing health care costs. The social and economic costs of tobacco are noteworthy. Economic cost-benefit analysis studies reveal that while government tax revenues from tobacco sales are substantial, they are largely outweighed by the costs attributable to smoking. Collishaw and Myers (1984) found that smoking led to a C$1.12 bi l l ion increase in Canadian health care costs during 1979, representing 0.4% of Canada's G D P . During 1991, Kaiserman (1997) estimated that the total costs of tobacco use in Canada were roughly C$15 bi l l ion ($2.5 bil l ion were attributable to additional health care costs) while tax revenue amounted to $7.8 bil l ion. Single, Robson, X i e , and Rehm (1998), meanwhile, approximate the 1992 economic costs of tobacco in Canada to be C$9.56 bil l ion ($2.68 bi l l ion for direct health care costs). A rationale for government intervention has been outlined. The Tobacco Products Control Act, and the Non-Smokers' Rights Act that was enacted the same day (i.e., June 28, 1988), were the first Canadian federal laws to regulate the tobacco industry since 1908 (Brown, 1991; Cunningham, 1996). Considering the amount of research and knowledge surrounding the health consequences of smoking and the concern about the ever-increasing health care costs, it is remarkable that federal laws regulating the tobacco industry took nearly eighty years to modify. The Tobacco Products Control Act (TPCA) prohibited tobacco product advertising, but the sponsoring of cultural and sporting events by tobacco companies was still permitted, a convenient loophole in that existing legislation. The Tobacco Act, including its amendments (Bi l l C-42), replaced the T P C A and w i l l prohibit sponsorship promotion as of October 2003. Canada's three major tobacco manufacturers 4 have challenged the legislation on constitutional grounds, with trial court upholding the Act (Denis, 2002). Key Principles of Advertising and Promotion In addition to assessing the implementation of Canadian federal tobacco control policies, this thesis addresses the various promotional plans and strategies employed by Canada's principal cigarette firms. Promotion involves the seller communicating information to a potential buyer, with a goal of influencing their attitudes and behaviour. The primary purposes of promotion are to inform, persuade, and remind. Informing is considered particularly essential for newly developed or 'introduced' products, in which communications efforts are meant to tell potential customers something about the product. Promotions with an aim of persuading often focus on reasons why one brand is better than competing brands. The promoter seeks to develop a favourable set of attitudes among customers so that they w i l l buy and keep buying the product. Persuasion-based promotions commonly link products with desirable attributes, images, and 'personalities'. Finally, promotions with the goal of reminding are typically directed toward buyers that already have positive, well-established attitudes about a product, including its price, features or availability (Shapiro, Wong, Perreault, & McCarthy, 2002). Promotional planning starts with a clear target market in mind, and the audience may consist of potential buyers, current users, those who make the buying decision, or those who influence it. Few products are promoted in an undifferentiated manner, where the total potential market is treated as a whole. Rather, promotions tend to be directed toward well-defined consumer groups according to dimensions such as age, gender, ethnicity, income, 5 occupation, religion, family life-cycle, place of residence, lifestyles, interests, and values. The message is typically encoded such that it has reasonably broad appeal (i.e., referring to 'popular' culture), yet at the same time w i l l be most salient to a specific cluster or segment. The target audience w i l l heavily affect communication decisions regarding what wi l l be said, how it w i l l be said, when it wi l l be said, where it w i l l be said, and who w i l l say it (Shapiro, Wong, Perreault, & McCarthy, 2002). Promotional planning, then, involves establishing advertising objectives and determining the target audience. There are several ways of communicating to consumers including advertising, event sponsorship, packaging, coupons, personal selling, sampling, contests, publicity, product placement, and public relations. For conventional advertising, the copy platform entails the formation of creative promises (i.e., communicating what benefits the product w i l l provide, or alternatively, what problems the product w i l l solve). According to Kapferer (1997, p.30), one primary consumer benefit served by brands is \"to have confirmation of your self-image or the image that you present to others.\" Brands can help provide an identity for consumers, making them feel as though they belong to a special group (D'Alessandro, 2001). When selecting a particular brand of cigarettes, a consumer is engaging in an act of distinction (i.e., it says something about them, much like the clothes they are wearing, the music they listen to, or the car they drive). Communicating brand image is considered particularly crucial for product categories such as cigarettes and beer, which consist of several brands possessing minimal product differentiation, yet have a high degree of social visibility. Such characteristics are the basis for these goods being coined \"badge products.\" 6 Several principles are considered cornerstones for successfully communicating brand identity or image. Repetition, both over time and across multiple media, lends to 'friendly familiarity' being created. A dense environment of cigarette promotion and imagery gives the impression that tobacco use is socially acceptable, desirable, and prevalent (Pollay, 2002). The persistence and pervasiveness of tobacco promotion is notable. In the United States, for example, more than US$10 bi l l ion is spent annually on marketing cigarettes (Mackay & Eriksen, 2002). Having a highly repetitious message obviously requires a considerable advertising budget. It is also considered important among marketers to have promotional messages that are continuous and consistent (Aaker, 1996; Ries & Ries, 1998; D'Alessandro, 2001). According to Wells, Burnett, and Moriarty: Because the effects of images advertising build up over time, consistency is critical to the process. Y o u can't say one thing today and something different tomorrow... every ad contributes to the image. The message must focus on what the image is supposed to be, and should be consistent over a long time. (1989, p.207) Marlboro serves as a good example of a brand/trademark that has been successfully linked to consistent imagery over a long period of time. Wernick (1991) explains that: .. .the meaning of any single message is modified by, and depends on, the ones that came before. The same is true for sub-campaigns, where even the launching of a new product may build on meanings previously achieved. During the 1980s, for example, ads for Marlboro Lights projected a soft focus version of the leathered cowboy which had already become ultra-familiar in previous advertising for its parent brand, (p.92) 7 Yet, while the image(s) communicated may remain the same, different symbols may be . employed to remain relevant, contemporary, and appealing to an ever-changing audience (D'Alessandro, 2001). The brand's identity or image is collectively constructed through the use of brand names, logos, taglines, typography, pictorials, and primary and secondary colours (Perry, 2003). In pursuit of effective communication, advertisers attempt to create a message that w i l l be simple, familiar, easily recognized, and comprehensible. Acknowledging that many ads get limited and indirect attention from the viewer amidst all o f the 'clutter', messages are often designed to draw attention or stand out, yet do not require large amounts of time and effort to understand. This is facilitated through visual imagery predominating in many ads, with its function illustrated by the aphorisms, \" A picture is worth a thousand words\" and \"Seeing is believing.\" Market research is typically conducted that both informs (i.e., pre-testing) and validates (i.e., post-testing) promotional planning efforts. Brand imagery is further reinforced or enhanced in ad visuals through the use o f lifestyle portrayals, which do not necessarily require depictions of people. Co-branding, event sponsorship, and endorsements exemplify three ways of enriching the symbolic value of brands or trademarks. Distinct trademark meanings (and implied product users) w i l l be communicated i f one ad features tickets for an opera performance on the dashboard of a Mercedes, while another ad depicts tickets for a stock car race on the dashboard of a Chevy pick-up truck. Product endorsement from a recent gold medallist at the Olympics would potentially associate a trademark with qualities o f nationalism, leadership, and high performance. The personality of the particular athlete might also be transferred to the 8 endorsed brand. Clearly, associating a brand/trademark with other objects, settings or people that are rich in meaning can effectively convey lifestyle imagery. Research Methodology Using an interdisciplinary and multi-method approach, this thesis examines Canadian tobacco industry practices that were responses to federal advertising regulations, with a specific focus on the role that sponsorship has played. Case studies are provided for the marketing strategies of Player's, Export ' A ' and Rothmans, thus a dominant trademark from each of Canada's three principal tobacco manufacturers is represented. These three trademarks are considered \"major\" brands, having had a substantial presence on the market for an extended number of years, with similar target markets (Player's and Export ' A ' are particularly comparable with male youth identified as the segment o f primary interest). Each case study employs a different framework to understand the persuasive elements of cigarette promotion and the importance of message repetition, continuity, and consistency. The effectiveness of Player's promotions are made clear by integrated marketing communication efforts, while Export ' A ' and Rothmans promotional activities are understood in terms of rhetoric and intertextuality, respectively. Studying the marketing strategies of tobacco firms is unique, considering that numerous internal corporate documents, previously confidential, are now available to the public as a result of litigation. For Part I of this thesis, internal tobacco industry documents were reviewed, accessible primarily as a result of two sets o f court proceedings in Canada: (1) the 1989 Canadian trial to decide the constitutionality of the 9 T P C A , and (2) the 2002 Quebec Superior Court trial in which Canada's three major tobacco manufacturers challenged the constitutionality of the Tobacco Act. Physicians for a Smoke-Free Canada and the Canadian Council for Tobacco Control (available at www.smoke-free.ca and www.cctc.ca, respectively) house many of the industry documents from these two trials. 4 A n analysis o f internal corporate documents is a well-suited research methodology for exploring how tobacco manufacturers have intentionally coordinated their communication efforts toward the target customer. The corporate documents provide valuable insight for encoding the production of promotional messages. The early 1960s were selected as the starting point of my historical analysis, reflecting that the Canadian Tobacco Manufacturers' Council ( C T M C ) was established in 1963 and adopted its first voluntary code on marketing practices in 1964. The C T M C was comprised of Canada's major tobacco manufacturers, with the purpose of establishing cigarette advertising guidelines and representing its members on matters of common interest pertaining to health, taxation, defining product standards, exchanging statistical information, support for technical research, liaison with other sectors of the tobacco industry, and international industry relations ( C T M C , 1994). Part II of the thesis utilizes content analysis as a research methodology, where it is assessed whether the content of Canadian tobacco promotions changed as a result of the T P C A ' s implementation. This thesis examines in what respects advertising content remained consistent or changed as the tobacco industry's marketing strategy shifted from traditional product advertising to sponsorship promotions (by comparing the content of promotions from the p re -TPCA era with those from the pos t -TPCA era). 10 A Thesis Overview Part I of the thesis consists of Chapters 2 through 5. In Chapter 2, internal corporate documents are reviewed to provide a historical overview of how sponsorship became an increasingly important component of the promotional mix for Canadian tobacco firms. It is revealed that sport and arts sponsorship became more prominently utilized by the industry during the early 1970s to compensate for cigarette advertising that had been voluntarily withdrawn from television and radio. Expenditures on sponsorship increased considerably during the late 1980s, which was an industry response to the implementation of the T P C A . Sponsoring sports and cultural events (and related promotions) provided tobacco companies with opportunities to circumvent existing advertising regulations. Moreover, the documentary evidence demonstrates that youth have been an important target market for the promotional campaigns of several cigarette brands, and that tobacco sponsorship promotions are best classified as 'lifestyle' advertising. A primary objective of sponsorship is to enhance or reinforce brand imagery, thus tobacco firms select to sponsor sports and cultural events possessing symbolic imagery or 'personalities' that are transferable to their respective brands. Chapter 3 examines how Imperial Tobacco Ltd. , Canada's largest tobacco manufacturer, effectively utilizes integrated marketing communications to construct brand imagery for Player's, a leading cigarette trademark in Canada. Internal industry documents show that Player's has been positioned to appeal to the attitudes and desires of male youth. Promotional taglines such as \" A Taste Y o u Can Ca l l Your Own,\" \" A Tradition of Excellence,\" and \"It's Your Wor ld ,\" complemented with visual images o f 11 outdoor settings and athletic activities such as windsurfing, mountain climbing and kayaking, consistently linked the trademark with images o f youthfulness, masculinity, freedom, independence, quality, and tradition. Auto racing is the major sponsorship property of Player's, with a core objective of enhancing the trademark's essence through images being transferred from the sport, participating athletes, and various co-sponsors. Chapter 4 assesses how conventional notions of masculinity have been exploited in promotional efforts to construct brand imagery for Export ' A ' , a Canadian cigarette trademark manufactured by JTI-Macdonald Corp. B y reviewing trade press and tobacco industry documents, it is indicated that Export ' A ' has been positioned to appeal to the attitudes and desires of male youth through three decades. Ut i l iz ing promotional slogans such as \" A Taste for Adventure\" and \"Go Your Own Way,\" the cigarette trademark has been linked with images of ruggedness, adventure, escapism, independence, and rebelliousness. In 1997, Export ' A ' began sponsoring an extreme sports, and it is demonstrated that this sponsorship property is an effective means o f communicating virtually the same symbolic meanings as previous traditional product advertisements no longer permitted according to legislation. Since persuasion is a basic objective o f promotion, Export ' A ' marketing activities have been analyzed in terms o f rhetoric. In Chapter 5, industry documents are reviewed to reveal the promotional strategies utilized for the Rothmans trademark. The concept of intertextuality, as proposed by Julia Kristeva (1984), is used to demonstrate how the meaning and identity of Rothmans has been inconsistently communicated to consumers. Rothmans, Benson and Hedges Inc., Canada's second largest tobacco manufacturer, has positioned the Rothmans trademark as an expression of internationalism, eminence, tradition, and 12 premium quality. These 'intentional' constructions o f Rothmans narratives do not correspond with other sources of consumer information, however. Wi th a dearth of promotional spending directed toward the trademark, Rothmans lacked visibili ty and was perceived by consumers as unpopular. Moreover, few updates were made to the package design, which led consumers to identify the Rothmans trademark as ' o ld ' and outdated. Ultimately, the profile of the typical Rothmans smoker became a retirement-aged person, thus those at the stages o f smoking onset did not perceive the trademark to be an appealing alternative. Rothmans, once a prominent trademark, has seen its market share decline substantially during the past few decades. Part II of the thesis consists of Chapter 6 and 7. In Chapter 6, content analysis is defined, and it is made clear that 'objectivity', 'systematization', and 'quantification' are important characteristics. A thorough and international review of English-language content analyses (that are specific to cigarette advertising) published in academic journals is provided. The studies reviewed for this chapter include content analyses of cigarette magazine ads and cigarette billboard ads. A general overview o f the reviewed studies is presented, with emphasis placed on research methodology and key findings. Common study limitations are identified and several directions for future research are offered. This literature review is meant to inform current and future content analysts about important methodological decisions and enable research on cigarette advertising to be further improved in the areas of objectivity, systematization, and quantification. Chapter 7 consists of a content analysis study that assesses whether the content or character o f Canadian tobacco promotions was shaped by the implementation of the T P C A . It is determined which advertising content dimensions remained consistent, or 13 alternatively which ones changed, as the Canadian tobacco industry's marketing strategy shifted from traditional product advertising to sponsorship. The data set consists of English-language print ads for Player's, Export ' A ' , and Rothmans that circulated from 1973 through 2002. Thus, the content analysis study findings could be substantiated with the disclosures of the internal corporate documents. It was found that 'lifestyle' advertising persisted despite the enactment of the T P C A , as sponsorship ads remained predominantly image-based and emphasized physical activity in outdoor settings. Little information was found to be present in both traditional product advertising and sponsorship promotions. Chapter 8 concludes the thesis with conclusions and recommendations regarding the regulation of tobacco advertising and sponsorship. It is argued that Canadian cigarette trademarks are successfully marketed to youth, including consumers who are classified as 'starters' or 'new smokers'. A primary objective o f tobacco promotion, as well as sponsorship, is the reinforcement or enhancement of brand image or 'personality'. Tobacco promotions are commonly dominated by visual imagery, with colour playing an important role in distinguishing trademarks and communicating both imagery and product characteristics. Message repetition is another important characteristic of promotion, whereby persistent and pervasive communication suggests social acceptability, popularity, and builds 'friendly familiarity'. Tobacco firms also recognize the importance of linking their cigarette trademarks with consistent images over a sustained period of time. Considering that 'lifestyle' advertising and sponsorship promotion w i l l be prohibited in October 2003, the future directions o f the tobacco industry are assessed. Suggestions for future study are also provided. 14 Endnotes 1 Stanbury (1979) effectively categorized 25 different definitions of the \"public interest.\" The first category o f definitions considered the public interest to be \"resulting from the aggregation, weighing and balancing of a number of special interests\" (p.213). Other definitions considered it to be \" i n terms of the common or universal interests which all (or at least almost all) members of society/nation/political unit share\" (p.214). 2 Common costs attributed to smoking include lost productivity among employees who are smokers (due to higher levels of absenteeism), the value o f life lost as a result of premature death (both in the workforce and household sectors), litter, fires, and health care expenditures. 3 The decision to use three different frameworks for understanding the promotional efforts of Player's, Export ' A ' , and Rothmans reflects that these case studies were adapted from three separately prepared manuscripts. A n adapted version o f the Player's chapter is currently under review at the Journal of Sport Management (i.e., a special issue on \"Issues in Sport Media\" called for manuscripts that included themes of integrated marketing communications). A modified version of the Export ' A ' case study is \" i n press\" as a chapter for the book entitled, Sexual Sports Rhetoric Globally. Sections o f the Rothmans case study were adapted from an \" in press\" manuscript (co-authored by Robert Sparks) that w i l l be found in an upcoming issue of the Journal o f Sport and Social Issues. 4 In 1998, as a result o f the trial involving the Minnesota State Attorney General (and insurers) as plaintiffs and the U . S . tobacco industry as defendants, the Guildford 15 Depository and the Minnesota Tobacco Document Depository were established. Academics, lawyers, as well as representatives from various government agencies and interest groups have conducted searches at both depositories, making many o f the collected documents accessible at Internet websites (e.g., www.librarv.ucsf.edu/tobacco, http://roswen.tobaccodocuments.orR, www.tobaccodocuments.org). Part I: Review of Industry Documents 16 C H A P T E R T W O Tobacco Sponsorship and Regulatory Issues in Canada Sponsorship of sports and cultural events has been a part of the promotional mix for Canadian tobacco manufacturers throughout the twentieth century. In 1903, for example, the Red Cross tobacco brand was a prominent sponsor of a high wire act, in which a person crossed the Montmorency waterfalls in the province of Quebec (Cunningham, 1996). RJR-Macdonald began sponsoring the Brier, a men's curling championship, sanctioned by the Canadian Curling Association, in 1929 (W. Proctor, personal cornmunication, October 18, 1994). A 1932 promotional campaign for Turret, a tobacco brand oriented toward the blue-collar class according to Imperial Tobacco documentation, offered cash prizes to those correctly estimating the number of goals that would be scored by National Hockey League ( N H L ) teams. Sweet Caporal, another brand offered by Imperial Tobacco, sponsored the first Canadian football radio broadcasts during the 1930s (Cunningham, 1996). Despite these examples from the earlier part of the twentieth century, there have been two later defining periods for when sports and cultural sponsorship became an increasingly important component of the promotional mix for Canadian tobacco manufacturers. First, tobacco sponsorship became more prominent during the late 1960s and early 1970s when cigarette advertising was voluntarily withdrawn from television and radio. Second, expenditures substantially increased during the late 1980s and early 1990s with the implementation of the T P C A , which severely limited conventional forms 17 of advertising. In the current political-legal context, key issues being debated are whether youth remain an important target of tobacco promotional activities and whether tobacco sponsorship promotions are most aptly classified as lifestyle or informational advertising. This chapter focuses on these issues by reviewing academic literature, trade press, annual reports and policies, as well as by examining internal industry documents publicly accessible as a result of Canadian court proceedings. A Shift to Sponsorship Once Broadcast Advertising Was Withdrawn (1972) It was not until the late 1960s and early 1970s that the sponsorship of sports and cultural events became an increasingly significant part of the promotional mix for Canadian tobacco manufacturers. The shift toward sponsorship reflected the reality that many other important means of promotion were no longer viable options. Starting in 1964, Canadian tobacco companies voluntarily agreed to confine cigarette advertising on television to hours after 9 p.m. ( C T M C , 1964). In 1970, incentive programs were a marketing practice discontinued in response to government pressure. Incentive brands -those offering redeemable coupons, prizes or gifts - had become an important marketing strategy considering that they accounted for greater than 50 percent of the total Canadian cigarette market in 1969 (Imasco Ltd. , 1970; Imasco Ltd., 1971). Effective January 1, 1972, cigarette advertising was voluntarily withdrawn from the broadcast media altogether ( C T M C , 1971). This industry initiative could be seen as a public relations move aimed at warding off impending government regulation, considering that the federal Health Minister at the time, John Munro, had introduced B i l l C-248 - an Act that 18 included stipulations to ban cigarette advertising - on June 10, 1971 (Imasco Ltd., 1972; Cunningham, 1996). Tobacco companies turned toward sponsoring broadcast sports and cultural events as a means to compensate for lost broadcast advertising exposure. Imperial Tobacco, Canada's largest tobacco manufacturer, recognized in its marketing plans that \"as a result of current and possible future T V restrictions, the opposition [competitors] has become increasingly active in the sponsorship of major events across Canada. Therefore we can as a company foresee the possibility of sponsoring most of these events with one of our own brands\" (1970, p.566628113). In a review of its competitors' promotional activities as of July 1970, Imperial Tobacco claimed, \"we can certainly see a big swing towards sponsorship of national or regional events\" (1970, p.566628132). Within this corporate document, Rothmans was used as an example of a competitor with prominence since it was the sponsor of special events caravans, Canadian Open Tennis Tournaments, and the Canadian Equestrian Team. Roughly four years later in an annual report, the parent company of Imperial Tobacco stated that, \"marketing activities, particularly for cigarettes, have changed significantly since the industry's withdrawal from broadcast advertising. More advertising is now associated with spectator events, although total advertising expenditures are down\" (Imasco Ltd., 1974, p.8). There were several examples of new or expanded sponsorship properties during the early 1970s. Anticipating the withdrawal from broadcast advertising, Imperial Tobacco signed a five-year contract with the Royal Canadian Golf Association during November 1970 in order to link its Peter Jackson cigarette brand with the Canadian Open Golf Championships, and the agreement also included the sponsorship of provincial tournaments. 19 Another Imperial Tobacco cigarette brand, Player's, began sponsoring auto racing in Canada in 1961, yet announced that this support would be extended toward regional events in 1971 (Imasco Ltd. , 1971). Also in 1971, the tobacco manufacturer formed the du Maurier Council for the Performing Arts to provide grants toward music, ballet, and theatre-related performing groups on behalf of their du Maurier brand (Imasco Ltd., 1972). Similar trends were being observed elsewhere. In fact, it has been noted that the tobacco industry's involvement in sports and cultural sponsorships during the 1970s and 1980s was a significant factor in the general development of sponsorship as a marketing discipline (Otker & Hayes, 1987; Meenaghan, 1991; Cornwell, 1995). Cigarette advertising was banned from the broadcast media in the United Kingdom in 1965, for example, and the resulting shift in tobacco industry promotional spending largely contributed toward the overall growth of sponsorship expenditures that was observed during the 1970s (Taylor, 1984; Marshall & Cook, 1992). In the United States, federal legislation - the Public Health Cigarette Smoking Act - stipulated that broadcast advertising for cigarettes was banned, commencing January 2, 1971. While the primary response of American tobacco manufacturers was to forward their advertising spending toward the print media, there was also a notable shift toward sponsorship (Feinberg, 1971; Teel, Teel, & Bearden, 1979; Warner, 1985; Stoner, 1992). Notably, Virginia Slims began sponsoring women's professional tennis in 1970, while Winston Cup auto racing and Marlboro Cup horse racing debuted in 1971 and 1973, respectively (Wichmann & Martin, 1991; Crompton, 1993; O'Keefe & Pollay, 1996). 20 The Tobacco Products Control Act Takes Effect (1989) The late 1980s and early 1990s marked a second defining period in which tobacco sponsorship of Canadian sports and cultural events grew substantially. The dramatic increase in sponsorship expenditures was primarily driven by the T P C A , which was enacted in 1988 and took effect on January 1,1989. U p until this time, the tobacco industry self-regulated by establishing a voluntary advertising code in 1964, with amendments being made to the code in 1971,1976, and 1984. Although the T P C A stipulated that tobacco product advertising was banned, sponsorship remained permissible i f the full name of the manufacturers (rather than the brand names) were placed on promotional materials. In response to the imposed advertising regulations, the Canadian tobacco industry invested heavily in the sponsorship of various events, and advertising and promotional support remained a significant component of this investment. Documentation from the C T M C (1987,1997) indicates that sponsorship contributions grew more than six-fold from 1987 to 1995. Sponsorship became an attractive promotional tool once conventional advertising was no longer permitted. \"Shell\" Companies Were Formed for Sponsorship Purposes The T P C A stipulated that tobacco products offered for sale in Canada were not to be advertised. The Act did go on to note, however, that this statement did not apply to foreign media or to the sponsorship o f sports and cultural events. Sponsorship was still allowable i f the name of a company was used on promotional materials. Thus, as the T P C A took effect, Canadian tobacco firms hastily registered their various brands or trademarks as separate corporate entities for sponsorship purposes. RJR-Macdonald formed Vantage Arts Ltd. and 21 Export ' A ' Inc. as subsidiary companies. Vantage and Export ' A ' are cigarette brands offered by RJR-Macdonald.1 4 Rothmans, Benson & Hedges formed Craven \" A \" Ltd., Belvedere Ltd., Rothmans Ltd., and Benson & Hedges Inc. as subsidiary companies for sponsorship purposes. Imperial Tobacco created four new companies for the purpose of advertising and promoting the sponsorship of sports and cultural events. The new corporate entities were Player's Ltd., du Maurier Ltd., Matinee Ltd., and du Maurier Council for the Arts Ltd. (Descoteaux, 1989). Player's, du Maurier, and Matinee are cigarette brands manufactured by Imperial Tobacco. In its 1988 annual report, Imperial Tobacco acknowledged that the TPCA was being challenged on constitutional grounds, but while awaiting the outcome they were \"prepared to make the best of the few communication opportunities that are still available. We have, for example, incorporated a number of associated entities that will serve as vehicles for the sponsorship of a variety of sports and cultural events\" (Imasco Ltd., 1989, p.8). During the same year, an internal document for Imperial Tobacco stated that \"in a very real sense the company's expectations are that [the TPCA] C-51 will not end tobacco marketing, but will bring about very major changes in how that exercise is conducted... Imperial's extensive sponsorship portfolio is being restaged under new corporate names that will allow them to continue exploiting the huge equity in their investments in this area\" (cited in Pollay, 2002, p . l l ) . In a press release, Imperial Tobacco's Chairman and CEO, Jean-Louis Mercier claimed that the entities were formed to eliminate an unfair disadvantage imposed upon the company by the TPCA. It was argued that RJR-Macdonald and Rothmans, Benson & Hedges had already formed corporate names that were directly related to one or more of its 22 major trademarks, thus Imperial Tobacco publicly justified the newly formed \"shell\" corporations as a means to remain competitive with the other two major tobacco manufacturers (Descoteaux, 1989). Despite their public position, internal documentation revealed that Imperial Tobacco was in fact the forerunner in this strategic move: \" I T L led the way with brand corporations under which to conduct sponsorship activities\" (Imperial Tobacco, 1994, p.38). This statement reaffirmed British American Tobacco's (1991) claim that RJR-Macdonald followed Imperial Tobacco's lead with respect to forming \"shell\" companies. It should be evident that sponsorship companies, such as Player's Ltd. and du Maurier Ltd., were easily identifiable as cigarette brand names. The omission of Ltd. or Inc. was often apparent in the news coverage of the events that were sponsored by tobacco companies. For example, newspaper articles (\"Canadians suffer,\" 1994; Davidson, 1994) routinely referred to the Matinee Ltd. International women's tennis tournament as the Matinee International. Such an omission might not seem significant, but by not including Ltd., this tournament appeared to be sponsored by a tobacco brand name product. This was not permitted according to the T P C A . Tobacco companies could obviously avoid responsibility for such omissions by stating that they were not responsible for journalists who were ignorant o f the regulations. The corporate entities that were formed by tobacco manufacturers for sponsorship purposes had logos, trademarks, typography, and colouring that closely resembled those on the brand name packages. Thus, the colours and designs used on particular tobacco brand packages became the key mechanisms of promotion under the restrictions of the T P C A . 23 Unlike conventional tobacco product advertising, health warnings were not necessary for ads promoting the sponsorship arrangements of tobacco \"shell\" companies. Brand Exposure Was Maintained on Television and Radio Sponsoring sports and cultural events provided tobacco companies with several opportunities to circumvent the existing advertising regulations. Sponsorships allowed tobacco companies to gain access to various media that, according to the TPCA, would otherwise not have been permitted. Sponsorship programs were considered critically important since they were a means of providing \"broadcast brand i.d.\" (RJR-Macdonald, 1996a, p.80150 3496). Export 'A' Inc., for example, sponsored The Skins Game that included premiere male professional golfers (such as Jack Nicklaus, Fred Couples, Greg Norman, Ernie Els, Nick Faldo, John Daly, Mike Weir, and David Duval) and was televised nationally by the Canadian Broadcasting Corporation (CBC). According to RJR-Macdonald (1996b), \"The Skins Game has become one of the top televised golf events in Canada over its three year existence... The event gives Export 'A' Inc. an unparalleled opportunity to 'own' five hours of weekend television and represents a leveragable [sic] advertising opportunity in the two to three month window leading up to the event\" (p.80151 3320). Imperial Tobacco (1993a) observed that, \"a solution to the quality of viewership [sic] is available through the hour or more broadcast of our major events. With appropriate on-site signage, they become one hour commercials\" (p.014723). Player's, a leading Canadian cigarette trademark, has been a long-time and prominent Championship Auto Racing Teams (CART) sponsor for Canadian Champ car racers such as Jacques Villeneuve, Greg Moore, Patrick Carpentier, Alex Tagliani, and 24 Paul Tracy. Villeneuve won the Indianapolis 500 and the 1995 C A R T championship while sponsored by the corporate entity, Player's Ltd. In recognition of Villeneuve's achievements in 1995, he was awarded the Lou Marsh Trophy for being voted Canadian athlete of the year. Exposure for Player's Ltd. was obviously not limited to television coverage of the auto racing events since photographs of the wirining driver often appear in magazines, newspapers, and television newscasts. Competing tobacco manufacturer, Rothmans, Benson & Hedges, observed that, \"over the past 23 weeks and with three quarters of the Player's season completed, Imperial [Tobacco] has amassed an average of 4 broadcast hours a week that does not include Player's advertising and news coverage\" (cited inPollay,2002,p.47). Imperial Tobacco monitored the minimum exposure that would be attained for Player's during CART races according to the placing of its drivers and whether they controlled the television coverage: Controlling Television: There is no magic to controlling television, the concept is quite simple and based on being in control of what the audience sees. As an outline let me review what a team sponsor might expect of a car running in 6 t h place. IndyCar will make all efforts to cover each team twice, once when they announce the race and once when the race is on. Clearly if you are not part of the lead group your coverage is limited. However, with control of television one can guarantee oneself the following: Promotional Bumpers by Network prior to race 4 minutes Show Opening Segment 1 minute Commercial Bumpers 1 minute 25 Leader Boards 1 minute In show segments 2 minutes Pit Action with dedicated cameraman 2 minutes Post Race Interviews with driver 2 minutes 13 minutes This then wi l l quadruple your coverage on the air. (cited in Pollay, 2002, p.47,48) Clearly, brand exposure is maximized in auto racing sponsorship i f the sponsored driver is a front-runner or successful. A n d as a primary sponsor of the Molson Indy C A R T races that are held annually in Toronto and Vancouver, Player's is able to exercise 'control' o f the television coverage for events located in Canada. Canadian tobacco companies continued to place their promotional efforts on radio as well . In 1993, Craven \" A \" Ltd. began sponsoring country music on 59 radio stations as a means of circumventing the ban on broadcast cigarette advertising. According to Rothmans, Benson & Hedges, \"Craven ' A ' Ltd. Today's Country was launched June 5 t h , 1993. Radio broadcast grid has 59 affiliates for a 90% placement rate, ahead of our 60% estimate for this point in time\" (cited in Pollay, 2002, p.46). Within a few years, Craven \" A \" Ltd.-sponsored country music concerts were being taped for rebroadcast on 74 radio stations across Canada (Rothmans Inc., 1995). Meanwhile, Belvedere Ltd. and Export ' A ' Inc. were sponsoring various rock concerts. Rothmans, Benson & Hedges proceeded with a music-based promotion of Belvedere by developing \"a media program for Quebec and the Maritimes to deliver on-going image and awareness to 18-24 target consumers through the use of mediums such as Musique Plus and Radiomutuel\" (cited in Pollay, 2002, p.40). RJR-Macdonald, through its Export ' A ' \"Smooth\" line extension, sought to \"reinforce brand 26 imagery by utilizing New Music Series... Innovative tour sponsorship which has supported 175 bands performing over 1000 shows across Canada utilizing the Smooth Plugged New Music tour bus... Media Support: Radio, Urban newspapers, Posters, Retail Pamphlets, Internet, Street Banners, Club advertising\" (cited in Pollay, 2002, p.40, 41). There are several reasons why it is desirable for tobacco manufacturers to have the broadcast media (most notably, television) persist as a part of the communications mix. Rothmans, Benson & Hedges explains the relevance of television coverage for its Belvedere Rock program: \"Television was chosen to form part of the total communication mix for this program. Its role was to provide the perception of mass, total reach and credibility\" (cited in Pollay, 2002, p.47). This is consistent with the commonly cited advantages of selecting television as a media choice for promotion. Television allows for extensive market coverage considering that 98% of American households have at least one television and an average viewing time of eight-and-a-half hours per day (Arens, 1999). The reach of television is also impressive, which is illustrated nicely by the fact that the attendance for the 2002 Formula One race held in Montreal was 117,000, yet the number of television viewers for each Formula One race is estimated to be 300 million (\"Race report,\" 2002; Hawaleshka, 2001). Television is perceived to be the most authoritative and influential medium, thus it offers advertisers a prestigious image. Although the production and airtime costs of television advertising are potentially quite high, promoting on television communicates to viewers that the brand is a 'major player'. Sight, sound, and movement may also be combined on television, which allows product use or brand imagery to be demonstrated with impact (Arens, 1999). 27 Additional Loopholes Exploited Within Existing Advertising Regulations Canadian tobacco manufacturers exploited several other loopholes among the stipulations of the T P C A . The T P C A did not permit tobacco companies to promote tobacco products on billboards, but sponsoring sports and cultural events allowed billboards to be constructed that supposedly advertised the tobacco company's association with the event. In some cases, however, sponsorship billboards continued nmning long after the events being promoted were finished. Imperial Tobacco was scrutinized in a newspaper article by Mellor (1992) for a du Maurier Ltd. billboard that promoted an equestrian event taking place in October yet remained standing the following January. The T P C A also did not allow the purchasers of tobacco products to be offered the right to participate in a related contest, lottery or game. B y sponsoring sports and cultural events, however, tobacco companies were able to stage contests. Export ' A ' Inc., through its sponsorship of a salmon fishing showdown in 1992, held a contest offering C$50,000 cash as 1 s t prize. Benson & Hedges Inc. serves as another example since its Symphony of Fire contest (reflecting the \"shell\" company's sponsorship of fireworks) gave contestants an opportunity to win a one-week cruise vacation. The T P C A prohibited the use of tobacco brand names on non-tobacco items, yet logos of the \"shell\" companies were placed on items such as T-shirts, hats, and towels. From 1994 to 1996, the Player's Ltd. Racing Team issued catalogues through direct mailings, magazines, bars, and various race events that enabled consumers to purchase branded items such as rugby and polo shirts, jackets, key chains, knapsacks, and cargo bags. The Canadian tobacco industry's voluntary advertising codes ( C T M C , 1964, 1971, 1976, 1984) stipulated that all models used in cigarette ads were to be at least 25 years old, yet 28 tobacco sponsorship promotions often depicted athletes, celebrities, or other event participants that were younger. The Tobacco Act: Another Policy Facing a Constitutional Challenge The Tobacco Act (Bi l l C-71) was implemented in 1997, and established as a replacement of the T P C A and the Tobacco Sales to Young Persons Act. The T P C A needed to be replaced because in September 1995, in a five-to-four decision, the Supreme Court of Canada ruled that it was unconstitutional. Imperial Tobacco and RJR-Macdonald had legally challenged the T P C A , claiming that it was an infringement of commercial expression as stated by Section 2(b) of the Canadian Charter of Rights and Freedoms. Following the Supreme Court ruling, the Canadian tobacco industry adopted a new voluntary advertising code and resumed conventional advertising in February 1996. Thus, conventional cigarette advertising continued for roughly one year before Canadian tobacco companies had to adhere to the stipulations of the Tobacco Act . The Tobacco Act places a ban on lifestyle advertising, while informational advertising remains permissible assuming that it is placed in adult establishments, in publications with a minimum adult readership of 85% or in mailings addressed to adults by name. This legislation has been adopted to protect a specific marketing segment (i.e., youth) that is deemed to be particularly vulnerable to manipulative advertising. With respect to tobacco sponsorship activities, amendments were made to the Tobacco Act (i.e., B i l l C-42) in December 1998, which stipulated a five-year transition period before a total tobacco sponsorship ban is imposed. Imperial Tobacco, JTI-Macdonald, and Rothmans, Benson & Hedges have challenged the constitutionality of the Tobacco Act, but a Quebec Superior 29 Court decision in December 2002 upheld the legislation (Denis, 2002). Canada's three largest tobacco manufacturers have appealed the Quebec court ruling, and it is expected that the Quebec Court of Appeal and the Supreme Court of Canada wi l l eventually hear the case. According to the tobacco industry, the Tobacco Act places unreasonable limits on commercial expression. Under the broader title of \"Fundamental Freedoms,\" Section 2(b) of the Canadian Charter of Rights and Freedoms guarantees everyone the \"freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication\" (Laskin et al., 1994, p.CA-1). Section 1 of the Charter, however, does allow reasonable limits or restrictions on expression assuming that the policy objectives are pressing and substantive. Wi th tobacco use representing the single most important cause of preventable illness and premature death in Canada, smoking is clearly a national health problem and a pressing and substantive concern. Establishing what are reasonable limitations on commercial expression proves more contentious. K e y issues before the court include (1) justification that youth currently remain a target of tobacco promotional activities and thus represent a specific segment worthy of protection; and (2) distinguishing between lifestyle advertising and informational advertising, as well as establishing which classification best describes sponsorship promotions (Sparks, 1997). Youth: A Key Target of Tobacco Promotional Activities The promotion of tobacco products has long been a hotly contested issue, stimulating questions about whether youth are targeted by specific marketing campaigns. The Tobacco Act states that one of its purposes is \"to protect young persons and others from inducements to use tobacco products and the consequent dependence on them.\" Thus, in a 30 legal context whereby it is being justified as appropriate and necessary to severely restrict tobacco promotions, attention is often drawn toward establishing whether many promotional activities are in fact directed toward youth. The rationale for why tobacco companies would direct their promotions toward youth is that the pivotal period for smoking initiation in Canada is 13 to 14 years of age, with very few smokers beginning beyond adolescence (Health Canada, 1996). According to industry research, \"Recall o f cigarette adoption among respondents suggests that peer pressure and image are/were the key motivational factors... Most indicated that they had their first cigarette between the ages of 10-15\" (Rothmans, Benson & Hedges, 1991, p.27365). Smokers are also known to be extremely brand loyal, so the brand choice of consumers during the early stages of their smoking 'careers' becomes crucial. Market research prepared for RJR-Macdonald recognized that \"smokers exhibit extremely high levels of brand loyalty,\" and \"loyalty to cigarette brands remains very strong... Only 3% of all smokers are considered 'convertable' [sic] \" (Harrod & Mi r l i n , 1995, p.80154 2410). In the United States, brand switching among smokers is less than 10% annually, with less than 8% switching companies (Cummings, Hyland, Lewit, & Shopland, 1997). Comparable estimates for Canada could likely be lower considering that merely three manufacturers account for more than a 99% share of the domestic cigarette market, and Imperial Tobacco alone, commands a 70% market share. Tobacco industry representatives have publicly denied that they market their products to youth, yet internal corporate documents indicate otherwise. Pollay and Lavack (1993), Cunningham (1996), Pollay (2000), and Dewhirst and Sparks (in press) reviewed Canadian tobacco industry documents that were manifest in proceedings 31 assessing the constitutionality of the T P C A and found that youth are a target of tobacco marketing activities. Internal documents from the British tobacco industry and its leading advertising agencies also reveal that youth are a key group for marketing purposes (Hastings & MacFadyen, 2000). Glantz, Slade, Bero, Hanauer, and Barnes (1996), Perry (1999), and Cummings, Morley, Horan, Steger, and Leavell (2002) have examined U . S . tobacco industry documents and reached similar conclusions. To cite some specific examples, two Imperial Tobacco planning documents from the early 1980s, Fiscal '80 Media Plans and Fiscal '81 National Media Plans, included the age segment 12-17 years old among the identified target groups for several of the company's brands and trademarks. Another Imperial Tobacco (1987) document, Overall Market Conditions-F88, states that: If the last ten years have taught us anything, it is that the industry is dominated by the companies who respond most effectively to the needs of younger smokers. Our efforts on these brands w i l l remain on maintaining their relevance to smokers in these younger groups in spite of the share performance they may develop among older smokers, (p.6). A few years later, another internal document for the company revealed that, \"I .T.L. has always focused its efforts on new smokers believing that early perceptions tend to stay with them throughout their lives. I.T.L. clearly dominates the young adult market today and stands to prosper as these smokers age and as it maintains its highly favourable youthful preference\" (cited in Pollay, 2002, p. 19). RJR-Macdonald, the third largest tobacco manufacturer in Canada, recognized that \"new smokers are critical to continued growth in the market\" (1989a, p.80108 9826) and \" in 32 order to make further inroads into the younger segment, we must continue to project an image that is consistent with the needs and values of today's younger smokers\" (1989b, p.80118 3934). The company claimed that, \"the younger segment represents the most critical source of business to maintain volume and grow share in a declining market. They're recent smokers and show a greater propensity to switch than the older segment. Export [the best-selling cigarette brand manufactured by RJR-Macdonald] has shown an ability to attract this younger group since 1987 to present\" (RJR-Macdonald, 1989, p. 80118 3930). Another internal document entitled, Export \"A \" Brand Long-term Strategy, included \"new users\" under the sub-title, \"Whose behaviour are we trying to affect?\" (RJR-Macdonald, 1987, p.800230290). Acting as an indicator that the marketing practices of Canadian tobacco manufacturers have not changed during more recent times, a Rothmans, Benson & Hedges document states that, \"a strong regular length business is key to attracting younger users and ensuring a healthy future franchise\" (cited in Pollay, 2002, p. 19). Another mid 1990s document for the firm recognized the need to \"identify products and activities which wi l l strengthen R B H ' s position among the key 19-24 age group to gain a much larger share of starters,\" and \"although the key 15-19 age group is a must for R B H there are other bigger volume groups that we cannot ignore\" (Rothmans, Benson & Hedges, 1996, p.002756, 002757). According to Imperial Tobacco documentation, \"marketing activities have historically been and continue to be targeted at younger smokers due to their greater propensity to change brands\" (1995, p.018110). When reviewing the corporate documents, it is important to recognize that references to 'younger smokers', 'younger users', 'starters', 33 'potential starters' or 'new smokers' indicates that adolescents are likely the age segment being discussed. Despite an obvious interest in recruiting new smokers, the tobacco industry maintains that their promotions do not influence overall consumption levels, but rather affect the market share of each brand. The basis for sponsorship expenditures, it is argued, is to defend existing share and to increase it at the expense of the competition. The C T M C (1997) states that, \"a 1% market share of the cigarette market is worth approximately $22 million in annual net sales revenue to a manufacturer, not including any federal or provincial tobacco or sales taxes\" (p.8). A British American Tobacco (1984) document, however, contradicts the argument that merely market share is.influenced by stating the company's objective is to expand industry volume by maximizing instances of starting through relevant products and attitude change. Attitude change can be facilitated through promotional activities, and academic research has shown that youth disproportionately smoke heavily advertised brands (Pollay et al., 1996). Even i f one was to accept the argument used by the tobacco industry that its promotions are only aimed at existing smokers - encouraging them to either remain loyal to the brand they smoke or to switch brands i f they are smoking a competitor's brand - it seems dubious that the promotions would reassure smokers about choosing a particular brand, yet not reassure them about continuing to smoke more generally. Overall consumption levels would be affected i f promotions encourage smokers to continue smoking rather than quit. The aforementioned documents reflect the marketing practices of the respective tobacco companies generally. It seems reasonable that numerous sponsorship campaigns would be directed toward similar targets as part of an integrated marketing plan. Inter-office 34 correspondence at RJR-Macdonald (1996b) indicates that, \"generally younger smokers tend to be more aware of sponsorship events than older smokers\" (p.80151 3337). Tobacco- t sponsored events that include young spectators, volunteers, and participants have been the source o f criticism. Dewhirst (1999) demonstrated that a mascot representative of the Craven A cigarette brand was interacting with young children at the Craven A-sponsored Just For Laughs comedy festival in Montreal. When Imperial Tobacco sponsored the Canadian Open tennis championships in Toronto and Montreal, teenage volunteers (including the ball-boys and ball-girls) wore uniforms bearing the trademarks, logos, and colouring o f cigarette brands (Physicians for a Smoke-Free Canada, 1996). Tobacco control groups and many health practitioners have also expressed concern about the age of the participants for events that are promoted by tobacco companies. In women's professional tennis, for example, many of the world's top players are teenagers. Jennifer Capriati was the champion of the 1991 Matinee Ltd. International tennis championships when she was 15 years old, and as a result was featured in several ensuing promotional materials. Interestingly, when Martina Hingis won the 1999 du Maur i er Open in Toronto at 18 years of age, she was still not of legal smoking age in the province o f Ontario. RJR-Macdonald (1996a, p.80150 3496) has described sponsorship as being equivalent to advertising, thus it is perhaps not surprising that studies indicate youth recognize advertising of tobacco-company sponsored events as advertising for tobacco products (Rootman & Flay, 1995; Health Canada, 1996). A study by Charlton, While, and Kel ly (1997) has found that English boys, aged 12-14, who enjoy watching Formula One auto racing, are nearly twice as likely to smoke compared to those who do not follow Formula One. Sparks (1999), after assessing the relative contribution of sponsorship to 35 brand awareness among 14 year old boys and girls in New Zealand and reviewing previous research on tobacco sponsorship and youth, concludes that \"selectively targeted cigarette sponsorships can help to build positive brand associations and awareness in the youth (starter) market and thereby contribute to the customer-based equity of the sponsoring brand\" (p.256). Tobacco Sponsorship Promotions: A Form of 'Lifestyle' Advertising A second key issue during recent court proceedings has been to make a distinction between lifestyle advertising and informational advertising. Making such a distinction was suggested by the decision of the Supreme Court of Canada during the T P C A trial. The majority judgement identified that by not distinguishing between \"brand preference\" and \"lifestyle\" advertising, it was not clear whether the objectives of the T P C A could have been met with less intrusive measures (Wyckham, 1997; Manfredi, 2002). The Tobacco Act defines lifestyle advertising as \"advertising that associates a product with, or evokes a positive or negative emotion about or image o f a way of life such as one that includes glamour, recreation, excitement, vitality, risk or daring.\" Informational advertising, meanwhile, is described as a promotion that provides factual information to a consumer about the product's characteristics, price or availability. Looking to leading marketing and advertising textbooks, lifestyle is considered to establish the ways in which one's time and money is spent and reflect which activities are most valued. Lifestyle is defined as a person's pattern of living that becomes manifest in their activities, interests, and opinions (Wells, Burnett, & Moriarty, 1989; Kotler, Armstrong, & Cunningham, 1999; Lamb, Hair, & McDaniel , 2000). Lifestyle advertising, 36 then, involves the association of products and brands with particular activities, interests and opinions, appealing to a specified segment of consumers. The activities, interests, and opinions that are depicted in an ad may reflect the actual or desirable lifestyles of either current or prospective consumers. According to Tuckwell (1988), in an attempt to match the lifestyle of the product user, one may appeal to \"their looking-glass self.\" Lifestyle advertising can be accomplished through the portrayals of people, settings, and objects (or combinations thereof). Some leading textbooks on marketing and advertising do not make extensive use of the phrase \"lifestyle advertising,\" however, preferring to use terminology such as \"image advertising\" and \"transformation advertising.\" Transformation advertising has an objective of building a product/brand personality or image and making the experience of consumption seem richer, warmer, and more enjoyable (Wells, Burnett, & Moriarty, 1989). Consumers often use the same terminology to describe brands and people, such that particular brands are perceived as expressing excitement, success, sophistication, ruggedness, and so on (Aaker, 1996). Brand imagery or personality has traditionally been seen by the tobacco industry as very important to communicate. According to an Imperial Tobacco document, 1971 Matinee Marketing Plans, \"without price differentials and without easily perceptible product differentiation (except for extremes, e.g. Matinee versus Player's) consumer choice is influenced almost entirely by imagery factors\" (1970, p.566628090). Roughly 25 years later, Rothmans, Benson & Hedges claims that, \" in the cigarette category brand image is everything. The brand of cigarettes a person smokes is their identity. Cigarettes tell others who they are as a person. There is a strong emotional connection to the brand, the image it projects about the smoker, not only to themselves but to others\" (cited in Pollay, 37 2002, p. 13). Another internal document indicates that the taste qualities o f cigarettes are developed only after an appropriate brand personality has been selected: \"Must think imagery/brand personality first and then develop the products with taste qualities/product and package attributes that reinforce image\" (cited in Pollay, 2002, p. 14). The role o f lifestyle, meanwhile, is to \"promote and reinforce the social acceptability among the peer group to smoking as a relaxing, enjoyable self-indulgence\" (Imperial Tobacco, 1979b, p.13). Contemporary Canadian industry documents indicate that the function o f many tobacco sponsorship promotions is consistent with the objectives of lifestyle advertising. Canadian firms recognize that lifestyle creative and imagery is conveyed by sponsorship communications. According to Imperial Tobacco, \"opportunities to utilize image advertising in Sponsorship communication should be exploited\" (1992, p.013870). Rothmans, Benson & Hedges (1993) considers sponsorship to be \"one of the few image-enhancing marketing tools available,\" and looks to \"use sponsorships as a means to establish and build upon lifestyle image associations through targeted selection, strong promotional programmes and professional execution, all of which reflect the desired character and image\" (p.005381). RJR-Macdonald recognizes that, \"our sponsorship approach must be consistent with our brand position to enhance image reinforcement\" (1996b, p.80151 3317). With conventional cigarette advertising severely restricted in Canada, tobacco manufacturers have directed their promotional dollars toward sponsorship, and attempted to have the content of the sponsorship promotions resemble their previous conventional ads as much as possible. In 1992, Imperial Tobacco acknowledged that, \"we have already begun the transition from event advertising to more image based advertising. We still need to fully exploit the communications value inherent in our sponsorship involvement. Unt i l further regulatory change, this is the means by which we w i l l replace traditional brand/trademark image advertising\" (cited in Pollay, 2002, p . l 1). To exploit a trademark's link with a particular image, several tobacco sponsorship promotions communicate that the cigarette trademark is a general supporter of an activity (i.e., Export ' A ' sponsors an extreme sports series, Player's sponsors auto racing, Matinee sponsors fashion, and du Maurier claims to sponsor music, photography, and nightlife) without specifying any details about the particular events being sponsored. Tobacco companies have found that a challenge with event sponsorship advertising is the duration that the accompanying promotional campaign can effectively run. In other words, i f a one-day event is being sponsored, it proves difficult to justify promoting the event throughout the year. RJR-Macdonald specified that sponsorship vehicles should be selected that \"spread throughout the year to provide continuity\" and \"support the brand sell message that is the same in non-event periods\" (cited in Pollay, 2002, p.9). Similarly, another document from the company included \"duration\" and \"timing seasonality\" as important criteria forjudging sponsorship opportunities (cited in Pollay, 2002, p.9). According to Imperial Tobacco, \" in terms of understanding, it is very clear that while the event itself is a communications vehicle, the true value is the amount o f targeted imagery communications which surround the event. It gives us the legitimate excuse to promote. In analyzing event operation costs, the goal w i l l be to identify expenditures which wi l l not effect our image, and re-channel to communications\" (1992, p.013835). The objectives and budgets sections of tobacco industry documents, pertaining to sport and cultural sponsorship, are dominated by the importance o f enhancing or reinforcing brand imagery. Reflecting on the implementation of the T P C A , Rothmans, Benson & 39 Hedges claimed that, \"today (1988-) sponsorship is the only means whereby company trademarks can be exposed to the public. The image of the activity and the broadcast exposure received in large part determine trademark awareness. I T C O event inventories are being streamlined and investment is being made in broadcast programming and broad scale image communication\" (1995, p.008593). One Imperial Tobacco (1993b) document bluntly states that the primary objective of sponsorship advertising is to communicate image, while selling tickets to the sponsored event is only a secondary objective: \"Specific Objective: To communicate relevant sponsorship imagery to its target group - national versus local. To maintain year-round presence of this imagery on a national basis... A secondary objective is to promote ticket sales for the events\" (p.014435). Canadian tobacco sponsorship promotions, in many cases, link a cigarette brand with a particular image at the expense of providing important information about the actual event being sponsored (i.e., neglecting to indicate which athletes or teams are participating, the cost of attending the event, where tickets may be purchased, or where the event is being held). Tobacco companies select to sponsor sports and cultural events possessing symbolic imagery or 'personalities' that are desirable to link with their respective brands. The objective is to have the image of a sports or cultural event transferred to the sponsoring brand: \"Borrowed Imagery: Association with sporting events creates a situation where, because of the perceived 'personality' of the sport, sponsoring corporations can 'borrow' imagery from that personality in order to strengthen their own public perception\" (cited in Pollay, 2002, p. 13). According to Imperial Tobacco: With regard to the brand or corporate image, the sponsor gives the impression of seeking to associate itself with the image of the event or of those who participate in 40 the event. When a company sponsors a tennis or golf tournament, a regatta or the classical arts, this is interpreted by the public as a kind of expression (by the sponsor) of the temperament o f the company. Depending on the event sponsored, the company appears young, self-assured, master of itself, classical, adventurous, etc. (cited in Pollay, 2002, p. 13) Don Brown, Chairman, President and C E O of Imperial Tobacco, claims that, \"sponsorship is still limited in the degree to which it delivers a specific product attribute message. The value lies in matching imagery of the event to that of the product or service\" (cited in Gross, 1994, p.67). To illustrate the matching of a cigarette brand and a sponsored sports event along imagery dimensions, du Maurier - the best selling cigarette brand family in Canada and described by an industry insider as \"a high quality, upscale, young brand in Canada with a solid image\" (Bingham, 1992, p.500028180) - has sponsored prestigious tennis, equestrian, and golf events. D u Maurier was the title sponsor for the professional men and women's Canadian Open tennis championships that alternated annually each summer between Toronto and Montreal (both o f these tournaments were categorized as top-tier tournaments, and only the four Grand Slam tournaments were considered to be of greater importance), as well as the Ladies Professional G o l f Association ( L P G A ) tournament held in Canada. The du Maurier Ltd. Classic represented one of the four major championships on the L P G A tour. The apparent objective with these sponsorship properties was for the upscale, aspiring, high quality, and classy dimensions associated with the events (and sports) to be transferred toward the du Maurier brand. 41 For the title sponsor, the process of image matching and transfer is also applicable to the event's participants (i.e., celebrities) and the co-sponsors. Popular auto racer Jacques Villeneuve, for example, has been characterized as a 'wi ld child ' , rebel, and daredevil (and meanwhile engages in a very high risk sport) which makes him a desirable person to link with cigarette brands that are marketed with such imagery (Dewhirst & Sparks, in press). When du Maurier sponsored the Canadian Open tennis championships, B M W was a co-sponsor, which exemplifies co-branding and image matching opportunities being exploited with sponsorship partners. D u Maurier and B M W complement one another with respect to how the brands are positioned in their respective product category. Such brand matching is consistent with McCracken's (1988) concept o f Diderot unities, which emphasizes that the meaning of goods is largely determined by their relationship to other goods. Player's, du Maurier, Matinee, Benson & Hedges, and Export ' A ' currently represent the Canadian cigarette brands most prominently depicted in sponsorship promotions. Player's continues to sponsor C A R T auto racing, while du Maurier supported 271 art groups during 2002. B y promoting grants that are provided to Canadian fashion designers, Matinee is linked with images of relaxation, youthfulness, self-expression, and indulgence (Imperial Tobacco, 1993a). Benson & Hedges sponsors the Gold Club Series, which features leading DJs performing in club settings. Export ' A ' , which is positioned according to dimensions of adventure, masculinity, and independence, sponsors an extreme sports series (Pollay, 2001; Dewhirst, in press). It has been observed that the extreme sports series consists of activities involving competitors who succeed because of 42 their willingness to take extreme risks, and the promotions for these events appeal to the viewer's desire for independence because the selected activities are all individual sports. It should be apparent that Canadian tobacco companies sponsor a diversity of events, yet the majority of sponsorship expenditures are toward sports events (the budgets for arts and cultural sponsorship are considerably less). This weighting reflects sponsorship spending generally, as it is estimated that sports events account for at least two-thirds of sponsorship expenditures (Linstead & Turner, 1986; Shanklin & Kuzma, 1992; Copeland, Frisby, & McCarvi l le , 1996). During the mid-1990s, the annual contributions by Player's toward auto racing accounted for roughly one-sixth o f the total sponsorship expenditures by all Canadian tobacco companies (Gross, 1994; C T M C , 1997). Conclusion It has been demonstrated here that tobacco sponsorship largely evolved once other elements of the promotional or communications mix were no longer permissible. Put simply, sponsorship became one of the best 'available' promotional options for Canadian tobacco companies. The industry quickly found that despite cigarette advertising being withdrawn from the broadcast media in 1972, cigarette brand exposure could persist on television and radio i f broadcast sporting and cultural events were sponsored. The Canadian trend toward sponsorship was consistent with the U.S . experience. Despite cigarette advertising becoming prohibited on U . S . television in 1971, B l u m (1991) and Siegel (2001) have illustrated that by sponsoring sports such as auto racing, U . S . tobacco 43 companies continue to receive millions of dollars worth of low-cost national television exposure. In 1988, the T P C A was legislated, which was significant due to the restrictions it upheld. Under the stipulations of this Act, tobacco product advertising was not permitted. Advertising that promoted the sponsorship arrangements of tobacco companies could not reveal tobacco products. The capacity of the T P C A to ban tobacco advertising in Canada was limited, however. This relative powerlessness was largely due to the Act 's numerous loopholes, the most notable of which allowed the formation of corporate entities by tobacco companies. Used for sponsorship purposes, these corporate entities employed logos and trademarks that closely resembled those found on the tobacco products (i.e., packages) of their parent companies. Obviously, such a strategy allowed the ostensibly prohibited dissemination of these logos and trademarks in the mass media. After reviewing internal tobacco industry documents from all of the major Canadian firms, it is revealed that the primary objectives for sponsoring sports and cultural events are to increase brand awareness (through continued brand exposure) and to enhance or reinforce brand image. Tobacco brands continue to gain widespread exposure on television through the sponsorship of sports and cultural events, and in effect circumvent supposed bans on broadcast advertising. In an attempt to enhance or reinforce brand imagery, tobacco companies identify sports and cultural events possessing complementary symbolic properties, with a common goal of having the image that is linked to the event transferable to the sponsoring brand. The Tobacco Act has replaced the T P C A , but like its predecessor, it faces a constitutional challenge. The Supreme Court of Canada is expected to eventually hear the 44 case. The Tobacco Act appears to place reasonable limits on expression considering that its objectives are pressing and substantive, while internal corporate documents reveal that youth remain a key target of tobacco promotional activities and many tobacco sponsorship promotions are a form of lifestyle advertising. If the Tobacco Act and its amendments (i.e., B i l l C-42) are upheld, tobacco sponsorship wi l l become banned in Canada, effective October 2003. 45 Endnotes 1 RJR-Macdonald Inc. was re-named JTI-Macdonald Corp. after Japan Tobacco Inc. purchased R . J . Reynolds International in 1999. 46 C H A P T E R T H R E E Player's\u00E2\u0084\u00A2 Cigarette Brand Marketing: Sponsorship as Part of an Integrated Marketing Communications Plan Introduction Imperial Tobacco Ltd. (ITL) is Canada's largest tobacco manufacturer, accounting for more than two out o f every three cigarettes sold domestically. The major trademarks owned by I T L are Player's and du Maurier, which collectively hold an impressive 60% share of the Canadian tailor-made cigarette market. Player's has a long and well-established history, predating the incorporation of I T L in 1912. The logo o f the Player's trademark dates back to the late 1800s, described in marketing trade press as \"a ship's life belt framing an antiquated naval scene depicting a 19 t h century sailor with a George V beard and the word 'Hero' on his hat\" (cited in The Manitoba Educational Research Council , 1966, p . l 10). During both the First and Second Wor ld War, a naval theme was apparent in much o f Player's advertising, in an effort to communicate patriotism, dependability, and strength of reputation. The naval motif is also evident in Player's packaging, where blue and white colouring is utilized in the form of marine flags. Today, the Hero logo remains in use, and the Player's brand family consists of Plain, Filter, Medium, Light, Light Smooth, Extra Light, Special Blend, John Player Special, and Silver. The market share held by ITL has impressively risen over the past 25 years, reflecting a strong performance by the Player's trademark. The company's market share 47 of the Canadian cigarette market was 37% in 1975, yet by 1988 it was 56%, and by 1998 surpassed 68% (ITL, 1989; Imasco Ltd., 1989, 1999). It is demonstrated here that I T L has linked its Player's trademark with images of youthfulness, masculinity, independence, freedom, tradition, and modernity on a consistent basis, effectively utilizing integrated marketing communications, which involves \"the intentional coordination of every communication from a firm to a target customer to convey a consistent and complete message\" (Shapiro, Wong, Perreault, & McCarthy, 2002, p.433). The emerging popularity of Player's is largely explained by ITL ' s well-integrated marketing communication efforts (relative to competitors' trademarks), and the firm's ability to appeal to the all-important youth market. The target customer of Player's has consistently been identified as males less than 25 years of age, with conventional ads commonly portraying sports scenes in an effort to communicate brand imagery and to appeal to the core consumer. Following the implementation of the T P C A , sport sponsorship became an increasingly important part o f the communications mix. The Target Market of Player's: Segmentation as a Marketing Tool Shapiro, Wong, Perreault, and McCarthy's (2002) definition of integrated marketing communications indicates that a target customer needs to be established for a product or service. Segmentation is a commonly used marketing strategy for determining who w i l l be targeted, in which specific audiences are identified for a product by dividing a mass market into subsets on the basis of variables such as demographics, geography, psychographics, and product benefits. Thus, many advertising campaigns (including those for Player's) are not meant to influence behaviour on a one-to-one basis nor have 48 mass appeal, but rather be directed toward a subgroup of consumers that are homogeneous according to factors such as age, gender, ethnicity, education, socio-economic status, marital status, lifestyle, and/or geographic location. The objective is to meet the needs typified by a specific group of consumers in an efficient manner, whereby the product's characteristics and promoted attributes can clearly match what is desired from the user(s) (Rothschild, 1987; Lamb, Hair, & McDanie l , 2000). B y reviewing internal documents of I T L from the mid 1970s through 1990s, it became apparent that the primary target audience for Player's promotions has been male youth who aspire to be masculine, independent, self-reliant, and modern. Demographic Dimensions Age and gender are the primary segmenting dimensions utilized in the marketing of Player's. The Player's trademark has been strategically positioned to appeal to the attitudes and desires of male youth, and in particular, those that are still at a fairly early stage o f their smoking 'careers.' During the late 1970s, the overall target market for the Player's brand family was identified as those less than 24 years old (Spitzer, M i l l s & Bates, 1977). Similarly, in 1980, the media target group was defined as \"young people under 35 years of age with major emphasis on under 24 year old males\" (ITL, 1980a, pp.684451897-912). While a primary objective of the advertising creative was to appeal to male youth, I T L did not want to exclude women. It was outlined that \"a l l Player's advertising w i l l focus their appeal on young males* 24 years old and under. *Activities and scenarios should be seen as inspired by the male but appropriate for feminine participation\" (Ibid., asterisk in original).' 49 To appeal to those less than 24 years old, I T L sought models for the ad creative that appeared to be among the same peer group. Under industry self-regulation, however, all models used in Canadian cigarette ads were to be at least 25 years old ( C T M C , 1976). Thus, internal documentation outlining creative guidelines for Player's indicated that an activity should be featured \"which is practiced by young people 16 to 20 years old or one that these people can reasonably aspire to in the near future\" (ITL, 1980b, p.000001), while casting requirements were \"models in Player's advertising must be 25 years or older, but should appear to be between 18 and 25 years o f age\" (Chacra, 1980, p. 17). During the early 1990s, the target market of Player's was male smokers aged 18 to 25 ( B A T , 1989). This segment seemingly reflected the prototype Player's consumer, as Player's was the most popular trademark among Canadian smokers aged 18 to 24 (The Creative Research Group, 1992). In 1988, 39% of smokers in this age group were consuming Player's, with Player's Light clearly the leading choice of the brand family. Beyond Player's, du Maurier and Export ' A ' (manufactured by competitor, R J R -Macdonald) were the only additional trademarks considerably consumed by this age segment, accounting for a 29% and 20% market share, respectively (The Creative Research Group, 1988). From 1990 through 1994, Continuous Market Assessment ( C M A ) data clearly revealed that Player's, du Maurier, and Export ' A ' were still the trademarks with the largest market share o f the key target group of smokers less than 25 years o f age. It was noted, however, that Player's consumers were now most commonly represented by the age segment 25-34, reflecting that its 'user share' was aging (ITL, 1994a). I T L did not 50 want to convey an image that its typical customer was getting older, considering it desirable to position Player's such that the \"trademark is to be perceived as the most youthful and popular and as having the most tradition\" (1995, p . l 1). The target market for Player's has remained nearly identical over a 25-year period. The rationale for why ITL would continuously select 'youth' as the desired target market is that the pivotal period for smoking initiation in Canada is 13 to 14 years of age, with very few smokers beginning beyond adolescence (Health Canada, 1996). I T L recognizes that, \"Generally, smokers start smoking in their teens and usage grows as a smoker ages\" (1993, p.017969). Industry insiders consider the initial brand choice to be crucial since smokers are known to be extremely brand loyal. One internal document of a competitor reveals, \"I .T.L. has always focused its efforts on new smokers believing that early perceptions tend to stay with them throughout their lives. I.T.L. clearly dominates the young adult market today and stands to prosper as these smokers age and as it maintains its highly favourable youthful preference\" [emphasis added] (cited in Pollay, 2002, p. 19). Geographic Dimensions Considering that Canada is a bilingual country, Player's is marketed to both French and English audiences. When developing ad copy, I T L does not merely translate from one language to the other. For the launch of ITL ' s Player's Go ld Leaf brand in 1963, for example, the communication process was described as, \"though title, package, and symbol are the same in both languages from there on in Imperial - and its major agency, M c K i m , Montreal - go to great pains to ensure the approach in French is purposeful and to the point without having to adhere to any English principles\" (cited in 51 The Manitoba Educational Research Council , 1966, p.82). Jacques Bouchard, head o f Montreal-based B C P Advertising Ltd. , explains that, \"French translations of English slogans not only don't make much sense most of the time; they don't se l l . . . Quebec is a different place, with different values, habits, instincts and tastes\" (cited in Fraser, 1977, p.69). For promotions that were to run in Quebec, a predominantly French-speaking province, \"Models used in our French advertising are selected to ' look' Quebecois. . Typically Anglo/American featured (blond, blue-eyes) models are avoided\" (ITL, 1980a, pp.684451897-912). Media plans outlining the target groups for various Player's brands commonly divide English- and French-speaking people into separate categories. I T L (1995, p.50) has also identified that, \"Quebec and the Atlantic continue to be full-flavoured markets; British Columbia and Ontario tend to be milder markets,\" indicating which line extensions are most favourably received in various regions of Canada. Although geographic segmentation has played a role in the advertising strategy o f Player's, I T L recognizes the importance o f communicating a consistent brand image: \"When image advertising is used in response to specific regional strategies, creative w i l l continue to reflect a lifestyle realization of youthful self-expression, independence and freedom, with subject matter that is particularly relevant to young males\" (ITL, 1980c, p.4). Psychographics (Lifestyle Dimensions) Psychographics, also referred to as lifestyle analysis, is another commonly used segmentation approach in which the personality, activities, interests, and opinions of the target market are considered. I T L advised that, \" A l l Player's advertising w i l l use 52 lifestyle imagery to relate product to its core target group by association\" (1980c, p.2). The Player's Family: A Working Paper document, among its media objectives, proposed developing a media program that specifically spoke to those with a youthful, masculine lifestyle. According to Spitzer, M i l l s & Bates, \"Relevant lifestyle is the key to the brand's positioning, and the youthful emphasis is a psychological not a chronological one\" (1977, p.12). Positioning is defined as the place a product, brand, or group of products occupies in consumers' minds (with respect to brand identity and value) relative to competing offerings (Aaker, 1996; Shapiro, Wong, Perreault, & McCarthy, 2002). According to a marketing firm representing ITL, \"Freedom and independence are at the core o f Player's positioning\" (Marketing Strategy & Planning, 1985, p.102692367). The first advertising objective for Player's Filter was \"to communicate that the brand is for those who make their own choice about what they do, for people who want to assert their own individuality, and who are seeking a more independent lifestyle\" (Spitzer, M i l l s & Bates, 1977, p . l 1). Marketing plans for the Player's trademark reveal that, \"Since 1971, ITL ' s marketing strategy has been to position Player's as a 'masculine trademark for young males.' It has been our belief that lifestyle imagery conveying a feeling of independence/freedom should be used to trigger the desire for individuality usually felt by maturing young males\" (ITL, 1983). Reflecting this positioning, the commonly used tagline for Player's ads during the late 1970s and through most of the 1980s was A Taste You Can Call Your Own. B y the mid-1990s, the brand positioning and communications strategy of Player's was defined as \"Male Mainstream achiever through success based on independence/heroism (knight of 90's)\" (Bateman et al., 1994, p.32). 53 Consumer research for I T L has revealed that self-reliance is a logical extension of the freedom and independence dimensions. \"Self-reliance seems to offer good opportunities for extending the positioning of Player's\" (Marketing Strategy & Planning, 1985, p.102692366). The Player's 1988 document states that, \"In order to move Player's Light up on the masculinity dimension, we w i l l continue throughout F '89 to feature creative which reflects freedom, independence and self-reliance in a relevant fashion for young males\" (ITL, 1987, p.4). Modernity, considered an expression of a brand's vitality, is another lifestyle dimension purposely communicated in Player's promotions. While modernity has multiple meanings, it is defined by I T L for marketing purposes as \"successfully adapting to the times\" (Saine Marketing, 1990, p. 18). I T L stipulates that, \" A l l Player's brands are to be seen as up-to-date\" (1995, p. 11). Reaffirming the Social Acceptability of Smokins Finally, an important objective of Player's promotions has been to reaffirm the social acceptability of smoking. The Player's Trademark F\"81 Advertising document indicates that lifestyle imagery is to be maintained for each of the Player's brands, with ads continuing to: . . . reflect the brand's popularity among young people, to demonstrate the social acceptability of these brands among the target consumer's peer group, and to place the products in scenarios and settings which invite the target consumer to easily associate a Player's brand with a pleasant lifestyle to which he w i l l identify (Chacra, 1980, p. 1). 54 Such promotional content counters the defensiveness that is often demonstrated by smokers (i.e., some feel the need to rationalize why they are doing a behaviour with severe health consequences). Some smokers perceive themselves as 'social pariahs,' considering that it is increasingly difficult to find a public space where it is legal to smoke. American tobacco firm, R.J . Reynolds (1985, p.80) claims that, \"While smokers may identify with each other on a personal level, they do not on a group level. A major factor inhibiting the development of group cohesion among smokers is a sense of shame. One way o f mitigating this sense of shame is to build on the positive functions of smoking.\" Many tobacco promotions, including those for Player's, serve the goal o f normalizing smoking (while many tobacco control efforts have 'denormalization' as an objective). Sport Portrayals in Advertising Creative To augment the desired lifestyle dimensions, sport portrayals have been commonly used in Player's advertising. Despite the irony of linking tobacco use with exercise, sports are an effective way to appeal to a target market of male youth and to communicate brand imagery. Player's creative was to portray sport activities that \"should not require undue physical exertion... the chosen scene should ideally depict a pause or moment o f relaxation... However, the scene may show participants in action i f the moment o f product consumption can be assumed to be close to the scene depicted\" (ITL, 1980b, p . l , 2). Athletic pursuits perceived as aerobically taxing or strenuous were avoided in an apparent effort to minimize counter-argumentation among readers (i.e., relative to those fishing, it is less believable or credible that runners are smokers or 55 smokers are runners). I T L is l ikely attempting to curtail the possibility o f reminding consumers about the link between smoking and shortness of breath or other respiratory problems. During the late 1970s, advertising creative commonly featured portrayals o f horseback riding, fishing, canoeing, and downhill skiing (see Appendix 1). Two people were usually depicted, one man and one woman, who were presumably a couple. B y the early 1980s, advertising executions began to frequently feature windsurfing, sailing, white-water rafting, hang-gliding, downhill skiing, cycling, hiking, and mountain climbing (see Appendix 2). Initially, the ads depicted people participating in largely individual sports, but identifiable in a group of four (two men and two women). During the latter part of the decade, the typical ad showed an individual male smoking a cigarette while taking a break from a sports activity (i.e., hikers were often shown smoking at the top of a mountain, which also signified cigarettes as a reward) (see Appendix 3). Wi th the ad depictions often being in the 'great outdoors' or wilderness settings, the notion offreedom was seemingly communicated. According to Saine Marketing (1990), the mountains and the sea are traditional Player's symbols. The creative guidelines for effectively communicating freedom and independence included the portrayal of people who were \"free to choose friends, music, clothes, own activities, to be alone i f he wishes\" (ITL, 1985, p.60). Overall, Player's ads have effectively communicated the values that are typically seen as important to male youth, including the desire for independence, the ability to exercise freedom, haying contact with nature, being adventurous and recognized for those efforts, and possessing a certain masculine vitality (Saine Marketing, 1990). 56 Product Benefit Dimensions E d Ricard, an executive in charge o f ITL ' s strategy and product development, explained during testimony at the 2002 Quebec Superior Court trial: \"Benefit segmentation really allows us to identify where there might be opportunities for either an improvement to a current brand or a new brand to be able to go in and satisfy those particular needs of those smokers\" (1379).1 I T L indicates that, \"Based on a segmentation study done in Boston and New York, we have found that consumers describe and differentiate between cigarette products in a very similar fashion to Canadians. Their basic needs are also similar as they are characterized by such attributes as youth, popularity, masculinity, full flavour, mildness, aftertaste and irritation\" (1993, p.017989). It has already been demonstrated that Player's is linked with youth, popularity, and masculinity. These dimensions may be classified as both lifestyle portrayals and product benefits (i.e., Player's may be seen by its user as an expression o f desirable imagery). The remaining product attributes that were listed are largely based on the physical or taste characteristics of cigarettes. Cigarette strength is a partial determinant of initial brand selection. According to I T L (1995), 79% of males less than 25 years of age prefer cigarettes falling in the strong, medium, or mid-light strength category (accompanied by suitable imagery), and Player's, du Maurier, and Export are the only trademarks perceived to match these desires. When discussing the deep-blue colour of Player's Filter packaging, Paul Pare, from ITL, claimed that the \"particular blue was chosen because respondents indicated it best represented good flavour\" (cited in The Manitoba Educational Research Council , 1966, \u00E2\u0080\u00A2 57 p.89). Male adolescents often choose brands with a high taste impact (i.e., full flavour) and express a strong dislike for brands with very low tar deliveries (Kwechansky Marketing Research, 1977). Strong tasting, higher tar brands are viewed as a way to display one's maturity and toughness, and to indicate among peers who are more experienced smokers. Player's has traditionally been recognized as having a strong taste and high tar content compared to most other Canadian cigarette brands. Player's Filter, for example, currently has a machine-measured tar delivery of 16 mg, while I T L ' s Matinee Ultra M i l d (with a target market of female smokers aged 18 to 35) delivers 2 mg. Relative to Player's, Export is the only cigarette brand family perceived by smokers as stronger. Taste does not appear to be the all-important factor for initial brand selection, however. According to Spitzer, M i l l s & Bates (1977, p.12): A t a younger age, taste requirements and satisfaction in a cigarette are thought to play a secondary role to the social requirements. Therefore, taste, until a certain nicotine dependence has been developed, is somewhat less important than other things. Indeed strength of taste has not been a factor in the current creative strategy for Player's Filter. The brand has been positioned so that people can apply their own taste qualities to it. Ultimately, it is the brand's image or essence that is considered paramount.2 A defining moment for I T L was the launch of the Player's Light line extension, which occurred in 1976. When Player's Light was entered into the marketplace, the primary target market was identified as those in their early twenties to early thirties, yet adolescents were also included in the marketing strategy: \"Very much a secondary 58 market, but one that we anticipate w i l l increase in importance as the brand gets better known is young people starting the smoking habit and looking to start with a milder mid-range brand\" (Spitzer, M i l l s & Bates, 1977). A n evaluation of the marketplace in 1978, however, revealed Player's Light had quickly become appealing to youth. \"Recent C M A data shows that the brand profile has settled considerably younger than was originally defined for the target market, and is one o f the youngest in the cigarette market as a whole\" (Spitzer, M i l l s & Bates, 1978). To reflect and reinforce the line extension's popularity among youth, marketing objectives for Player's Light were revised to \"more firmly establish Player's Light as a milder version of Player's Filter - the brand for modern young smokers\" (Spitzer, M i l l s & Bates, 1978, p.689451877). In I T L ' s (1979) Fiscal '80 Media Plans for Player's Light, the age segment receiving the highest media placement weighting among English-speaking groups was 12-24. Although Player's has been largely positioned as a 'masculine' trademark, Player's Light proved popular among young women as well . According to Kwechansky Marketing Research (1982, p.66): The single most popular brand, and the one that seems to have become the customary badge among young males in particular, but among females very commonly too, was Player's Light. That this brand went from introduction to this incredibly lofty posture in so relatively few years is truly a marketing success story... Players, which by virtue o f Players Light has moved from malestream to mainstream. Player's Light was regarded as a ' light' cigarette (even though it had a higher tar delivery than most cigarettes in the 'light' category) that still offered acceptable taste and flavour. 59 Player's Light and Player's Filter are clearly the strongest performers of the brand family. Player's Medium, meanwhile, has under-performed since its 1988 launch, with its market share eroding largely due to the popularity of Export Medium among those under 25 years old in the province of Quebec. I T L claims, \"Player's has suffered somewhat recently due to a perceived lack of taste relative to strength... Player's Medium has been ineffective in competing in the medium strength segment which accounts for 20% of the market\" (1994a, p.502596051). Player's Extra Light also does not meet the expectations o f ITL , as smokers resist shifting from the Light brand to the Extra Light line extension (Saine Marketing, 1990). Thus, I T L is worried that 'health concerned' smokers contemplating a switch to a lower tar deliver brand might choose a competitor's offering. Other line extensions in the Player's brand family include Special Blend, Light Smooth, and Silver. 3 Player's Special Blend is perceived as exotic relative to the other brand family members (Qualitative Science Inc., 1994). Player's Light Smooth was introduced by I T L in 1992, attempting to fulfill the commonly expressed consumer desire for reduced irritation or harshness in the cigarette brand they smoke. I T L claims that, \"Reduced irritation, reduced aftertaste, the area of health remain key product needs where we must be innovative\" (1993, p.017966). Within a year of its launch, the Light Smooth line extension possessed a market share of more than 2%, but it quickly stagnated. E d Ricard, testifying at the 2002 Quebec Superior Court trial, clarified that the meaning o f smooth is meant to be \"reduced irritation,\" although the message is not wel l understood by consumers relative to product descriptors such as 'light' and 'extra light.' Anticipating that Tight' and 'm i ld ' descriptors may no longer be permitted for tobacco 60 products in Canada, Player's Silver was launched in March 2002, possessing a machine-measured tar delivery o f 8 mg (ITL, 2002). Again, the positioning of Player's has remained remarkably consistent over time. The Player's 1980 positioning objective was \"to maintain the trademark's established image as a full-flavoured, masculine trademark for young people,\" and in 1995, ITL ' s stated objective was to \"establish image as a popular, masculine trademark with a tradition of offering the highest quality, full flavoured products for young male smokers.\" Sport Sponsorship: A n Increasingly Important Communications Tool During the late 1980s and early 1990s, sponsorship of sports events became an increasingly important component of the promotional mix for Canadian tobacco manufacturers. From 1990 through 1994, ITL had registered or applied for 54 tobacco trademarks, with 24 of them being sponsorship-related (ITL, 1995). One subsidiary company formed by ITL for sponsorship purposes was Player's Ltd., which utilized colours, designs, typography, and logos closely matching what was found on brand packaging. 4 Complementing the formation of 'shell ' companies was a substantial increase in sponsorship expenditures, such that sponsorship contributions by Canada's three major tobacco manufacturers grew more than six-fold from 1987 to 1995 ( C T M C , 1987, 1997). The major sponsorship property of Player's Ltd. was open-wheeled, 'Champ car' auto racing, otherwise known as the Championship Auto Racing Teams ( C A R T ) series. Rothmans, Benson & Hedges (1995), Canada's second largest tobacco manufacturer, estimated that I T L spent C$34.2 mil l ion during 1995 on the Player's Ltd. racing sponsorship, which was a significant contribution considering that the total sponsorship 61 expenditures of Canadian tobacco companies that year was roughly C$62 mil l ion ( C T M C , 1997). 5 Brand Image Enhancement: A Key Sponsorship Objective of I T L A key objective o f sport sponsorship is to reinforce or enhance brand image (Irwin & Asimakopoulos, 1992; Copeland, Frisby, & McCarvi l le , 1996; Cornwell & Maignan, 1998). Indeed, the objectives and budgets sections of ITL documents pertaining to sport sponsorship are dominated by the importance of reinforcing or enhancing brand imagery. According to ITL, \"In terms of understanding, it is very clear that while the event itself is a communications vehicle, the true value is the amount of targeted imagery communications which surround the event\" (1992, p.013835). ITL documentation also reveals that imagery dimensions play a particularly important role in appealing to the age segment 18-24, which is the primary target market of Player's, and that \"the brands which are the most developed among the 18-39 have the strongest and clearest image platforms and personality attributes attached to them\" (Bateman et al., 1994, p.8). Player's and the Image(s) Transferred From Sponsoring Sports Events To fulfil the objective o f enhancing or reinforcing brand imagery, I T L selects to sponsor sports events possessing symbolic imagery or 'personalities' that are desirable to link with their respective brands. The objective is to have the image of a sports event transferred to the sponsoring brand. Don Brown, Chairman, President and C E O of ITL, claims that, \"Sponsorship is still limited in the degree to which it delivers a specific product 62 attribute message. The value lies in matching imagery of the event to that of the product or service\" (cited in Gross, 1994, p.67). 6 According to ITL: With regard to the brand or corporate image, the sponsor gives the impression of seeking to associate itself with the image of the event or of those who participate in the event. When a company sponsors a tennis or golf tournament, a regatta or the classical arts, this is interpreted by the public as a kind o f expression (by the sponsor) of the temperament of the company. Depending on the event sponsored, the company appears young, self-assured, master of itself, classical, adventurous, etc. (cited in Pollay, 2002, p. 13). When speaking about image matching, Player's and auto racing were described as \"an excellent fit.\" Under the sub-title Brand Image Association, I T L claimed that: \u00E2\u0080\u00A2 Racing is ideally consistent with existing imagery and desired brand profile: strong, masculine, young, adventurous, human \u00E2\u0080\u00A2 It symbolizes man on his own, independent, self-reliant (ITL, 1992, p. 011811). Another internal document reveals several questions that were asked about the sponsorship campaigns of the Player's trademark: \"How do the sponsorship events contribute to Player's image (stressing modernity) and lifestyle portrayal (freedom and independence)? How does the advertising for the sponsorship events contribute to Player's image and lifestyle portrayal (freedom and independence)?\" (Saine Marketing, 1990, p.9). Auto racing sponsorship secures an association with a dynamic and exciting sport, and it was anticipated by Canada's second largest tobacco manufacturer that Player's would \"update pack graphics in keeping with the chevron and the speed and emotion o f their new sponsorship promotion pieces\" (Rothmans, Benson & Hedges, 1996, p.002744). 63 The spectators o f auto racing also nicely match the lifestyle interests of the Player's target market, as the \"speed and excitement suggests a young, adventurous audience. Skews male both in terms of imagery and interest\" (Qualitative Science Inc., 1994, p.24). 7 Player's and the Image(s) Transferred From Sponsoring Athletes/Celebrities The process of image matching and transfer is also applicable to the participants (i.e., athletes, celebrities) of a sponsored event (Kahle & Homer, 1985; McCracken, 1989; Misra & Beatty, 1990; Kamins & Gupta, 1994; Lynch & Schuler, 1994). Player's initiated its sponsorship of auto racing in 1961 with the Player's 200, and the event allowed I T L to utilize celebrity appeals. Stirling Moss and Olivier Gendebien, both world-renowned auto racers, were featured in Player's promotional materials during the early 1960s (The Manitoba Educational Research Council , 1966). In 1975, Jackie Stewart was the spokesperson of the Player's Challenge Series (Schuler, 1997). During the past decade, Player's has sponsored Canadian Champ car racers such as Jacques Villeneuve, Greg Moore, A lex Tagliani, Patrick Carpentier, and Paul Tracy. Facilitated by the strong results of its sponsored drivers, Player's leadership and quality images are reinforced. Villeneuve was the rookie of the year in 1994, a victor o f the Indianapolis 500 in 1995, and the C A R T series champion in 1995. In recognition of his achievements, Villeneuve was the 1995 recipient of the L o u Marsh Trophy, being voted Canadian athlete of the year. Villeneuve moved on to Formula One auto racing in 1996. Meanwhile, Greg Moore won the Indy Lights circuit in 1995, winning a record 10 of 12 races; He entered C A R T in 1996, and during the following season, became the youngest driver to ever win a race. Moore died in 1999, however, as a result of a crash during the 64 Marlboro 500 race. During the 2002 season, Player's sponsored Patrick Carpentier and Alex Tagliani, with Carpentier placing third in the overall C A R T standings. Player's has signed Paul Tracy for the 2003 season and he represents the fifth Canadian driver to race for the C A R T team. When Player's Ltd. signed Villeneuve, he was identified as \"the next Canadian hero.\" In the eyes o f I T L , Villeneuve and Player's Ltd. possessed consistent imagery along leadership, independence, self-confidence, and youth dimensions. The Villeneuve father-son association (Gilles Villeneuve, Jacques' father, is regarded as Canada's best known car racer, and he raced in the Player's Challenge Series during the early part of his career) was also seen as an appropriate match for the Player's theme, Tradition of Excellence ( ITL, 1992). Moreover, in the popular press, Villeneuve has been characterized as rebellious and adventurous (and meanwhile engages in a very high risk sport) which makes him a desirable person to link with a brand marketed with such imagery (Dewhirst & Sparks, in press). The TeamPlayers.ca website communicates that when drivers such as Tagliani and Carpentier are \"off track,\" they still enjoy participating in adventurous and thrilling activities. Current print ads for Player's sponsorship of auto racing have the tagline, It's your world, and feature the various racing-team members in their 'spare time,' shown kayaking, white-water rafting, mountain biking, and rock climbing (see Appendix 4 and 5). Player's and the Image(s) Transferred From Sponsorship Partners Finally, the process of image transfer can occur between the title sponsor of an event and various co-sponsors or partners (Dewhirst & Hunter, 2002). B y forming partnerships 65 and strategic alliances through the sponsorship of sports events, co-branding opportunities are facilitated, whereby the functional or symbolic value of cigarette brands may be enriched. Co-branding is defined as placing two or more brand names on a product, its package, or additional elements of the promotional mix (Lamb, Hair, & McDanie l , 2000). One form o f co-branding is complementary branding, which involves marketing that suggests how products may be used together (Aaker & Joachimsthaler, 2000; Lamb, Hair, & McDanie l , 2000). Brands can be 'complements' on the basis of functional or symbolic properties. A promotion featuring Player's cigarettes and a Zippo lighter would exemplify functional complements, and a purposeful attempt to borrow upon one another's reputation as quality products. Wi th respect to symbolic properties, Budweiser beer would represent a suitable complement to Player's since alcohol consumption and smoking often happen concurrently and both brands have consistently been linked with masculine and independent dimensions. Such brand-matching observations are consistent with McCracken's concept of Diderot unities, which emphasizes that the meaning of goods is largely determined by their relationship to other goods. According to McCracken, \"The meaning of a good is best (and sometimes only) communicated when this good is surrounded by a complement of goods that carry the same significance. Within this complement, there is sufficient redundancy to allow the observer to identify the meaning of the good\" (1988, p.121). Considering the high operating costs that characterize the sport of auto racing, multiple sponsors are nearly inevitable. Player's and Molson Canadian (a popular brand of beer in Canada) are primary sponsors of the Molson Indy C A R T races held annually in 66 Toronto and Vancouver. The partnership between Player's and Molson Breweries, Canada's largest producer of beer, has facilitated cross-promotional/co-branding opportunities. A Molson Take Care advertising campaign from 1998, for example, depicted Greg Moore in his Player's racing,gear and the Player's racing car alongside, with the ad copy stating, \"This isn't a racing poster. It's a don't be stupid poster. Don't drink & drive\" (see Appendix 6). Other Take Care responsible drinking campaigns illustrated Moore in his Player's uniform with the tagline, \"Moore common sense: Don't drink and drive.\" Player's potentially gains credibility from such promotional campaigns by linking itself with a 'responsible' message. More recently, Player's and Molson Canadian were partners for the canada.com Ultimate Racing Challenge, which gave contestants a chance to w in authentic racing gear or a trip to Australia with drivers Patrick Carpentier and A l e x Tagliani. Player's and Molson Canadian can be considered suitable complements in many regards. The two brands have corresponding target markets since \"the entry level drinker\" is considered the main source o f volume in Canada's beer market. Promotional campaigns for Molson Canadian have been directed at those aged 19-24, with the objective of being \"the voice that defines the identity of Canadian youth\" (Durnan, Fitzgerald, & Davis, 1997, p.2). MacLaren McCann, the ad agency handling the Molson Canadian account during the late 1990s, claimed that, \"a compelling brand image is all-important - because beer drinkers wear their brand choice as a badge\" (Ibid.). The ad agency strategically positioned Molson Canadian as an expression of personal discovery, self-confidence, patriotism (toward being Canadian), and popularity. 67 These kinds o f cross-promotional opportunities can 'leverage' the value o f sponsorship for any single firm. The objectives of co-branding are best realized i f the link between the two brands is long-standing and well promoted. According to the president of Co-Options, an American firm that specializes in seeking out co-op marketing prospects, \"We want to be matchmakers for brands, but we don't want these to be flings. Our goal is to build long-term relationships that last at least a few years, giving brands maximum benefit from connections with other marketers and brands\" (Fitzgerald, 1994, p.30). Both Molson and Player's have been long-time sponsors o f Champ car racing in Canada. The inaugural Molson Indy was held in Toronto in 1986, and the Player's Challenge was included as one of the feature races (Schuler, 1997). Discussion From a managerial and marketing perspective, I T L can be considered a success story. The firm's market share has impressively increased during the past 25 years, with ITL now possessing nearly a 70% share of the Canadian tailor-made cigarette market. In explaining why I T L was the unmistakable market leader, Canada's second largest tobacco manufacturer claimed I T L was consistent in their communication, recognized the importance o f product benefits and imagery, and placed heavy investment behind everything that was done (Rothmans, Benson & Hedges, 1996). Reviewing internal industry documents has also revealed that ITL, largely through its Player's trademark, has been successful in appealing to the all-important 'youth' segment. Player's Light is clearly the preferred choice among the brand family, and is representative o f the Player's trademark for most consumers. 68 Building a Strong Brand With Consistency A key guideline for building strong brands is to have an identity, position, and execution that are consistent over time (Aaker, 1996; Ries & Ries, 1998). The Player's trademark has consistently been linked with youthful, masculine, independent, traditional, and modern (up-to-date) dimensions. For example, a 1939 print ad for Player's depicted a naval battleship, with ad copy indicating that, \"the sailor trademark on the package is recognized as a guarantee that its contents w i l l uphold Player's traditional standard\" (Stephenson & McNaught, 1940, p.201). During the mid 1990s, promotions for Player's sponsorship of auto racing utilized the tagline, A Tradition of Excellence. While the conventional ad from 1939 implies that Player's is a quality product with an extensive history, the sponsorship promotion communicates that the trademark has a long-standing association with auto racing, accompanied by quality events and impressive performances by sponsored drivers. I T L attempts to ensure that the various line extensions of the Player's brand family have cohesive meanings. Saine Marketing (1990) explored the extent in which the Player's trademark was \"unified,\" and considered the role of elements such as the use of the colour blue, the Hero logo, and additional graphics. According to ITL , \"The hero, the word Player's, and the colour blue should continue to be aggressively exploited\" (1995, p.9). The entire Player's trademark is associated with blue, thus a \"visual w a l l \" is created at retail, which gives the brand a degree of prominence and visual impact. In harmony with the target market of the Player's trademark, market research indicates that the Player's package design is appreciated overall, although more so by males. \"Player's 69 was often described as having a young sporty look\" (Qualitative Science Inc., 1994, p.29). The sailor on the Player's package, known as Hero, also contributes to consumer perceptions that the brand is positioned primarily toward males (Kwechansky Marketing Research, 1982). I T L has also communicated a consistent, complementary message to the target consumer through different elements of the promotional mix. While the T P C A was in effect, I T L claimed that: Traditional communication channels are now virtually eliminated. We must be more aggressive in exploiting new ways of communicating trademark images and new brand launches... Corporate entities, database marketing, trademark diversification, unique packaging formats must all be reviewed in the context of our new marketing environment (1993, p.017966-7). The corporate entity, Player's Ltd. , was successfully utilized as I T L focused its sponsorship portfolio toward auto racing (i.e., a few other sponsorship properties were discontinued). Recent sponsorship promotions, however, feature Player's racing-team members doing activities such as white-water rafting and rock climbing, where the ad creative is very reminiscent of conventional advertising seen during much of the 1980s. The promotion of auto racing sponsorship became an effective means of communicating virtually the same symbolic meanings as previous traditional product advertisements no longer permitted according to legislation. Sponsoring auto racing has also enabled I T L to collect demographic data from target consumers that is usable for direct marketing or relationship marketing purposes. Databases have been generated by staging contests, with entrance ballots accessible 70 through magazine ads or at the sponsored event sites. Direct marketing is increasingly used by I T L to continue communicating with its consumers. I T L has launched a lifestyle magazine called Rev, which contains content that is indicative of M a x i m or Gear, but is specifically designed to improve the brand profile of Player's. The magazine, which is glossy and roughly 100 pages thick, is distributed by direct mail to those on the tobacco company's database list (McLaren, 2001). Finally, package design and retail signage remain important parts of the communications mix. The Importance of Capturing the Youth Market Those representing I T L have publicly denied that their firm markets its products to youth. For example, both the chairman and president of ITL , in reporting to shareholders, claimed that, \"In the marketing of its products the Company has always regarded smoking as an adult custom. Its advertising has been directed to adults and has been unfailingly motivated by good taste\" (ITL, 1964, p.6). Under the heading smoking and youth, the website o f I T L currently states that, \"Imperial Tobacco Canada believes the choice tp smoke should be made only by adults.\" Contradicting these statements, internal documentation for I T L (1995, p.50) indicates that, \"Marketing activities have historically been and continue to be targeted at younger smokers due to their greater propensity to change brands.\" Moreover, The Creative Research Group (1988, 1990, 1991), on behalf of I T L , extensively researched males and females, aged 13 to 24, assessing their attitudes, values, lifestyle interests (music and sport preferences), media watching habits, ownership of items, and discretionary income. Rothmans, Benson & Hedges (1995, p.45), when assessing the 71 primary strengths that contribute to the success of ITL, stated that I T L \"owns the 14-17 age segment with over 90% of consumers smoking du Maurier or Player's.\" Player's is a widespread choice among youth and regarded as the dominant brand among male smokers. Player's advertising is considered \"suited to young people, because it shows one or more young people who like their independence and who live their lives confidently\" (Saine Marketing, 1990, p.29). Male youth demonstrate a need for independence, such that, \" Y A M [young adult male] members want to be 'on their own' . . . and they view independence as a key sign of their transition to adulthood\" (Ibid.). Linking the Player's trademark with independent dimensions reflects the market research findings for ITL: The adolescent seeks to display his new urge for independence with a symbol, and cigarettes are such a symbol since they are associated with adulthood and at the same time adults seek to deny them to the young. B y deliberately flaunting out this denial, the adolescent proclaims his break with childhood, at least to his peers (Kwechansky Marketing Research, 1977, p.i , i i ) . Ultimately, adolescents use cigarettes as a symbol (as they may also use coffee, alcohol, marijuana, or other drugs) to indicate they are growing up and ready to make the transition from childhood to adulthood. Focus group research, regarding Player's creative depicting horses, generated responses such as \"You ' re in the country, aye, and you're free and everything's going all right. I f you buy Player's that could be you\" (Kwechansky Marketing Research, 1977, p.566627927). Among the various ads reviewed, Kwechansky Marketing Research (Ibid.) found that the one with horses was \"perceived as the most teen-oriented cigarette 72 ad, and as teen-oriented as any other ad.\" Interestingly, ads for Player's Filter repeatedly featured horses in 1979. I f I T L was genuine about marketing its products only toward an adult market, it seems inappropriate to have independence and freedom at the core o f Player's positioning. Player's as a Big Player Player's is also purposely positioned to be the trademark with the highest perceived popularity relative to all other cigarette trademarks in Canada (Armada, 1999). Popularity is an additional dimension important to youth who are contemplating which brand to smoke. According to market research conducted for Rothmans, Benson & Hedges, \"Many respondents saw the brand [Player's Light] as the starter brand for youth due to its popularity generally\" (Segmentation Marketing Ltd., 1993, p.28). According to Kwechansky Marketing Research (1982, p.59), \"The boys almost exclusively pick Players Filter or Players Light (the latter because it is so popular, not because it is ' l ight ' .\". Contributing to the success of Player's is the fact that, \"The overwhelming and most influential factor in determining a brand's ' cool ' is perceived popularity. Nothing makes a brand seem uncool [sic] more than the perception that nobody smokes it, especially young people\" (Rothmans, Benson & Hedges, 1999, p. 18). Young consumers typically select cigarette brands that are strong sellers and mainstream, taking minimal social risks. Firms often spend a large proportion of advertising expenditures on one or two leading brands (i.e., those that have demonstrated popularity). I T L is no exception, spending a considerable proportion of its promotional dollars on the Player's and du 73 Maurier trademarks (Rothmans, Benson & Hedges, 1996). Thus, high volume communication for the Player's trademark is likely, with high quality ads also to be expected. Since the Player's trademark is advertised with such frequency, consumers are l ikely to believe that it is popular and a better buy relative to other brand offerings. Nelson (1974, p.50) claimed that: The consumer is right in his belief that advertised brands are better. The better brands have more incentive to advertise than the poorer brands... Those brands that get a lot of repeat purchases find it more profitable to advertise than brands that w i l l not get repeat purchases. Simply put, it pays to advertise winners rather than losers. In consequence, the amount o f advertising gives consumers a clue as to which brands are winners and which brands are losers. 9 Similarly, academic research findings by Kirmani and Wright (1989) suggest that consumers often consider newly introduced products to be higher quality i f accompanied by large advertising expenses. Sponsorship has increasingly become an important element of the promotional mix for Canadian tobacco manufacturers, and the trademark leader in sponsorship spending is clearly Player's. There are no trademarks from competing manufacturers that rival the promotional expenditures of Player's. During 1995, Player's sponsorship spending was estimated at C$34.2 mil l ion, while the top-spending trademarks for Canada's second and third largest tobacco manufacturers had expenditures of C$6.8 mil l ion and C$10 mill ion, respectively. In outlining I T L ' s strengths, Rothmans, Benson & Hedges (1995, p.46) listed \"high profile sponsorships with big ticket ($84 mil l ion) .\" Meanwhile, RJR-Macdonald (1996, p.13) observed that, \"Sponsorship dollars invested 74 do not correlate with market share performance. DuMaurier/Players disproportionately overspend relative to Rothmans & R J R Macdonald.\" Market research also found that the majority of respondents anticipated Player's as the most likely cigarette brand to sponsor a 'b ig ' event (Segmentation Marketing Ltd., 1993). In a separate study, it was discovered that, \"Males also see and appreciate Player's as the most active brand in the sponsorship arena\" (Rothmans, Benson & Hedges, 1999, p.20). When RJR-Macdonald reviewed various sponsorship properties, the Player's Racing Series was regarded as the most established and highest profile (Harrod & M i r l i n , 1996). Not surprisingly, Player's sponsorship properties were well recognized, with the Player's Racing Series generating the highest levels of awareness across all age segments. Awareness of the Player's Racing Series was highest among the age group 19-24, which was consistent with the overall finding that, \"Generally younger smokers tend to be more aware of sponsorship events than older smokers\" (RJR-Macdonald, 1996, p.801513337). 1 0 Player's Light has become the Player's Trademark Due to regulatory stipulations, tobacco sponsorship advertising in Canada has typically featured a trademark (i.e., Player's Ltd. , Player's) rather than a branded line extension such as Player's Light or Player's Extra Light. Interestingly, however, it was discovered by I T L during the early 1980s that, \"The Players family identification now seems to be centered upon the image of Players Light\" (Kwechansky Marketing Research, 1982, p.66, 67). Over the years, consumer perceptions about the Player's trademark have not changed much in this respect. In-depth interviews with males and females from Toronto, primarily under the age of 25, revealed that, \"Player's Light is the 75 pivot and anchor of the entire trademark\" (Marketing Strategy & Planning, 1986, p. 18). During focus group research, consisting o f Quebecois males aged 16-24, it was found that, \"When these young men think of Player's, they think of Player's Light first\" (Saine Marketing, 1990, p.61). For focus group respondents in Toronto, Vancouver and Montreal, ranging from 18 to 21 years of age, Qualitative Science Inc. found that, \"When smokers think o f the various trademarks they tend to think of a particular version. Wi th Player's they are most likely to bring to mind the Player's Light\" (1994, p.18). Finally, in a market review and business assessment for 1994/1995, Bateman et al. (1994, p.28) concluded that, \"Parent brands which inherently embody the core values of the trademarks remain very important to the young smokers. Player's Light can be judgmentally considered as the Parent brand of the Player's trademarks.\" Conclusion This chapter has reviewed the marketing activities of ITL for their major cigarette trademark, Player's. Internal industry documents reveal the segmentation strategies that have been utilized, in which Player's promotions have consistently targeted Canadian male youth who aspire to be masculine, self-confident, independent, and modern. To effectively meet the desires of the target consumer, conventional ads for Player's have commonly portrayed sports activities in outdoor settings, while sponsorship properties are largely centered on auto racing (albeit Player's racing team members are still shown rock climbing, hiking, camping, and so on). I T L is increasingly utilizing other marketing strategies (such as trademark diversification and direct marketing campaigns) in an attempt to continue 76 comrnuriicating that the Player's trademark is an expression of youthfulness, popularity, masculinity, freedom, independence, quality, tradition, and modernity. This case study of Player's brand marketing demonstrates that the connection between cigarettes and sport is well established. In an editorial on sport sponsorship, Wenner (1993) termed such an association, \"patently oxymoronic,\" claiming that, \"Athleticism and smoking clearly do not go together\" (p. 146). Indeed, the statements that smokers are mountain bikers and mountain bikers are smokers seem odd. Interestingly, however, visual associations of athleticism and cigarette brands appear to undergo less counter-argumentation from readers than verbal claims that make the link. Given the seriousness o f the public health issue, any promoted association of tobacco with athleticism seems inappropriate. 77 Endnotes 1 This quote is taken directly from court transcripts, thus the statements reflect precisely how it was phrased during legal proceedings. Wi th the exception of 1985-1986, a period dubbed the \"price war,\" price segmentation has not seemed to be a differentiating factor in consumer decision-making for Canadian cigarettes (Audet-Lapointe, 1991). 3 Player's Premiere was a line extension fully launched early in 1997, but was withdrawn from the Canadian marketplace in 2001 due to lingering sales (Pollay & Dewhirst, in press). 4 Player's is presently identified as the sponsor of various properties rather than Player's Ltd. 'Shell ' companies were no longer required for sponsorship purposes once the Supreme Court of Canada ruled the T P C A unconstitutional in September 1995. 5 It is difficult, however, to ascertain whether the same criteria were used for the sponsorship figures cited. The C$34.2 mil l ion estimate for Player's Ltd. was based on advertising, merchandising, and operation costs related to auto racing sponsorship. The C T M C did not specify how the C$62 mil l ion figure was determined. Nevertheless, auto racing attracts a large proportion of sponsorship spending. Turco (1999) claims that American tobacco firms allocate more than 90% o f their sport sponsorship budgets toward motor-sports. 6 Gladden and Wolfe (2001) describe image matching as \"the extent to which a sponsored sport property and a sponsoring corporation project consistent images\" (p.41). The notion of a sporting event's image being transferred to a brand through event 78 sponsorship promotional activities is consistent with the academic research findings of Ferrand and Pages (1996), Mi lne and McDonald (1999), Musante, Mi lne , and McDonald (1999), and Gwinner and Eaton (1999). 7 C A R T audience demographics are defined as 78% male (J. Horwood, personal communication, August 10, 2001). 8 Player's was classified as a \"starter brand\" by Pollay (2000), after he reviewed internal industry documents that were the basis of an expert witness report prepared for the proceedings assessing the constitutionality of the T P C A . 9 Given the addictiveness of tobacco and the fact that many cigarette brands have minimal product differentiation, the applicability of Nelson's argument is unclear. 1 0 Crompton (1993) notes that a central issue surrounding tobacco sponsorship of sport is whether it enables tobacco firms to penetrate the youth market. 79 C H A P T E R F O U R Export ' A ' \u00E2\u0084\u00A2 Cigarette Brand Market ing Male Youth, Extreme Sports, and the Gendering of Smoking Manufactured by JTI-Macdonald Corp., Export ' A ' is currently the third best-selling cigarette brand in Canada, holding an approximate 11 percent share of the Canadian cigarette market. Considering that all o f the tobacco manufacturer's cigarette brand offerings combined account for less than 13 percent o f the total Canadian market, the success of Export ' A ' is extremely important toward JTI-Macdonald's bottom-line (Physicians for a Smoke-Free Canada, 2002). This chapter examines how conventional notions o f masculinity have been exploited in promotional efforts to construct an identity for the cigarette brand, Export ' A ' . 1 The cigarettes were provided to Canadian troops fighting in the Second World War, and by 1945, one-half of the Canadian forces were smoking Export (\"The Blue Book of Canadian Business,\" 2002). Today, the Export brand name identifies plain cigarettes, while the Export ' A ' brand family consists of filtered full flavour, medium, mi ld , light, extra light, and ultra light line extensions. During the early and mid 1980s, advertising for Export ' A ' began to link the brand with adventurous, action-based sports, and more recently, it has been the title sponsor of an extreme sports series. To further understand how the tobacco firm sought to make the brand attributes and 'personality' o f Export ' A ' both masculine and appealing to male youth, trade sources and internal industry documents have been 80 analyzed. The industry documents provide valuable insight for encoding the production of promotional messages, and in this chapter, both the written and visual persuasive aspects of advertising for Export 'A' are explored. Advertising and Rhetoric Rhetoric, a term derived from the Greek rhetor, which literally means 'orator', is defined as the art of persuasion, using language to influence other people either in terms of their future actions or beliefs (Edgar & Sedgwick, 1999; Danesi, 2000). Rhetorical analysis has traditionally been limited to assessing the use of written or verbal language; yet, studying the persuasive aspects of advertisements needs to also account for the importance of visual imagery. Cigarette advertising studies using content analysis as a methodology, for example, have typically assessed both visual and verbal assertions. Ringold (1987) and Ringold and Calfee (1989) represent notable exceptions by not considering pictures and images for coding purposes (i.e., coding was limited to the information conveyed in headlines, subheadings, and ad copy), considered a significant omission since the majority of cigarette advertising layouts are visually oriented. In discussing how rhetorical figures are defined, McQuarrie and Mick (1999, p.39) state, \"Nothing in the fundamental definition of a figure either requires a linguistic expression or precludes a visual expression.\" Durand (1987) was a pioneer in proposing a comprehensive list of visual figures, and since that time research has increasingly accounted for visual elements of rhetoric with an application to advertising (McQuarrie & Mick, 1999), responding to Scott's (1994) call for developing a theory of visual rhetoric when studying advertising imagery. 81 Leading textbooks on marketing and advertising recognize that persuasion is a basic objective of promotion (Arens, 1999; Semenik, 2002). According to Shapiro, Wong, Perreault, and McCarthy (2002, p.435), \"When competitors offer similar products, the firm must not only inform customers that its product is available but also persuade them to buy it. A persuading objective means that the firm w i l l try to develop a favourable set of attitudes so that customers w i l l buy\u00E2\u0080\u0094and keep buying\u00E2\u0080\u0094its product.\" Persuasion is a particularly important promotional objective for product categories such as cigarettes, where differences among various brands are often very intangible. According to documentation from Imperial Tobacco (1970, p.566628090), Canada's largest tobacco manufacturer, \"Without price differentials and without easily perceptible product differentiation (except for extremes, e.g. Matinee versus Player's) consumer choice is influenced almost entirely by imagery factors.\" Roughly 25 years later, Rothmans, Benson & Hedges, the second largest tobacco manufacturer in Canada, maintained that, \"In the cigarette category brand image is everything. The brand of cigarettes a person smokes is their identity. Cigarettes tell others who they are as a person. There is a strong emotional connection to the brand, the image it projects about the smoker, not only to themselves but to others\" (cited in Pollay, 2002, p. 13). Abstract qualities are commonly associated with products, such that brands are considered to have personalities much like people. Particular brands are depicted as expressions of success, sophistication, femininity, rebellion, and so on. Marlboro, for example, represents masculine, rugged, tough, and no-nonsense qualities, while Virginia Slims tends to be feminine, sexy, and glamorous in comparison. Using a non-cigarette 82 example, Harley-Davidson is a symbol of personal freedom, patriotism (toward being American), rugged individuality, and being macho (Aaker, 1996). Those critical o f advertising often express concern about how products become arbitrarily associated with attributes that are socially desirable. Some advertisements present something magical since it is suggested that the featured product w i l l do something special for those who purchase it (Leiss, Kl ine , & Jhally, 1997). According to Gossage (1967), who worked al l his life in the advertising industry, the appeal o f advertisements became more basic as there was an attempt to persuade more people of the worthiness o f the promoted product. He stated: When, in addition, the product advertised is virtually identical with its competitors, or when the product's value to its user is largely subjective, the appeals become so basic that they slide away from fact as we know it. They go beyond reason into something even more basic, the most common denominator of all , magic (1967, p.364). Leo Burnett, the creator of the Marlboro man, explains the work process for those at an advertising agency: After all the meetings are over, the phones have stopped ringing and the vocalizing has died down, somebody finally has to get out an ad, often after hours. Somebody has to stare at a blank piece of paper. This is probably the very height of lonesomeness. Out of the recesses of his mind must come words which interest, words which persuade, words which inspire, words which sell. Magic words. I regard him as the man of the hour in our business today (cited in Twitchell, 1999, p.56; Simpson, 1964, p.83). 83 A magical spell is evident since the consumer is promised to get great results from purchasing the advertised product (Williamson, 1978). Advertising may instil in consumers the belief that satisfaction is obtained through purchasing commodities, yet those defending advertising counter that people want and desire the symbolism of advertising since they do not consider products simply for their use. People interpret their interactions with products symbolically, and the need and desire for symbolism is considered a defining quality of human nature (Leiss, Kl ine , & Jhally, 1997). Systems of signification are necessary and inevitable. Persuasion is also considered to be a common part of our society, thus some question why advertising deserves special attention. Whether the consumer has the same ability to detect truth among products of differing characteristics and complexities remains contentious, however. Export ' A ' : Segmentation as a Marketing Tool A s discussed in Chapter 3, segmentation is a commonly used strategy in marketing, in which specific target audiences are identified for a product or service by dividing a mass market into subsets on the basis of variables such as demographics, geography, psychographics, and product benefits. B y reviewing internal documents of RJR-Macdonald from the 1980s and 1990s, it became apparent that the primary target audience for Export ' A ' promotions has been male youth who aspire to be independent, adventurous, and rugged. 84 Male Youth as a Primary Target Market A document concerning the long-term strategy o f Export ' A ' included the subtitle, \"Whose behaviour are we trying to affect?\", with the following statements in response: Young adults who are currently in the process o f establishing their independence and their position in society. They look for peer group acceptance in their brand selection, and may often be moderate or conservative in their choices. A s young adults they look for symbols that w i l l help to reinforce their independence and individuality (RJR-Macdonald, 1987, p.800230290). In this document, young adults were identified as predominantly 18-24 year old males. \"New users\" were also listed as a subgroup in which the company was hoping to have an influential effect on behaviour, a reference that is meaningful and significant by indicating that the age segment likely being discussed is those in the early or mid-stages of adolescence. RJR-Macdonald (1982a, p.395552) hypothesized that, \"Very young starter smokers choose Export ' A ' because it provides them with an instant badge o f masculinity, appeals to their rebellious nature and establishes their position amongst their peers.\" Younger smokers are considered essential to the Export market (RJR-Macdonald, 1982b). RJR-Macdonald (1989, p.801183934) stated that, \"In order to make further inroads into the younger segment, we must continue to project an image that is consistent with the needs and values of today's younger smokers.\" Within the same document, the company asserted that: The younger segment represents the most critical source o f business to maintain volume and grow share in a declining market. They're recent smokers and show 85 a greater propensity to switch than the older segment. Export has shown an ability to attract this younger group since 1987 to present (Ibid.). In addition to its appealing brand imagery dimensions, Export ' A ' is likely to be desirable to the youth segment because it is a strong-selling, popular brand. Qualitative Science Inc. (1994, p.800939376), a marketing research firm representing RJR-Macdonald, acknowledges that, \"When young consumers first experiment with smoking they are prone to select a brand which they perceive as having an image which is 'mainstream, youthful'.\" Reinforcing and Enhancing Export ' A ' Brand Imagery The Export Family Strategy Document (RJR-Macdonald, 1982a, p.395552) contains a section entitled \"How We Want Consumers to V i e w the Brand,\" indicating that image dimensions for Export ' A ' w i l l appeal to \"the breed o f men who are masculine, independent, adventurous and possess the qualities of natural leadership... Women are attracted to these men because o f their youthful virili ty, independence and spirit o f adventure.\" In the Export Family\u00E2\u0080\u0094Marketing Strategy document, it states that, \"The objective of the family's copy strategy is to convince males that only the Export Family o f cigarettes provides the highest degree o f smoking satisfaction in the range o f desired strength levels for individuals who aspire to be masculine, rugged, self-determined and independent\" (RJR-Macdonald, 1980, p.45676). The Beaumont Organization, Ltd. (1982) recommended that RJR-Macdonald explore symbols that integrate the themes of masculinity, peer group acceptance (among younger adult males), and pleasure into a single, comprehensible message. Export ' A ' and competing brand, Player's have a very similar brand positioning, and when 86 comparing the user image of the two brands during the mid 1980s, an ad agency representing Imperial Tobacco found that \"the common ground shared by Export ' A ' and Player's Filter is lone masculinity in the context of outdoor, physical activity\" (Marketing Strategy & Planning, Inc., 1985, p.000019). In the late 1980s, RJR-Macdonald looked to \"seek and exploit innovative promotional opportunities... based on Export's 'escapist' brand image\" (cited in Pollay, 2002, p.8). To summarize, over roughly the past 30 years, RJR-Macdonald (and now JTI-Macdonald) has positioned Export ' A ' to be an expression of masculinity, independence, adventure, ruggedness, and escapism. Export ' A ' Brand Marketing (1972-2002) From 1972 to 1988, the print advertising o f Export ' A ' progressed through four highly integrated periods. The year 1972 is used as a starting point o f analysis since Canadian tobacco companies increasingly directed their advertising spending toward the print media once they had voluntarily withdrawn advertising from the broadcast media, effective January 1, 1972. Wi th the enactment of the T P C A in 1988, Canadian tobacco firms, such as RJR-Macdonald, invested heavily in the sponsorship of sports and cultural events. Export ' A ' as Man's Best Friend (Early 1970s) During the early 1970s, an advertising campaign for Export ' A ' had the slogan, Good Companions placed in bold text, the subtext claiming, \"Take time to relax. Take time for quiet companionship. Export ' A ' Kings . . . a good companion any time\" (see Appendix 7). The ads were predominated by visual imagery, in which a man was shown 87 seated outdoors in the midst o f a leisurely activity with a dog, smoking an Export ' A ' Kings cigarette. The conveyed message was that smoking Export ' A ' cigarettes provided independent males with an opportunity to relax. Wi th the ad depictions being in the wilderness or at a cottage, the notion of 'escapism' was seemingly communicated. Export ' A ' cigarettes were presented as though they had a relationship with their user, much like a relationship one would have with a pet or a person. The exclusion of women as potentially good companions is particularly noteworthy. Big Flavour and Export ' A ' Satisfaction (Mid and Late 1970s, Early 1980s) From 1975 to 1978, the slogan for Export ' A ' advertising campaigns was Export, eh?, containing the subtext, \"When you know what you l ike\" (see Appendix 8). Again, individual males were depicted smoking a cigarette while engaged in a passive leisure activity. During the late 1970s and early 1980s, the emphasis o f the advertising campaigns seemed to shift increasingly toward the high impact flavour o f Export ' A ' , one ad claiming, \"I L ike it! B i g Flavour. Export ' A ' Regular and K i n g Size.\" Advertising campaigns during 1979 and 1980 featured the slogan, \"Feeling Satisfied!\" and the ad copy, \"Export ' A ' Satisfaction\" (see Appendix 9). The slogan for the 1981 campaign was \"Man , that's satisfaction\" with the remainder o f the ad copy stating, \" Y o u know what you want and know where to get it. That's why you reach for the big bold taste of Export ' A ' . \" A green package was depicted in these ads, which was representative of the full flavour version of Export ' A ' . When a blue package was portrayed, representing 'medium', the slogan was \"That's smooth! That's satisfaction!\" with the remainder o f the ad copy claiming, \"That's what you want and you know where 88 to get it. So you reach for the satisfying smoothness o f Export ' A ' medium. Man , that's satisfaction.\" For these advertising campaigns, individual males were commonly featured who were loggers and truckers (i.e., blue-collar occupations), shown smoking a cigarette in wilderness settings, presumably taking a break while on the job. In 1981 and 1982, ads featured the slogan, \"Satisfaction Country\" and visuals that were similar to previous campaigns. 3 A Taste For Adventure and Aspiring Imagery (Mid 1980s) In an attempt to revitalize the brand, ad campaigns from 1983 to 1985 ran with the slogan, \" A Taste for Adventure\", and featured males in 'active' pursuits such as white water kayaking and canoeing, hydroplaning, downhill skiing, and windsurfing (see Appendix 10). RJR-Macdonald also signed a five-year agreement with the Canadian Ski Association ( C S A ) in 1983. 4 The modified approach reflected the recommendations of The Beaumont Organization, Ltd., who found that the Export ' A ' user was becoming commonly associated with older, macho, blue-collar men. To gain market share, they claimed Export ' A ' must be dissociated from such overtones and become more appealing to the younger adult male peer group. While the blue-collar truck driver image o f Export ' A ' may have been accurate, it did not capture the aspirations of youth (Bang & K i m , 2001; Pollay, 2001). When assessing the lifestyle interests o f many Export ' A ' users, market research found that they were likely to enjoy attending rock concerts and sports events, as well as spend time in bars and clubs. It was recommended that Export ' A ' should be positioned toward younger males who were sports-oriented, drink beer, enjoy 89 popular music, and commonly wear blue jeans and T-shirts (McCann-Erickson Advertising o f Canada Ltd. , 1986). Meeting the Entire Export ' A ' Brand Family (Late 1980s) In 1987, an advertising campaign for Export ' A ' extra light had the slogan, \"Export ' A ' . Where you want to be\", with visual imagery of a city skyline and nightfall beckoning. Interestingly, this ad campaign was digressing from the traditional positioning o f the trademark, yet the ad platform was more consistent the following year. Export ' A ' extra light ads in 1988 featured \"Export Y o u r s e l f as ad copy along with visuals of a computer with a wilderness scene and flying eagle projected on the screen. A 1988 advertising campaign presented Export ' A ' ultra light as a new product line extension. Other ads utilized the slogan, \" ' A ' Smooth New Look,\" and depicted packaging of the parent full flavour brand, as well as the medium, mi ld , light, and extra light members o f the brand family. 5 Emphasis on all o f the various product line extensions l ikely reflected that 1988 was the final year of conventional tobacco product advertising being permissible in Canada, in accordance to the T P C A . Sponsorship: Becoming an Important Part of the Tobacco Marketing Plan During the late 1980s and early 1990s, sponsorship of sports and cultural events became an increasingly important component of the promotional mix for Canadian tobacco manufacturers, shifts in promotional spending toward sponsorship largely reflecting that conventional forms of promotion were no longer viable options due to legislative stipulations. In addition to improving corporate/brand awareness, a key objective of 90 sponsorship has been to enhance or reinforce brand imagery. According to RJR-Macdonald (1996, p.801542472): Associative marketing allows us to associate the brand with images which we are prevented from using in brand advertising. In other words, the actual sponsorship is simply the price we pay in order to feature a particular image in our advertising. Although there are many additional benefits to traditional sponsorship programmes such as promotional extensions, our primary concern is with the image advertising potential around the sponsorship. We are attempting to alter a brand's image and, in our view, this is best achieved through advertising which we control. Considering that Export ' A ' has a long-standing association with masculinity, independence, adventure, ruggedness, and escapism, sponsoring an extreme sports series would seem to be a particularly apt way of enhancing or reinforcing the desired brand image. Export ' A ' : Sponsor of the Extreme Sports Series Beginning in 1997, Export ' A ' became the title sponsor of an extreme sports series, which has included activities such as free-skiing, skier-cross, snowboarding, snowmobile racing, mountain biking, jet-skiing, windsurfing, kayaking, water-skiing, wakeboarding, hydroplaning, and motor-cross and auto racing (see Appendix 11).6 The aforementioned list of sponsored sports reflects that RJR-Macdonald wanted to select events that were distributed throughout the year. During 2001 and 2002, snowboard and free-ski events for the Export ' A'-sponsored extreme sports series continued to occur during the summer months as part of the Molson Canadian Snow Jam. 91 Interestingly, RJR-Macdonald also found it desirable to depict sports that were not perceived as too aerobically taxing, a 1997 document indicating that, \"Associations of Export ' A ' with (non-cardio) demanding sports with a social celebration angle is most credible and motivating\" (cited in Pollay, 2002, p.45). This observation might explain why mountain biking was discontinued as an event after the inaugural year of the extreme sports series. To augment the 'independent' brand image of Export ' A ' , the extreme sports series features exclusively individual sports, despite the popularity o f (non-cardio) team/group sports such as baseball or white-water rafting (Pollay, 2001). For the advertising campaigns running from 1997 to 2000, independence, individualism, and self-reliance dimensions were further conveyed by the slogan o f the Export ' A ' extreme sports series, which was \"Go Your Own Way.\" The promotions have typically depicted one person in primary focus. Notions of self-expression, freedom, and rebellious youth are communicated since many of the extreme sports events epitomize resistance to dominant sport cultures. Several of the extreme sports events are distinctive from 'mainstream' sports according to their less rigid and controlled rule structure. For example, downhill ski racing requires participants to take a prescribed route (that is, the path is dictated by the gates that are set) and each run is timed, whereas free-skiing participants may negotiate their own pathway (among couloirs, cliffs, and non-groomed terrain that is typically 35 to 50 degrees steep) and are judged according to their fluidity, route selection, aggressiveness, and dynamic turns (Lee, 2002). 92 'Adventure' is expressed since the extreme sports series activities involve competitors who succeed and generate attention because o f their willingness to take extreme risks. According to Webster's dictionary, extreme means \"going to great or exaggerated lengths; exceeding the ordinary, usual, or expected.\" For the Big Hit Contest, which was sponsored by Export ' A ' and Ripzone at the 2002 Molson Snow Jam, it was emphasized that C$1,000 would be awarded to the first place finisher, yet nothing in recognition for second place. Wi th simply first place being rewarded, and only the best ' trick' among several attempts being counted (that is, scores were not tallied cumulatively), participants were encouraged to 'push their limits. ' Moreover, with the sponsorship of extreme sports by Export ' A ' , the link of cigarette smoking and its own role toward risk-taking is seemingly implied. For adolescents who are smoking in defiance o f their parent's wishes, initial experiences with cigarettes can be exciting or thrilling. Smoking may constitute a rebellious act that is meant to undermine adult authority (Dewhirst & Sparks, in press). i During 2001 and 2002, promotions for the extreme sports series have included the following edgy slogans: \"Gravity is for sissies\"; \"If it ain't broke, you're not going fast enough\"; \"Sure it's all a blur. That's the point\"; \"Competitors kiss their loved ones before every race. Usually on the hood\"; \" I f you can't wrap it around a tree, what's the point?\"; \"It's loud. It's fast. It's hard. Questions?\"; \"Real men don't cry. They bleed\" (see Appendix 12). These slogans further imply the need for thrill seeking, 'going to the edge', and the desirable stimulation resulting from an adrenaline-rush. The narratives relating to Export 'A'-sponsored extreme sports also communicate alpha 93 masculinity, whereby it is implied that these events provide male youth with opportunities to express ruggedness and machismo. Women, however, have a contrasting role in the subculture of Export ' A ' -sponsored extreme sports. They have been largely unrecognized in Export ' A ' promotions, in which most readers o f the ads would be unaware that women are also participating in the extreme sports events. During the few occasions that women are depicted in Export ' A ' promotions, they are often presented as 'eye candy', framed as sexually desirable in spectator or after-event socializing roles. L e i f Zapf-Gilje, a competitor in the free-skiing event, claims that, \"The rewards in extreme skiing are women, money and free beer, but not necessarily in that order\" (cited in Andreeff, 2000, p. 107). During the Big Hit Contest at the 2002 Molson Snow Jam, lots of free items were distributed to help maintain the crowd's enthusiasm. To attain a few o f the coveted prizes, male event participants required women in the audience to reveal their breasts. Discussion Youth as a Target of Export ' A ' Promotional Activities One key issue during tobacco-related legal proceedings has been the determination of whether youth are targeted by specific marketing campaigns. Tobacco industry representatives have publicly denied that they market their products to youth; for example, John Wildgust, the director of corporate affairs for JTI-Macdonald, has stated: \"To set the record straight, JTI-Macdonald Corp. never undertakes any activity that 94 would encourage anyone to take up smoking. Yes, we do compete for brand share among current adult smokers. But we do not market to children, period\" (2002, p.A17). Contradicting Wildgust's statements, JTI-Macdonald ran television ads in support of the Export ' A ' extreme sports series events held from February to November 2000, in which adventure and sex were central themes (i.e., internal documents indicated that a premise relating to adventure or sex was an effective way to appeal to those in their early teenage years). One o f the 30-second ads commenced such that a bed mattress and springs could be seen moving up and down, while a bedside alarm clock was depicted with the time flashing and a television showing merely static. The vibration resulting from the 'bed activity' caused a drink on the bedside table to spill , while other objects (including a set of downhill skis) fell to the floor. The 'dominant/preferred' meaning of this ad seemed clear; during the initial segment of the ad, most viewers were l ikely to assume that the bed mattress was moving due to sexual activity. It was later revealed, however, that a young adult male was jumping up and down on his bed, practicing ski moves that he would presumably use in the extreme sports series. The question, \"Feeling a little extreme?\" was then posed, as well as the Export ' A ' slogan, \" G o your own way.\" The ad aired even though market research testing for an Export ' A ' campaign in 1996 indicated, \"Respondents believed certain executions were more l ikely than others to appeal to the younger set, i.e. those under the age of 19. Generally speaking, ads that identified with 'adventure or sex' were said to more likely appeal to the teen and even pre-teen segment\" (cited in Pollay, 2002, p.44). JTI-Macdonald is also a client of M A S E V Communications, who claim to be an authority on youth lifestyle and culture, and who specialize in communicating brand 95 messages through experiential events and promotion-based marketing. According to the website of M A S E V Communications (see www.masev.com/corporate.html, which lists JTI-Macdonald as a client and discusses its involvement with organizing the Molson Snow Jam), \"Wi th over 18 years involvement in music, action sports and events M A S E V has become the credible marketing message conduit for corporations to reach the vibrant 12-35 youth lifestyle demographic.\" When looking at the marketing strategies of other sponsors for similar action sports events, teenagers (including those that are still not of legal smoking age) are commonly among the target market. For example, the Telus World Ski and Snowboard Festival is a 10-day event held in Whistler, Canada that includes ' B i g A i r ' snowboarding contests, which closely resemble the Export ' A ' sponsored ' B i g H i t ' contests held at the Molson Snow Jam. Explaining why Telus (a large telecommunications firm based in British Columbia) is the title sponsor of the event, the company's marketing executive, John Mikkelsen, states: \"To be honest with you, we are trying to be a little cool\" (cited in Ward, 2002, p . B l ) . He claims that the snowboarding and ski festival is a way of promoting the Telus brand to a target market of young males, who are aged 16 to 24 (Ward, 2002). John Stouffer, editor o f Transworld's snowboarding magazine, also spoke about the audience at these events, and their familiarity with some of the competitors' stances and sponsors: Most o f these kids down there are snowboarders. They're in the pipes, in the terrain parks. They are participants and so they know what it takes... These kids w i l l dial their set-up just like their favourite pro. They ' l l want to wear their 96 beanie [toque] just at right angle with the logo showing. That's how infectious this is (cited in Ward, 2002, p.B2). Interestingly, Export ' A ' logo emblazoned items - such as key-chains, screen-saver C D -R O M s , baseball caps, T-shirts, and C D cases - have been distributed at the extreme sports series events held at the Molson Snow Jam. Glen Plake, a skier infamous for his cliff-jumping abilities, represents another early icon of extreme sports. In describing Plake, Powder magazine indicates that he \"smokes cigarettes, opens beer bottles with his ski edges and [spreads the word] that... the point of skiing [is] not wholesome family fun, but to be extreme\" (cited in Chrzanowski, 1990, p.36). B y sponsoring extreme sports events, it seems dubious that JTI-Macdonald is genuine about only marketing toward an adult market of current smokers. 9 Even if, implausibly, this were the only intent, it would not be the only effect. Export ' A ' and the Exclusion of Women In early research on masculine-feminine dimensions of cigarette brands, V i t z and Johnston (1965) found that consumer perceptions about the 'masculinity rating' of brands was influenced by physical characteristics of the product, as wel l as the advertised image. Cigarettes positioned for a predominantly male target audience, or meant to convey 'masculine' or rugged dimensions, are commonly higher in tar content, strong tasting, regular length, and/or wide (McCracken, 1992; Pollay & Dewhirst, 2001). Conversely, cigarette brands positioned for primarily female use, or meant to be expressions of 'femininity' or glamorous dimensions, are typically those that have very low tar deliveries, longer lengths (that is, 100 mm), or are billed as ' s l i m ' . 1 0 Recognized as a 97 relatively high tar and strong tasting brand, the Export ' A ' brand family consists o f various line extensions, yet the full flavour and medium versions are the most commonly promoted. Longer length, slim, or menthol cigarettes are not offered as part of the Export ' A ' brand family. Women are rarely depicted in Export ' A ' advertising. One marketing firm (Marketing Strategy & Planning, Inc., 1985, p.000007), when comparing the brand imagery o f Export ' A ' and Player's, indicated, \"Player's Filter has a considerably softer masculine image than Export ' A ' which carries masculinity to an extreme... The Player's Filter image acknowledges and accomodates women. The Player's Lights image actually incorporates women. The Export ' A ' image excludes women.\" Although a small number of entries for the Export ' A ' extreme sports series are women, their participation seems largely unwelcome and uncelebrated. Export ' A ' promotions are placed in media most l ikely to be consumed by males. For example, television ads supporting the sponsorship of the 1999 extreme skiing and snowmobile events aired on The Sports Network (TSN) - Canada's equivalent to E S P N -from February 15 t h until Apr i l 5 t h . Auto Racing, Gallagher Live, Leafs Hockey, NBA Basketball, and Sports Desk (the program is now known as Sports Centre) were T S N programs that featured Export ' A ' extreme sports ads. Nielsen Media data indicates that the gender breakdown of T S N Sports Desk, Leafs Hockey, and N B A Basketball audiences is 73%, 70%, and 74% male, respectively (L. Cameron, personal communication, September 25, 2001). Recent Export ' A ' promotions for the extreme sports series are to be found in Canadian magazines such as Shift (chronicles the impact of technology on culture), Cycle Canada (stories about motorcycling in Canada), and 98 Urban Male Magazine (its content is indicative of M a x i m , Gear, or Stuff magazine), which have a male-skewed readership. The Importance of Integrated Marketing Communications While Export ' A ' is presently the third-best selling cigarette brand i n Canada, its market share has been in continuous decline over the past few decades. In 1978, the market share of Export ' A ' was 17%, yet in 1988 it had dropped to 12% and in 1997 rested at 11% (Pollay, 2001; Physicians for a Smoke-Free Canada, 2002). One explanation for the waning popularity of Export ' A ' is that its marketing communication efforts have not been wel l integrated, involving \"the intentional coordination of every communication from a firm to a target customer to convey a consistent and complete message\" (Shapiro, Wong, Perreault, & McCarthy, 2002, p.433). Considering that Export ' A ' has been promoted as an expression o f masculinity, independence, adventure, ruggedness, and escapism for a lengthy duration, the objective of its manufacturer should be to select events for sponsorship purposes that w i l l consistently link the brand with such symbolic imagery or 'personalities'. Sponsoring an extreme sports series seems to offer great potential for being integrative with previous Export ' A ' marketing communication efforts. Overall, however, the brand identity o f Export ' A ' has been inconsistently communicated through the selected sponsorship mix, which has also included The Skins Game (featuring premiere male professional golfers such as Jack Nicklaus, Greg Norman, and David Duval), The New Music Series (presenting rock music concerts by several promising Canadian bands), and salmon fishing showdowns. This inconsistency 99 with the selected sponsorship properties became particularly crucial once sports and cultural sponsorship took an increasingly important role in the promotional mix. The Export ' A ' package colour emphasized in the artwork of promotions has also shifted over time. The green-colour package, which signify full flavour cigarettes, was prominent in many Export ' A ' promotions during the 1970s and 1980s (including the \" A Taste for Adventure\" campaign that depicted various action sports), yet promotions for the extreme sports series (from 1997 to the present) have utilized the blue hue that resembles the packaging for medium cigarettes. The shift in colour emphasis may reflect that smokers from the past decade have elevated health concerns, and are l ikely to prefer a brand extension without the highest tar content and delivery. It is probable that RJR-Macdonald (and JTI-Macdonald) has prominently depicted blue - as opposed to maroon, gold, silver, and baby blue - since the mi ld , light, extra light, and ultra light line extensions do not convey an image that is nearly as macho or rugged. The drawback o f using Export ' A ' medium as the headline extension, however, is that the competing Player's cigarette brand is already wel l established with a similar shade of blue. Pollay (2001, p.73) argues: Export ' A ' is adopting a 'me too' positioning, known to be a difficult task. B y imitating and extending that which has been so successful for Player's, they risk falling into the perceptual shadow of Player's brand image, with some viewers confusing their effort with ads for Player's. The deteriorating market share of Export ' A ' has been largely at the expense of Player's, a competing trademark with similar positioning. It was mentioned that the market share 100 of Export ' A ' dropped six percent from 1978 to 1997, yet the market share of Player's improved an impressive 14% during the same period. Conclusion This chapter has reviewed the marketing activities of RJR-Macdonald (and JTI-Macdonald) for their flagship cigarette brand, Export ' A ' . Internal industry documents reveal the segmentation strategies that have been utilized, in which Export ' A ' promotions have been largely designed over the past 30 years to appeal to male youth who aspire to be masculine, rugged, self-determined, independent, rebellious, and adventurous. When the tobacco firm attempted to invigorate the brand during the early and mid 1980s, promotions utilized the slogan, \" A Taste for Adventure\" and began to commonly show individual males involved in adventurous, alternative, action-oriented sports. Once conventional tobacco brand advertising was no longer permissible in Canada, the tagline \"Go Your Own Way\" was embraced, and sponsoring extreme sports events provided a valuable opportunity to maintain and enrich the brand identity of Export ' A ' . 101 Endnotes 1 Export cigarettes - first known as Brit ish Consol Export - were introduced in 1928. \" 2 See the criticisms provided by Cohen (1989), Pollay (1989), Dewhirst and Pollay (2001). 3 The creative for these ads seem to lack imagination, as the slogan, \"Satisfaction Country\" is highly similar to the very familiar and successful \"Marlboro Country\" campaigns. 4 A t the time, downhill ski racing was a high profile sport in Canada due to the impressive achievements of the 'Crazy Canucks', which included Olympic and World Cup downhill skiers K e n Read (who retired in 1983), Steve Podborski, Dave Irwin, J im Hunter, and Dave Murray. Wi th the racers consistently reaching speeds o f more than 65 mph on an icy course, many viewers perceived them as daredevils and adventurous. 5 E d Ricard, an executive in charge of Imperial Tobacco's strategy and product development, clarified the meaning of smooth as \"less irritation and less harshness\" during testimony at the 2002 Quebec Superior Court trial. 6 Export ' A ' was the title sponsor.rather.than Export ' A ' Inc. 'Shel l ' companies , were no longer required for sponsorship purposes once the Supreme Court o f Canada ruled the Tobacco Products Control Ac t unconstitutional in September 1995. Snow Jam, which is the largest action sports and music festival in Canada, features competitions in snowboarding, free-skiing, B M X freestyle, in-line demos, and skateboarding. The event is very reminiscent of the popular U.S.-based, X-Games. 102 8 Health Canada has issued 16 different health-warning messages, in which one of the warnings must be printed on each package of cigarettes sold in Canada. Interestingly, it was observed that all o f the 2001 and 2002 Export ' A ' extreme sports series ads depicting a health warning message, and that were located in the History of Advertising Archives, utilize the same Health Canada warning: \" C I G A R E T T E S L E A V E Y O U B R E A T H L E S S . Tobacco use causes crippling, often fatal lung diseases such as emphysema.\" In the context of extreme sports, this warning is l ikely to have multiple meanings to the reader of the ad. Breathless can be defined as gasping for breath, or alternatively, as taking one's breath away (i.e., breathless speed). 'Thr i l l ing ' is a synonym for 'breathtaking'. 9 Pollay and Lavack (1993), Cunningham (1996), Pollay (2000), and Dewhirst and Sparks (in press) have reviewed internal documents from RJR-Macdonald that were manifest in proceedings assessing the constitutionality o f the T P C A , and have also concluded that youth are a target of the firm's marketing activities. 1 0 Market research for British American Tobacco (1985, p.26) indicates, \"There is little question that a slimmer product, by its physical dimensions, clearly communicates style-fashion-distinctive-female imagery.\" 103 C H A P T E R F I V E Rothmans\u00E2\u0084\u00A2 Cigarette Brand Marketing: Intertextuality and the Decline of a Flagship Trademark Rothmans, Benson & Hedges Inc., formed in 1986 following the merger o f Benson & Hedges (Canada) Inc. and Rothmans of Pal l M a l l Canada Limited, is Canada's second largest tobacco manufacturer. Benson & Hedges Inc. had initially entered the Canadian market in 1906, becoming a Philip Morris subsidiary in 1954. Rothmans of Pal l M a l l , meanwhile, moved into Canada in 1957. The merged firm, which remains a Philip Morris and Rothmans International B . V . affiliate, enjoys an approximate 17% market share of Canada's cigarette industry (Cunningham, 1996).1 The key trademarks of the tobacco company include Rothmans, Craven ' A ' , Benson & Hedges, Dunhil l , Belvedere, Belmont, Viscount, and Canadian Classics. Internal industry documents are reviewed in this chapter to reveal the promotional tactics of Rothmans, Benson & Hedges (RBH) for the Rothmans trademark, which consists o f Rothmans K i n g Size, Rothmans Special M i l d , Rothmans Light, Rothmans Extra Light K i n g Size, Rothmans Extra Light Quality Blend, and Rothmans Ultra Light (see Appendix 13). The machine-measured tar delivery for each o f these brand family members is 15 mg, 12 mg, 11 mg, 8 mg, 8 mg and 6 mg, respectively. Overall, the Rothmans trademark has been promoted as a symbol of internationalism, premium quality, upward status, and tradition. 104 The market share held by R B H has been eroding notably, however. The combined market share of Benson & Hedges (Canada) Inc. and Rothmans of Pal l M a l l Canada Limited was 43% in 1975 (Cunningham, 1996). In 1988, the year that the T P C A was legislated, the market share of R B H was approximately 28% (Rothmans Inc., 1988). R B H ' s market share is now 17%. It is demonstrated here that the declining market share is largely explained by the firm's inability to capture the highly valued youth segment. Largely due to deficient promotional expenditures and a lack of packaging design updates, Rothmans was perceived by consumers as unpopular, ' o ld ' , and lacking contemporaneous. Rothmans has been inconsistently linked with particular imagery, thus the firm has failed to effectively exploit intertextual relations in its promotions. Intertextuality and the Nearly Infinite Promotional Chain The term intertextuality is wel l recognized among literary scholars, but only recently has its use become more common in marketing communications and consumer studies literature. Developed and popularized by French literary critic, theorist and textual analyst, Julia Kristeva, intertextuality is defined as the passage o f one (or several) sign system(s) into another. According to Kristeva, this signifying process: .. .involves an altering of the thetic [sic] position\u00E2\u0080\u0094the destruction of the old position and the formation of a new one. The new signifying system may be produced with the same signifying material... Or it may be borrowed from different signifying materials... If one grants that every signifying practice is a field of transpositions of various signifying systems (an inter-textuality), one then understands that its 'place' o f enunciation and its denoted 'object' are never 105 single, complete, and identical to themselves, but always plural, shattered, capable o f being tabulated. (1984, p.59, 60) In other words, there are manifold ways in which any one text is inseparably inter-involved with other texts. For Kristeva, each text represents a site in which a myriad of other texts are intersecting. Thus, every text is in fact an \"inter-text\" (Abrams, 1993; Wernick, 1991). To determine the meanings generated from advertising, the concept of intertextuality suggests that ads should not be studied in isolation. Current promotions have an inter-textual relationship with all past, present, and future ads of their respective campaigns. Brand name, logo or trademark elements, as well as the designs and colours found on packaging, are l ikely to reappear in nearly all o f the ads. Similarly, taglines often carry over from campaign to campaign. N o single ad can yield the reader a complete and full understanding of the promoted product's intended and actual symbolic meaning. Advertising is better understood i f the reader has exposure to an entire ad campaign, and best understood i f there is familiarity with a long history of ad campaigns for a particular brand or trademark. Meanings are produced through an interactive process between the texts (i.e., ads) and the reader, a process in which the reader is actively involved. The advertising text is considered 'unstable', indicating that multiple readings of the same text are possible. Indeed, all texts are polysemic to some extent, demonstrating that signs or texts have multiple, hierarchical meanings. Readers may respond differently to the same textual stimuli, resulting from their varying personal and cultural histories or simply how much time they spend reading the ad (Brown, Stevens, & Maclaran, 1999). Nevertheless, ads 106 are intentionally encoded with 'dominant' meanings, which are l ikely best understood by readers that exemplify the target market of the brand being promoted. It is important to recognize that consumer knowledge, beliefs, attitudes and opinions are regularly and extensively researched by marketers, and i f pre-testing efforts of a proposed ad campaign revealed that the 'dominant' meaning was not being 'realized', the ads would presumably be further refined or never put forward. To assist with creating an effective, relevant or understood message, the producers o f ads refer to market research findings, indicating that ad production does not occur within a closed system. In many ways, consumers are both the source and receiver o f a message. Another point to make about the meaning and symbolism of brands is that consumers have multiple sources of information. Wernick claims that, \"the promotional chain to which any product and 'its ' ad belongs is endless\" (1991, p.94). The meaning generated from an ad is affected and further understood by the readings o f preceding ads from both current and previous campaigns, as well as through additional sources such as word-of-mouth communication, direct experience, novels, and portrayals in television programming and fi lm (Fiske, 1989, 1990; Hirschman, Scott, & Wells, 1998). The credibility of particular ad appeals partially depends upon whether or not they are consistent with the consumer's other sources of meaning. 107 A n Overview of Rothmans\u00E2\u0084\u00A2 Promotional Strategies Rothmans King Size A historical overview reveals that the Rothmans trademark has been positioned as an expression of upward status, premium quality and British heritage, which has international or global recognition. Rothmans also had an initial aura o f innovation and contemporaneousness, as its 1957 introduction by Rothmans o f Pal l M a l l represented the first king-size filter cigarette in the Canadian marketplace (The Manitoba Educational Research Council , 1966). The king-size descriptor drew attention to a longer length cigarette. A t the time, most cigarettes in the marketplace were 72 mm in length, yet Rothmans K i n g Size was 84 mm (\"Plenty of smoke bil lows,\" 1962). 2 Value or 'more bang for your buck' was the implied product benefit, with consumers getting an 84 mm cigarette for the same price as a regular-length one. Substantial promotional spending supported the introduction of Rothmans K i n g Size. One year after Rothmans of Pall M a l l entered the Canadian market, advertising spending by the Canadian tobacco industry more than doubled (Thomas, 1958). According to a spokesperson from competing tobacco firm, Tabacofina, Rothmans o f Pal l M a l l ' s entry \"hit like a bomb... We think they must have spent about $10 mil l ion on their initial sales efforts\" (Baxter & Touche, 1958). Rothmans K i n g Size is considered the trademark's parent brand. During the early and mid 1960s, Rothmans was regarded by its account executive as: .. .a cigarette which achieved a very considerable success in the market, is sold throughout the entire socio-economic scale... but primarily in the 'upper-middle' 108 or 'lower-upper' income groups. It is described as 'an adult's cigarette' and 'a mature cigarette'; that is, it does not have a 'youthful' image and is generally smoked in higher proportions in the older age range, (cited in The Manitoba Educational Research Council , 1966, p. 140) During 1964, Rothmans was advertised as \" K i n g Size at regular price,\" \"the greatest name in cigarettes,\" \"best tobacco money can buy,\" and \"finer filter for better flavour. N o coupons.\" In 1965, ad copy claimed, \"Canada's Best K i n g Size Cigarette,\" \"Canada's Fastest Growing Cigarette,\" and \"first in Canada, first in the world.\" The brand essence of Rothmans was universality, dependability, and market leadership. The advertising appeals exuded self-confidence, and were largely informational, comparative, and competitive in style (The Manitoba Educational Research Council , 1966). In 1970, the slogan, \"The Greatest Name in Cigarettes\" was maintained, with the remainder o f the ad copy stating that, \" A l l over the world the swing is to Rothmans K i n g Size. Rothmans extra length, finer filter and the best tobacco money can buy, give you true K i n g Size flavour. Rothmans K i n g Size really satisfies. World 's largest selling -most wanted - K i n g Size Virginia .\" The Rothmans crest was prominently featured, as well as the right hand of a male (with the cuff of a white-collar shirt apparent) holding a package with cigarettes extended. For ads that circulated in Maclean's during 1974, 'Brit ish heritage' was seemingly communicated as part of the brand's personality (see Appendix 14). Rothmans of Pall M a l l was declared as \"Wor ld Famous For Quality Since 1890\" and \"Blenders of fine cigarettes through six reigns.\" The visual image o f a horse and coach in the streets o f London, England was supported by ad copy, which alleged, \"Every day, from Pal l M a l l through the West End of London, Rothmans still deliver their 109 world-famous cigarettes to select Clubs and Embassies by coach and footmen. This time-honoured custom is a tradition of the House of Rothmans of Pal l M a l l . \" In 1974 and 1975, the tagline for Rothmans K i n g Size advertising was \"The best tobacco money can buy,\" and a Rothmans package (open with cigarettes extended) was shown in the foreground with a set of tobacco leaves as a backdrop (see Appendix 15). The ad copy pointed out that, \"Rothmans cigarettes are sold today in over 160 countries, on more than 100 airlines and 150 shipping lines.\" Similar visuals were used by 1978, with the prevailing ad copy stating, \"Rothmans is Canada's favourite K i n g Size by far.\" Other written claims included, \"Good taste and outstanding quality,\" and \"When you light up a Rothmans K i n g Size, you treat yourself to the smooth, satisfying flavour of Canada's favourite king size.\" Rothmans advertising in 1980 featured the right hand of a male airline pilot (apparent with part o f the jacket sleeve shown) holding a package with cigarettes extended. The superseding written text stated, \"Wor ld Leader,\" followed by the assertion \"Good taste and outstanding quality... these are the reasons for Rothmans success throughout the world.\" During the early 1980s, ads portrayed a man at sea on a sizable sailboat, with ad copy claiming, \"Rothmans K i n g Size Really Satisfies.\" Consumers were encouraged to enjoy the great taste. When references are made to 'flavour', it is worthwhile noting that Rothmans K i n g Size is generally perceived by consumers as relatively strong tasting, making it nearly comparable to Player's Filter and Export Fu l l Flavour ( R B H , 1998). n o Rothmans Special M i l d During the late 1970s, the brand family simply consisted of Rothmans K i n g Size and Rothmans Special (Audet-Lapointe, 1991). Ads for Special M i l d encouraged consumers to \"Enjoy the great taste of Rothmans in a special mi ld cigarette,\" with the tagline, \"extra Special extra M i l d . \" The visual images were predominated by two packages, with cigarettes extended. B y the early 1980s, Special M i l d ads showed 'before and after' sequences, with the principal image portraying couples doing an activity together (such as sailing, cross-country skiing or riding a sled pulled by horses), with a secondary image showing two couples (four people) meeting afterwards to recap their day, smoke cigarettes, and enjoy coffee at a marina or lodge setting (see Appendix 16). In 1983, ads portrayed two men and a woman cross-country skiing (taking a break from the activity) or on a yacht at sea, with the ad copy, \"Great taste... and they're mi ld . \" In 1984, one ad featured two men and a woman that were going hot-air ballooning. The principal colours in the ads were red, blue, and white (including the clothing worn, equipment portrayed, etc.). The tagline, \"The best tobacco money can buy\" was still used, but sparingly. A 1985 ad showed one couple with tennis gear, meeting with another couple that were seated in a fine, antique-style convertible. The ad copy indicated that Special M i l d was also available in \"New 100's\" and \"Menthol 100's.\" Rothmans Extra Light, Light, and Ultra Light Relative to other Canadian cigarette trademarks, Rothmans has traditionally offered a short line of extensions. So-called 'light' products were not introduced until the I l l 1980s. According to an annual report of Rothmans Inc., \"ongoing research has helped to identify opportunities, and has led to a number of well received initiatives including our expanded offering of lighter cigarettes\" (1991, p.4). The Extra Light line extension was introduced in 1980, while Light and Ultra Light entered the marketplace in 1986 and 1990, respectively (Audet-Lapointe, 1991). Another annual report indicated that, \"Responding to consumer demand for a full-flavoured light product, Rothmans Light K i n g Size was introduced nationally in early A p r i l 1988\" (Rothmans Inc., 1988, p.4). Rothmans advertising presented Rothmans Extra Light (both regular and king-size) as \"new\" in 1981, claiming that it was \"the first extra light cigarette with real Rothmans flavour,\" and that Rothmans was a \"world leader in quality and good taste.\" Visuals depicted middle-aged people (two men and one woman) playing golf or tennis. During 1982, advertising for Extra Light placed the reader's attention toward the package and extended cigarettes, claiming, \"Enjoy Rothmans K i n g Size satisfaction in an Extra Light cigarette.\" In 1984, Rothmans Extra Light advertising showed four people (2 couples) enjoying food on a yacht, which was docked in a marina (see Appendix 17). The ad copy stated, \"Great taste... in an extra light cigarette.\" Ads for Rothmans K i n g Size Lights asserted, \" A t Last, Full-Flavour Lights!\" while other ads simply stated, \"The Full-Flavour Lights!\" A l so in 1984, Rothmans advertising introduced a 100 mm product line. The ad copy stated that, \"Rothmans goes to great lengths to give you great taste,\" and indicated that mentholated versions were available. In 1985, Rothmans 100's (Extra Light) were still being 'introduced', with the claim that they were \"longer, smoother, milder.\" The lozenge symbol, placed at the bottom of the ad, enclosed a representation of 112 the Rothmans o f Pal l M a l l coach, which was known from previous advertising to deliver Rothmans cigarettes to select Clubs and Embassies in London, England. Rothmans Plus Rothmans Plus was entered into the Canadian marketplace during the early autumn of 1996, and reported to shareholders as a new product initiative in 1997 (Rothmans Inc., 1997). This line extension was supported by a conventional advertising campaign, possible since the Supreme Court of Canada had ruled the T P C A unconstitutional in September 1995. Advertising for this brand followed voluntary guidelines set by the industry, which stipulated that the use of human figures was forbidden. A d copy gave specific late-hour times (e.g., 1:15 a.m.) and used the slogan, \" A Prime State of Plus\" (see Appendix 18). Predominant visual images included a pool table in a bar setting and the exterior of a diner-style restaurant. Although no people were depicted in the ads, the presence of human figures was still implied. The photos were framed such that readers would l ikely presume someone was present at the scene, yet simply cut from the picture (e.g., the front of a car was shown in the foreground of one ad, but cropped at the windshield so that people could not actually be seen). The photos, it could also be argued, took the perspective of someone that was present at the scene. Regardless, Rothmans Plus was on the market for merely nine months. The brand was pulled after failing to capture more than a one-tenth of one percent market share (Physicians for a Smoke-Free Canada, 1998a). 113 Rothmans Sponsorship Properties During the late 1980s, sponsorship became an increasingly important part o f the communications mix for tobacco manufacturers. Like Canada's other two major tobacco firms, internal documents for RBH reveal that they sought to sponsor events in which the activity was perceived to be anaerobic in nature: \"Sporting activities considered least logical/appropriate for tobacco sponsorships are those most linked with health/aerobics and strenuous physical activity (tennis and skating vs. auto/horse racing)\" (cited in Pollay, 2002, p.45). The major sponsorship properties of Rothmans were motor racing (including Formula One), thoroughbred horseracing, and film festivals held in Toronto, Vancouver and Montreal. RBH placed particular emphasis on the international aspects of these events, with an objective of having this image transferred to the Rothmans trademark.4 According to Rothmans Inc. (1993, p.5), Rothmans \"will always be known as an international, world-class cigarette.\" It was felt that \"the term International communicates quality and status. It suggests a strong product, possibly with a distinct aroma and flavour, which may or may not be desired. However, the superior quality and social status inherent in international travel appear generally appealing to almost all smokers\" (RBH, 1995c, p. 19). A competitor saw the creative positioning Of Rothmans sponsorship advertising as: \"To adults 25-49 seeking international quality entertainment, Rothmans brings cultural events of world-class stature\" (Armada, 1999, p.26011). Motor Racing During the late 1980s, prior to the implementation of the TPCA, Rothmans sponsored multiple motor-racing events and teams. In 1987, for example, Rothmans and 114 Honda were sponsorship partners for motorcycle racing properties in Canada. A t the time, the Rothmans Porsche Turbo Cup series, consisting of eight events, was also held at major racing tracks in Canada (Rothmans Inc., 1989). Conventional advertising for Rothmans, depicting packaging for K i n g Size, Special and Extra Light, leveraged this sponsorship initiative since the ads were dominated by visuals o f the Rothmans Porsche racing cars (see Appendix 19). Porsche was likely a highly desired sponsorship partner for Rothmans since the car is well recognized for being expensive and its high quality, upscale, class appeals. The typical demographics of a North American Porsche owner have been identified as male, forty-something years old, a university graduate, and earning more than US$200,000 per year. Market research also reveals that the taxonomy of Porsche buyers consists of five personality types: Top Guns (i.e., driven, ambitious types who seek power and control), Elitists (i.e., old-money blue bloods), Proud Patrons (i.e., the car is a trophy for their hard work), Bon Vivants (i.e., worldly jet setters and thrill seekers), and Fantasists (i.e., the car represents an escape) (Lamb, Hair, & McDanie l , 2000). A s a result of a decision by British-based Rothmans International, the Rothmans trademark was also the primary sponsor of Jacques Villeneuve during the auto-racer's first Formula One season. Notably, Rothmans owned the most 'real estate' on the car of the Rothmans Will iams Renault Formula One racing team. In 1997, Villeneuve won the Formula One world championship and was recognized as Canada's athlete o f the year for the second time, receiving the L o u Marsh trophy (\"Canada's best,\" 1997). Rothmans obviously benefited from the media coverage that Villeneuve generated during his successful season, with brand visibility even produced from third party advertisers such 115 as Castrol motor o i l (see Appendix 20). Winfield, another R B H brand, replaced Rothmans as the sponsor during the following season.5 Formula One auto racing is seen as an appealing sponsorship property due to its ability to generate brand visibility and reinforce brand imagery. According to Barrie G i l l , who was Chief Executive of a major sporting consultancy operation called Championship Sports Specialists Ltd. (CSS), \"After football, it 's [motor racing] the Number One multinational sport. It's got total global exposure, total global hospitality, total media coverage and 600 mil l ion people watching it on T V every fortnight... They're there to get visibility. They're there to sell cigarettes\" (cited in Taylor, 1984, p. 101, 103). A t the time of G i l l ' s statements, CSS had several tobacco accounts. More recently, it has been estimated that the television viewer-ship for each Formula One race is 300 mil l ion (Hawaleshka, 2001). The ethos of Formula One is associated with several images, many of which are considered highly desirable for linkages to various brands or trademarks. Market research prepared for Marlboro reveals that Formula One is \"highly targeted, heroic and international\" (Philip Morris , 1990, p.5). According to Richard West (1993), marketing director of Wil l iams Grand Prix Engineering, \"Formula 1 provides an ideal 'prestige image',\" and is \"recognised as high performance, glamorous, exciting and aspirational [sic]\" (p.303593873, 303593875). The sponsorship positioning o f Formula One is considered higher class or more upscale relative to other forms of auto racing (RJR-Macdonald, 1996). Other dimensions associated with Formula One sponsorship include power, popularity (worldwide recognition), innovation (technological expertise), contemporaneousness, self-confidence, masculinity, and independence. When drivers are 116 successful, as Villeneuve demonstrated while sponsored by Rothmans, Formula One can link a trademark with dimensions such as leadership, quality, and excellence. Rothmans Racing promotions commonly depicted Jacques Villeneuve, particularly as the date of the Canadian Grand Prix race in Montreal approached, proclaiming him to be \"Our Hero\" (see Appendix 21). The ad copy called attention to Villeneuve's Canadian background and his successful performance on the international Formula One circuit. The tagline, \" A New World To Conquer\" was utilized. The top portion of the ads featured various flags representing different countries, presumably representing the origins of the various drivers on the circuit. A website, www.connect.ca, was also listed. Thoroughbred Horseracing Another high profile Rothmans sponsorship property was a horseracing event held annually at the Woodbine Racetrack in Toronto. Known as the Rothmans Ltd. International Turf Classic, entries competed for C$1,000,000 in prize money during 1992, which was coined a \"r ich\" purse in Rothmans Inc.'s annual report. According to promotional material, this race represented the only Canadian event on the Grade 1 stakes calendar, and offered more prize money than any other thoroughbred horserace in Canada. The race was televised nationally on C B C . In 1992, Rothmans Ltd . expanded its sponsorship of thoroughbred racing into western Canada, with International Days taking place at Assiniboia Downs in Winnipeg and The Track in Vancouver (Rothmans Inc., 1992). 117 Promotions for the Rothmans Ltd. International claimed that it was \"a world class event.\" Jockeys were shown in action, with a 'world map' visual as the backdrop. Sweepstakes were also held for an opportunity to win 'world class' holidays to international horse races. To enter the contest, contestants were required to f i l l out entry forms with their name, address, telephone number, and age. In a competitive market analysis prepared for RJR-Macdonald, the Rothmans horse racing sponsorship (and related advertising support) was considered to communicate the following images: \"regal, traditional, British roots, older, established, gender neutral\" (Harrod & M i r l i n , 1996b, p.80150 2756). Annual reports of Rothmans Inc. repeatedly described the thoroughbred racing events as \"prestigious,\" \"world class,\" and featuring the \"world's finest thoroughbreds.\" Fi lm Festivals and Special Screenings Rothmans sponsored international film festivals held annually in Toronto, Montreal, and Vancouver. Ads promoting Rothmans' sponsorship o f the fi lm festivals used the tagline, \"Pictures that move the world\" (see Appendix 22). The dates of the fi lm festival (e.g., October 4-20), and the fact that it was international in scope, were the only 'informational' aspects of the promotions. The lozenge symbol was portrayed, with \"Rothmans International F i l m \" stated within. Reminiscent of the trademark's positioning from the mid 1960s to early 1980s, subsequent campaigns also included the ad copy, \"The best the world has to offer,\" accompanied by depictions of a globe and a superseding strip of film. The lozenge symbol included \"Rothmans Wor ld f i lm\" within. 118 To leverage the link with film, Rothmans also sponsored an After Dark Series, which gave viewers an opportunity to have advance sneak previews o f various movies including The Matrix (see Appendix 23). The movies were shown at midnight, with the screen time complemented by the slogan, \"The Wor ld Comes Out After Dark.\" Ballot submission was required for the chance to win preview tickets, with requisite information being the contestant's name, address, daytime phone number, and evening phone number. The screenings were restricted to those 18 years o f age and older, even though the legal smoking age in many provinces, including Ontario, was 19. This promotional initiative was in partnership with M I X 99.9, Now, and Warner Bros. Some o f the After Dark Series initiatives were also in partnership with various movie video outlets, where tickets to the advance screenings could be acquired. Interestingly, during the same year of this initiative, R B H observed that, \"Targeting Y A M [young adult market] consumers may be increasingly difficult as their tastes (e.g. music) and leisure pursuits become more fragmented and less fad driven. A n exception may be movie rentals . . . with video stores emerging as a unique point-of-contact opportunity\" (1999, p.4). Static in the Communication Process Traditionally, communication theory proposes a sender\u00E2\u0080\u0094message\u00E2\u0080\u0094receiver framework, and a communication system involves the encoding and decoding of messages. The 'intended' message consists of the aims, goals and objectives of the sender (i.e., R B H ) , as wel l as the purpose behind their ad agency's creation. The 'intention' of a message may include linking imagery or 'personalities' with a particular 119 brand/trademark, as wel l as providing information about a product or service. The message itself represents the character of the promotion as it is transmitted via a selected channel. Finally, the 'received' message epitomizes the receivers' interpretations, perceptions, and meaning construction. The meaning of the message is arbitrary rather than given since the meaning constructed at each level of the communicative system is not necessarily equivalent. The encoded message and the decoded message do not necessarily correspond, thus the sender and the receiver do not always arrive at common meanings (Ang, 1996). According to H a l l (1980): The codes of encoding and decoding may not be perfectly symmetrical. The degrees o f symmetry - that is, the degrees of 'understanding' and 'misunderstanding' in the communicative exchange - depend on the degrees of symmetry/asymmetry (relations of equivalence) established between the positions of the 'personifications', encoder-producer and decoder-receiver, (p.93) Ha l l later stated that, \"since there is no necessary correspondence between encoding and decoding, the former can attempt to 'pre-fer' but cannot prescribe or guarantee the latter\" (p.100). The 'intended' message surrounding the Rothmans trademark has been clearly demonstrated by reviewing internal corporate documents, annual reports, and describing the various advertising campaigns. Rothmans has been positioned over the past forty years as symbolizing premium quality (i.e., \"the best tobacco money can buy\"), international prominence (i.e., in 160 countries on six continents), upward status, and British heritage. Despite the 'intentions' o f R B H ' s message(s) about Rothmans, 120 consumers have developed an incongruous understanding o f the trademark. Project Ringo and Project Starr were qualitative research initiatives put forward by R B H during the late 1990s with an objective of identifying how to rejuvenate or restore vitality to the Rothmans trademark. The rapidly declining market share of Rothmans reflected the trademark being generally perceived by consumers as ' o ld ' , and lacking popularity, contemporaneous, and recognition. The Principal Image Perception of Rothmans is ' O l d ' It was discovered that, among Canadian youth, Rothmans is considered to be what their grandparents or a retired person would select to smoke. Market research found that \"the ultimate descriptor o f Rothmans was ' o l d ' \" ( R B H , 1998, p. 14). According to R B H (1998), \" R K S and the Rothmans trademark in general continue to suffer from an exceptionally old, outdated image\" (p.3). Unappealing and non-desirable imagery was the fundamental reason for the declining standing of Rothmans as opposed to product characteristics: \" R K S was known primarily for strength and secondarily for harshness... but product issues overall appeared less contentious than the brand's aging, conservative character\" ( R B H , 1998, p. 13).. Rothmans is largely perceived as 'un-cool' among the young adult market. According to R B H (1999), cool is best defined as \"an absence o f image negatives,\" while un-cool is considered \"unpopular, old-fashioned and tacky\" ( R B H , 1999, p. 18). Market research reveals that, \"Perceptions o f Rothmans among Y A M smokers were decidedly negative. While the brand was viewed negatively as somewhat harsh and extremely old 121 fashioned, the key issue is its apparently entrenched uncool image because o f its perceived utter lack o f popularity among younger smokers\" ( R B H , 1999, p.24). The perception that Rothmans smokers tend to be older is accurate. During the late 1980s, the overall market share of Rothmans was precisely five percent, but when figures were provided for different age segments, it was revealed that its market share was merely 0.8% among 18-24 year olds, 3.0% among 25-34 year olds, and 8.2% among those 35 and older. While the greatest proportion o f sales are generated from the 25-34 age segment for Player's, Export ' A ' and du Maurier (the best-selling trademarks in Canada), a whopping 77% of the Rothmans franchise fall in the 35 and older age category (RJR-Macdonald, 1990). Very few starter smokers select Rothmans, but for the few that do, many do not remain loyal to the trademark. \"Rothmans' retention level is definitely lower among younger smokers. This partly explains the declining market share,\" and among smokers under 25 years old, \"Rothmans is losing about 50% of its franchise and none is going to another Rothmans' [sic] brand\" (Audet-Lapointe, 1991, p.48, 49). During 2000, the entire Rothmans brand family possessed a market share of less than one percent among the highly coveted 19 to 24 year old age segment (\"Consumer shares,\" undated). Rothmans Lacks Popularity and Visibility The qualitative research findings revealed that Rothmans now has an exceptionally low presence among consumers, possessing a low 'share of mind ' . The trademark is presumed to be unpopular, and many focus group participants reported never seeing the brand. \"The brand lacks presence or profile. Most see it as unpopular 122 and doing little to change that... no package change and, among English Canadians, no promotion or sponsorship. Lack of profile appears critical to 19 to 24 year olds who seek the confidence and peer approval inherent in a more popular, visible brand\" ( R B H , 1998, p.3). When discussing the profile of Rothmans, \"a l l groups indicated it needs to advertise/promote/sponsor more. In fact, non-users were emphatic that a repositioned Rothmans Light would have to be more active/competitive in sponsorship and promotion... suggesting that 19 to 24 year olds want the sense of legitimacy and peer approval that comes with a higher profile brand\" ( R B H , 1998, p. 12). Many of the Rothmans sponsorship initiatives had been discontinued, with Winfield becoming the Formula One sponsor o f Jacques Villeneuve and Benson & Hedges taking over as a lead sponsor o f the Vancouver and Toronto international film festivals. According to a detailed report prepared for Imperial Tobacco, which outlined the evolution of switching among brands with varying machine-measured tar deliveries, Rothmans was best classified within the \"Unpopular O l d \" segment (Audet-Lapointe, 1991, p.24). This classification was based on information gathered from consumers during the late 1980s regarding image perceptions. Such perceptions were not apparent in 1977, thus there was roughly a ten-year window where the Rothmans image declined dramatically. This image change partially reflected the advertising resources being placed toward the trademark. When Rothmans o f Pal l M a l l (the initial manufacturer o f Rothmans) first entered the Canadian marketplace, the firm was recognized for its large promotional expenditures. Over time, this strategy was abandoned, and it was Imperial Tobacco that became known for its substantial advertising budgets, largely directed toward the Player's and du Maurier trademarks. 123 Few Modifications Were Made to the Packaging Design The 'premium quality' image of Rothmans suffered since R B H does not effectively demonstrate integrated marketing communications, nor effectively exploit inter-textual relations. Despite communicating quality dimensions in conventional advertising and sponsorship properties, few modifications were made to the package design. Rothmans was considered \"uniquely static in package design... virtually the only brand to have remained unchanged in a market noted for packaging updates\" ( R B H , 1998, p. 15). Thus, market research findings revealed that: Most consumers tend to view Rothmans' packaging as out of date and unappealing... A s a result, the R K S image is linked with a lower quality, very strong, harsh product and a profile distinctly lacking in fun, contemporaneity [sic] and personality in general. The legacy of the brand's international cache, heritage and premium quality status appears lost among younger smokers and any associations of Rothmans with status appear linked more with the past than the present. ( R B H , 1998, p . l ) The lack o f packaging updates was a significant oversight on the part o f R B H considering the integral role of cigarette packaging in communicating brand image, particularly in an increasingly regulated advertising environment (Wakefield, Morley, Horan, & Cummings, 2002). Rothmans packaging also lacks visual impact, giving the trademark less presence at point-of-purchase. The large body o f white space that characterizes the packaging design contributes a generic or clinical essence. Moreover, the white-dominated 124 packaging contradicted the taste characteristics that are associated with Rothmans since \"stronger cigarettes were more associated with brighter, darker, more visually impactful [sic] packaging... light/milder cigarettes were more linked with white\" ( R B H , 1999, p. 16). According to John Digianni, designer and vice-president at Gianninoto Associates Inc., \"red packs connote strong flavor, green packs connote coolness or menthol and white packs suggest that a cigaret [sic] is low-tar. White means sanitary and safe\" (cited inKoten , 1980, p.22). RJR-Macdonald (undated, p.80099 2050) observed that competing Canadian trademarks, most notably Player's and du Maurier, have \"maintained the same colour across their families, depicting strength by varying the amount of white space on their packaging,\" with lighter strength products possessing greater use o f white on the package. Interestingly, all o f the Rothmans brands have packaging that is predominated by white, including the parent brand, which is known to be a relatively strong tasting and high tar delivery cigarette. Focus group participants for R B H indicated that a more colourful design would be desirable for the Rothmans brand family. More specific aspects of design were also assessed. The lozenge, for example, was not wel l received. This shape was \"described as old, feminine and conservative... the lozenge emerged as second only to 'white' as a target among package critics\" ( R B H , 1998, p. 15). There were a few design elements that did generate some favourable feedback, however. The signature script of Rothmans was generally accepted by consumers: \"While occasionally criticized as too feminine, most appreciated its sense of style and class and viewed it as inextricably linked with Rothmans\" ( R B H , 1998, p. 15). The gold crest, depicting a crown and an ' R ' , was found to be \"an important signal of 125 quality and status, even to younger non-users,\" and \"most embrace the heritage and tradition communicated by the crest. It communicates positive product messages (higher quality) as well as a sense of respect\" ( R B H , 1998, p.5, 15). User Profile is Non-Aspiring Another factor of Rothmans' declining market share is that the perceived user profile is non-aspiring (particularly to youth), and does not resemble the lifestyles conveyed in advertising. \"Findings suggest that those more involved in the Rothmans franchise are more physically and intellectually sedentary\" ( R B H , 1998, p.8). Relative to smokers of other trademarks, Rothmans consumers are not regarded as being trendsetters or contemporary: Non-users appeared much more aware of and ' into' popular trends... current and former Rothmans smokers were, for example, much more l ikely (usually half or more) to cite a traditional mainstream beer (e.g. Canadian or Budweiser) as their beverage of choice in a social setting. B y comparison, these brands were favoured by only a small minority among younger non-users who were much more l ikely to cite beverages which are currently ' i n ' . . . vodka, rum, M i k e ' s Hard Lemonade, micro beers and European imports. ( R B H , 1998, p.8) In this study, non-users were defined as smokers of any non-Rothmans brand, although most were in fact du Maurier consumers. Rothmans smokers were found to be over-represented in the psychographic category Empty Nesters in a segmentation analysis study conducted for competitor, R J R -Macdonald. Empty nesters were defined as primarily women over 50 years old, with less 126 than average education and income, who held modern views about women in the workplace and maintaining financial independence, yet traditional views toward sex and men being the head of households (Harrod & M i r l i n , 1996). R B H (1998) confirmed that, \"The user was more often linked with females than males but was invariably described by both segments as very o ld\" (p. 14). Wi th the 'older' profile, Rothmans smokers were associated with a lifestyle that entailed frequenting bingo halls, bowling alleys, and donut shops. The 'o ld ' perception was even found to transcend to the taste qualities o f Rothmans, with some focus group participants claiming that the cigarette is perceived as dry and stale, as though it has past its due date. Finally, the Rothmans user profile is inconsistent with common perceptions about who smokes strong-tasting, higher tar-delivery cigarettes. The market share held by Rothmans is higher among females than males, yet \"the brands most often linked with excessive strength (Player's Filter, Rothmans, Dunhi l l and especially Export) were most often linked with older males\" ( R B H , 1999, p. 16). Similarly, after reviewing internal documents from all of the major U . S . firms, Pollay and Dewhirst (2002) found that women are more l ikely to manifest health concerns. Rothmans, with its relatively high tar delivery, is not positioned as a trademark for those demonstrating elevated health concerns. Rothmans Line Extensions Don't Carry Their Weight During the eariy 1990s, it was noted by competitor, Imperial Tobacco, that smokers o f Rothmans K i n g Size were particularly loyal to their brand, with little switching occurring toward the trademark's lower (machine-measured) tar delivery line 127 extensions (Audet-Lapointe, 1991). 6 One problem noted by R B H (1998) was that consumers were generally ignorant o f mild even appearing on the package of Rothmans Special M i l d , referring to it as merely, Rothmans Special. 7 When R B H assessed why smokers left Rothmans K i n g Size for another brand, it was also found that \"almost none of these former Rothmans smokers appeared aware of lighter line extensions such as Rothmans Light, Extra Light and Ultra Light. Those who cited a variety o f 'product' reasons (too strong, too dry) had not considered a lighter version o f Rothmans... usually because they were unaware of them\" (1998, p.9). Those switching brands often selected the parent versions of Player's, Export, Dunhil l or du Maurier, which are all regarded as relatively strong tasting, popular, and have fairly established or appealing image platforms. The lack of familiarity with Rothmans line extensions was important since \"downswitching to lighter tar levels is a key dynamic in the cigarette market\" (RJR-Macdonald, undated, p.80099 2050). R B H observed that many smokers \"appear to have gone through a lengthy, gradual and conscious move down the T & N ladder, often starting within a stronger, mainstream parent and making several stops/switches along the way. They do not know what determines T & N level (process, blend, filter) but that appears unimportant. They simply want a healthier product\" (cited in Pollay, 2002, p.26). According to market research prepared for Imperial Tobacco, the most common reason for smokers switching brands is they are looking for a \"milder/lighter\" product (Audet-Lapointe, 1991, p.465043182). Similarly, market research prepared for RJR-Macdonald found that, \" K e y switching motivators include wanting a 'lighter/milder' 128 cigarette, 'other cigarette too strong', 'wanted less tar and nicotine' and 'good taste'\" (Harrod & M i r l i n , 1996, p.80150 2738). The Rothmans parent brand was not able to maintain its current franchise when current users contemplated switching brands. R B H (1998) observed that, \"Compared to R K S and Rothmans in general, Rothmans Light was a virtual unknown. Indeed, few outside the franchise had ever seen the brand and many seemed surprised that lighter line extensions existed under the Rothmans trademark\" (p. 12). One explanation for the lack of familiarity was R B H ' s slow introduction of so-called 'light' products for Rothmans. Rothmans Light entered the Canadian marketplace a full ten years after Player's Light and Export ' A ' Light (Audet-Lapointe, 1991). A n d strangely, Rothmans Extra Light was introduced six years before Rothmans Light. R B H seemingly missed an important market opportunity. Market research for Imperial Tobacco points out, \"it is important to correctly monitor the consumers and to identify any new need. The first company to identify such opportunities and to answer smokers' needs, w i l l have a competitive advantage over the others just like when Player's Light was introduced, answering a need for a lighter cigarette in a major trademark\" (Audet-Lapointe, 1991, p.15). Discussion and Conclusions The concept of intertextuality provides a theoretical context for illustrating the image development of particular cigarette brands, helping to explain the continually declining market share of Rothmans, which was onCe the flagship trademark of Canada's second largest tobacco manufacturer.8 Conceptually, intertextuality insists that suggestive relationships exist between texts (Brown, Stevens, & Maclaran, 1999). 129 According to Fiske, \"the theory o f intertextuality proposes that any one text is necessarily read in relationship to others\" (1987, p.108), and that \"the meanings generated by any one text are determined partly by the meanings of other texts to which it appears similar\" (1990,p.l66). The interdependence of texts applies to both the encoding and decoding parts of the communications process, which helps explain the particular usefulness o f considering the process in inter-textual terms (O'Donohoe, 1997). The concept of integrated marketing communications is constructive and important, but applies to the encoding aspects of a message. Thus, to think of messages in inter-textual terms is all embracing and more comprehensive. While advertising has an objective o f reinforcing brand imagery, it is necessary to recognize that several additional sources o f meaning lend to the message being perceived as credible or reasonable. Although a Rothmans ad featuring Jacques Villeneuve in his Formula One racing car is likely designed to communicate elements o f youthfulness and contemporaneousness, these meanings may be quickly dispelled according to the different experiences and understandings a reader brings to the text (i.e., the reader may be aware o f Rothmans' 'static' packaging, only know retirement-aged people that consume the trademark, and recently attended a comedic performance in which the trademark was mocked). Nevertheless, the case history of Rothmans promotional activities underscores the importance of continuity, consistency, and repetition in advertising. N o single text in fact stands alone with respect to the meaning that is generated by readers. It is demonstrated in this chapter that inter-texts are not always complementary; rather it is possible for conflict or contradictions. For example, mentholated and extra length (i.e., 100s) 130 versions of Rothmans were promoted, which is seemingly contradictory with the strong taste qualities of the trademark. This contradiction l ikely lends confusion among consumers about whether the trademark is more aptly described as 'feminine' or 'masculine'. Strong-tasting, highly flavourful brands, such as Player's and Export ' A ' , which are offered by Canada's other two principal tobacco manufacturers, do not include mentholated or 100 mm line extensions, thus contributing to their 'masculine' reputations. Rothmans clearly lacks visibility, which speaks to the importance o f substantial promotional spending toward maintaining the social acceptability of both smoking and a particular brand. Wi th its present market share, Rothmans would be directly observed less often than other trademarks. When observed, Rothmans is l ikely reinforced as a cigarette that older people select. Since many younger smokers perceive Rothmans to be unpopular and lacking a mainstream image, they sense their peers would be disapproving of the trademark. Among the young adult market, Rothmans is looked upon as 'un-cool' . R B H regards the \"young adult market\" potential of Rothmans as quite limited. The two main typologies o f cigarette consumers used by cigarette firms are \"new users\" (young starters) and \"latent quitters\" (the reassurance of concerned smokers) (Pollay, 2000). Having brands appeal to either of these segments has the important objective o f maintaining or increasing market volume. Yet, the Rothmans trademark has been ineffectively positioned, such that it does not appeal to smokers who are at the stage of initiation (i.e., typically adolescents) nor those who are more established smokers who are contemplating quitting. Rothmans is strongly (and accurately) perceived as an older person's cigarette. Moreover, the machine-measured tar delivery o f the parent Rothmans 131 brand is relatively high, making it unappealing to smokers that are regarded as \"concerned\" or \"health conscious.\" For those smoking the parent brand and contemplating a switch toward a lower yield product, they rarely select a line extension from within the Rothmans brand family. Rather, another trademark is selected. Ads claimed that Rothmans was \"Canada's Best K i n g Size Cigarette\" and \"Canada's Fastest Growing Cigarette\" during the mid 1960s. Today, such claims are not even remotely applicable. 132 Endnotes 1 Phil ip Morris has a 40% ownership stake in Rothmans, Benson & Hedges, while Rothmans Inc. - a Canadian holding company that is owned by Netherlands-based Rothmans International B . V . - has a 60% stake (Cunningham, 1996). 2 The 84 mm length reflected the fact that higher federal excise taxes were designated toward cigarettes with a length of 85 mm or more. These taxes were later rescinded, paving the way for the introduction of even longer length cigarettes, such as Benson & Hedges 100s. King-size cigarettes accounted for a 10% market share in 1960, but by 1966 this product category made up roughly 40% of the cigarette market (\"Plenty of smoke bil lows,\" 1962; \"Now Craven A , \" 1966). 3 Benson & Hedges (Canada) Inc. purchased Tabacofma in 1962 (\"Why small cigarette firms,\" 1962). 4 According to R B H (1998), the perceptions toward the Rothmans trademark generally appear to reflect viewpoints about the Rothmans parent. Thus, sponsorship ads featuring a trademark (i.e., Rothmans Ltd. , Rothmans) made most consumers/readers think o f Rothmans K i n g Size. 5 Prior to Winfield's Formula One sponsorship of Jacques Villeneuve, the trademark was not distributed in Canada. More than nine mil l ion Winfield cigarettes were shipped to the province of Quebec, however, during the three months leading up to the 1998 Formula One race held in Montreal (Physicians for a Smoke-Free Canada, 1998b). Villeneuve is now racing for the British American Racing ( B A R ) Honda 133 Formula One team, and British American Tobacco (the parent company of Imperial Tobacco) is the primary sponsor. 6 Line extensions are defined as \"new products introduced under existing brand names\" (Wells, Burnett, & Moriarty, 1989, p.74), where attempts are made to leverage the already established reputation of the parent brand. 7 According to R B H , \"This and other studies have consistently indicated that 'light' is viewed as a signal of strength while 'm i ld ' is viewed as a signal of smoothness\" (1998, p.10). 8 h i terms of sales volume, Craven ' A ' was identified as R B H ' s top trademark in 1988, with Rothmans and Benson & Hedges ranking second and third, respectively (Rothmans Inc., 1988). The 1993 annual report for Rothmans Inc., however, referred to Rothmans as \"our largest trade-mark\" (p.5). In its 2002 annual report, Rothmans Inc. noted that the market share of the Rothmans trademark continues to decline, and that \" R B H ' s approach to building market share is to focus on brands that have demonstrated increased consumer acceptance within the marketplace\" (p. 19). Benson & Hedges is now regarded as their most promising trademark. Part II: Content Analysis of Tobacco Promotions 134 C H A P T E R SIX Content Analyses of Cigarette Advertising: A Critical Review of the Literature The second part of this thesis utilizes content analysis as a research methodology. Chapter 6 reviews the existing content analysis literature that assesses the character of cigarette advertising in the print media. Chapter 7 offers an original content analysis study in which the content of Canadian conventional cigarette print ads from the pre-T P C A era (1973-1988) are compared with the content of Canadian tobacco sponsorship ads from the pos t -TPCA era (1989-2002). More specifically, the trademark strategies of Player's, Export ' A ' , and Rothmans are examined, whereby it is explored whether the results o f the content analysis study correspond with the findings from the internal corporate documents. Content analysis is a well-established method of studying advertising messages and quantifying textual elements (Leiss, Kl ine , & Jhally, 1997; Dyer, 1982). Content analysis is defined as: .. .a systematic technique for analyzing message content and message handling - it is a tool for observing and analyzing the overt communication behavior of selected communicators. (Budd, Thorp, & Donohew, 1967, p.2) .. .any technique for making inferences by objectively and systematically identifying specified characteristics of messages. (Holsti, 1969, p.14) .. .a research technique for the objective, systematic, and quantitative description of the manifest content of communication. (Berelson, 1971, p.18) 135 A s indicated by the definitions listed above, objective, systematic, and quantitative are distinguishing characteristics of content analysis. To ensure \"objectivity,\" coders undergo training and are presented with precise and detailed operational definitions, categories, rules, and procedures for analyzing communication content. Content analyses ideally involve multiple coders (including some separate from the authors) that assess communication content independently. Thus, judgements should not entail consultation with the researchers or other coders. Applying the same operational definitions, categories, rules and procedures to the same data set, the various coders should secure highly replicable and reproducible results and arrive at similar conclusions. Inter-coder reliability is calculated to indicate the level of agreement among coders (Holsti, 1969; Berelson, 1971; Kassarjian, 1977; Kolbe & Burnett, 1991; Leiss, Kl ine , & Jhally, 1997). \"Systematization\" implies that developed definitions, categories, rules, and procedures are to be applied to the data set consistently. To ensure a proper basis for comparison, descriptive categories should not be added after the coding process has commenced and definitions are not to be modified. The coders should proceed through the data set in a different and ideally counter-balanced sequence to help minimize any potential learning and maturation effects. Biases may be further reduced by having the data analyzed by coders who are unaware of the hypotheses established by the researchers. Ideally, but not always possible, the data set is pre-selected through recognized randomizing procedures and represents an unbiased, representative sample of defined communication content. In other words, the researchers should not actively search for ads ideally supportive of their arguments. Content analyses are systematic 136 since scientific problems or hypotheses are examined (Holsti, 1969; Kassarjian, 1977; Kolbe & Burnett 1991; Leiss, Kl ine , & Jhally, 1997). The description o f communications content resulting from content analysis should have general applicability and theoretical relevance (Holsti, 1969; Berelson, 1971; Kassarjian, 1977). \"Quantification\" is another defining characteristic of content analysis. Content analyses are utilized to measure the extent that analytic categories appear in communication content (i.e., relative emphases and omissions are determined) and the data generated is to be amenable to statistical methods. While researchers often equate quantification with strict frequency counts and the assignment of numerical values, quantitative words such as \"more,\" \"always,\" \"increases,\" and \"often\" may also be used (Berelson, 1971; Kassarjian, 1977). Having defined content analysis and its accepted methodological standards, a thorough and international review of English-language content analyses (that are specific to cigarette advertising) published in academic journals w i l l now be provided. The studies reviewed for this chapter (see Appendix 24) include content analyses of cigarette magazine ads and cigarette billboard ads, with the studies pertaining to American cigarette magazine advertising content being categorized into two sections: (1) assessing the impact of health/smoking controversy events and the broadcast advertising ban; and (2) assessing the impact o f Federal Trade Commission (FTC) regulation. Studies specific to non-American cigarette advertising content have been placed in a separate category. Within each section, the studies under review are placed in chronological sequence (according to date of publication). A general overview of the reviewed studies is presented, with emphasis placed on research methodology and key findings. Common 137 study limitations are identified and several directions for future research are offered. This literature review w i l l hopefully inform current and future content analysts about important methodological decisions and enable research on cigarette advertising to be further improved in the areas of objectivity, systematization, and quantification. Literature Review U.S. Magazine Advertising Assessing the Impact of Health'Smoking Controversy Events and the Broadcast Advertising. Ban The first known study to use content analysis specific to cigarette magazine advertising was done by Weinberger, Campbell, and DuGrenier (1981). They utilized content analysis, as well as media spending and market share data, to examine possible shifts in U . S . tobacco industry advertising tactics resulting from various regulatory measures. Several dimensions were accounted for, including advertising volume, physical aspects, major appeals, models employed, product attributes, and brand extensions. Using Newsweek, Sports Illustrated, and Ladies Home Journal as sources for cigarette ads, three separate sampling years were selected (1957, 1967, 1977) to determine the possible impact of various influential events and surrounding circumstances. First, beginning in the early 1950s, scientific and popular articles more commonly associated lung cancer with smoking, leading smokers to become increasingly \"health concerned.\" Second, in 1964, the first Surgeon General's report specific to smoking was released. Third, cigarette advertising was removed from the broadcast 138 media in 1971. The authors were particularly interested in changes that might be apparent in magazine advertising as a result of the broadcast ban. Weinberger, Campbell, and DuGrenier's three-year sample included a total o f 251 cigarette ads. The authors observed that, taken as a percentage o f the total advertising spending by the six major U .S . tobacco manufacturers, the amount of cigarette advertising found in magazines dramatically increased from 1967 to 1977. Tobacco manufacturers also responded to the broadcast media ban by placing more resources toward print media advertising, evident by more frequent use of special positioning, colour, and full-page or double-page ads. Ads were typically placed on right-side pages, and during the observed period (1957 to 1977) increasingly located on the back covers of magazines. They noted, however, that some o f the observed changes, such as the increased use o f colour, might be reflective of innovations being utilized by magazine advertisers in general. Weinberger, Campbell, and DuGrenier observed several changes in the use of product attributes and appeals during the 1957 to 1977 period. Claims related to low tar, implied health benefits, and cigarette length became more frequent in magazine advertising, whereas social appeals were uncommon in the 1967 sample and the use of taste claims diminished from 1967 to 1977. The decline among taste claims was not statistically significant, however. Meanwhile, claims or appeals became increasingly directed toward a female audience, evident by the emergence o f \"slimness\" claims, greater use of female models, and the heavy upswing of low tar claims found in Ladies Home Journal. Although females were more frequently used as models, overall use of human models declined with emphasis being placed on the cigarette package only. In the 139 1977 sample, brand comparisons were much more common and most o f these comparative ads were for low tar brands. Finally, brand extensions became more consistently used over the observed twenty-year period. There are study limitations that should be noted, however. For content analysis purposes, all magazine issues from the first three months of each sampling year were used, thus indicating a seasonal limitation for the ads examined. During the first few months of the year, the frequency of cigarette advertising tends to be higher, and ads may more often account for the common New Year 's resolution to quit smoking or display activities that are specific to the winter season (Basil, Basi l , & Schooler, 2000). While Weinberger, Campbell, and DuGrenier found that cigarette advertising was increasingly directed toward women, this finding was largely based on their sample o f ads from Ladies Home Journal. A greater representation of women's magazines would be useful for confirming this finding. In addition, intra- and inter-coder reliability was not reported for this study. Although the authors outlined all of the measured dimensions within a table, definitions for the various dimensions and themes were not provided. Thus, it was unclear what did or did not constitute a major appeal such as health. Advertising claims such as filter, mildness, low tar, and menthol were accounted for under the coding category \"product attribute,\" but it remained ambiguous whether such claims were also classifiable under the \"major appeal\" category pertaining to health. Discussion later in this chapter w i l l make it apparent that defining health-related claims is a matter o f great debate. Warner (1985) also used content analysis to assess how the cigarette industry had responded to various significant events that publicized the association between smoking 140 and various health consequences. He sought to identify whether tobacco advertisers' approaches had changed over time with respect to addressing the health concerns of consumers. Several significant periods related to the publicity of health concerns were identified within the study. First, the American Cancer Society released a major study linking smoking with lung cancer in 1951. Second, popular press articles, most notably in Reader's Digest, emerged in 1953 discussing the relationship o f smoking with cancer. The first Surgeon General's report on smoking and health was published in 1964 and Fairness Doctrine anti-smoking messages were prominently shown on television from 1968 to 1970. In 1978, Health and Education Welfare (HEW) secretary, Joseph Califano, announced a new anti-smoking campaign. Moreover, the Surgeon General released a tVi report in 1979 that was given special attention since it marked the 15 anniversary o f the 1964 report. Warner used selected issues of Time magazine from 1929-to 1984 to generate a total sample of 716 cigarette ads. According to Warner: .. .we examined all cigarette ads, recording for each the brand name, whether or not the cigarette was filtered, the cigarette's length, tar category (regular, low, ultra-low), whether or not cigarettes were extracted from packs, whether or not lit, whether or not smoke was present, how the cigarettes were held (hands, mouths, suspended in air), presence (and sex and number) or absence o f models, nonhealth [sic] themes (e.g., modern design, humor, rugged individualism, sophistication, romance, sex appeal, emancipation, femininity, nature, athletics, entertainment, expertise, fame), health theme (extent, from nonexistent to predominant message), and mix o f written and pictorial material. (1985, p . l 16, 117) 141 Data was provided for the number of cigarette ads found per issue, as well as the percentage of ads containing filter-tipped cigarettes, low tar cigarettes, a predominant or exclusively health-related theme, mostly or all words, visible smoke, no models, and modern design. Like Weinberger, Campbell and DuGrenier (1981), Warner discovered that the number of cigarette ads found in each magazine issue dramatically increased during the 1970s, largely the result of the broadcast ban. Warner also found that during periods of elevated health concerns, cigarette manufacturers responded by increasingly utilizing ads that included health themes and \"technological fixes.\" Filter-tipped cigarettes were introduced in the 1951 sample and became prominent. By 1968 and thereafter, all sampled ads were for filtered cigarettes. Meanwhile, low tar cigarettes initially became present in the 1967 sample and became more and more commonplace over time. Low tar cigarette advertising was often positively correlated with the use of health-related claims, but the 1984 sample represented a notable exception. Interestingly, visible smoke became less frequently observed in ads (with no observations after 1975), even in cases in which a cigarette was clearly lit. With the possible exception of the 1964 Surgeon General's report, cigarette manufacturers engaged in \"responsive\" advertising for each smoking-and-health controversy event by more frequently using health-related claims (most notably during the 1953 sample). \"Responsive\" ads were less likely to feature models and employed more verbal content. Ads conveying a health message relied more heavily on words as opposed to visual images. It also held true that ads placing greater emphasis on pictorial aspects typically had de-emphasized health themes. Warner concluded that through advertising \"cigarette companies 'talk' with consumers about the 142 health issue, but only when 'necessary' (i.e., to counter visible adverse publicity)\" (1985, p.124, 125). Once again, a formal listing o f coding definitions was omitted for the study and although it was acknowledged that the use of health-related claims was measured according to a rating scale ranging from 1 (indicating no health content) to 5 (indicating message virtually all health-oriented), the necessary criteria for each rating remained unclear. In addition, intra- and inter-coder reliability was not reported, nor was the number of people coding or the training procedures. Coding for the presence of lit cigarettes and visible smoke in advertising content was a unique contribution. It would be interesting to know whether or not this dimension was difficult to code. Rogers and Gopal (1987) examined cigarette ads from over a fifty-year period, 1936 to 1986, using Time and Life as sources for the data set. Ads were taken at five-year intervals for the first and sixth year of each decade being assessed (i.e., 1936,1941,1946, and so on), thus 11 points in time were studied. To reduce potential seasonality effects, ads were taken from A p r i l , September, and December magazine issues. A total o f 216 ads were analyzed. The objective of the study was to determine whether the content of cigarette advertising was influenced by factors such as World War n, the development of filtered and lower machine-measured tar delivery products, the broadcast advertising ban, and the requirement to include the Surgeon General's health warning. They considered whether the following eight ad appeals were present: taste/flavour, mildness or tar/nicotine/acid content, filter, good for health, experience of manufacturer, bandwagon/endorsement, gift/present, and other. Other measured dimensions included layout type, ad size, colour use, visual style (i.e., illustration or photograph), human 143 presence, target audience (i.e., men, women, both, can't say), and smoking depictions. The key findings were that 1971 marked a dramatic increase in the frequency of cigarette advertising in magazines, and a smaller proportion of cigarette advertising was found in the December magazine issues. The peak years for filtered cigarette ads were 1956 and 1961, while ads calling attention to mildness and low tar yields rose sharply in 1971. Appeals relating to taste and flavour were quite persistent. A number of study limitations are apparent, however. Both intra- and inter-coder reliability were not calculated, thus it is difficult to ascertain whether the results are highly replicable or reproducible. The number o f coders used for the study is unknown and the training procedures undergone were not stated. Furthermore, no definitions were provided for the various data-coding instrument items. \"Good for health\" was classified as a separate category from \"filter\" and \"tar/nicotine content\" even though all three categories imply elements of healthfulness. The categories were not mutually exclusive, but without coding definitions being put forward, it is indeterminable how explicit health claims had to be for placement in the \"good for health\" category. The list o f ad appeals does not appear to be comprehensive considering that the \"other\" category was applicable to roughly 60% of the ads in 1966. The authors concede that \"other\" refers to primarily lifestyle platforms, but further specification is not given. The failure to monitor 'lifestyle' dimensions seems to be a major omission. Overall, the data-coding instrument was simplistic (e.g., ads were measured for human presence, yet the number of people and gender of models in each ad were not determined). The \"smoking depictions\" item consisted of three categories (actually smoking, holding, and not smoking), thus cigarettes placed in an ashtray were not 144 accounted for, nor were open cigarette packages. The \"ad size\" item was futile since Rogers and Gopal admitted that, \"a content analysis was conducted using full page advertisements\" (1987, p.258). Airman, Slater, Albright, and Maccoby (1987) used content analysis to determine whether tobacco industry tactics differed among magazines with a youth or women-oriented readership. The sample was comprised o f ads from eight magazines with varied readership demographics, namely Rol l ing Stone, Cycle World, Mademoiselle, Ladies Home Journal, Time, Popular Science, T V Guide, and Ebony. One issue o f each magazine was randomly selected for the years 1960 through 1985, with safeguards undertaken to avoid generating a sample that had limited seasonal variation. The sample analyses were largely limited to image-based ads with settings or models present (ads for low tar and low nicotine cigarettes were exempted from this requirement and represented 22% o f the ads in the data set). In total, 778 ads were coded. According to Altman, Slater, Albright and Maccoby, four coders underwent training and overall inter-coder reliability was 95%.' Inter-coder reliability calculations were based on a 5% sample of coded ads, and intra-coder reliability was not reported. Data was collected for seven variables that fell under three broad categories: act of smoking, presence of low tar and low nicotine theme, and vitality of smoking. More specifically, coding was done for the presence of visible smoke, cigarettes being held or consumed, cigarettes within the ad photo, explicit low tar or low nicotine appeals, and appeals pertaining to adventure/risk, recreation and romance/eroticism. Coding for additional variables (i.e., number and type of people featured, setting characteristics, 145 props used, activity featured) was conducted, but the data was not examined systematically. Formal definitions were not provided for the various coding variables. It was found that beginning in the late 1960s, visible smoke became more and more infrequent among ads from all eight sampled magazines. In fact, the 1984 and 1985 samples contained no ads with visible cigarette smoke. Although ads with portrayals of people smoking or holding a cigarette also declined over time, it remained a feature for 68% of the ads in the 1985 sample. Up until 1979, low tar and low nicotine cigarettes were increasingly emphasized. Since then, some ads began to emphasize additional brand extensions, such as those that were differing lengths or mentholated. Relative to youth magazines, a greater percentage of women's magazine ads were with a low tar or low nicotine theme. Health and vitality were increasingly associated with cigarette smoking. While risk/adventure, recreation, and erotic images were more frequently depicted over time in both youth and women-oriented magazines, recreation and adventure/risk-taking images were particularly emphasized in youth readership magazines and erotic images were particularly emphasized in women's magazines. The findings supported all three hypotheses\u00E2\u0080\u0094during the observed 1960 to 1985 period, the act of smoking was featured less frequently, low tar and low nicotine cigarettes were increasingly emphasized, and vitality became a theme more prominent in advertising content. The authors alleged that through advertising the tobacco industry differentially targets women and youth, and portray misleading images to reassure \"health concerned\" smokers. 2 K i n g , Reid, Moon , and Ringold (1991) analyzed visual aspects of cigarette magazine advertising during the period 1954-1986. The magazines sampled included 146 Time, Ladies Home Journal Vogue, Sports Illustrated, Popular Mechanics, Redbook, Esquire, and Playboy. The selected magazines represented diverse readerships with regard to age, gender, and hobbies/interests. Like the study by Airman, Slater, Albright, and Maccoby (1987), one issue of each magazine was randomly selected for each year examined, with safeguards being undertaken to avoid generating a sample that had seasonal biases. The resulting sample for analysis included 1,100 cigarette ads. The authors divided the 33-year period being studied into three distinct \"event eras\": the pre-broadcast ban era (1954-1970), the post-broadcast ban era (1971-1983), and the anti-smoking ideology era (1984-1986). It was anticipated that important changes in visual content would be observed as tobacco manufacturers shifted their advertising resources from broadcast to print media. The coding instrument accounted for ad size, ad position, pictorial representations (i.e., use of colour, type of illustration, and visual/verbal balance), presence and gender of human models, and activities portrayed (i.e., adventure, erotic/romantic, individualistic/solitary, recreation, sociability, and work). The methodology employed was thorough\u00E2\u0080\u0094procedures were undertaken to minimize order bias for coding, all variables had an inter-coder reliability score of at least .80, and the overall inter-coder reliability was disclosed as .89. K i n g , Reid, Moon , and Ringold found that the number o f cigarette magazine ads increased following the broadcast ban and over the three event eras the ads became larger in size and were more often printed in colour. Meanwhile, artwork was used less frequently and photos became the preferred ad illustration class. Pictures, as opposed to words, became the predominant means of communicating to consumers. The presence of human models did not steadily increase over the three event eras (models were present 147 for 82% of pre-broadcast ban ads, 65 % o f post-broadcast ban ads, and 84% o f anti-smoking ideology ads). Male models were more frequently depicted than female models during each event era, with male models typically appearing in ads placed in men's magazines and female models usually appearing in ads placed in women's magazines. Models were increasingly engaged in activities over the three event eras (reaching nearly 83%o during the \"anti-smoking ideology\" era), although the typical type o f activity portrayed differed among eras. During the three event eras, portrayals o f adventure and work steadily increased, while portrayals of eroticism or romance declined. Overall, individualistic/solitary and recreation themes were depicted most frequently. The authors accounted for magazines with diverse audience orientations and concluded that youth-oriented magazine issues did not contain a greater number of cigarette ads or \"targeted\" attention with respect to the common themes portrayed. Some study limitations are worthwhile noting, however. First, the establishment o f three event eras (1954-1970, 1971-1983, 1984-1986) resulted in analyzed ads being categorized in a misrepresentative manner. Among the total sample o f 1,100 ads, 842 were represented within the one \"post-broadcast ban\" category. Meanwhile, the 1984-1986 grouping, referred to as the anti-smoking ideology era, contained 163 ads. A n additional study limitation was that ads generated for the sample were not studied entirely as originals. Rather, some ads were accessed through black and white microfilm and thus not examined according to how initially seen by readers. This procedure is problematic because the copy and imagery featured in these ads would be less detailed as a result of photocopying. It is difficult to ascertain the quality of the ads that were viewed from black and white microfilm, but some may have been excluded for analysis purposes 148 considering that the tables accounting for the visual/verbal balance o f ads reported base samples of 986 rather than 1,100. In other cases, data was not generated for the entire sample due to disagreements among those coding. Such data, however, is typically not dismissed because the senior investigator w i l l attempt to minimize coding disagreements by having coders resolve disputes. In the event that coders remain unable to reach agreement, the senior investigator may review the coding in question and attempt to act as a tiebreaker. The researchers also could have developed a more comprehensive list o f activity portrayals. A category was not developed to account for visuals pertaining to success, high status, and sophistication. It also remained unclear which activities the authors regarded as health-related. The activity category recreation, for example, may include visuals with diverse meanings intended (e.g., someone pictured reading may express comfort and relaxation, while a portrayal of beach volleyball may express physical activity and sociability). Finally, the variable ad position was defined according to three categories (inside page, back cover/third cover, and inside front cover). Alternatively, it would have been interesting to code for use o f left-sided and right-sided pages and page number location (i.e., front-half/back-half o f magazine). Pollay (1991) accounted for both the verbal and visual content of cigarette ads, measuring the frequency of claims made for 12 different attributes (i.e., wel l made, good deal, enjoy, female, male, bold/lively, glamour/luxury, health/safety, relax, official, popular, and pure scene). A n additional variable called healthiness was established to integrate the health/safety, lively/bold, and pure scene claims. Definitions were provided for each o f the coding categories. The categories monitored the product-cost-benefit 149 reasons for purchase, sex role modeling, lifestyle portrayals, the consequences o f consumption, the nature of social support for smoking, and associated physical environments. The sample was generated from 108 available back copies o f Life (from 1938 to 1983) and 26 back copies of Look (from 1962 to 1971), yielding an overall sample of 567 ads. While the total sample included the representation of 57 different brands, 14 of these brands (Camel, Chesterfield, Kent, K o o l , L & M , Lucky Strike, Marlboro, Old Gold, Pal l M a l l , Philip Morris, Salem, Herbert Tareyton, Viceroy, and Winston) accounted for 75% of the ads studied. Seventeen people were recruited for coding purposes (two additional people specifically coded for health/safety claims) and all intra- and inter-coder reliability calculations indicated a minimum agreement of 80%, with the majority over 90%. Pollay found that throughout the 46-year period studied, healthiness was a manifest theme in the vast majority of American cigarette magazine ads. During the late 1950s and early 1960s, however, the typical means o f communicating healthiness shifted from verbal to visual. The frequency of health/safety claims notably increased during the 1973-1983 period. Overall, wel l made and enjoyment claims were also featured in the majority of ads. Moreover, data was generated for the joint appearance of two attributes, with statistical tests being run to indicate which pairs were highly improbable by chance alone. It was revealed that official and health/safety assertions appeared hand-in-hand (\"well made\" was also commonly associated with both of these assertions). While female-oriented ads were more likely to feature claims pertaining to enjoyment, glamour/luxury and popularity, male-oriented ads were commonly associated with bold/lively lifestyle portrayals. Lively/bold and pure scene assertions were often 150 associated, suggesting that such ads often displayed people engaged in l ively and robust activities in outdoor, nature settings. Pollay provided a thorough account o f the study methodology, including the measures, the sample, judges for coding, and the procedures undertaken. Ads were numbered and divided into subsets to enable the judges to code in a varied order sequence. The randomization of coding sequence and starting-points protected the data from being impacted by learning and fatigue effects experienced by the judges during coding. There remain, however, some study limitations that should be noted. Available back copies o f Life and Look were used to generate the ad sample, and while data was provided about the number o f issues used among various eras, it remained unclear whether seasonal variations were accounted for. In other words, it was ambiguous whether or not the available magazines utilized for the study were evenly represented according to the month o f the issue. Finally, the number of ads generated for the 1973-1983 era does not appear representative of the entire sample. Despite the continued growth o f cigarette advertising, 112 ads were studied for the 1970-1972 era, while 51 ads were studied for the 1973-1983 era (i.e., the data is more heavily weighted in the pre-1972 era). Assessing the Impact of FTC Regulation For each of the U . S . studies reviewed thus far, none of the authors included within their rationale of selected sampling years how F T C regulations might impact the cigarette 151 advertising content being studied. 3 This was a factor, however, accounted for by Ringold (1987) and Ringold and Calfee (1989). Ringold (1987) examined ads for six cigarette brands (Camel, Chesterfield, Old Gold, Viceroy, Lucky Strike, and Kent) using magazine issues dated 1926 to 1985. Ringold sought to determine the extent to which cigarette advertising was informative, focusing on the use of health-related claims and the type typically employed. Information categories were established pertaining to taste, cigarette construction, health, pleasure, exhortation to purchase/slogans, tar and nicotine figures, reduced tar and nicotine, the Surgeon General's warning, hedonic satisfaction, price/availability, coupons/contests, and celebrity/athlete endorsements. Ringold's content analysis was a novel contribution since its sample was brand specific. Time was the preferred source for the ad sample, but when an ad for the desired brand could not be found, The New Yorker, The Saturday Evening Post, or Life were used. Referring to the Time issue dated closest to July 1 s t for each year of the 60-year period under review, one ad for each brand was sought. Presumably, it was assumed that advertising content was not influenced by the July 4 t h Independence Day holiday. Two coders examined a total o f 211 ads and inter-coder reliability was found to be .81. It was found that among the 211 ads, an average of 5.6 claims were made (12 different claims were measured). From 1926 to 1954, the most frequently observed claims were health, taste, and cigarette construction. Health was the most common type o f claim during the 1926-1929, 1930-1939, and the 1950-1954 periods. Following 1955, as a result o f implemented F T C guidelines for cigarette advertising, health claims no longer remained a consistent and dominant advertising theme. Health claims continued 152 to be scarce until 1969 and in the meantime claims related to taste, cigarette construction, and tar and nicotine were most frequent. However, Ringold's statement that the \"commercial presentation of health related information was virtually prohibited\" (1987, p.273) for the period 1955-1969 should be interpreted with caution. Health-related claims, for the study, were exemplified by reference to reduced throat irritation, reduced coughs, and protection against adverse health effects. Yet, several of the separately measured claims (i.e., cigarette construction, exhortation to purchase/slogans, tar and nicotine figures, reduced tar and nicotine, athlete endorsements, and even taste) were also l ikely meaningful for smokers with health concerns. Since the Ringold and Calfee (1989) study presents similar conceptual concerns, this issue w i l l be discussed more thoroughly at a later point in this chapter. Ringold's study presents additional limitations. First, the overall inter-coder reliability o f .81 was low since according to Kassarjian (1977) it should preferably exceed .85. Second, coding was done only for information conveyed in headlines, subheadings, and ad copy. A s a result, pictures and images were not considered for coding purposes and information revealed in small print was overemphasized. The dismissal of pictures and imagery from the coding process should be considered a significant omission since the majority o f cigarette advertising layouts are visually oriented. Ringold failed to recognize an important element of cigarette advertising, an element that may have changed in character or importance over the time period under consideration. Third, the sample was primarily restricted to Time magazine. Relative to several other magazines, Time readership is l ikely characterized by a higher proportion of 153 smokers that are classified as \"health concerned,\" more highly educated, and elevated socio-economic status. Ringold and Calfee (1989) analyzed 568 cigarette ads that were sampled primarily from Time magazine. Ads for seven American cigarette brands were analyzed from 1926-1986, using eight mid-decade cross-sections (ads were sampled twice during the 1950s to assess the impact of mounting cancer concerns and the implementation of the 1955 F T C cigarette advertisement guidelines). The authors established 51 coding categories, including health claims, cigarette construction claims, exhortation to buy, slogan, taste, price, availability, competitive advantage, endorsements, premiums/contests, utility, and miscellaneous. In addition, 27 general ad characteristics were coded (e.g., size of ad, whether a cigarette is portrayed). The depicted meanings in pictures and imagery were omitted from the coding process. Rather, coding was limited to the ad's headline, subhead, and copy. Ringold and Calfee found that health claims highlighting the negative health aspects of smoking were prominent in ads, except during periods in which such claims were not permitted due to regulation. Health-related claims emphasized and addressed the most common concerns o f the time among consumers, most notably health protection and tar and nicotine content commencing during the early 1950s and \"smoker's cough\" and throat irritation prior to the 1950s. A s previously mentioned, the early 1950s represent a significant distinguishing period because scientific and popular articles presenting lung cancer research findings became more commonplace and initiated what the tobacco industry referred to as a \"health scare.\" According to Ringold and Calfee, health-related claims peaked during the early 1950s (in response to consumers being 154 increasingly concerned about the potential health risks incurred from smoking), but no longer remained a consistent and dominant advertising theme once F T C guidelines for cigarette advertising were imposed in 1955. The authors suggested that the F T C guidelines were i l l advised since tobacco manufacturers (particularly smaller ones) had typically engaged in advertising with a \"negative\" approach that first reinforced consumer fears about smoking and then provided their product as a less harmful alternative. Such advertising claims, they argued, would hurt competitors and the industry as a whole, but benefit a small manufacturer i f its increased market share more than offset its portion of the decrease in total industry sales. Solow (2001), however, has convincingly countered that it was industry collusion rather than F T C regulation that led to dramatic changes in the number of health claims observed in cigarette advertising during the early 1950s. The study and arguments put forward by Ringold and Calfee drew considerable attention. Cohen (1989) criticized the content analysis conducted by Ringold and Calfee on three accounts. First, Cohen disagreed with Ringold and Calfee's argument that advertising using health-related claims reinforces consumer fears about the health consequences of smoking. Rather, he argued that proposing a particular brand as a safer alternative is misleading and in fact reduces fear among consumers. Since nicotine is highly addictive and quitting is not an easy option, switching to a lower yield, or a supposedly safer, cigarette became an attractive and reassuring alternative for many \"health concerned\" smokers. According to Cohen, \"the only way for an advertiser to reinforce consumer health fears is to correctly inform consumers that no brand of cigarettes is safe\" (1989, p.25). 155 Cohen continued his first argument by stating that consumers were l ikely to assume governmental agencies would not permit use of deceptive health claims. Meanwhile, tobacco manufacturers utilized F T C test results for tar and nicotine yields in ad copy in attempts to gain a competitive advantage. For example, ads claimed that among all cigarettes \"Carlton is lowest\" by referring to the most recent U . S . Government Laboratory test (Pollay & Dewhirst, 2002). With F T C accreditation, consumers were likely to perceive the tar and nicotine ratings as precise even though tobacco manufacturers acknowledged within internal corporate documents that the F T C testing procedures were inaccurate.4 Cohen's (1989) second criticism about Ringold and Calfee's content analysis concerned the limitations of the study. In seeking objectivity and replication, Ringold and Calfee did not want coders to make any interpretations when analyzing the ads. According to Ringold and Calfee, the coders were instructed \"to take claims literally, focusing on the words actually used, and were not to speculate on possible interpretations. This meant that some implied claims which nowadays would be considered 'obvious' might be lost, but we made this sacrifice in order to maximize reliability\" (1989, p.8). Cohen argued that the ads were analyzed as though inferences did not exist or were unimportant. Mildness and filtration claims were not classified as being related to health. Rather, \"mildness\" was regarded to be a claim about taste or flavour and \"filters best\" was considered to be a cigarette construction claim. It is worthwhile to note that cigarette construction claims represented the largest single category o f claims recorded i f voluntary and mandated health claims were considered as separate categories. 156 Cohen (1989) presented a third criticism, claiming that the purpose of content analysis must be reconsidered. When conducting content analyses: One could simply decide to count the number of times a certain word, or scene or type of person appears in advertising. If, say, we are interested in minority group representation in advertising, we could start by counting such instances. But, i f our hypotheses include role portrayals, how people are interacting, whether there is a dominance relationship among the people, etc., the coder must interpret the ads. I f one hypothesizes that certain themes (e.g., materialism, conformity) are becoming more prevalent in advertising, there can be no escaping the need to record the latent as wel l as manifest content of the material... it really comes down to deciding whether one is more interested in determining what was said or what was communicated. (Cohen, 1989, p.28) Cohen pointed out that in an attempt to increase inter-coder consistency, Ringold and Calfee limited their coding to merely the direct verbal assertions made within the ad copy. Thus, pictures and images o f athletic people engaged in sporting activities would not be recorded as health claims. L ike Cohen, Pollay (1989) argued that health-related claims in cigarette advertising were largely underestimated in Ringold and Calfee's content analysis due to their conservative definition o f \"health.\" Cigarettes introduced with newly constructed filters or described as mild, light, fresh, smooth, clean, pure, soft, and natural were not coded as health claims. Prominent athletes commonly endorsed cigarette brands from the 1930s until the 1960s, yet such endorsements were not coded as portraying healthfulness. Moreover, Pollay considered it a significant oversight to dismiss from the coding process 157 the messages communicated by pictures and images since the majority o f cigarette advertising layouts are visually oriented. Pollay also voiced disapproval toward Ringold and Calfee classifying the Surgeon General's warning as a \"health\" claim. Such classification misrepresents the intent o f the advertiser and credits the \"advertisement\" with the content of the warning. Pollay observed that the data generated for Ringold and Calfee's content analysis focused on seven American cigarette brands over a 60-year period, a time span in which the market position of the brands studied changed dramatically. Chesterfield and Old Gold, for example, moved from positions of dominance to near extinction. L ike Ringold and Calfee, Pollay noted that \"top dogs\" and \"underdogs\" face different advertising challenges, including the extent in which health-related claims are to be used. This observation raises the question of whether researchers should be constant in the brands included in a longitudinal sample, or alternatively, i f it is more appropriate to analyze the brands occupying particular market positions. Finally, Pollay stated that headlines and fine print should not be coded as equally important and claim repetitions should not be ignored. Ringold and Calfee (1990) responded to the critiques provided by Cohen and Pollay, addressing several methodological issues and maintaining their position that there was little reason to regard mildness or filter assertions as health claims, unless they were accompanied by explicit references to health. It was also still argued that smaller firms utilizing health-related claims in their advertising would do so for their own benefit, yet suppress the overall demand for cigarettes in the process. In an effort to induce consumers to switch brands, competition may lead to cigarette advertising that reminds 158 consumers about their own fears of smoking. A d copy placing emphasis on tar and nicotine yields, for example, might remind consumers there is something to fear in cigarettes. According to Ringold and Calfee, their study results supported the notion that advertising serves a useful informational role and more stringent regulation of tobacco advertising is unlikely to improve the market. A key methodological issue in the critique of Ringold and Calfee's study was whether it would have been preferable to analyze \"manifest\" or \"latent\" content. For manifest content analysis, coders limit their analysis to explicit advertising content rather than also interpreting the likely meaning conveyed by advertising messages. Ringold and Calfee defended their use of manifest content analysis, stating it is validated in a straightforward manner using objective criteria. As an example, they offered, \"Was 'cough' stated or not within the advertisement?\" Latent content analysis, they argued, would have been problematic for their study since it encompassed 60 years of cigarette advertising. Validation problems would have occurred because: However thoroughly the coders were schooled in evolving historical circumstances and shifting audience characteristics, their judgments would have to be validated using a naturally occurring audience (the only kind of audience that can assess what ad content means to an ordinary consumer in the informational context of the day). Such an audience is essentially unobtainable for the cigarette market of past decades. (Ringold & Calfee, 1990, p.31) While latent content analysis is useful for analyzing contemporary ads, it is very difficult to employ for a historical assessment of ads that span over several decades. \u00E2\u0080\u00A2 159 Cohen (1992) responded to the issues raised by Ringold and Calfee (1990), stating that to overcome the challenge of utilizing latent content analysis to interpret ads covering an extended time period, it is the researchers rather than the coders that require background and preparation concerning evolving historical circumstances. Onus rests with the researchers to clearly define for the coders the requirements o f category membership. Cohen asserted that Ringold and Calfee overstated the responsibilities of coders for latent content analysis. Once again, Cohen (1992) challenged the exclusion of several mildness and filtration assertions from being coded as health-related claims. He charged that by considering mildness assertions to be distinct from health-related claims, Ringold and Calfee's understanding o f such assertions were considerably different from the tobacco industry. Indeed, trade sources and tobacco industry documents publicly accessible through various court proceedings reveal that mildness and filtration claims are meant to communicate health-related messages to consumers. For yet one example, according to a 1977 British American Tobacco ( B A T ) document, communication strategies: . . .should be directed towards providing consumer reassurance about cigarettes and the smoking habit. This can be provided in different ways, e.g. by claimed low deliveries, by the perception of low deliveries and by the perception of 'mildness'. Furthermore, advertising for low delivery or traditional brands should be constructed in ways so as not to provoke anxiety about health, but to alleviate it, and enable the smoker to feel assured about the habit and confident in maintaining it over time. (Short, 1977, p.3) 160 A n additional B A T document claims that, \"opportunities exist for filter and cigarette designs which offer the image of 'health reassurance'\" (1976, p.6). Finally, a Loril lard document assessed whether consumers perceived their Kent brand to have the best filter, stating '\"best filter' is undoubtedly considered in terms of many different benefits including the taste the filter delivers, ease o f drawing, mi ld taste, as wel l as health\" (Kieling, 1964, p. 12). The selected industry document quotations are not exhaustive and clearly demonstrate that it was a dubious decision for Ringold and Calfee to not classify references to mildness and filtration as health-related claims. U.S. Billboard Advertising Altman, Schooler, and Basi l (1991) analyzed 901 billboards in San Francisco, California, with specific attention being given to assessing how tobacco and alcohol billboard advertising differs in Asian, black, Hispanic, and white neighbourhoods. Billboard data collected included location, size, and advertising content (i.e., language, ethnicity of models featured, and theme). The theme categories were romance, recreation, adventure/risk, health, product quality, price, comparison, style/packaging, fashion, other, and no theme. Unfortunately, the researchers did not provide the data-coding instrument or definitions for the theme categories. Among all product or service billboard ads, tobacco was the most heavily advertised, representing 19% of al l billboard ads. The most prevalent theme among tobacco billboards was product quality, followed by no theme, romance, and recreation. Romantic themes were notably more common among billboards located in black neighbourhoods. Moreover, black neighbourhoods had proportionately more tobacco ads 161 (per 1000 population) compared to Asian, Hispanic, and white neighbourhoods. The inter-coder reliability among the four coders was 92% (based on a random sample of 18% of the coded ads). Intra-coder reliability was not reported. Taylor and Taylor (1994) conducted a content analysis of 705 billboards located along federally funded highways in the state of Michigan. The study assessed the information content present in billboard advertising and the types of corporations or businesses using this medium. The data-coding instrument included six product classes, 29 product categories (tobacco products were identified as one o f the product categories), and 30 information content categories (e.g., price, variety, quality, availability/location, components/content, taste). On the basis of the product or service promoted, billboards were classified into six different classes. \"Manufactured goods\" represented one of the six product classes and tobacco billboards were included within this product class. While the average billboard contained 2.04 information dimensions, manufacturer billboards possessed 0.95 information dimensions. The figure 0.95 represented the lowest information content among the six product classes. Taylor and Taylor concluded that overall the assessed billboards contained a substantial amount of information. However, this conclusion was based on the criterion that an ad was informative i f at least one information cue was present. Taylor and Taylor (1994) also found that alcohol and cigarette billboards represented a small proportion of the total billboards located along federally funded highways. Combined, alcohol and cigarette billboards accounted for only 8% of all billboards. Although these findings may appear to contradict the results of Altaian, Schooler, and Basi l (1991), neighbourhood-located billboards and those regulated by 162 local municipalities were not included for analysis. Taylor and Taylor did find that tobacco billboards were much more common in urban locations compared to rural locations. Thus, for studies focusing on the content of tobacco billboards, it may be a notable oversight to exclude neighbourhood-located billboards from the samples. Studies Analyzing the Content of Non-American Cigarette Advertising Chapman (1986) made a unique contribution by using content analysis to analyze cigarette and hand-rolling tobacco ads from seven nationally distributed Australian magazines and three New South Wales newspapers. The magazines sampled were Australian Women's Weekly, New Idea, Australasian Post, Cleo, Australian Playboy, Bulletin, and Wheels. The Sydney Morning Herald, Dai ly Telegraph, and Dai ly Mirror represented the newspapers included in the sample. The 10 selected publications had diverse readerships. Nearly all issues from the year 1983 were used to gather the sample;-resulting in a total of 1,026 ads in which 23 different cigarette brands were represented. Five people were responsible for coding, with the senior investigator participating as one of the coders. Coding was done for the following 17 themes: leisure; sport; high cultural activity; science/technology; sex; nature; prestige; history/nostalgia; product only; product and price claim; identification; identification sex role stereotype; identification celebrity; romance; adventure; taste, strength or flavour; and humour. Definitions were provided for each theme and data was also generated for the size o f the ad, the presence of human models, the sex and estimated age of any models depicted, and whether the brands advertised were regular or low tar. Inter-coder reliability was reported to be .82. 163 Chapman found that nearly 54% of the ads were exclusively for \"regular strength\" cigarettes, 31% were for low tar or mi ld blend cigarettes, and 14% presented brand families or extensions that included both regular and mi ld varieties. Human models were featured in 62% of ads and couples or \"men only\" were the usual depictions. \"Women only\" were featured in merely 14% of the ads in which models were present. M e n were much more l ikely to be depicted smoking compared to women. On average, 1.9 themes were recorded per ad and among the 17 different coding themes, nine accounted for 87% of the instances. The leading nine themes were price, prestige, nature, taste, leisure, humour, sex-role stereotypes, identification, and romance. Leisure, nature, prestige, sex-role identification, romance, and taste were themes more common in women-oriented magazines. Humour and sport appeals were uncommon in women's magazines. Price, humour, and product themes were more frequently found in newspapers. Magazines that were classified as down-market featured a high concentration of value appeals and a notable absence of leisure, prestige, and taste appeals. Two key study limitations are apparent. First, only 20 ads were randomly selected from the sample for the purpose o f calculating inter-coder reliability. While the mean inter-coder reliability was .82, the reliability coefficient was .57 for seven o f the 20 selected ads. To calculate inter-coder reliability, the five coders were asked to score the number of themes present in each of the 20 ads. Comparisons were then made among the coders regarding the frequency of different themes coded. A s a result, when discrepancies were apparent for the number o f themes found present, it remained unclear whether the discrepancies were repeatedly based on assessing a particular theme. A second limitation is that the sample included several \"repeat\" ads. Among the total 164 sample of 1,026 ads, only 214 different ads were represented. Thus, the study results largely reflect advertising content for a select number of brands that possess the largest advertising budgets. One virtue of such a study design, however, is that ads in the analyzed data set were weighted according to their actual frequency and relative advertising budgets (i.e., share of voice), reflective of the consumer's typical experience. Pollay (1990) also made a unique contribution by using content analysis to analyze Canadian cigarette ads. Using 14 different magazines with varying editorial styles, target audiences and reach, a total of 394 ads from 1987 were analyzed. Canada's three major tobacco manufacturers were each wel l represented by ads in the sample. To assess the information content of the ads, dimensions were developed that measured availability, competitive advantage, contents/features absent, contents/features present, directions, guarantees/warranties, nutrition, packaging/product design, performance, personal safety, premiums/contests, price, product variations, product quality, research, and testimonials/endorsements. Pollay (1984) had previously developed these information dimensions to determine how informative advertising was for any product and with the exception of nutrition, the 16 information dimensions were all deemed appropriate for analyzing cigarette advertising. Using the same 12 themes as previously mentioned during the review o f Pollay (1991), Pollay (1990) measured for the presence or absence of associated themes and images in Canadian cigarette advertising. A new variable, healthfulness, was established to integrate the information gathered from the pure scene, bold/lively, and health/safety themes. Pollay provided coding definitions for each o f the advertising themes. 165 Pollay found that overall there was little information conveyed in Canadian cigarette advertising during 1987. Ten of the sixteen information dimensions were not found among any of the sampled ads. The only information dimensions observed with relative frequency were contents/features absent (55% of ads), performance (58% of ads), product variations (52% of ads), and product quality (40% of ads). Ultra light and extra mild exemplified common claims about contents absent and references to taste accounted for nearly all assertions about product performance. Although product variations (e.g., indications of regular and king size) were evident in roughly half of the ads, the information was often conveyed in fine print or on displayed packages. Meanwhile, visual imagery accounted for the bulk of ad space and attention with healthfulness being a theme communicated in 68% of the ads (pure scene, bold/lively, and health/safety were themes present in 19%, 36% and 41% of the ads, respectively). Enjoy (featured in 59% of ads) and luxury (featured in 47% of ads) were the other most prevalent themes. Pollay concluded that four styles of cigarette ads prevailed: (1) portrayals of autonomy and independence; (2) outdoor, nature settings with people engaged in lively and robust behaviour; (3) professionals or affluent people to indicate success, wealth and sophistication; and (4) little depiction of human models, but use of descriptors such as light and extra mild to present the product as a safer alternative. Some study limitations must be acknowledged. Fourteen magazines were used to generate ads for the study, but they were not named. While Pollay recognized that nutrition was a non-applicable information dimension for cigarette advertising, it remained listed as information coded. Finally, the ad sample was not representative of each Canadian tobacco manufacturer's market share during 1987. Approximately 39% of 166 the observed ads were from Imperial Tobacco and precisely one-half o f the observed ads were from Rothmans, Benson and Hedges. Yet, according to the annual reports o f Imasco Ltd. and Rothmans Inc., Imperial Tobacco possessed a 55% market share while the market share for brands manufactured by Rothmans, Benson and Hedges had stabilized at approximately 28% (Imasco Ltd., 1987; Rothmans Inc., 1988). Discussion Summary of Research Approaches A total of 12 content analysis studies specific to cigarette advertising were critically reviewed in this chapter. Studies focusing on the information content of magazine advertising have typically examined content changes in relation to major \"smoking and health controversy\" events, assessed the repercussions o f F T C regulations, or focused on the impact o f the Public Health Cigarette Smoking Act that banned cigarette advertising in the U . S . broadcast media starting in 1971. Among the 10 content analyses pertaining to magazine advertising, all eight studies of U . S . media utilized a longitudinal or historical approach while the two studies (Chapman, 1986; Pollay, 1990) restricting its data set to non-U. S. media or cigarette brands employed a cross-sectional approach. Chapman (1986) represents the only study conducted thus far that includes newspapers in the data set. Content analysis has also been used to study cigarette billboard advertising, but all studies to date have been situated in the U . S . and limited to American brands. Taylor and Taylor (1994) assessed billboards in federally funded highway settings, while 167 Airman, Schooler, and Basi l (1991) examined billboards sited in different ethnic neighbourhoods to determine whether information content might differ according to the intended target group or segment. The sample for both studies was not limited to tobacco advertising (i.e., additional product categories were analyzed). A s stipulated by the U . S . Tobacco Settlement, billboard advertising was banned commencing in A p r i l 1999 (Lubove, 1999). Thus, future content analyses specific to billboard advertising must either be historical in nature, originate outside of the United States, or consider other forms of billboard advertising by tobacco firms (e.g., advocacy, sponsorship). Key Study Findings Despite reviewing studies with contrasting sample frames, sample sizes, time frames, data-coding instruments and coding category definitions, several robust findings are apparent. Several content analysis studies confirm that the number o f cigarette ads found in American magazines dramatically increased during the 1970s, largely the result of the U . S . broadcast ban in 1971. Also evident were more frequent use of special positioning, colour, and full-page or double-page ads. Healthfulness has been a consistent theme in magazine advertising content, reflecting attempts by the tobacco industry to reassure smokers concerned about potential health risks. Low-yie ld products have become more frequently promoted over time and health-related themes are increasingly communicated through visual imagery. When cigarettes are depicted (even in cases where cigarettes are obviously lit), current ads are less l ikely to feature visible smoke. Prior to implementation of the U . S . Tobacco Settlement, tobacco billboards were more commonly found in urban settings. 168 Methodological Issues and Interpretation of Research Results A review of the literature reveals several lessons are to be learned from previous content analyses specific to cigarette advertising. Ringold and Calfee's (1989) content analysis o f cigarette advertising, and the criticism and discussion that followed by Cohen (1989), Pollay (1989), Ringold and Calfee (1990) and Cohen (1992), indicates careful consideration must be given to how mildness and filtration claims are coded. Ringold and Calfee did not classify mildness as a health-related claim, yet other researchers such as Pollay (1990) have done so. Mildness and filtration exemplify claims that may have multiple meanings (i.e., claims that are potentially classifiable under more than one information category). Mildness, for example, may be classified as a claim about taste/flavour and \"health.\" Content analysis is an effective method for assessing the common themes conveyed in a \"universe\" of cigarette advertising. However, it appears contentious to conclude whether or not ads are informative based on the number o f dimensions conveyed, and thus may depend on the medium analyzed. Taylor and Taylor (1994) considered billboard ads to be informative i f at least one information cue was present, while Pollay (1990) concluded that cigarette advertising was not informative despite finding three information dimensions present in the majority of ads. Pollay's conclusion was based on the number of information dimensions that were not present (i.e., 10 of the 16 information dimensions measured were not found among any o f the sampled ads). Since cigarettes are a highly hazardous product, forthcoming information is considered to be especially important for consumers. Rather than merely attempting to pursue the 169 quantity of information depicted, or alternatively how much information could have been conveyed, establishing a standard number for the dimensions required for deeming an ad \"informative\" would be useful. It is acknowledged that different standard numbers o f \"informativeness\" may be appropriate for the various advertising media. For example, fewer dimensions may be necessary for billboard promotions to be considered informative relative to magazine ads. It is also relevant to consider whether certain themes or information dimensions should be considered o f greater importance and thus given greater weight or merit. Furthermore, while the frequency of themes is meant to indicate the intention o f the text's producer and the extent of information provided within the message, content analysis does not indicate judgement about the accuracy or truthfulness of claims (i.e., coded information may include deceptive claims). To counterbalance and minimize learning and maturation effects, coders should not proceed through the data set in the same, strict chronological order. If possible, it is advisable to avoid using photocopies or reductions for coding purposes. Researchers should attempt to ensure that ads still resemble their original state and appear as initially seen by readers. For ad samples, coding should not be limited to direct verbal assertions. Since the majority of cigarette advertising layouts are visually oriented, pictures or visual imagery should be included in the coding process. When measuring visual aspects, researchers must account for how ads with multiple images w i l l be coded. Consideration should be given to whether headlines and small print w i l l be given similar emphasis in the coding process. Moreover, it must be pre-determined whether claim repetitions w i l l be acknowledged and how mandatory health warnings w i l l be coded. While Ringold and 170 Calfee (1989) classified the Surgeon General's warning as a \"health\" claim, Pollay (1989) argued that such classification misrepresents the intent of the advertiser. Although it is important to code for the presence o f mandated health warnings, the Surgeon General's warning and the ad copy should not be considered equivalent assertions. Study limitations commonly identified in the reviewed literature include using a sample of one magazine from the entire universe, using a sample that had limited seasonal variation, failing to calculate or report intra-coder reliability, categorizing ads in a misrepresentative manner, and failing to provide coding definitions for the themes and dimensions measured.5 Overall inter-coder reliability was commonly reported, but in some cases calculations were based on a small sample. In addition to reporting overall inter-coder reliability, it is helpful to provide reliability calculations for each of the various dimensions or variables. To maximize overall inter-coder reliability, coders were instructed in some of the reviewed studies to take claims literally and avoid making interpretations about the likely meaning conveyed in advertising messages. Reliability measures may be misleading because data that generates high percentage agreement among coders may reflect a simplistic data-coding instrument and mundane information collected. Great attention has been given toward maintaining a high inter-coder reliability and debate remains about the level of percentage agreement needed to reflect adequate coding definitions. While 70% is conventionally used as an acceptable criterion for reliability, Kassarjian (1977) stated that 85% should be arrived at for content analysis. 171 Directions For Future Research A review o f the content analysis literature should inform methodological approaches undertaken for future studies. For example, future research should utilize different magazines (i.e., that may differ according to segmentation, price, recent popularity, or editorial approach) from those used in previous studies. People, G O , Glamour. Cosmopolitan. Elle , U S , Spin, Gear, Entertainment Weekly, M a x i m , and Penthouse exemplify popular magazines that have not been included in previous analyses. In addition, when developing future studies pertaining to cigarette advertising, researchers may choose to utilize sampling frames and methodological approaches consistent with previous studies. Thus, study results may be compared with greater confidence. Two reviewed studies (Chapman, 1986; Pollay, 1990) analyzed the advertising content of non-American cigarette brands and magazines. Further research that informs about cigarette advertising content from a greater array of countries is needed. The absence of studies from the United Kingdom and developing countries is particularly noteworthy. Meanwhile, recent F T C reports indicate that U . S . domestic cigarette promotional spending is at record levels. Given the persistence of cigarette promotion and continuing public health concerns, more content analysis research is inevitable and desirable. One domain of particular importance is the need for evaluating the effectiveness of policies. Content analysis can be used to assess public policies that place restrictions on cigarette promotion by comparing the character o f ads that precede a policy's implementation with those that follow. 172 Endnotes 1 According to Ringold and Calfee (1989), the first three authors were among those coding for the study, thus raising concerns about whether the data yielded from the coding instrument were replicable. Using the same eight magazines and similar methodology, Albright, Altaian, Slater, and Maccoby (1988) determined that the average number of cigarette ads per magazine issue increased substantially from 1960 to 1985 (most notably following the broadcast ban in 1971). Initiating in the late 1960s and early 1970s, a greater proportion of cigarette ads were placed in magazines with a women and youth-oriented readership. The study accounted for the frequency o f cigarette ads found in magazines with varied readerships, but coding was not reported for the information content o f the sampled ads. The F T C issued a set of voluntary guidelines in 1955, which were meant to curtail the tendency by cigarette manufacturers to associate smoking with deceptive \"health claims\" in their advertising. U.S.. cigarette manufacturers were to abstain from using claims that suggested medical approval of smoking, as wel l as those related to the effects of smoking on energy levels or various parts of the body (including all parts of the respiratory tract, digestive system, and nerves). Claims concerning lowered tar and nicotine yields (i.e., due to the length of the cigarette, the addition of a filter, etc.) were prohibited unless they were substantiated by competent scientific proof. Assuming that the claim was true, the yield cited was also to be regarded as a significant difference. The F T C and U . S . cigarette manufacturers reached an informal agreement in which tar and nicotine content claims became banned altogether from 1960-1965, but the F T C 173 rescinded this stipulation in 1966 (Ringold & Calfee, 1989; McAul i f fe , 1988; Kluger, 1997). 4 Cigarette papers and filters were developed that enabled smoke to be \"air-conditioned\" and the smoke column to be diluted through the entry of side-stream air. These vents were placed in locations of the cigarette commonly obstructed by a person's fingers or lips once being smoked. Thus, tar and nicotine yields generated for cigarettes smoked by machines during F T C testing were appreciably lower than yields delivered by those smoked by actual people (Canova, Myers, Smith, & Slade, 2001; Koz lowski & O'Connor, 2002). F T C test results were inconsistent with actual tar and nicotine yields since the machines did not initially account for the compensatory behaviour demonstrated by people. To satisfy their addiction, smokers often compensate when smoking lower yield cigarettes (Cohen, 1989). Compensatory behaviour includes smoking the cigarette closer to the butt, taking deeper puffs from the cigarette, increasing the number o f puffs taken while smoking the cigarette, and smoking more cigarettes per day. 5 The common omission of coding definitions for measured themes and dimensions may reflect editorial decisions and word length limitations for many journals. In such cases, however, it is advisable that content analysts make it clear that their data-coding instruments are available upon request. 174 C H A P T E R S E V E N Content Analysis Study Comparing Traditional Product Advertising and Sponsorship Promotions: Persistent Lifestyle Messages in the Face of Regulation Despite intense effort by tobacco control groups during the past few decades, legislation to control tobacco promotions in Canada has appeared limited in its. effectiveness. Following the passage of the T P C A in 1988, expenditures on event sponsorship flourished, with advertising and promotional support remaining a significant component o f the industry's investment. This chapter consists o f an original content analysis study in which the character o f Canadian cigarette print ads from the p re -TPCA era is compared with those from the post -TPCA era. The data set is comprised o f ads for the Player's, Export ' A ' and Rothmans trademarks, thus allowing the findings to be validated with the disclosures of the internal corporate documents. The objectives of the study reported in this chapter are to assess whether the character of the content o f Canadian cigarette print ads was appreciably changed by the implementation o f the T P C A , as well as examine and compare the promotional strategies utilized for the Player's, Export ' A ' , and Rothmans trademarks over a 30-year period. It is determined which advertising content dimensions remained consistent, or alternatively which ones changed, as the Canadian tobacco industry's marketing strategy shifted from traditional product advertising to sponsorship. In evaluating whether tobacco promotional messages have been continuous or changing with respect to several content 175 dimensions, conclusions can be drawn about the effectiveness of the T P C A . Content analysis is regarded as an appropriate research methodology for determining whether public policies have been effective in changing the content of advertising (Kassarjian, 1977). Hypotheses Variation Over Time The review of internal corporate documents revealed that sponsorship was viewed by the industry as a way of maintaining image-based portrayals in their advertising. It is also known that communicating brand image has historically been a primary objective of cigarette promotional efforts. Thus, it is expected that, for many of the measured content dimensions (particularly those relating to lifestyle), the character of Canadian cigarette ads w i l l not differ significantly when making comparisons between those from the pre-and pos t -TPCA eras. It is anticipated that tobacco promotional content w i l l persist among several themes, dimensions, and categories despite regulation. However, some observable content changes are anticipated, reflected in the hypotheses listed below. Hypothesis 1 During the p re -TPCA era, the depiction of tobacco products and packaging was allowable. In his cross-sectional analysis of Canadian cigarette ads circulating in magazines during 1987, Pollay (1990) found that \"contents/features absent,\" \"performance,\" and \"product variations\" were the only information dimensions present in 176 the majority of the sample. Pollay noted that indications o f product variations were often conveyed on displayed packages, thus it seems reasonable to expect that tobacco products, including packaging, were depicted in p re -TPCA ads with relative frequency. The T P C A , meanwhile, \"permitted promotion of sponsorships using a corporate name, provided this was not done ' i n association with a tobacco product,' with the onus on the advertiser to show no such association\" (Cunningham, 1996, p.97). To adhere to the stipulations o f the T P C A , it is presumed that ads from the pos t -TPCA era did not include the depiction of either tobacco products or packaging. HI: Tobacco products, including packaging, will not be depicted in post-TPCA promotions. Hypothesis 2 Pollay (1990) found that 55% of Canadian cigarette ads, circulating in 1987, made claims about \"contents/features absent\" (e.g., ultra light or extra mild products), while 52% of the ads contained assertions about \"product variations\" (e.g., regular and king size products). Thus, it is anticipated that ads from the p re -TPCA era w i l l often include information about various brand types or line extensions (e.g., Player's Light) rather than merely presenting the trademark or brand family name (e.g., Player's). The T P C A , however, stipulated that the full name of the manufacturer was required on promotional material as opposed to a tobacco brand name. A s discussed in previous chapters of this thesis, Canada's three principal tobacco firms responded by hastily registering their various trademarks as separate corporate entities. Ads from the pos t -TPCA era w i l l 177 presumably present either a corporate entity (e.g., Player's Ltd.) or a trademark (e.g., Player's). 1 H2: Brand types or line extensions will not be depicted in post-TPCA promotions. Hypothesis 3 Prior to the implementation of the T P C A , a process of industry self-regulation was in effect, with voluntary advertising codes ( C T M C , 1976, 1984) stipulating that, \"No advertising w i l l use, as endorsers, athletes or celebrities in the entertainment world.\" A s Canadian tobacco firms shifted their promotional spending toward sponsorship, it is anticipated that celebrities were increasingly depicted in ads, in an effort to have the image(s) o f event participants transferred to the sponsoring brand. Specifying who is participating, performing or competing in an event is also seemingly important information to convey to potential event attendees. H3: The presence of celebrities in tobacco promotions will significantly increase following the implementation of the TPCA. Hypothesis 4 Amendments to the Tobacco Act , B i l l C-42, stipulate that tobacco sponsorship w i l l be banned in Canada, commencing October 2003. Other means o f promotion such as direct marketing remain permissible, however. It is anticipated that Canadian tobacco firms w i l l increasingly turn toward direct mailings as a way to continue communicating to consumers once tobacco sponsorship is no longer permitted. Methods o f collecting .178 demographic data from target consumers (to further generate databases that are usable for direct marketing or relationship marketing purposes) include staging contests, utilizing mail-in offers, creating event guest lists, and forming magazine subscription lists. Reflecting the anticipated increase in the frequency of contests, it is expected that copy volume w i l l be greater in post -TPCA ads, since contest promotions are typically accompanied by participant requirements and stipulations. H4a: The frequency of contests will significantly increase following the implementation of the TPCA. H4b: Copy volume will significantly increase in post-TPCA promotions. Hypothesis 5 The Internet was not widely accessible to consumers during the p re -TPCA era, thus it is expected that no website listings w i l l appear on conventional product ads. When reviewing internal corporate documents and describing the various advertising campaigns pertaining to the post -TPCA era, however, it was observed that website listings were provided in some cases for Player's, Export ' A ' and Rothmans. Moreover, it is anticipated that relationship marketing (including the use o f toll free numbers) has become an increasingly important component of the promotional mix for Canadian tobacco firms. H5: Phone number and website listings will significantly increase following the implementation of the TPCA. Hypothesis 6 Sponsorship objectives include cross-promotional/co-sponsorship opportunities, as well as enhancement o f trade relations and goodwill (Irwin & Asimakopoulos, 1992; Irwin & Sutton, 1994; Copeland, Frisby, & McCarvi l le , 1996). Tobacco companies are seldom the exclusive sponsors of an event. Through the formation o f partnerships and strategic alliances, there are opportunities to associate cigarette trademarks with other . products that possess complementary symbolic elements. It has also been observed that sponsorship partnerships are desirable for tobacco firms since cigarette trademarks may in some cases gain credibility by being linked with less contentious products (Dewhirst & Hunter, 2002). Thus, it is hypothesized that sponsorship w i l l facilitate an increase in the strategic use of cross-promotion and co-branding. H6: The use of cross-promotion (i.e., identification of co-sponsors or additional advertisers) will significantly increase following the implementation of the TPCA. Hypothesis 7 Voluntary advertising codes ( C T M C , 1976, 1984) stipulated that all Canadian cigarette print advertising would display the following: \" W A R N I N G : Health and Welfare Canada advises that danger to health increases with amount smoked - avoid inhaling.\" Health warnings are not mandatory for sponsorship promotions in the pos t -TPCA era, however. A s a result, it is expected that the frequency o f health warning depictions w i l l decline substantially. Health warnings are still to be expected in some pos t -TPCA ads though, as it was observed in Chapter 4 that several Export ' A ' extreme sports series promotions utilize the Health Canada warning: \" C I G A R E T T E S L E A V E Y O U 180 B R E A T H L E S S . Tobacco use causes crippling, often fatal lung diseases such as emphysema.\" H7: The frequency of health warning depictions will significantly decrease in post-TPCA promotions. Variation by Cigarette Trademark/Brand Hypotheses 8 and 9 The primary basis for anticipating content variations according to cigarette trademark is the review of internal industry documents that was offered in Chapters 2 through 5. Player's has been successfully promoted toward male youth, symbolically expressing independence, freedom, self-expression, and masculinity. Export ' A ' has been similarly positioned, such that the trademark has been linked with images o f ruggedness, machismo, adventure, escapism, independence, and rebelliousness. . Rothmans has been traditionally promoted as a premium quality trademark that is internationally renowned, with its consumers often presented as upscale and worldly. The Rothmans trademark has a positioning that is not overtly 'masculine' or 'feminine', using appeals with cross-gender relevance. Wi th 'masculinity' being a central theme communicated in the positioning of Player's and Export ' A ' , it is anticipated that models w i l l tend to be exclusively males. H8: When models are portrayed, males exclusively will be the most common depiction for Player's and Export 'A', yet both males and females will be more frequent for Rothmans. 181 H9a: The predominant lifestyle dimensions for Player's will be independence, adventure/excitement, and ruggedness. H9b: The predominant lifestyle dimensions for Export 'A ' will be independence, adventure/excitement, and ruggedness. H9c: The wealth/prosperity lifestyle dimension will prevail for Rothmans. Hypothesis 10 The market share of Player's has grown impressively over the historical period of analysis outlined for this study. Player's, it has been argued, is a market leader due to the consistency demonstrated in the trademark's communication, particularly with respect to brand imagery. While Export ' A ' and Rothmans remain prominent trademarks for their respective manufacturers, the market share of both brands has been in continuous decline over the past few decades. One explanation for the diminishing popularity o f both trademarks is that their marketing communication efforts are inconsistent relative to Player's. H10: Player's will demonstrate the greatest consistency with communicating lifestyle dimensions that correspond to its positioning. Methodology The Data Set The data set consists of ads that were entirely acquired from The History of Advertising Archives, which is located at the University of British Columbia in 182 Vancouver, Canada. The Archives were considered a suitable data set source since it is purported to house the largest collection of Canadian tobacco ads (Stueck, 1999). The data set was limited to English-language ads for three Canadian cigarette trademarks -Player's, Export ' A ' , and Rothmans - representing each of Canada's three major tobacco manufacturers. The three trademarks are comprised of flagship or dominant brands, having had a substantial presence in the Canadian marketplace for a considerable number of years. These trademarks are competitors, and in the case of Player's and Export ' A ' , have been strategically positioned to appeal to a similar target audience. The data set represents a near census of print ads held by the Archives for the pre-selected Canadian cigarette trademarks, which circulated from 1973 through 2002. The starting point of this historical analysis is 1973 because print advertising became an increasingly important part of the promotional mix once the Canadian tobacco industry voluntarily withdrew cigarette advertising from radio and television in 1972. The archival search was exhaustive and no ads meeting the outlined selection criteria were excluded from the data set.2 The entire data set was initially retrieved from general circulation magazines and newspapers, although several of the ads have been widely disseminated and featured in multiple media (i.e., point-of-sale ads, transit ads, billboards). There were no repeat ads included in the data set. Campaigns, as wel l as unique ads, were represented, with the final data set consisting of 219 ads (see Table 1). 183 Table 1. Longitudinal Data Set (1973-2002) Sample Sizes. Cigarette Trademark (Manufacturer) Number of Traditional Product Ads (1973-1988) Number of Sponsorship Promotions (1989-2002) Total Number of Ads in Data Set Player's (Imperial Tobacco Ltd.) 51 42 93 Export ' A ' (JTI-Macdonald Corp.) 39 43 82 Rothmans (Rothmans, Benson & Hedges Inc.) 30 14 44 Total 120 99 219 Proportionately, the data set representation of the three cigarette trademarks roughly reflects their market share, popularity, and promotional expenditures (i.e., Player's would be expected to have the largest number of ads represented, with Export ' A ' and Rothmans placing second and third, respectively). Once the data set was compiled, the 219 ads were randomized, with the order determined by lottery. The random sequencing of ads was undertaken to minimize any learning effects that might be observed i f coders proceeded through the data set chronologically or trademark by trademark. Following the lottery process, each ad in the data set was given an identification number. The ads were then separated into 11 binders for coding purposes (i.e., Binder A contained the ads identified as 1-20, Binder B contained the ads identified as 21-40, and so on). 184 Coding Categories To assess the content o f the 219 ads, a data-coding instrument was developed that consists o f 65 items, with each item having accompanying categories and definitions (see Appendix 25). The coding items are listed as Descriptor Dimensions (Q1-Q13), Social Portrayal Descriptors (Q14-Q26), Information Categories (Q27-Q43), and Value/Lifestyle Analysis Items (Q44-Q65). The 'descriptor dimensions' section, consisting of 13 items, includes those pertaining to ad copy volume, the display of cigarette packaging, product visibility, and demonstration of product use. There are also 13 'social portrayal descriptor' items, accounting for factors such as the ad's seasonal setting, the number of models portrayed, and what proportions o f models are male or female. The 'information category' section is comprised of 17 potential claims, including those relating to taste/flavour, price, availability, and product variations. Finally, 'value/lifestyle analysis' items include 22 potential claims or depictions, such as relaxation, nationalism, prosperity, independence, and adventure. The content categories and definitions were largely adapted from a data-coding instrument previously established by Pollay (1984), in which he measured the level of information contained in general advertising (i.e., the measured dimensions were suitable for a variety o f products). Modifications were made to the instrument so that it would have specific relevance for tobacco promotions; nutrition exemplified an information dimension in Pollay's instrument that was deemed unsuitable for analyzing cigarette advertising. When preparing operational definitions, it proved helpful to refer to dictionaries and thesauri. 185 To determine the adequacy of the data-coding instrument, it was pilot-tested using a sample of 40 cigarette ads that included both American and Canadian trademarks, but excluding those for Player's, Export ' A ' , and Rothmans. Two doctoral marketing students were responsible for coding, and some minor revisions were made to the instrument as a result o f the pilot-test. For example, the physical activity dimension was modified since the basis for coding discrepancies was common. One coder consistently included auto racing as a depiction of physical activity, while the other coder did not make such an interpretation. The modified definition for physical activity includes depictions of auto racing since it is an activity that requires intense stamina, coordination, alertness, and quick reaction time. Coding Procedures Following the recommendation of Holsti (1969), a recording unit (also referred to as level of analysis) and a context unit were identified for the study. Holst i defines the recording unit as \"the specific segment of content that is characterized by placing it in a given category... the item is the recording unit when the entire article, f i lm, book, or radio program is characterized\" (1969, p . l 16, 117), while the context unit is considered to be \"the largest body o f content that may be searched to characterize a recording unit\" (1969, p . l 18). The item was identified as the recording unit because virtually the entire ad was characterized. Coding was done for all information conveyed in headlines, subheadings, ad copy, pictures and images, with only mandatory health warnings (i.e., from Health Canada) being excluded from the coding process. The presence or absence of health 186 warnings was still monitored (i.e., Q13 of the data-coding instrument), but the decision that Health Canada warnings and ad copy should not be considered 'equivalent' assertions followed the argument put forward by Pollay (1989). Three undergraduate marketing students were hired for coding purposes. The three coders underwent a one-day training session, in which they were presented with a booklet of operational definitions, categories, rules, and procedures for analyzing the content of tobacco promotions. To become familiar with the coding process, the coders practiced with an orientation sample of 20 cigarette ads (all o f the ads in the orientation sample featured different trademarks from the ones being assessed in the final data set). During the training session, difficult coding decisions were discussed and attempts were made to resolve any questions that arose. For examining the final data set, the coders were instructed to consistently apply the operational definitions, categories, rules, and procedures to each o f the 219 ads. Ensuring a proper basis for comparison, no descriptive categories were added after the coding process had commenced and definitions were not modified. The coding process was done independently, thus judgements did not entail further consultation with the researcher or with the other coders. In situations that coders were doubtful about the appropriate response for a particular question or item, they were instructed to use their best judgement for which category best described the content of the ad being examined. The coders were unaware o f the hypotheses established for the study. The coders were told to work at an unhurried pace and to take breaks whenever refreshment was needed. They were encouraged to code on successive days to ensure 187 that they would remain familiar with the process and continue to consistently adhere to the operational definitions, categories, rules, and procedures. The data set sequence varied for each o f the coders, in an effort to counterbalance and minimize any learning and maturation effects that might be observed. Potential changes in coding behaviour resulting from fatigue were also diminished through utilization of this process. The first coder analyzed the ads in strict numerical order (i.e., coding started with the ad identified as 001 and finished with 219). The second coder examined and coded the ads in reverse order (i.e., coding began with the ads in Binder K , then the ads in Binder J, and so on). The third coder began the coding process with the ads in Binder F , proceeding through the data set in chronological order to Binder K (i.e., coding the ads identified as 101-219), and then coded the ads in Binders E through A . To summarize the data set sequence for each of the coders: Coder 001 Binders A - K Coder 002 Binders K - A Coder 003 Binders F - K , Binders E - A Coding was done question-by-question (item-by-item) for the prescribed sequence o f ads in each binder (i.e., coding for all o f the ads within a designated binder was done for Q l , then for Q2, and finished with Q65). The coders were provided with data set coding forms for recording their responses to each of the questions or items (see Appendix 26). The responses were later entered into both Excel and SPSS for data analysis purposes. The three sets of responses were recorded for calculating inter-coder reliability, but one entry (representing each item of the data-coding instrument for every ad in the data set) was utilized for testing the hypotheses and conducting further analyses. 188 The one entry was determined on the basis o f majority rules (i.e., when coder discrepancies were apparent, the response put forward by two coders was selected). To deal with the 20 situations in which a different response was recorded by each coder, a meeting was arranged once the independent coding task was complete, and the coders were asked to review and discuss these few ads and codings to collectively identify the most suitable coding decision. Most of these coding discrepancies were based on the Q7 item, which instructed the coders to decide whether package colour was depicted in the artwork of the ad scene. It was revealed by the coders that this determination was difficult for several reasons. First, package colours have evolved over time as packaging updates were made (e.g., the hue of blue utilized for Player's packaging has been modified), making it uncertain about which colour (or hue of a particular colour) was ' i n effect' for each ad under analysis. Second, the coders indicated that there was confusion about which colour(s) represented the Export ' A ' trademark (i.e., different colours are utilized for each member o f the Export ' A ' brand family, which consists o f filtered full flavour, medium, mild, light, extra light, and ultra light). Third, the coders expressed that it was unclear how precise the colour match should be when comparing the packaging and artwork of the ad scene. This dilemma would be exemplified by a Player's ad depicting a noticeably clear, blue sky in which the hue of blue approximated but did not exactly match what was conveyed on the packaging. Reliability Measures The three coders made judgements for all dimensions and all ads of the data set so that inter-coder reliability could be measured thoroughly. Overall inter-coder reliability, 189 based on a criterion of reproducibility, was calculated to indicate the level of agreement between the three coders. Reliability calculations were also made for each of the measured items so that it would be apparent whether coding discrepancies were repeatedly based on assessing a particular dimension. Moreover, inter-coder reliability was calculated for each ad in the data set. Applying the same operational definitions, categories, rules and procedures to the same data set, the coders should secure highly replicable and reproducible results and arrive at similar conclusions. While Krippendorff (1980) and Hughes and Garrett (1990) note that it is difficult to set a minimum standard for acceptable inter-coder reliability estimates, Kassarjian (1977) claims that overall inter-coder reliability should preferably exceed .85. Percentage agreement, the most commonly used inter-coder reliability measure in content analyses, was used for this study (Perreault & Leigh, 1989; Hughes & Garrett, 1990).3 Results Reliability Measures The overall inter-coder reliability score was .97 for the ,42,705 coding decisions made by the coders. Reliability scores were also calculated for each ad in the data set. A l l ads in the data set had reliability scores >93% (see Appendix 27). Moreover, reliability scores were determined for coding decisions based on each o f the 65 data-coding instrument items. It was found that 53/65 coding items had reliability scores >95%, 60/65 coding items had reliability scores >90%, and 64/65 coding items had reliability scores >85% (see Appendix 28). 190 Q7 (package colour), with a reliability score of .70, represented the only data-coding instrument item that did not meet the .85 criterion. A s mentioned, once the coding task was complete, the coders were approached collectively to determine why the Q7 item had a substantially higher number of disagreements relative to the other coding items. With the exception of Q7, reliability scores were wel l above acceptable levels. Confirmation of Hypotheses For tests of significance, Fisher's Exact Test was used for items that generated coder responses falling into one o f two mutually exclusive categories (i.e., for coding items producing one of two possible responses). Such items are exemplified by Q4 (colour) of the data-coding instrument that categorizes ads as either black and white or with colour, as well as all information categories and value/lifestyle analysis measures (Q27-Q65) that involve determination of whether a dimension is present or not present. Conversely, Pearson Chi-Square was used for items that generated three or more possible responses from coders. A s predicted in Hypothesis 1, tobacco products, including packaging, were not depicted in pos t -TPCA ads. There were no instances of cigarettes or packages being displayed in the 99 pos t -TPCA ads, which are evident by the coding decisions for items Q6, Q8, Q9, and Q10 (SPSS data analyses pertaining to all data-coding instrument items and hypotheses are presented in Appendix 29; see Table 2 for a summary o f findings for hypotheses 1-7). Hypothesis 2 - brand types or line extensions w i l l not be depicted in pos t -TPCA ads - is also fully supported. Coding item Q32 measured depictions of 191 product variations, and there were no post -TPCA ads where this item was found to be present. In support of Hypothesis 3, the presence of celebrities in tobacco ads significantly increased following the implementation of the T P C A (Q39 of the data-coding instrument measured celebrity depictions). Celebrities were not depicted in any of the p re -TPCA promotions, yet 43% (43/99) of post-TPCA promotions featured celebrities. When testing this hypothesis according to each trademark, very strong support was found for both Player's and Export ' A ' (p<.001), with significance also found for Rothmans (p<.01). Celebrities were depicted in 50% (21/42) o f Player's ads from the pos t -TPCA era, while 42% (18/43) of Export ' A ' and 29% (4/14) of Rothmans ads contained celebrities in the pos t -TPCA era. Support was shown for Hypothesis 4 (measured by coding items Q l and Q38), as contest frequency significantly increased following the implementation of the T P C A (p<.001), which corresponded with an increase in copy volume. The Canadian tobacco industry voluntarily withdrew incentive programs in 1970, and apparently adhered to this guideline since no contests were evident in p re -TPCA era ads. During the post -TPCA era, however, 13% (13/99) of ads featured contests, reflecting a means of collecting personal data on consumers that can be used toward direct marketing mailings. Fi rm-specific analyses reveal that for Player's, Export ' A ' and Rothmans pos t -TPCA promotions, 17% (7/42), 7% (3/43) and 21% (3/14) featured contests, respectively. Statistical significance was found for the Player's (p<.01) and Rothmans (p<.05) trademarks, while marginal significance is apparent for Export ' A ' (p=.139). 192 In addition, copy volume significantly increased in pos t -TPCA ads (p<.001). For p re -TPCA ads, 72% (86/120) contained less than 15 words, while 10% (10/99) of post-T P C A ads were in this copy volume category. The most common copy volume categories for pos t -TPCA ads were \"15-30 words\" (47%, 47/99) and \"more than 75 words\" (23%, 23/99), respectively. Merely 2% (2/120) of p re -TPCA ads contained more than 75 words. Statistical significance for copy volume differences between p re -TPCA and pos t -TPCA promotions was found for all three trademarks (p<.001 for Player's and E x p o r t ' A ' ; p<.01 for Rothmans). Measured by coding item Q42, toll-free phone number and website listings significantly increased following the implementation of the T P C A (p<.001). Website listings and toll free numbers, which were not apparent in any p re -TPCA era ads, were found in 48% (48/99) of post -TPCA era ads. Most notably, 76% (32/42) o f sponsorship ads for Player's listed a website (i.e., www.teamplayers.ca). Imperial Tobacco and Rothmans, Benson & Hedges have built their databases by establishing websites promoting various sponsored cultural and sporting events (Yakabuski, 1998). E x p o r t ' A ' ads characteristically listed a toll-free phone number rather than a website. A s predicted in Hypothesis 6, overall, the use of cross-promotion significantly increased following the implementation of the T P C A (p<.001). Coding for the Q12 item indicates that 2% (2/120) of the ads during the p re -TPCA era identified co-advertisers, partners or supportive organizational bodies, while 46% (46/99) of ads during the post-T P C A era were coded for the presence of this item. The two ads utilizing cross-promotion during the p re -TPCA era were for the Rothmans trademark. During the post-T P C A era, 48% (20/42) of Player's promotions, 53% (23/43) of Export ' A ' promotions, 193 and 21% (3/14) o f Rothmans promotions identified partners. When testing Hypothesis 6 according to each trademark, however, strong support was found for both Player's and Export ' A ' (p<.001), but confirmation was not found for the Rothmans trademark (p=.151). Coding item Q40 also measured the frequency of endorsements in cigarette promotions. None of the ads from the pre -TPCA era were found to include a seal of approval by outside groups, nor assertions regarding product performance from named groups or informed experts. During the post -TPCA era, however, 37% (37/99) of promotions included the identification of other supporting sponsors and organizations. Confirming Hypothesis 7, which is measured by coding item Q13, the frequency of health warning depictions significantly decreased in pos t -TPCA promotions (p<.001). A s expected, all ads from the pre-TPCA era contained the mandated health warning. For the post -TPCA era ads, with the depiction of health warnings being optional, merely 3% (3/99) contained a warning. The three post -TPCA ads depicting a health warning were for the Export ' A ' trademark, utilized for its 2001 and 2002 extreme sports series advertising campaign. Table 2. Confirmation of \"Variation Over Time\" Hypotheses Hypotheses Operational Measures Tests of Statistical Significance H I Q6 (package display), Q8 (product visibility), Q9 (product depiction), Q10 (product use demonstration) Overall: p<.001*** Player's: p<.001*** Export 'A ' :p< .001*** Rothmans: p<.01** H2 Q32 (product variations) Overall: p<.001*** Player's: p<.001*** Export ' A ' : p<.001*** Rothmans: p<.01** 194 H3 Q39 (celebrity depiction) Overall: p<.001*** Player's: p<.00T*** Export 'A ' :p< .001*** Rothmans: p<.01** H4a Q38 (purchase incentives, contests, premiums, coupons) Overall: p<.001*** Player's: p<.01** Export 'A ' :pvalue=.139 Rothmans: p<:05* H4b Q l (copy volume) Overall: p<.001*** Player's: p<.001*** Export ' A ' : p<.001*** Rothmans: p<.01** H5 Q42 (phone number/website listing) Overall: p<.001*** Player's: p<.001*** E x p o r t ' A ' : p<.001*** Rothmans: p value=.096 H6 Q12 (identification of partners, co-advertisers or supportive organizational bodies), Q40 (endorsements) Overall: p<.001*** Player's: p<.001*** Export ' A ' : p<.001*** Rothmans: p value=.151 H7 Q13 (health warning depiction) Overall: p<.001*** Player's: p<.001*** Export 'A ' :p<.001*** Rothmans: p<.001*** Hypothesis 8, which considered whether the most common model portrayals were exclusively males for Player's and Export ' A ' , and most frequently both males and females for Rothmans, was measured by Q21 (sex of people featured) of the data-coding instrument. Hypothesis 8 was supported for the Player's and Export ' A ' trademarks, but not confirmed for Rothmans. Notably, 91% (85/93) of Player's ads featured models. Among the 85 ads with people shown, 52 of them had exclusively males visible, while 33 featured both men and women. Interestingly, 32 of the ads portraying both men and 195 women were classified in the p re -TPCA era. Merely 2% (1/42) of the Player's ads in the post -TPCA era showed females, however. A s predicted, when people were portrayed in Export ' A ' ads, males exclusively was the most common depiction. People were present in 76% (62/82) of Export ' A ' ads, with females appearing in merely 6% (4/62) of the ads. For the four cases those women did appear in Export ' A ' advertising, the ads were classified in the post -TPCA era. In Rothmans ads, people were portrayed in 59% (26/44) of the time. Among the 26 Rothmans ads that included models, males exclusively were featured in 13 of them, while both males and females were represented in the other 13 ads. There were no Rothmans ads in the data set that exclusively featured women. Hypothesis 9, accounting for the predominant lifestyle dimensions that were expected for each trademark, was not fully supported. While adventure/excitement, independence, and ruggedness were themes with relative prevalence in Player's and Export ' A ' ads, \"friendship\" and \"physical activity\" were the value/lifestyle items that occurred with the highest frequency. \"Adventure/excitement\" was a theme evident in 45% (42/93) o f Player's ads and 30% (25/82) of Export ' A ' ads, while \"independence\" was apparent in 30% (28/93) of Player's ads and 32% (26/82) of Export ' A ' ads. \"Ruggedness\" was communicated in 16% (15/93) of Player's ads and 24% (20/82) of Export ' A ' ads. Notably, the \"friendship\" appeal was found to be present in 60% (56/93) of Player's ads, which was a distinguishing dimension since it was manifest in 4% (3/82) of Export ' A ' ads. \"Relaxation/escapism\" occurred in 26% (24/93) of Player's ads, although most of these occurrences were in the p re -TPCA era. \"Physical activity\" was the leading theme in both Player's and Export ' A ' ads, appearing 83% (77/93) and 51% (42/82) times, respectively. Surprisingly, no Rothmans ads were found to communicate 196 \"wealth/prosperity.\" The most common value/lifestyle theme expressed in Rothmans ads was \"tradition,\" although this dimension was only evident in ads from the p re -TPCA era. Hypothesis 10, which outlined that Player's would demonstrate the greatest consistency with communicating lifestyle dimensions that corresponded to its positioning, was not fully supported. A marked increase in \"ruggedness,\" \"adventure/excitement\", and \"tradition\" appeals is apparent during the pos t -TPCA era for the Player's trademark (p<.001). \"Independence\" and \"nationalism\" portrayals also increased with statistical significance in Player's sponsorship ads (p<.05). \"Physical activity,\" however, has been a predominant theme over time (i.e., apparent in 83% of Player's ads), evident in ads from both the pre- and post-TPCA eras. For Export ' A ' , a notable change in appeals is apparent for \"physical activity\" (p<.001). This dimension was evident in 31% (12/39) of Export ' A ' p re -TPCA ads, yet communicated in 61% (30/43) of post-TPCA ads. From the p re -TPCA era to the post-T P C A era, the \"relaxation/escapism\" appeal declined with statistical significance for the Export ' A ' trademark (p<.05). \"Independence,\" \"adventure/excitement,\" and \"ruggedness\" were communicated lifestyle themes in Export ' A ' advertising, although there were no noteworthy changes in frequency from the p re -TPCA era to the pos t -TPCA era. Despite upward status being an important component o f the Rothmans trademark identity, the \"wealth/prosperity\" dimension was not evident in Rothmans advertising. The \"tradition\" appeal that was found in the majority of p re -TPCA Rothmans ads was not evident in any post -TPCA sponsorship ads. 197 Discussion and Conclusions This chapter makes a unique contribution to the academic literature since it consists of the first content analysis study to examine non-American cigarette advertising with a longitudinal sample. To this date, Pollay (1990) provided the only comprehensive content analysis study analyzing the ad content of Canadian cigarette brands and magazines, and the design was cross-sectional. To my knowledge, the content analysis study reported in this chapter is the first to draw comparisons between the content of traditional product ads and sponsorship ads for tobacco products. When making general comparisons between ads from the pre- and pos t -TPCA eras, it is revealed that tobacco sponsorship ads are persistent in communicating lifestyle imagery. Physical activity portrayals, for example, have endured in Player's sponsorship ads and emerged as a prevalent theme in Export ' A ' promotions. Despite sports, athletic pursuits and outdoor activities prevailing in many post -TPCA tobacco promotions, health warnings are not mandatory and rarely depicted. The only occurrences of health warnings being shown in advertising from the pos t -TPCA era were entirely accounted by Export ' A ' ads circulating for the 2001 and 2002 extreme sports series. It was pointed out in Chapter 4 that this might be an orchestrated effort by JTI-Macdonald to exploit the alternative meanings of \" C I G A R E T T E S L E A V E Y O U B R E A T H L E S S \" when presented in the context of extreme sports and risk-taking behaviour. This endeavour may also inoculate against the primary meaning of the health-warning message. It is also evident from this study that brand types or line extensions are no longer explicitly depicted in the post -TPCA era ads. Rather, trademarks are utilized for tobacco sponsorship ads. The review of internal corporate documents revealed, however, that 198 when thinking about a trademark, consumers typically identify with a specific brand or line extension. Industry-cornmissioned research indicates that the brand family identification of Player's is concentrated on Player's Light, while Rothmans K i n g Size is the focus o f the entire Rothmans trademark. Interestingly, the 'identifying' brand of the Export ' A ' trademark is undergoing a transition. During the p re -TPCA era, the Export ' A ' trademark was best known for its parent full flavour brand, yet 'medium' has become the distinguishing line extension in many of its sponsorship promotions. This point illustrates the importance and value of establishing which primary and secondary colours are utilized in the artwork of promotions. While several p re -TPCA ads for Export ' A ' depicted a full flavour package or portrayed green as a border colour, blue is the primary colour used in the artwork of extreme sports series ads (see Appendix 30). The border enclosing the Lassie logo and the statement \"sponsored by: Export ' A ' \" closely resembles the packaging design. With each line extension of the Export ' A ' brand family bearing a different colour, the colour that predominates in ad visuals is rich in meaning. For Export ' A ' , green communicates full flavour, while blue communicates medium. 4 The shift toward sponsorship advertising has prompted the presence o f celebrities. During the p re -TPCA era, there was not a single ad that depicted a celebrity, reflecting that rule nine of the voluntary advertising code(s) set by the industry was being followed. Celebrities were portrayed in 43% of post -TPCA promotions, however. Reflecting that cigarette products are not depicted in sponsorship advertising, featured celebrities are not engaging in testimonials where direct product claims are made. Nevertheless, linkages between cigarette trademarks and celebrities may reflect both social approval (particularly in cases where the celebrity is a smoker) and efforts to have the presented 199 persona of the person transferable to the sponsoring brand. Recent sponsorship ads of Player's, for example, specify the names of auto racing team members even when they are depicted in non-auto racing settings (see Appendix 4 and 5). In addition to \"celebrity depiction,\" \"purchase incentives and contests\" and \"phone number or website listings,\" two other information dimensions notably emerged during the pos t -TPCA era: \"endorsements\" and \"availability.\" It is revealed that partner or other supporting sponsors and organizations are identified in roughly one-third of sponsorship promotions. Presenting co-sponsors may enhance the social acceptability of tobacco products, and reflect efforts toward strategically using cross-promotion and co-branding. The growth of availability appeals shows that approximately one-quarter o f sponsorship ads include assertions regarding when and where sponsored events can be watched (i.e., mention is made about which television network w i l l broadcast the event or the date and time of the broadcast). Meanwhile, statements about when, where or how cigarettes could be purchased were very infrequent during the p re -TPCA era. Finally, the shift to sponsorship advertising in the pos t -TPCA era has led to the disappearance o f taste appeals, which was once a central theme o f ad copy in cigarette advertising. Claims related to taste or flavour were evident in 84% of ads from the pre-T P C A era, yet no such appeals were observed in the pos t -TPCA era. While this change in the frequency of taste appeals is noteworthy, it is expected that taste or flavour would not be communicated in sponsorship ads since the depiction of cigarette products and packaging is disallowed. The implementation o f the T P C A seemed to prompt less information being communicated about product characteristics such as taste and line 200 extensions offered, with sponsorship advertising serving the primary purposes of associating trademarks with particular lifestyles or imagery. Wi th respect to analyzing the various trademark strategies, Player's ads in the post -TPCA era have seemingly become more image-driven, with appeals central to the trademark's positioning (i.e., independence, ruggedness, adventure, and tradition) more frequently communicated. The identity of Player's is increasingly masculine, with women rarely shown in sponsorship ads. Success and nationalism appeals have also emerged, l ikely reflecting Player's sponsorship of auto racers that are Canadian in origin, who have largely performed very well . I f nationalism were becoming a part o f the Player's trademark identity, the partnership with Molson Canadian would appear very desirable for Imperial Tobacco. Physical activity portrayals were always common in Player's advertising, thus visual depictions of sports in the era o f sponsorship advertising were not a transitory ad platform, nor jarring to the target consumer. The transition to sponsorship advertising may not have been so smooth for Export ' A ' , however. The frequency of physical activity depictions nearly doubled from the pre-T P C A era to the pos t -TPCA era. It is not clear from the data analyses whether the transition to sports portrayals largely occurred due to the implementation of the T P C A , or alternatively shows a competitive ('me-too') response to Player's. The review of internal corporate documents suggests that it was a competitive response, commencing with the \" A Taste for Adventure\" campaign. Success has emerged as a theme in sponsorship ads for Export ' A ' , while pleasure has declined. This observation seems to reflect that sponsorship ads often consist of more competitive narratives (e.g., athletes wearing bibs with numbers, recognition of the winner with a trophy) than physical activity depictions 201 seen in conventional advertising. Pleasure was an appeal commonly associated with the flavour of cigarettes. The small sample size of Rothmans sponsorship advertising echoes the low promotional spending and presence of the trademark, as revealed in the internal industry documents. Rothmans ads during the pre-TPCA era placed particular emphasis on the worth o f the product, with 43% (13/30) including quality appeals and 90% (27/30) making taste or flavour claims. A s the transition to sponsorship advertising occurred in the late 1980s, direct product assertions were no longer permissible and the image platform of the Rothmans trademark was already in crisis. Admittedly, the overall results regarding Hypotheses 9 and 10 (i.e., determining predominant trademark images and assessing whether Player's demonstrated the greatest consistency with images being communicated) were not as compelling as anticipated. The results may reflect a key limitation of this study. Information categories and value analysis items listed in the data-coding instrument were assessed for mere presence, thus claim repetitions were not acknowledged. Consequently, an ad utilizing the tagline, \" A taste you can call your own\" and portraying one person doing an individual sport (yet wearing a knapsack revealing a sewn badge of the Canadian flag) would likely be coded as depicting both \"independence\" and \"nationalism.\" Attempts w i l l be made to address this 'weighting' issue i f further content analysis studies are conducted. Nevertheless, content analysis remains a useful research method since it allows qualitative data to be put in quantitative terms, providing objective and systematic answers to the question, \"What is going on in the studied ads?\" This research method also helps identify patterns within large amounts of data that would be difficult to detect otherwise. Semiotics is an alternative method that is valuable for studying texts, albeit this methodology comes its own set of limitations (i.e., it is problematic for studying large quantities of ads). Leiss, Kline, and Jhally (1997) propose a combined semiological/content analysis approach, which is worth pursuing further. 203 Endnotes 1 Ads from 1989 to 1995 will presumably use a corporate entity as a sponsorship identifier. Once the Supreme Court of Canada ruled the TPCA unconstitutional in September 1995, however, corporate entities were no longer required for sponsorship purposes. 2 The only ads held by the Archives that were excluded from the data set were traditional product ads that circulated in 1996. Following the 1995 Supreme Court of Canada ruling that declared the TPCA unconstitutional, the Canadian tobacco industry embraced a new voluntary advertising code and resumed conventional advertising in February 1996. Conventional cigarette ads circulated for roughly one year before the Tobacco Act was implemented. These ads, which amounted to roughly one ad campaign per trademark, were not included in the data set because the objective of the study was to compare the content of traditional product advertising from the pre-TPCA era with sponsorship ads from the post-TPCA era. Cohen's kappa, while not without its own limitations, is an alternative estimation approach that explicitly recognizes the likelihood of chance agreement between coders and withdraws it from consideration. However, this estimation approach is designed for measuring agreement between two coders. 4 Kool, an American trademark, employs a similar strategy of using colour and sponsorship properties to communicate particular brands or line extensions. Kool sponsored two auto racing drivers competing in the CART circuit during 2002. One 204 K o o l racing car was painted green (suggestive of 'Filter Kings ' ) while the other was painted blue (indicative o f the ' M i l d s ' line extension). 205 CHAPTER EIGHT Conclusion This thesis has utilized an interdisciplinary and multi-method approach. The review of internal corporate documents reveals that the two key typologies of cigarette consumers are \"starters\" (i.e., initiation typically happens during adolescence) and \"pre-quitters\" (i.e., the reassurance of existing smokers). Player's and Export 'A' represent two Canadian trademarks that have been effectively positioned toward the \"starters\" segment. Like most commodities, cigarettes are not a generic item since they are designated trademark and brand names that compete with one another for specific market segments. By choosing to smoke a particular brand, an individual engages in an act of distinction. Player's is widely seen a symbol of masculinity, youthfulness, independence, freedom, self-reliance, tradition, and modernity. The trademark attributes and personality of Export 'A' are masculinity, ruggedness, independence, self-determinedness, adventurousness, and escapism. While the two competing trademarks have similar image platforms, Player's is considered to have a softer masculine image (holding appeal for women despite its target market of male youth) and adventure themes that are more understated. Export 'A' has an edgier identity, with its users largely perceived as exclusively male who in some cases are daredevils. The advertising budgets of Player's are far superior to those demonstrated for Export 'A', thus reinforcing and elevating consumer perceptions about the popularity and social acceptability of Player's. Popularity is considered to be a crucial factor in brand desirability among youth. 206 Rothmans, on the other hand, exemplifies a Canadian trademark that is non-appealing to youth. The trademark has been promoted as an expression of internationalism, premium quality, upward status and tradition, yet youth widely perceive Rothmans as unpopular, ' o ld ' , and lacking contemporaneous. Rothmans has a low presence or 'share o f mind ' among consumers, and youth generally (and accurately) consider Rothmans smokers to be remarkably older. The ineffectiveness of Rothmans promotional efforts partially reflected that advertising budgets were insufficient and few modifications were made to the packaging design. The lack of package updates was in conflict with the trademark's supposed premium quality identity and contributed to the 'o ld ' perception. Consumers have multiple sources of information about the meaning of brands and trademarks, and in the case of Rothmans, the ad appeals were not consistent with consumer experiences. Wi th its relatively high tar delivery and strong taste, Rothmans is also a non-desirable selection for those demonstrating elevated health concerns and seeking 'reassurance'. The case studies of Player's, Export ' A ' , and Rothmans also lend insight about how Canadian cigarette trademarks are uniquely positioned toward males and females. Both product features and the advertised image largely determine the masculine-feminine dichotomy o f Canadian cigarette brands and trademarks. Brands offering relatively high tar content and strong taste or full flavours are commonly perceived as 'masculine', which often correspond with promotional appeals that have an action, exciting, and adventurous orientation. Conversely, low tar, mi ld tasting, mentholated, longer length, and slim cigarettes are characterized as 'feminine' product characteristics, which often carry image platforms relating to relaxation, stress relief, self-indulgence, and inactive 207 pursuits.1 Canadian trademarks and brands with cross-gender positioning often employ-promotional appeals about upward status and being upscale. Player's and Export ' A ' clearly portray a masculine identity with their image platforms and the exclusion o f mentholated and 100 mm length line extensions (although so-called ' l ight ' products are offered for Player's and Export ' A ' , the machine-measured tar deliveries are higher relative to other competing light brands in the Canadian marketplace). Rothmans, however, communicates contradictory messages about whether the target consumer is male, female or both. The trademark is recognized as a high tar and strong tasting cigarette yet includes 100 mm and mentholated line extensions. The identity of Rothmans is supposedly gender neutral, but industry research indicates that Rothmans consumer demographics are over-represented by women aged over 50. The content analysis study reported in Chapter 7 revealed that when models were portrayed in Rothmans advertising, males exclusively were depicted in one-half o f the ads, while both males and females were represented in the other half. Interestingly, no Rothmans ads featured exclusively women. Another point made known in the review of internal corporate documents is that colours characterizing a trademark and its packaging design are an important component o f the trademark's identity and essence. In the Canadian marketplace, hues o f blue are used for a remarkable number of brands and trademarks, including Player's, Export ' A ' , Rothmans, Belvedere, Belmont, and Canadian Classics. According to Rothmans, Benson & Hedges, \"Blue may be becoming over-used, especially in Quebec\" (1998, p.5). Blue is the predominant colour for the core brands o f the Player's, Export ' A ' and Rothmans trademarks, which are Player's Light, Export ' A ' Medium and Rothmans K i n g Size, 208 respectively. In the trademark battle toward being 'top of mind ' among consumers with the colour blue, Player's seems to be the clear winner due to its popularity and wel l -established history. In addition, hues of blue are utilized for the entire Player's brand family, unlike the Export ' A ' and Rothmans trademarks which use an assortment of colours for the various line extensions. Rothmans Special M i l d uses red as a defining colour, which seems unusual since red normally communicates strong flavour with cigarettes (or spiciness with food). Reflecting industry responses to the Canadian regulatory environment, sponsorship became a key component of the promotional mix for Canadian tobacco firms. Trademarks, rather than brand line extensions, were used to identify tobacco sponsorships. Industry research shows, however, that a particular brand comes to mind when consumers think of a cigarette trademark. To maintain the credibility of the message and prevent counter-argumentation among consumers, tobacco firms purposely sponsor events or portray activities within sponsorship ad visuals that are not too aerobically taxing. The symbolic value of brands or trademarks maybe enriched through sponsorship by selecting events, partnerships or celebrities (i.e., event participants) possessing symbolic imagery or 'personalities' that are transferable to the respective trademarks. The enhancement or reinforcement of brand imagery is clearly a primary objective of tobacco sponsorship promotions. Manufactured by Canada's largest tobacco firm, Imperial Tobacco, the Player's trademark is particularly recognized for its effective, integrated communication strategies, exploiting inter-textual relations in its promotions. Effective communication of brand image is based on principles relating to message repetition, continuity, consistency, and relevance. First, the repetitious principle is served 209 by the large advertising budget of Player's, contributing to a persistent and pervasive communications mix that evokes social acceptability, popularity, and builds 'friendly familiarity'. Auto racing, a sport characterized by high operating budgets and the ability to generate high visibility for sponsors, is the cornerstone of Player's sponsorship properties. Second, Player's has a long, well-established history, predating the incorporation of Imperial Tobacco in 1912, which adds to the continuous tenet. Third, Player's has been linked with consistent images (i.e., masculinity, independence, freedom, self-reliance, tradition, modernity) in multiple media over a sustained period of time. The content analysis study indicates that post -TPCA era sponsorship ads for Player's have become even more image-driven. Individual sports are depicted to communicate independence and self-reliance, while freedom is expressed by portrayals of vast, outdoor settings. Fourth, while the image platform has been steady, the ways in which the images are communicated have been relevant and contemporary. The Player's packaging design, for example, has been updated frequently and the trademark is associated with present-day athletes through its sponsorship initiatives. Canada's other two major tobacco manufacturers have demonstrated inferior image platforms with respect to continuity and consistency for their flagship trademarks. Following the implementation of the T P C A , the major sponsorship properties for Export ' A ' (manufactured by JTI-Macdonald Corp.) have been an extreme sports series, The Skins Game (feahrring premiere male professional golfers such as Jack Nicklaus, Greg Norman, and David Duval), The New Music Series, and salmon fishing showdowns. The Skins Game, which was once the highest profile event sponsored by Export ' A ' in terms of promotional expenditures, did not communicate images of adventure, a central part of the 210 trademark's essence. Moreover, many of the golfers associated with Export ' A ' were unlikely to be inspiring for the trademark's target audience of males less than 25 years old. Similarly, salmon fishing showdowns would be unappealing to many male youth. The New Music Series was comprised of several bands performing at indoor settings, which conflicted with the independence and escapism (i.e., in the 'great outdoors') notions of Export ' A ' . The extreme sports series, which is now the focal point of Export ' A ' sponsorship plans, seems to be an effective way of communicating the trademark's identity. Beyond Export ' A ' , JTI-Macdonald does not offer any trademarks that possess a substantial market presence. Rothmans, once the flagship trademark of Rothmans, Benson & Hedges, has seen its market share decline considerably and sponsorship expenditures for the trademark are now nearly non-existent. During 1997, Rothmans sponsored Jacques Villeneuve in the Formula One racing circuit. The objectives of this high profile sponsorship initiative seemed unfulfilled since the association between Villeneuve and Rothmans was short-lived, lasting merely one season. It is hard to imagine that the link between Rothmans and Villeneuve would linger in consumers' minds, especially since the auto racer has been sponsored by several additional cigarette trademarks during his career, including Player's, Winfield, 555, and Lucky Strike. In addition, Villeneuve and Rothmans did not project complementary images. During the season that Villeneuve was sponsored by Rothmans, Donaldson (1997) observed that: On the track, Villeneuve's highly developed fighting spirit and daredevil approach to his profession invariably enlivened the proceedings... His penchant for speaking his mind was matched by a colourful approach to his personal 211 appearance that seemed more suitable for a rock star than a race driver. Wi th his hair dyed blond and his high grunge clothing, Villeneuve might have upset elements of the conservative F l establishment, but such antics further endeared him to a legion o f younger fans and his popularity soared. (p.C9) While Villeneuve's youthful, 'w i ld child ' and rebellious image is a desirable match for several cigarette trademarks, it does not seem credible for Rothmans. Moreover, the Rothmans trademark is recognized for symbolizing British heritage, which contrasts with Villeneuve's French-Canadian background. Finally, sponsorship ads depicting Villeneuve utilized the lozenge logo, which communicates old, feminine and conservative images according to industry-commissioned research. A total Canadian tobacco sponsorship ban w i l l be imposed in October 2003, however, in accordance to amendments that were made to the Tobacco Act (i.e., B i l l C -42). The Tobacco Act was implemented in 1997 and established as a replacement of the T P C A . The Act w i l l set a ban on lifestyle advertising, yet factual, informational advertising on the basis o f the product's characteristics, price or availability w i l l remain permissible assuming that it is placed in adult establishments, in publications with a minimum adult readership of 85% or in mailings addressed to adults by name. Canada's three largest tobacco manufacturers have challenged the constitutionality of the Tobacco Act, appealing a Quebec Superior Court decision that upheld the legislation. It is likely that the Quebec Court of Appeal and the Supreme Court of Canada w i l l eventually hear the case. The future of tobacco promotion in Canada remains uncertain. Assuming lifestyle advertising and tobacco sponsorship are banned, Canadian tobacco firms w i l l undoubtedly utilize other marketing strategies in an attempt to 212 continue communicating imagery for their respective trademarks. Richard Pollay, a marketing professor at the University of British Columbia, remarks, \"it 's like squeezing a balloon. Y o u can shut down one media, but the problem just moves somewhere else\" (cited in Herring, 1999, p.2). This point is echoed by Saffer and Chaloupka (2000), who argue that a limited set of advertising bans do not slow down advertising output, but rather lead to shifts in media spending by the tobacco industry. In other words, when one media form is prohibited, the tobacco industry simply finds media \"substitutes.\" It was observed that the exclusion or withdrawal o f other forms of promotion precipitated the shift to sponsorship. A sponsorship ban is expected to bring about further emphasis on point-of-sale strategies, trademark diversification, and direct marketing campaigns. The tobacco industry's proven ability to circumvent the 'spirit' o f policies also points to the necessity for monitoring whether future factual information-based ads comply with the stipulations o f the Tobacco Act . Pol icy initiatives are l ikely to be ineffective i f proper enforcement procedures and resources are not put in place. Future Directions for Research This thesis has limited its focus to federal tobacco control policies in Canada. One direction for future research is to account for provincial tobacco control policies and government jurisdictional issues. According to the Economic Counci l o f Canada, \"the growth of government regulatory activity is the growth of the interdependence between the federal and provincial governments... There are few areas of policy making where one government acts alone\" (1985, p.166). Schultz and Alexandroff (1985) observe that as the function of regulation has evolved, intergovernmental conflict has either ensued or 213 increased. This intergovernmental conflict may take the form of federal-provincial or inter-provincial dissent. Issues that are classified as both federal and provincial jurisdiction may result in \"duplication, overlap, inconsistency, and confusion in regulatory requirements imposed on individuals and firms in the private sector and in the regulatory activities of the two levels of government\" (Economic Council o f Canada, 1985, p. 166). These arguments certainly have relevance and applicability for tobacco control policies, as there are examples in which federal and provincial policies are not necessarily consistent with one another. For example, the federal Tobacco Act stipulates that 18 and over are the legal ages to smoke, while some provinces such as British Columbia and Ontario require smokers to be at least 19 years old. A s inconsistencies such as these develop, intergovernmental conflict is more likely. Another policy-oriented area of study that would have been interesting to expand upon, but was not within the realm o f my thesis, includes utilizing a public policy analysis framework to examine the Tobacco Act. Such a framework would account for the determinants, the content, and the implementation of a policy. Finally, a comparative analysis of Canadian and Australian tobacco control policies would prove useful. Under the provisions o f the Tobacco Advertising Prohibition Act of 1992 (and amendments made in 1995), tobacco advertising and sponsorship is now virtually prohibited in Australia. Considering Australia exemplifies a jurisdiction at a more advanced stage of advertising and sponsorship regulation, there are presumably many lessons to be shared about effectively implementing policy, as well as anticipating how tobacco firms w i l l l ikely respond to their regulatory environment. 214 Reviewing tobacco industry documents presents several additional opportunities for further research. For this thesis, the reviewed internal corporate documents were accessible primarily as a result o f the 1989 Canadian trial to resolve the constitutionality of the T P C A and the 2002 Quebec Superior Court trial to decide the constitutionality of the Tobacco Act . More Canadian tobacco industry documents are l ikely to be disclosed in time as a result of further litigation. In addition, two document depositories - The Guildford Depository and the Minnesota Tobacco Document Depository - were established in 1998 as a result of a court judgement in Minnesota, a trial involving the Minnesota State Attorney General (and insurers) as plaintiffs and the U . S . tobacco industry as defendants. The Guildford Depository, located in Guildford, England, possesses over six mil l ion pages of internal documents (40,000 files) from British American Tobacco ( B A T ) and its subsidiaries. The Minnesota Tobacco Document Depository is located in Minnesota, U S A and houses more than 33 mi l l ion pages of documents, representing the largest public collection of tobacco industry documents in the world. A l l documents in the Minnesota depository are from US-based tobacco manufacturers such as Philip Morris , Brown & Williamson, RJ Reynolds, and Lorillard. It should be obvious that a wealth o f information awaits researchers wishing to use industry documents as a resource. This thesis has provided case studies of the marketing strategies employed for the Player's, Export ' A ' , and Rothmans trademarks. Priorities for future industry document research include an examination of the marketing histories of additional trademarks that are prominent in the Canadian marketplace, including du Maurier, Matinee and Benson & Hedges. Considering that the focus of this thesis has been Canadian cigarette trademarks 215 that are primarily positioned toward male youth, exploring the marketing strategies of trademarks with a female target market would be useful for comparative purposes. The industry documents may also inform about the role of other elements o f the promotional mix (i.e., packaging, point-of-sale signage, direct marketing, and branding elements such as logos and trademark names), the meaning of various product descriptors (i.e., light, mild , smooth, natural), the implications of different product features and categories (i.e., menthol or s l im cigarettes, products supposedly offering harm reduction), and additional segmentation strategies being utilized (i.e., approaches according to the consumer's sexual orientation or primary language spoken). A l l o f these research suggestions put forward have applicability beyond Canada. Content analysis is another research methodology that may be utilized for future research. Additional Canadian-based studies could include more trademarks in the data set (particularly trademarks with a female target market) and a greater number of eras under examination. B y extending the assessment of cigarette advertising in the Canadian print media to the 1950s and 1960s, conclusions could be drawn about the impact of various health/smoking controversy events and the industry's withdrawal from the broadcast media in 1972. In addition, content analysis can be used to measure tobacco use portrayals in television, film, and literature. Following the recommendations o f Leiss, Kl ine , and Jhally (1997), a combined semiological/content analysis approach could be explored. 216 Endnotes 1 While menthol cigarette smokers are primarily women in Canada, this is inconsistent with consumption figures in the United States. Mentholated cigarettes are particularly popular among African American smokers, with brands such as K o o l having 'masculine' image platforms. 217 References Aaker, D . A . (1996). Building strong brands. New York: The Free Press. Aaker, D . A . & Joachimsthaler, E . (2000). Brand leadership. N e w York : The Free Press. Abrams, M . H . (1993). A glossary of literary terms (6 t h edition). Toronto, O N : Ffarcourt Brace College Publishers. Albright, C . L . , Altman, D . G . , Slater, M . D . , & Maccoby, N . (1988). Cigarette advertisements in magazines: Evidence for a differential focus on women's and youth magazines. Health Education Quarterly, 15, 225-233. Altman, D . G . , Schooler, C , & Basi l , M . D . (1991). Alcohol and cigarette advertising on billboards. Health Education Research, 6, 487-490. Altman, D . G . , Slater, M . D . , Albright, C . L . , & Maccoby, N . (1987). H o w an unhealthy product is sold: Cigarette advertising in magazines, 1960-1985. Journal of Communication, 37, 95-106. Andreeff, M . (2000, November). Huck off. Ski Canada, 29, 106-108. Ang , I. (1996). Living room wars: Rethinking media audiences for a postmodern world. N e w York : Routledge. Arens, W.F . (1999). Contemporary advertising (7 t h Edition). Boston, M A : Irwin/McGraw-H i l l . Armada. (1999, Apr i l ) . 1998 Project Nike: A n analysis of competitive advertising in the tobacco category. Prepared for Imperial Tobacco Ltd. Exhibit D-208, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. 218 Audet-Lapointe, C . (1991, August). Switching analysis. Prepared by Market Analysis Group for Imperial Tobacco Limited. Exhibit D-179, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates N o . 465043159-465043217. Bang, K . M . & K i m , J .H. (2001). Prevalence of cigarette smoking by occupation and industry in the United States. American Journal of Industrial Medicine, 40, 233-239. Basi l , M . D . , Basi l , D . Z . , & Schooler, C . (2000). Cigarette advertising to counter New Year 's resolutions. Journal of Health Communication, 5, 161-174. Bateman, M . , Higgins, Jr., T., Girault, E . , Hunt, A . , John, D . , & Salisbury, A . (1994, October 24/25). 1994/95 business assessment. Prepared for Rothmans, Benson & Hedges Inc. Exhibit D-177, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 003773-004112. Baxter, C. & Touche, R. (1958, March 22). Fight has only started in cigarette sales war. The Financial Post. The Beaumont Organization, Ltd. (1982, Apri l ) . Cigarette market strategy research 1981. Prepared for RJR-Macdonald Inc. Exhibit RJR-196, RJR-Macdonald Inc. v. Canada (Attorney General). Berelson, B . (1971). Content analysis in communication research. N e w York: Hafher Publishing Company. 219 Bingham, P . M . (1992, July 27). Canada: Du Maurier [letter addressed to J. Rembiszewski, D . Brown, and R. Cooper]. British-American Tobacco Company Limited. Bates No. 500028180. The Blue Book o f Canadian Business. (2002). Blue book's featured corporate profile: JTI-Macdonald Corp. Online http://www.bluebook.ca/profiles/338lbody.html#. Accessed August 20, 2002. B lum, A . (1991). The Marlboro Grand Prix: Circumvention of the television ban on tobacco advertising. The New England Journal of Medicine, 324, 913-917. British American Tobacco Co. (1976, March 29). The product in the early 1980s: British American Tobacco Co. (1984, October 1). Potential areas of investigation activity. Bates No. 102694881. British American Tobacco Co. Circa 1985. R&D/marketing conference. Bates N o . 100501581-100501658. British American Tobacco Co. Circa 1989. Player's 1990 marketing plans.. Prepared for Imperial Tobacco Ltd. British American Tobacco Co. Circa 1991. Canada. Bates No. 500118797. Brown, B . (1991). We wi l l advertise: Imperial Tobacco jumps on Canadian ruling. Advertising Age, 62, p.9. Brown, S., Stevens, L . , & Maclaran, P. (1999). I can't believe it's not Bakhtin!: Literary theory, postmodern advertising, and the gender agenda. Journal of Advertising, 28,11-24. Brown-John, C L . (1981). Canadian regulatory agencies. Toronto, O N : Butterworths. 220 -Budd, R . W . , Thorp, R . K . , & Donohew, L . (1967). Content analysis of communications. New York : The Macmil lan Company. Canada. (1988). Tobacco Products Control Act. Ottawa, O N : Dept. of National Health and Welfare. Canada. (1997). Tobacco Act. Ottawa, O N : Public Works and Government Services Canada. Canada. (1998). An Act to Amend the Tobacco Act. Ottawa, O N : Queen's Printer for Canada. Canada's best. (1997, December 17). The Toronto Star, p .E lO. Canadian Tobacco Manufacturers' Council . (1964). Cigarette advertising code of Canadian tobacco manufacturers. Ottawa, O N . Canadian Tobacco Manufacturers' Council . (1971, September 15). Cigarette advertising code of the Canadian Tobacco Manufacturers'Council. Ottawa, O N . Canadian Tobacco Manufacturers' Council . (1976). Cigarette & cigarette tobacco advertising and promotion code of the Canadian Tobacco Manufacturers' Council. Ottawa, O N . Canadian Tobacco Manufacturers' Council . (1984). Cigarette & cigarette tobacco advertising and promotion code of the Canadian Tobacco Manufacturers' Council. Ottawa, O N . Canadian Tobacco Manufacturers' Council. (1987). A brief to the Legislative Committee of the House of Commons on Bill C-51. Ottawa, O N . Canadian Tobacco Manufacturers' Council . (1994). Canadian Tobacco Manufacturers' Council. Ottawa, O N . Canadian Tobacco Manufacturers' Council. (1997). Brief on Bill C- 71 to the Standing Senate Committee on Legal and Constitutional Affairs. Ottawa, O N . Canadians suffer fall at windy Matinee. (1994, August 16). The Globe and Mail, p.C7. Canova D . , Myers, M . L . , Smith, D . E . , & Slade, J. (2001). Changing the future of tobacco marketing by understanding the mistakes of the past: Lessons from \"Lights\". Tobacco Control, 10, i43-i44. Chacra, A . (1980, M a y 5). Player's trademark F'81 advertising. Prepared for Imperial Tobacco Ltd. Exhibit A G - 3 5 , RJR-Macdonald Inc. v. Canada (Attorney General). Chapman, S. (1986). Great expectorations: Advertising and the tobacco industry. London: Comedia. Charlton, A . , While, D . , & Kel ly , S. (1997, November 15). Boys ' smoking and cigarette-brand-sponsored motor racing. Lancet, 350, 1474. Chrzanowski, P. (1990, December). Degrees of extreme. Vancouver, p.36. Cohen, J .B. (1989). Counting advertising assertions to assess regulatory policy: When it doesn't add up. Journal of Public Policy and Marketing, 8, 24-29. Cohen, J .B. (1992). Research and policy issues in Ringold and Calfee's treatment of cigarette health claims. Journal of Public Policy and Marketing, 11, 82-86. Collishaw, N . E . & Myers, G . (1984). Dollar estimates of the consequences o f tobacco use in Canada, 1979. Canadian Journal of Public Health, 75, 192-199. Consumer shares by brand family (19-24 year olds). No date. Exhibit P - l 12, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Copeland, R. , Frisby, W. , & McCarvi l le , R. (1996). Understanding the sport sponsorship process from a corporate perspective. Journal of Sport Management, 10, 32-48. Cornwell, T . B . (1995). Sponsorship-linked marketing development. Sport Marketing Quarterly, 4, 13-24. Cornwell, T . B . & Maignan, I. (1998). A n international review o f sponsorship research. Journal of Advertising, 27, 1-21. The Creative Research Group Ltd. (1988, August 15). Tracking study: 1988. Prepared for Imperial Tobacco Ltd. Exhibit D-090, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates N o . 016172-016250. The Creative Research Group Ltd. (1990, September 19). 3 N 1990. Prepared for Imperial Tobacco Ltd. Exhibit D-091, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. The Creative Research Group Ltd. (1991, September 9). 3 N 1991. Prepared for Imperial Tobacco Ltd. Exhibit D-093, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. The Creative Research Group Ltd. (1992, September 2). 3 N study 1992 (Volume 1: Highlights) [also referred to as \"Youth Target 1992\" study]. Prepared for Imperial Tobacco Ltd. Exhibit D-094, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 018273-018383. Crompton, J .L. (1993). Sponsorship of sport by tobacco and alcohol companies: A review o f the issues. Journal of Sport and Social Issues, 17, 148-167. Cummings, K . M . , Hyland, A . , Lewit, E . , & Shopland, D . (1997). Discrepancies in cigarette brand sales and adult market share: Are new teen smokers fill ing the gap? Tobacco Control, 6, S38-S43. Cummings, K . M . , Morley, C P . , Horan, J .K. , Steger, C , & Leavell, N - R . (2002). Marketing to America's youth: Evidence from corporate documents. Tobacco Control, 11, i 5 - i l 7 . Cunningham, R. (1996). Smoke and mirrors: The Canadian tobacco war. Ottawa, O N : International Development Research Centre. D'Alessandro, D.F . (2001). Brand warfare: 10 rules for building the killer brand. New York: McGraw-Hi l l . Danesi, M . (2000). Encyclopedic dictionary of semiotics, media, and communications. Toronto, O N : University of Toronto Press. Davidson, J. (1994, August 17). Graf humbles American, contemplates life without tennis. The Globe and Mail, p.C5. Denis, A . (2002, December 13). Quebec Superior Court decision rendered by Judge Andre Denis. JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Descoteaux, M . (1989, February 16). Imperial Tobacco incorporates four new corporate entities. Montreal: press release by Imperial Tobacco Limited. Dewhirst, T. (1999). Tobacco sponsorship is no laughing matter. Tobacco Control, 8, 82-84. Dewhirst, T. (in press). Male youth, extreme sports, and the gendering o f smoking: A case study of Export ' A ' \u00E2\u0084\u00A2 brand marketing in Canada. In L . Fuller (Ed.), Sexual Sports Rhetoric Globally. Binghamton, N Y : Haworth Press. Dewhirst, T. (in press). Smoke and ashes: Tobacco sponsorship of sports and regulatory issues in Canada. Chapter in L . R . Kahle & C . Ri ley (Eds.), Sports Marketing and the Psychology of Marketing Communication. Mahwah, N J : Lawrence Erlbaum Associates, Inc. Dewhirst, T. & Hunter, A . (2002). Tobacco sponsorship of Formula One and C A R T auto racing: Tobacco brand exposure and enhanced symbolic imagery through co-sponsors' third party advertising. Tobacco Control, 11, 146-150. Dewhirst, T . L . & Pollay, R . W . (2001). Content analyses o f cigarette advertising: A critical review of the literature. In C R . Taylor (Ed.), The Proceedings of the 2001 Conference of the American Academy of Advertising (pp.46-51). Vil lanova, P A : American Academy of Advertising. Dewhirst, T. & Sparks, R. (in press). Intertextuality, tobacco sponsorship of sports, and adolescent male smoking culture: A selective review of tobacco industry documents. Journal of Sport and Social Issues. Donaldson, G . (1997, December 28). Formula year: Jacques the rebel kept it interesting up to w i ld final. The Toronto Star, p.C9. Durand, J. (1987). Rhetorical figures in the advertising image. In J. Umiker-Sebeok (Ed.), Marketing and Semiotics: New Directions in the Study of Signs for Sale (pp.295-318). New York: Mouton. . 225 Durnan, B . , Fitzgerald, M . , & Davis, R. (1997). Canadian advertising success stories: Molson Canadian 'I AM'. Toronto, O N : Canadian Congress o f Advertising. Dyer, G . (1982). Advertising as communication. New York: Methuen. Economic Council o f Canada. (1985). Intergovernmental relations and regulation. In K . Kernaghan (Ed.), Public Administration in Canada (5 t h edition) (pp. 166-170). Toronto, O N : Methuen. Edgar, A . & Sedgwick, P. (1999). Key concepts in cultural theory. London: Routledge. Feinberg, B . M . (1971). Content analysis shows cigarette advertising up twofold in 14 magazines. Journalism Quarterly, 48, 539-542. Ferrand, A . & Pages, M . (1996). Image sponsoring: A methodology to match event and sponsor. Journal of Sport Management, 10, 278-291. Fiske, J. (1987). Television culture. New York: Routledge. Fiske, J. (1989). Understanding popular culture. New York: Routledge. Fiske, J. (1990). Introduction to communication studies (2 n d edition). New York : Routledge. Fitzgerald, K . (1994, October 17). Brands wed marketing plans. Advertising Age, 65, p.30. Fraser, G . (1977, June 13). Advertising: The French have a word for it. Maclean's, p.68-69. Gladden, J . M . & Wolfe, R. (2001). Sponsorship of intercollegiate athletics: The importance o f image matching. International Journal of Sports Marketing & Sponsorship, 3, 41-65. Glantz, S.A., Slade, J., Bero, L . A . , Hanauer, P., & Barnes D . E . (1996). The cigarette papers. Berkeley: University of California Press. Gossage, H . L . (1967). The gilded bough: Magic and advertising. In F . W . Matson & A . Montagu (Eds.), The Human Dialogue: Perspectives on Communication (pp.363-370). New York: The Free Press. Gross, G . (1994). The feeling's mutual. Racquet Sports Magazine, 9, 66-71. Gwinner, K . P . & Eaton, J. (1999). Building brand image through event sponsorship: The role of image transfer. Journal of Advertising, 28, 47-57. Ha l l , S. (1980). Encoding/decoding. In S. Hal l , D . Hobson, A . Lowe, & P. Wi l l i s (Eds.), Culture, media, language: Working papers in cultural studies, 1972-79 (pp.128-138). London: Hutchinson. Harrod & M i r l i n Ltd. (1995, December 12). Export \" A \" franchise: Advertising strategy recommendations. Prepared for RJR-Macdonald Inc. Exhibit D-178, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Harrod & M i r l i n . (1996, September 9). Export \" A \" 1997 communications plan. Prepared for RJR-Macdonald Inc. Exhibit D-183, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 801531139-801531162. Harrod & M i r l i n . (1996, A p r i l 26). R J R Macdonald Inc.: Vantage strategic recommendation. Exhibit D-203, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 80150 2730-80150 2771. 227 Hastings, G . & MacFadyen, L . (2000). A day in the life o f an advertising man: Review of internal documents from the U K tobacco industry's principal advertising agencies. British Medical Journal, 321, 366-371. Hawaleshka, D . (2001, June 25). Grand prix wizardry. Maclean's, 114, 40-41. Health Canada. (1996). Youth smoking survey, 1994: A technical report. Ottawa, O N : Minister o f Supply and Services Canada. Herring, H . B . (1999, A p r i l 25). Signs of bygone days. The New York Times, Sec.4, p.2. Hirschman, E . C . , Scott, L . , & Wells, W . B . (1998). A model of product discourse: Linking consumer practice to cultural texts. Journal of Advertising, 27, 33-50. Holsti , O.R. (1969). Content analysis for the social sciences and humanities. Don M i l l s , O N : Addison-Wesley Publishing Company. Hughes, M . A . & Garrett, D . E . (1990). Intercoder reliability estimation approaches in marketing: A generalizability theory framework for quantitative data. Journal of Marketing Research, 27, 185-195. Imasco Ltd. (1970). Annual report 1969. Imasco Ltd. (1971). Annual report 1970. Imasco Ltd. (1972). Annual report 1971. Imasco Ltd. (1974). Annual report 1973. Imasco Ltd. (1987). Annual report 1987. Imasco Ltd. (1989). Annual report 1988. Imasco Ltd. (1999). 1998 annual report. Imperial Tobacco Company of Canada, Limited. (1964, February 17). Annual report 1963 [Report to the shareholders by E . C . Wood & J . M . Keith]. 228 Imperial Tobacco Ltd. Circa 1970. 1971 Matinee marketing plans. Exhibit AG-204 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1979a. F i sca l ' 80 media plans. Exhibit ITL-13, RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1979b. Creative guidelines. Exhibit A G - 2 9 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1980a. Player's trademark strategy. Imperial Tobacco Ltd. Circa 1980b. Player's filter '81 creative guidelines. Exhibit A G -35, RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1980c. Alan\u00E2\u0080\u0094creative guidelines F '82 . Exhibit A G - 2 9 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1983. F'85 marketing plan for Player's. Imperial Tobacco Ltd. Circa 1985. Project Stereo final report. Exhibit A G - 2 7 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1987. Player's 1988. Exhibit AG-210 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1987. Overall market conditions\u00E2\u0080\u0094F88. Exhibit AG-214 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. 1989. The Canadian tobacco market at a glance. Exhibit A G - 3 1 , RJR-Macdonald Inc. v. Canada (Attorney General). Imperial Tobacco Ltd. Circa 1992. Autosport Player's Ltee racing. Exhibit D - l 8 1 , JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 011802-011849.' Imperial Tobacco Ltd. (1992). Sponsorships - communications plans 1992. Exhibit D -196, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Imperial Tobacco Ltd. Circa 1993. 1993 communications plans. Exhibit D - l 9 7 , JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Imperial Tobacco Ltd. (1993, July 12). 1994 communications plan. Exhibit D-198, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Imperial Tobacco Ltd. (1993, April) . Broadstrokes - 1994. Exhibit D-086, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 017959-017997. Imperial Tobacco Ltd. (1994, Apr i l 25). Market update. Exhibit D-088, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 502596045-502596055. 230 Imperial Tobacco Ltd. (1994, March 4). Imperial Tobacco Ltd. competitor analysis. Exhibit D-173, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 018506-018591. Imperial Tobacco Ltd. (1995, January). Trademark strategies and projects 1995. Exhibit D -176, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 018061-018112. Imperial Tobacco Ltd. (2002, September 30). Management report and interim financial statements. Third Quarter Report. Irwin, R. & Asimakopoulos, M . (1992). A n approach to the evaluation and selection of sport sponsorship proposals. Sport Marketing Quarterly, 1, 43-51. Irwin, R . L . & Sutton, W . A . (1994). Sport sponsorship objectives: A n analysis of their relative importance for major corporate sponsors. European Journal of Sport Management, 1, 93-101. Kahle, L . R . & Homer, P . M . (1985). Physical attractiveness of the celebrity endorser: A social adaptation perspective. Journal of Consumer Research, 11, 954-961. Kaiserman, M . J . (1997). The cost of smoking in Canada, 1991. Chronic Diseases in Canada, 18, 13-19. Kamins, M . A . & Gupta, K . (1994). Congruence between spokesperson and product type: A matchup hypothesis perspective. Psychology and Marketing, 11, 569-586. Kapferer, J. (1997). Strategic brand management: Creating and sustaining brand equity long term (2 n d edition). London: Kogan Page. 231 Kassarjian, H . H . (1977). Content analysis in consumer research. Journal of Consumer Research, 4,8-18. Kie l ing , R .F . (1964, August 31). Implications for Kent. K i n g , K . W . , Reid, L . N . , Moon , Y . S . , & Ringold, D.J . (1991). Changes in the visual imagery of cigarette ads, 1954-1986. Journal of Public Policy and Marketing, 10, 63-80. Kirmani , A . & Wright, P. (1989). Money talks: Perceived advertising expense and expected product quality. Journal of Consumer Research, 16, 344-353. Kluger, R . (1997). Ashes to ashes: America's hundred-year cigarette war, the public health, and the unabashed triumph of Philip Morris. New York : Vintage Books. Kolbe, R . H . & Burnett, M . S . (1991). Content-analysis research: A n examination of applications with directives for improving research reliability and objectivity. Journal of Consumer Research, 18, 243-250. Koten, J. (1980, February 29). Tobacco marketers' success formula: Make cigarets [sic] in smoker's own image. Wall Street Journal, p.22. Kotler, P., Armstrong, G. , & Cunningham, P . H . (1999). Principles of marketing (4 t h Canadian edition). Scarborough, O N : Prentice Hal l Canada Inc. Koz lowski L . T . & O'Connor, R .J . (2002). Cigarette filter ventilation is a defective design because of misleading taste, bigger puffs, and blocked vents. Tobacco Control, 11, i40-i50. Krippendorff, K . (1980). Content analysis: An introduction to its methodology. Beverly Hi l l s , C A : Sage Publications. 232 Kristeva, J. (1984). Revolution in poetic language. N e w York : Columbia University Press. Kwechansky Marketing Research Inc. (1977, October 18). Project 16. Prepared for Imperial Tobacco Ltd. Exhibit AG-216, RJR-Macdonald Inc. v. Canada (Attorney General). Kwechansky Marketing Research Inc. (1982, M a y 7). Project plus/minus. Prepared for Imperial Tobacco Ltd. Exhibit AG-217 , RJR-Macdonald Inc. v. Canada (Attorney General). Lamb Jr., C . W . , Hair Jr., J.F., & McDanie l , C . (2000). Marketing (5 t h edition). Cincinnati, O H : South-Western College Publishing. Laskin, J., Greenspan, E. , Rosenberg, M . , Penny, M . , & Henein, M . (1994). The Canadian Charter of Rights, annotated. Aurora, O N : Canadian Law Book Inc. Lee, J. (2002, February 18). Former champion embraces free skiing. The Vancouver Sun, p .B7. Leiss, W. , Kl ine , S., & Jhally, S. (1997). Social communication in advertising (2 n d edition). New York: Routledge. \u00E2\u0080\u00A2 Linstead, S. & Turner, K . (1986). Business sponsorship of the arts: Corporate image and business policy. Management Research News, 9, 11-13. Lubove, S. (1999, August 9). Brand power. Forbes, 164, 98-104. Lynch, J. & Schuler, D . (1994). The matchup effect of spokesperson and product congruency: A schema theory interpretation. Psychology and Marketing, 11, 4X1-445. Mackay, J. & Eriksen, M . (2002). The tobacco atlas. Geneva: Wor ld Health Organization. Manfredi, C P . (2002). Expressive freedom and tobacco advertising: A Canadian perspective. American Journal of Public Health, 92, 360-362. The Manitoba Educational Research Council . (1966). A study of cigarette advertising. Winnipeg, M N : A Report Submitted to the Department of National Health and Welfare. Marketing Strategy & Planning, Inc. (1985, February). \u00E2\u0080\u00A2 Projects stereo/phoenix final report. Prepared for Imperial Tobacco Ltd. Exhibit A G - 2 7 , RJR-Macdonald Inc. v. Canada (Attorney General). Bates No. 102692361-102692470. Marketing Strategy & Planning, Inc. (1986, Apri l ) . Project stereo advertising evaluation (qualitative research). Prepared for Imperial Tobacco Ltd. Marshall, D . W . & Cook, G . (1992). The corporate (sports) sponsor. International Journal of Advertising, 11, 307-324. McAul i f fe , R . E . (1988). The F T C and the effectiveness of cigarette advertising regulations. Journal of Public Policy and Marketing, 7, 49-64. McCann-Erickson Advertising of Canada Ltd. (1986, November). RJR-Macdonald Inc. brand family and smoker segmentation study ('85): K e y findings and communications implications. Exhibit RJR.-115, RJR-Macdonald Inc. v. Canada (Attorney General). McCracken, G . (1988). Culture and consumption: New approaches to the symbolic character of consumer goods and activities. Bloomington: Indiana University Press. McCracken, G. (1989). Who is the celebrity endorser?' Cultural foundations of the endorsement process. Journal of Consumer Research, 16, 310-321. McCracken, G . (1992). \"Got a smoke? \" A cultural account of tobacco in the lives of contemporary teens. Research Report Prepared for the Ontario Ministry of Health. McLaren, L . (2001, July 14). Canadian lifestyle magazines are really smoking. The Globe and Mail, p .L3 . McQuarrie, E .F . & M i c k , D . G ; (1999). Visual rhetoric in advertising: Text-interpretive, experimental, and reader-response analyses. Journal of Consumer Research, 26, 37-54. Meenaghan, T. (1991). The role of sponsorship in the marketing communications mix. International Journal of Advertising, 10, 35-47. Meier, K . J . (1985). Regulation: Politics, bureaucracy, and economics. N e w York: St. Martin 's Press. Mellor , C . (1992, January 10). Outdated D u Maurier billboard burns up anti-smoking group. The Chronicle-Herald, p .A6. Mi lne , G.R. & McDonald , M . A . (1999). Sport marketing: Managing the exchange process. Sudbury, M A : Jones and Bartlett Publishers. Misra, S. & Beatty, S.E. (1990). Celebrity spokesperson and brand congruence: A n assessment of recall and affect. Journal of Business Research, 21, 159-173. Musante, M . , Mi lne , G.R. , & McDonald , M . A . (1999). Sport sponsorship: Evaluating the sport and brand image match.. International Journal of Sports Marketing & Sponsorship, 1, 32-47. 235 Nelson, P J . (1974). The economic value of advertising. In Y . Brozen (Ed.), Advertising and Society (pp.43-66). New York: New York University Press. N o w Craven A goes king-size. (1966, September 9). Marketing, p. 12. O'Donohoe, S. (1997). Leaky boundaries: Intertextuality and young adult experiences of advertising. In M . Nava, A . Blake, I. MacRury, & B . Richards (Eds.), Buy This Book: Studies in Advertising and Consumption (pp.257-275). N e w York: Routledge. O'Keefe, A . M . & Pollay, R . W . (1996). Deadly targeting of women in promoting cigarettes. Journal of the American Medical Women Association, 51, 67-69. Otker, T. & Hayes, P. (1987). Judging the efficiency of sponsorship: Experience from the 1986 soccer World Cup. ESOMAR Congress, 15, 3-8. Perreault, Jr., W . D . & Leigh, L . E . (1989). Reliability o f nominal data based on qualitative judgments. Journal of Marketing Research, 26, 135-148. Perry, A . (2003). Before the brand: Creating the unique DNA of an enduring brand identity. New York: McGraw-Hi l l . Perry, C L . (1999). The tobacco industry and underage youth smoking. Archives of Pediatrics and Adolescent Medicine, 153, 935-941. Philip Morris . Circa 1990. Marlboro 1990. Accessible at www.pmdocs.com/getallimg.asp?DOCID=2504034844/4875. Physicians for a Smoke-Free Canada. (1996, March). Tobacco advertising and sponsorship increase children's smoking. Ottawa, O N . 236 Physicians for a Smoke-Free Canada. (1998a). Abandoned brands. Filter tips: A review of cigarette marketing in Canada. Published as an \"E-zine\" at http://vv ,ww.smoke-free.ca/filteii:ips 001. Physicians for a Smoke-Free Canada. (1998b). Winfield red rebels. Filter tips: A review of cigarette marketing in Canada. Published as an \"E-zine\" at http://www.smoke-free.ca/filtertips 001. Physicians for a Smoke-Free Canada. (2002). Cigarettes in Canada: Market share of leading brand families of manufactured cigarettes: 1994 - 1997. Online www.smoke-free.ca/factsheets/pdf/marketsharel.PDF. Accessed October 10, 2002. Plenty of smoke billows in battle of the \"Kings\". (1962, March 24). The Financial Post. Pollay, R . W . (1984). Twentieth century magazine advertising: Determinants of informativeness. Written Communication, 1, 56-77. Pollay, R . W . (1989). Filters, flavors.. .flim-flam, too! On \"health information\" and policy implications in cigarette advertising. Journal of Public Policy and Marketing, 8, 30-39. Pollay, R . W . (1990). Information and imagery in contemporary Canadian cigarette ads. In P. Stout (Ed.), Proceedings of the American Academy of Advertising, 1990 (pp.74-79). Austin, Texas: University of Texas. Pollay, R . W . (1991). Signs and symbols in American cigarette advertising: A historical analysis o f the use o f \"pictures o f health\". In H . H . Larsen, D . G . M i c k , & C . Alsted (Eds.), Marketing and Semiotics: Selected Papers from the Copenhagen Symposium (pp. 160-176). Copenhagen: Handelshojskolens Forlag. Pollay, R . W . (2000). Targeting youth and concerned smokers: Evidence from Canadian tobacco industry documents. Tobacco Control, 9, 136-147. Pollay, R . W . (2001). Export \" A \" ads are extremely expert, eh? Tobacco Control, 10,11-74. Pollay, R . W . (2002). H o w cigarette advertising works: R ich imagery and poor information [Expert report prepared for JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada; Quebec Superior Court]. Toronto, O N : Ontario Tobacco Research Unit. Pollay, R . W . & Dewhirst, T. (2001). Marketing cigarettes with low machine-measured yields. Chapter 7 (pp. 199-235) in Risks Associated with Smoking Cigarettes with Low Machine-Measured Yields of Tar and Nicotine, Smoking and Tobacco Control Monograph No . 13. Bethesda, M D : U . S . Department of Health and Human Services, National Institutes of Health, National Cancer Institute. Pollay, R . W . & Dewhirst, T. (2002). The dark side o f marketing seemingly \"Light\" cigarettes: Successful images and failed fact. Tobacco Control, 11, i 18-131. Pollay, R . W . & Dewhirst, T. (in press). A Premiere\u00E2\u0084\u00A2 example of the il lusion o f harm reduction cigarettes in the 1990s. Tobacco Control. Pollay, R . W . & Lavack, A . M . (1993). The targeting o f youths by cigarette marketers: Archival evidence on trial. Advances in Consumer Research, 20, 266-271. Pollay, R . W . , Siddarth, S., Siegel, M . , Haddix, A . , Merritt, R . K . , Giovino, G . A . , & Eriksen, M . P . (1996). The last straw? Cigarette advertising and realized market shares among youths and adults, 1979-1993. Journal of Marketing, 60, 1-16. Proctor, W . (1994, October 18). Personal communication. Qualitative Science Inc. (1994, August). A n investigation o f factors contributing to the growth of duMaurier. Prepared for RJR-Macdonald Inc. Exhibit D-171, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Race report [Report: Canada]. (2002, July). Fl Racing, p. 146. Ricard, E . (2001, January 25). Interrogatory in the Quebec Superior Court. Imperial Tobacco Ltd v. Attorney General of Canada, (500-05-031332-974), p.15-17. Ries, A . & Ries, L . (1998). The 22 immutable laws of branding. New York : HarperCollins Publishers. Ringold, D.J . (1987). A preliminary investigation of the information content of cigarette advertising: A longitudinal analysis. In M . Wallendorf & P. Anderson (Eds.), Advances in Consumer Research, 14, 269-273. Ringold, D .J . & Calfee, J .E. (1989). The informational content o f cigarette advertising: 1926-1986. Journal of Public Policy and Marketing, 8, 1 -23. Ringold, D .J . & Calfee, J.E. (1990). What can we learn from the informational content of cigarette advertising? A reply and further analysis. Journal of Public Policy and Marketing, 9, 30-41. R.J . Reynolds. (1985, March 1). Anti-smoking: The organized movement and individual orientations. Bates No . 513888089-513888177. Accessible at http://legacy.libraiy.ucsf.edu. RJR-Macdonald Inc. Circa 1980. Export family\u00E2\u0080\u0094marketing strategy. Exhibit RJR-174, RJR-Macdonald Inc. v. Canada (Attorney General). RJR-Macdonald Inc. (1982a, March 22). Export family strategy document. Exhibit A G -222, RJR-Macdonald Inc. v. Canada (Attorney General). RJR-Macdonald Inc. (1982b, August). Mul t i brand switching study. Exhibit A G - 3 4 , RJR-Macdonald Inc. v. Canada (Attorney General). RJR-Macdonald Inc. (1987, October 21). Export \" A \" brand long-term strategy. Exhibit A G - 1 5 , RJR-Macdonald Inc. v. Canada (Attorney General). RJR-Macdonald Inc. (1989, February 1). Competitive business development strategies for the 1990s. Exhibit D-226, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. RJR-Macdonald Inc. (1989, October 5). Export A family brand positioning statement. Prepared by J. Walter Thompson A d Agency. Exhibit D-175, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. RJR-Macdonald Inc. Circa 1990. Export source of business/product review. Exhibit D -227, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 80090 8826-800908840. RJR-Macdonald Inc. (1996, March 25). Sponsorship objectives and strategies [inter-office correspondence from Nancy Marcus to Patrick Mispolef]. Exhibit D-200, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 801513317-801513338. RJR-Macdonald Inc. (1996, September 12). Export \" A \" Inc. - how do we build a success story, 1997 & beyond. Exhibit D-192, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. RJR-Macdonald Inc. (1996, September 16). Re: Export ' A ' brand - 1997 communications in Quebec. Exhibit D-193, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. RJR-Macdonald Inc. (Undated). Export family packaging design brief. Exhibit D-225, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 80099 2048-80099 2051. Rogers, M . & Gopal, A . (1987). Up. in smoke: Fifty years of cigarette advertising in . America. In T. Nevett & S.C. Hollander (Eds.), Marketing in Three Eras: Proceedings of the Third Conference on Historical Research in Marketing (pp.257-272). East Lansing, M I : Michigan State University. Rootman, I. & Flay, B .R . (1995). A study on youth smoking: Plain packaging, health warnings, event marketing and price reductions. Toronto, O N : Ontario Tobacco Research Unit. Rothmans, Benson & Hedges Inc. (1991, February). Segmentation - phase 1: Focus group research Ontario/Quebec. Prepared by Johnston & Associates. Exhibit D-201, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1993, October). Strategic plan - 1994/95: Sales and marketing. Exhibit D-195, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1995, October). Strategic plan - 1996/97: Sales and marketing. Exhibit D-174, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1995, August). Craven \" A \" Ltd. 1995 review. Exhibit D- l94 , JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1995, June). Project 21. Exhibit D-230, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1996). Strategic plan - 1997/98: Sales and marketing. Exhibit D - l 7 0 , JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans, Benson & Hedges Inc. (1998, Apri l ) . Project Starr. Exhibit D-231, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. 242 Rothmans, Benson & Hedges Inc. (1999, Spring). Project Apol lo : A qualitative study o f opportunities in the Y A M segment. Exhibit D-236, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Rothmans Inc. (1988). Annual report 1988. Rothmans Inc. (1989), Annual report 1989. Rothmans Inc. (1991). Annual report 1991. Rothmans Inc. (1992). Annual report 1992. Rothmans Inc. (1993). Annual report 1993. Rothmans Inc. (1995). Annual report 1995. Rothmans Inc. (1997). Annual report 1997. Rothmans Inc. (2002). Annual report 2002. Rothschild, M . L . (1987). Marketing communications: From fundamentals to strategies. Toronto, O N : D . C . Heath and Company. Saffer, H . & Chaloupka, F. (2000). The effect of tobacco advertising bans on tobacco consumption. Journal of Health Economics, 19, 1117-1137. Saine Marketing. (1990, M a y 8). Project spur 2/trident: Qualitative French research. Prepared for Imperial Tobacco Ltd. Exhibit D-180, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No . 013219-013311. Schuler, A . (1997, February). Rothmans trade-mark 1957-1997. Prepared for Rothmans, Benson & Hedges Inc. Exhibit D-211, JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Schultz, R. & Alexandroff, A . (1985). Economic regulation and the federal system. Toronto, O N : University o f Toronto Press. Scott, L . M . (1994). Images in advertising: The need for a theory o f visual rhetoric. Journal of Consumer Research, 27,252-273. Segmentation Marketing Limited. (1993, April) . Qualitative research report for Rothman's Benson and Hedges Ltd re Belvedere Ltd. rock sponsorship. Exhibit D - l 9 0 , JTI-Macdonald Corp., Imperial Tobacco Canada Ltd., and Rothmans, Benson & Hedges Inc. v. The Attorney General of Canada. Quebec Superior Court. Bates No. 011024-011076. Semenik, R J . (2002). Promotion and integrated marketing communications. Beauceville, Q C : South-Western, Thomson Learning. Shanklin, W . L . & Kuzma, J.R. (1992). Buying that sporting image: What senior executives need to know about corporate sports sponsorship. Marketing Management, 1, 58-67. Shapiro, S J . , Wong, K . , Perreault, W . D . , & McCarthy, E J . (2002). Basic marketing: A global-managerial approach (10 t h Canadian edition). Toronto, O N : M c G r a w - H i l l Ryerson Limited. Short, P .L . (1977, A p r i l 14). Smoking & health item 7: The effect on marketing. Prepared for B . A . T . Co. Ltd. Exhibit 10,585, State of Minnesota and Blue Cross and Blue Shield of Minnesota v. Philip Morris, Inc., et al. Siegel, M . (2001). Counteracting tobacco motor sports sponsorship as a promotional tool: Is the Tobacco Settlement enough? American Journal of Public Health, 91, 1100-1106. Simpson, J .B. (1964). Contemporary quotations. New York: Crowell . Single, E . , Robson, L . , X i e , X . , & Rehm, J. (1998). The economic costs of alcohol, tobacco and il l ici t drugs in Canada, 1992. Addiction, 93, 991-1006. Solow, J .L. (2001). Exorcising the ghost of cigarette advertising past: Collusion, regulation and fear advertising. In T . H . Witkowski (Ed.), Milestones in Marketing History: Proceedings of the 10th Conference on Historical Analysis and Research in Marketing (pp. 155-163). Durham, N C : Association for Historical Research in Marketing. Sparks, R . E . C . (1997). B i l l C-71 and tobacco sponsorship o f sports. Policy Options, 18, 22-25. Sparks, R. (1999). Youth awareness of tobacco sponsorship as a dimension of brand equity. International Journal of Sport Marketing and Sponsorship, 1, 236-260. Spitzer, M i l l s & Bates Ltd. (1977, March 25). The Player's family: A working paper. Prepared for Imperial Tobacco Ltd. Exhibit A G - 3 3 , RJR-Macdonald Inc. v. Canada (Attorney General). Spitzer, M i l l s & Bates Ltd. (1978, January 20). Player's Light: Creative development. Prepared for Imperial Tobacco Ltd. Exhibit A G - 2 8 , RJR-Macdonald Inc. v. Canada (Attorney General). Stanbury, W.T . (1979). Definitions of the \"public interest\" [Appendix D] . I n D . G . Hartle (Ed.), Public Policy Decision Making and Regulation (pp.213-216). Toronto, O N : Butterworth & Co. (Canada) Ltd. Stephenson, H . E . & McNaught, C. (1940). The story of advertising in Canada: A chronicle of fifty years. Toronto, O N : The Ryerson Press. Stoner, R . H . (1992). 200 mph cigarette ads: A comparison of international restrictions on tobacco sports sponsorship. Hastings International and Comparative Law Review, 75,639-670. Stueck, W . (1999, September 7). Professor smokes out tobacco ads. The Globe and Mail, p.B\3. Taylor, P. (1984). Smoke ring: The politics of tobacco. London: The Bodley Head. Taylor, C R . & Taylor, J . C (1994). Regulatory issues in outdoor advertising: A content analysis of billboards. Journal of Public Policy and Marketing, 13, 97-107. Teel, S.J., Teel, J .E., & Bearden, W . O . (1979). Lessons learned from the broadcast cigarette advertising ban. Journal of Marketing, 43, 45-50. Thomas (1958, October 3). How a king-sized sales campaign... Marketing, p.36-37. Tuckwell, K . J . (1988). Canadian advertising in action. Scarborough, O N : Prentice-Hall Canada Inc. Turco, D . M . (1999). The state of tobacco sponsorship in sport. Sport Marketing Quarterly, 8, 35-38. 246 Twitchell, J .B. (1999). Lead us into temptation: The triumph of American materialism. New York: Columbia University Press. U . S . Department of Health and Human Services. (1992). Smoking and health in the Americas. Atlanta. V i t z , P .C . & Johnston, D . (1965). Masculinity of smokers and the masculinity o f cigarette images. Journal of Applied Psychology, 49, 155-159. Wakefield, M . , Morley, C , Horan, J .K. , & Cummings, K . M . (2002). The cigarette pack as image: New evidence from tobacco industry documents. Tobacco Control, 11, i73-i80. Ward, D . (2002, A p r i l 15). Sponsors try to get ' cool ' in the snow. The Vancouver Sun, p . B l . Warner, K . E . (1985). Tobacco industry response to public health concern: A content analysis o f cigarette ads. Health Education Quarterly, 12, 115-127. Weinberger, M . G . , Campbell, L . , & DuGrenier, F . D . (1981). Cigarette advertising: Tactical changes in the pre and post broadcast era. In H . K . Hunt (Ed.), Advertising In A New Age, Proceedings of the Annual Conference of the American Academy of Advertising, 1981 (pp. 136-141). Provo, Utah: American Academy of Advertising. Wells, W. , Burnett, J., & Moriarty, S. (1989). Advertising: Principles and practice. Englewood Cliffs, N J : Prentice Hal l . Wenner, L . A . (1993). Tidings for the New Year: On sport sponsorship without the smoke. Journal of Sport and Social Issues, 17, 146-147'. Wernick, A . (1991). Promotional culture: Advertising, ideology and symbolic expression. Newbury Park, C A : Sage. West, R. (1993, June 11). Sponsorship of major sporting events as a way to cross cultural divides. Prepared for British American Tobacco Co. Accessible at http://galen.libraty.ucsf.edu/tobacco/batco/html/500/559/omerp Why small cigarette firms plan merger. (1962, M a y 19). The Financial Post. Wichmann, S.A. & Martin, D .R. (1991). Sports and tobacco: The smoke has yet to clear. The Physician and Sportsmedicine, 19, 125-131. Wildgust, J. (2002, September 17). Bashing big tobacco's a nasty habit. The Globe and Mail, p.All. Williamson, J. (1978). Decoding advertisements: Ideology and meaning in advertising. New York: Marion Boyars. Wyckham, R . G . (1997). Regulating the marketing of tobacco products and controlling smoking in Canada. Canadian Journal of Administrative Sciences, 14, 141-165. Yakabuski, K . (1998, September 22). Tobacco giants use Web to collect personal data. The Globe and Mail, p .A4. it 248 ^^^^^^^ 1 W'mf Player I Players filter, taste you can call your own. Warning: Health and Welfare Canada advises that danger to health increases with amount smoked-avoid inhaling Av. per cigarette: Regular Size: 19mg \"tar\". 1.3mg nicotine. Players. A taste you can call your own. Warning: Health and Welfare Canada advises that danger to health increases with amount smoked - avoid inhaling Av. per cigarette: Players Filter: King Size: 17 mg \"tar\", 1.1 mg nicotine. Reg: 17 mg \"tar\", 1.2 mg nicotine 2 Player's \u00C2\u00A9 A taste you can call your own \u00C2\u00BB50 ISS; Hea'th 3S W D e 1 ' 1 a r e C f n a 1 d a ^ d v , s e s , h a t d a n g e ' t 0 heaHh increases with amount smoked - avoid inhaling Average per cigarette: Player s Light: Regular-13 mg \" t a r , 1.0 mg nicotine; King Size-14 mg -tar\". 1.1 mg nicotine. / - V a y w \u00E2\u0080\u00A2 K3\u00C2\u00A3.' is WHEN TO SHIFT GEARS. HP STEVE MOORE, 253 luxe <^are. ' 8 -a> I p l i a racing poster, It's a dort't b e s t u p i d . MOLSON isf Take Care. 254 Good Companions Take time to relax. Take time for quiet compamonship. Export 'A' Kings... a good companion any time. Warning: The Department of National Health and Welfare adv.ses that danger to health Ureases w\u00E2\u0080\u009Eh a m o u n t e d \" Export, eh? 255 When you know what you like. Warning. Health and Welfare Canada advise that danger to health increases with amount smoked -avoid inhaling Averageperagarette: Regular: 18mg \"tar\", 1.2mg nicotine. King Size: 19mg \"tar\", 1.3mg nicotine. /377 256 257 i - t -My Macdonakt ^ [ EXPORTAH EXPORT K Q neaiwi increases wun amount smoked -c v ^ \" * \" i - \u00E2\u0080\u009E i , . o i l u a n y B i l u n e d i u i k e d - avoid nhal na Averaae oer Ciaarpttp-I b'Hht P-^ 'g1- ta!\" 1 0 0 m 9 - nicotine0.8mg. King Size \"tar\" 10.0 mg., nicotine0 8 m g n n a \" \" r a 9 e per ugarerte Export A Extra Light Regular \"tar\" 8.0 mg.. nicotine 0.7 mg. King Size \"tar\" 9.0 m g , nicotine 0 8 mg. /MS 258 259 W A R N I N G CIGARETTES LEAVE YOU BREATHLESS Tobacco use causes crippling, often fatal lung diseases such as emphysema. Health Canada EXTREIHE SPORTS seriES sponsored by: EXPORT A* / LUNG CANCER Every cigarette you smoke iocreases your chance of getting lung cancer. Health Canada KING SIZE CIGARETTES t - ^ - . ^ w CIGARETTES CAUSE LUNG CANCER Every cigarette you smoke increases your chance of getting lung cancer. Health Canada \u00C2\u00AB1 DON'T POISON US Second-hand smoke contains carbon monoxide, ammonia, formaldehyde, benzo[a|pyrene and nitrosamines. These chemicals con harm your children. Health Canada SPECIAL ' : # ; C I G A R E T T E S ARE A HEARTBREAKER Tobacco use can result in the clogging of arteries in your heart. Clogged arteries cause heart attacks and can cause death. damaged hull muscle \u00E2\u0080\u0094 mull al clogged arteiy Health Canada L I G H T 260 IDLE BUT DEADLY Smoke from a lit cigarette contains toxic sub-stances like hydrogen cyanide, formaldehyde and benzene. Second-hand smoke can cause death from lung cancer end other diseases. Health Canada Second-hand smoke contains carbon monoxide, ammonia, formaldehyde, benzo[aIpyrene and nitrosamines. These chemicals can harm your children. Health Canada ft DON'T POISON US A1 ffteth mcuiA \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 ' M T i m r l i U 261 Every day, from Pall Mal l through the West End of London, Rothmans still deliver their world-famous cigarettes to select Clubs and Embassies by coach and footmen. This time-honoured custom is a tradition of the House of Rothmans of Pall M a l l , WORLD FAMOUS FOR QUALITY S INCE 1S?0 Warning: The Department ol National Health and Welfare advises that danger to health increases with amount s i n u i v t EXTRA LIGHT Warning: Health and Welfare Canada advises that danger to health increases with amount smoked - avoid inhaling Average per cigarette - Regular and King Size: \"Tar\" 8 mg Nic 0.8 mg. /SS^-U J M 1H[ WORLD COMES OUT AfTER DARK* S P E C I A L P R E V I E W T I C K E T S TO THE ADVANCE SNEAK PREVIEW OF LxJ \u00E2\u0080\u00A2 _ MIDNIGHT MARCH 27 THE FAMOUS PLAYERS PLAZA CINEMA HUDSON'S BAY CENTRE To win one of only 50 pairs of tickets to the After Dark Series film. The MATRIX, E-Mail us at contest@now.com to: After Dark \"The MATRIX Contest\" or drop off your ballot c/o NOW Magazine, 150 Danforth Ave. Toronto, Ont. M4K INI, All ballots must be received no later than 5pm, March 24. Opens March 31st in theatres everywhere. I. M/X99A* rll m V W V T A TIME WARNER ENTTJ N O W ^ WARNER BROS. ENTERTAINMENT COMPANY Rothmans WORLD FILM i Ii Donations accepted r support olIheJoronB WBtiaoonalHmFeaMl Group i and the tWfenfJrn Centre. SpecdthantaB famous Pbywslnc Restricted B18 yean or oktec \" Name 1 I Address Day Phone Eve. Phone. Winners will be contacted by phone. One ballot per person. 271 Reviewed Content Analysis Studies Authors, Year of Publication Sample Size (Cigs) Time Span of Sampling Period Media Sources Number of Coders Inter-coder Reliability Weinberger, Campbell, and DuGrenier (1981) 251 1957-1977 Newsweek, Sports Illustrated, Ladies Home Journal N R * * N R Warner (1985) 716 1929-1984 Time N R N R Rogers and Gopal (1987) 216 1936-1986 Time, Life N R N R Altman, Slater, Albright, and Maccoby (1987) 778 1960-1985 Roll ing Stone, Cycle World, Mademoiselle, Ladies Home Journal, Time, Popular Science, T V Guide, Ebony 4 95% King , Reid, Moon , and Ringold (1991) 1,100 1954-1986 Time, Ladies Home Journal, Vogue, Redbook, Sports Illustrated, Popular Mechanics, Esquire, Playboy Teams of 3 89% Pollay (1991) 567 1938-1983 Life, Look 17 >80% Ringold (1987) 211 1926-1985 Time (primary source) 2 81% Ringold and Calfee (1989) 568 1926-1986 Time (primary source) 2 89% Altman, Schooler, and Basi l (1991) 171 1985-1987 Billboards (San Francisco, California) 4 92% Taylor and Taylor (1994) 32 1992 Billboards (federal highways, Michigan) 2 >85% Chapman (1986) 1,026 1983 Australian Women's Weekly, New Idea, Cleo, Australasian Post, Australian Playboy, Bulletin, Wheels, 3 New South Wales newspapers 5 82% Pollay (1990) 394 1987 14 unnamed magazines N R N R **NR=Not Reported 272 Content Analysis Coding Categories and Definitions Section 1: Descriptor Dimensions Q01 Copy Volume 01- More than 75 words 02- 46-75 words 03- 31-45 words 04- 15-30 words 05- Less than 15 words Exclude copy volume of health warning and cigarette package. A l l cigarette brand families (e.g., du Maurier, Export ' A ' , Craven A ) are to be counted as one word. Descriptors of product type (e.g., mild, light) are to be counted as a separate word. Places such as La Malbaie are to be counted as 2 words; dates such as June 27-29, 1997 are to be counted as 3 words. Do not count dates i f they have been added for trial purposes. Q02 Illustration 01- Single image 02- Multiple images or pictures Cigarette packages placed in a separately bordered photograph are to be considered as a second image. Q03 Layout 01- Artwork/illustration(s) consumes <50% of advertisement 02- Artwork/illustration(s) consumes 50-74% of advertisement 03- Artwork/illustration(s) consumes 75% or more of advertisement 09-N/A, typographical only A ruler may be used i f necessary. Q04 Colour 01- Black & white 02- Colour in all or part o f advertisement 273 QOS Trademark and Logo Use 01- Use of trademark/logo 09-N/A (no use of trademark/logo) The logo is to be specific to a tobacco product or tobacco sponsorship company (e.g., Export 'A' Lassie or red \"X\" symbol, Player's Hero or chevron, Rothmans coat of arms or rectangle with curved exterior that resembles what is found on package). Trademark/logo may be featured on cigarette package. Do not include trademark/logo of co-sponsors or organizations producing events. Do not include stylized text of tobacco brand. Q06 Package Display 01 -One package displayed 02- Multiple packages displayed (i.e., same brand family, same brand type) 03- Multiple packages displayed of brand family (i.e., same brand family, different brand type) 04- Multiple packages displayed, including a competitor's brand (i.e., comparative advertising) 09-No packages displayed Export 'A', Player's, and Rothmans exemplify brand families. Filter, full flavour, mild, light, and ultra light exemplify brand types. Q07 Package Colour 01- Present (only in package) 02- Present (featured in both display of package and artwork of ad scene) 03- Present (in artwork, no package featured) 09-N/A (no package colour featured) Refer to the cigarette packages that have been provided to confirm whether the package colour is depicted in the artwork of the ad scene. Q08 Product Visibility 01- Package present, closed 02- Package present, at least one open with cigarettes extended 09-N/A (package not present) Q09 Product Depiction 01- Product shown only 02- Product and people both depicted 03- People shown only 04- No product, no people shown Product includes packaging, components, and accessories. i 274 Q10 Demonstration of Product Use 01- Cigarette in hand, away from mouth 02- Cigarette in mouth 03- Cigarette in ashtray 04- Cigarette package only featured in potential usage environment 05- Multiple demonstrations of product use (combination of 01, 02, 03, and/or 04) 09-N/A Qll Depiction of Complementary Items or Products 01- Magazine or book displayed (i.e., implied that it is in use, being read) 02- Beverage displayed 03- Matches, lighter, or ashtray displayed 04- Multiple complementary items displayed (i.e., combination of 01, 02, and/or 03) 09-N/A Q12 Identification of Partners. Co-Advertisers or Supportive Organizational Bodies 01- One partner identified 02- Multiple partners identified 09-N/A (no partners identified) Include logos/trademarks of partners or supportive organizations (e.g., Tennis Canada, Ladies Professional Golf Association), broadcasters (e.g., TSN, CBC), and co-sponsors (e.g., Goodyear may be visible on auto-racing car). Q13 Health Warning Depiction 01- Mandated health warning is present 02- Health warning is absent (no warning is provided) Section 2: Social Portrayal Descriptors If there are multiple images or pictures (e.g., before-after demonstrations; one illustration showing people doing an activity, and a second illustration showing people relaxing after completing the activity), code up to 3 images. Please code the dominant image first. Q14 Location 01- Interior (indoors) 02- Exterior (extension of building or structure such as patio, boat deck, or balcony) 03- Exterior (outdoors; e.g., nature scene) 09-N/A 275 Q15 Institutional Setting 01- Residential 02- Workplace (offices, factories and stores, but does not include work settings listed below) 03- Recreational (bars, restaurants, country clubs, art galleries, sports events) 09-N/A (no institution present, e.g., beach, camping) Q16 Time of Day 01- Daytime scene 02- Transitional time (dusk, twilight, sunrise, sunset) 03- Night scene 09-N/A (product only featured; studio setting that is artificially lit) Include indoor scenes in which time of day is noticeable through windows. Q17 Seasonal Setting 01- Winter (presence of snow, activity such as skiing featured, reference made to Christmas, Santa Claus or New Year's) 02- Non-Winter 09-N/A Q18 Activity Featured 01- Domestic work (including child care, shopping, cleaning, home maintenance for self) 02- Work for wages (activities normally financially compensated by employment including maids and other domestics, self-employment, and professional athletes or artists in competition or specified training) 03- Leisure or recreation (showing what someone does during their free time) 09-N/A (no artwork/illustration or model only\u00E2\u0080\u0094no activity) Q19 Depiction of Activity Featured 01- Active (physical exertion for people engaged in activity) 02- Passive (relaxation, taking break or resting from activity) 09-N/A Q20 Number of People in Primary Focus (in Predominant Image) 01- None 02- One (alone) 03- Two (couples, friends, competitors) 04- Three to six (small group) 05- More than six (e.g., bar scene, larger crowd) DO NOT include depictions of people that are in the distant background (i.e., hardly legible, blurred) and not considered to be interacting or associated with those in primary focus. If in primary focus, the person does not have to be visible in entirety (e.g., a close-up photo of a hand that holds a cigarette) to be counted. 276 If 01 was coded for Q20, please skip to Q26 (i.e., Non-Humans Represented) Q21 Sex of People Featured 01- Male, alone 02- Multiple males 03- Female, alone 04- Multiple females 05- M i x e d 09-N/A For Q21-Q25, continue referring to only those people that are considered in primary focus (or in the predominant image). Q22 Product Use by Males and Females 01 -Male with cigarette 02- Female with cigarette 03- Both male and female with cigarette 09-N/A Q23 Aee Spread of People Featured 01- 1 generation; roughly the same age, peers, siblings 02- Multiple generations; e.g., weddings with proud parents, family gatherings 09 -N/A (includes images showing just one person) Q24 Relationship of People Featured 01 -Family (multi-generation context) 02- Supportive peers (friends, colleagues, teammates, neighbours) 03- Competitive peers (competitors in sport or occupational context) 04- Strangers (large social gathering or scene) 05- Occupational scene depicting superior or supervisor 09 -N/A (no artwork or illustration, person featured alone) Q25 Racial and Ethnic Diversity of People Featured 01- N o evidence of racial or ethnic minorities 02- Presence of a racial or ethnic minority member, alone 03- Presence of one racial or ethnic minority member, in a couple or group of people 04- Presence of more than one racial or ethnic minority member 09-N/A Racial or ethnic minority may include Latino, Black, East Indian, and Asian. 277 Q26 Non-Humans Represented 01- Animals, Pets 02- Mythic figures 03- Mascots 04- Cartoon character 05- More than one type of non-human representation (a combination of 01, 02, 03, and/or 04) 09-N/A D O N O T include logos and trademarks. Section 3: Information Categories Code information categories that are featured in visual and written text, but DO NOT include content of the mandated health warning. If an information category is not present, please code 00. If an information category is present, please code 01. Q27 Quality Assertions regarding product characteristics, features, or properties resulting from engineering, workmanship, attention to detail in manufacture, excellence o f ingredients (e.g., \"the best tobacco money can buy\"), and superiority of personnel. Event features top-ranked/world class performers, classified as a top-tier event, first-rate facilities, superior personnel, or some other distinguishing attribute that communicates stature, merit, or excellence. Q28 Taste/Flavour Reference made to mildness, smoothness, or good taste. The sensation o f flavour perceived in the mouth. Includes assertions regarding the enjoyment or satisfaction derived from using the product. Q29 Product Content Assertions regarding the features present or absent in the product (e.g., additives, tar levels, natural ingredients, \"light\" cigarettes). Include assertions that emphasize the cigarette's filter. Q30 Price/Cost Value Assertions regarding the price of the product, the suggested retail price, special sale or limited time price offer, or reference to getting good value or the \"best buy for your money\". Include comparative pricing (e.g., \"less expensive than other brands\") and the price in reference to the number of cigarettes per package (e.g., \"more bang for your buck\"). Assertions about the cost of event tickets, as well as discounts or special offers i f tickets are purchased before a particular date. 278 Availability Assertions regarding when, where, or how the product can be purchased (e.g., mail, phone, all fine stores, at participating convenience stores). The hours of operation for retailers may be mentioned. Manufacturer's address does not constitute \"availability\" information. Assertions regarding when, where, or how event tickets can be purchased (e.g., mentioning that tickets for the event are available at all Ticketmaster outlets) or when and where the event can be watched (e.g., mentioning which television network w i l l broadcast the event, and the date or time of the broadcast). The mere mention of the city and date for the sponsored event does not count as \"availability\". Product Variations Assertions regarding families of product (varying flavours, strengths, lengths, or other features o f cigarettes). Product descriptors may be used such as 100s, slims, menthol, king size, regular, filter, full flavour, medium, mild, light, extra light, and ultra light (i.e., mentioning du Maurier Ultra Lights as opposed to du Maurier). Includes differing colour o f packaging for product family\u00E2\u0080\u0094e.g., Export ' A ' mi ld and Export ' A ' light have different colour packaging. Popularity Assertions regarding the product being well-known (commonly liked) and the preferred choice among consumers (e.g., \"now the best-selling brand in Canada\"). Include assertions regarding the previous attendance for events. Claims about tickets or the product \"selling fast\" should be included. Packaging Features Assertions regarding the container (i.e., carton) or package as a product feature (e.g., choice o f package size, new flip-top box, protective exterior, or maintaining freshness of product). Include mentions that the product is available in different sizes (i.e., packages of 20 or 25 cigarettes). Do not include merely package being shown. Health Concerns and Reassurance Assertions regarding the product being a safer alternative (e.g., this brand is better for you; reduced coughing and irritation; health advantage by switching to a filtered or lower tar cigarette). Reference made to the absence of additives or tar and nicotine levels. Health benefits implied (i.e., smoking is a source of energy or helps with digestion, controlling body weight and dieting for health reasons). Do not include mere use of the terms \"light\" and \"mild\" . Guarantees A n assurance that certain conditions w i l l be fulfilled, especially that the product wi l l be of a specified quality. Assertions regarding post-purchase assurances that accompany the product (e.g., i f you're not satisfied, product or ticket can be returned for an exchange or refund). May refer customers to a toll-free number i f they are dissatisfied. 279 Q37 Competitive Advantage Assertions regarding advantages o f the product over specific competing products. The reference may or may not mention the name o f the competing brand, but might say \"o f the three best selling brands...\" or depict \"Brand X \" in an identifiable or familiar-looking package. Q38 Purchase Incentives, Contests, Premiums, and Coupons Assertions regarding competitions, contests, gifts and prizes, free instruction booklets, inserts for send-away information or memorabilia, and coupons. Purchase of product does not have to be required. A d copy may request ballots to be submitted with demographic and contact details. Q39 Celebrity Depiction A famous (identified or recognizable without identification) person is depicted. Prominent athletes, musicians, or performers featured in promotion. Celebrities do not have to make product claims (i.e., testimonials). Q40 Endorsements Assertions regarding product performance or satisfaction from named groups or informed experts. Includes seal o f approval by groups. For event sponsorship, includes the identification of other supporting sponsors and organizations. Q41 Research Assertions regarding results o f data collected by the company on product performance or satisfaction (and using the data to establish facts or reach conclusions). Data collected by independent agencies, except i f appearing above in Q40 as endorsements from those external agencies. Q42 Phone Number/Website Listing Include reference to phone numbers, toll-free numbers, or websites in which additional information may be sought. Q42b New Product New offering, introduction, bringing into use for the first time, bringing to the consumer's attention for the first time, announcing a new product, a recent introduction, \"better than before\", renewed, reformed. 280 Section 4: Value/Lifestyle Analysis Code information categories that are featured in visual and written text If an information category is not present, please code 00. If an information category is present, please code 01. Q43 Beauty/Glamour Good appearance (including references made to being sl im and maintaining an ideal body weight), pretty, attractive, fashionable, stylish, glamorous. Applicable to people rather than places (buildings/nature scenes). Q44 Exotic Belonging to another part of the world (i.e., not native), strikingly different, unusual (yet fascinating). Q45 Relaxation/Escapism Comfort, restoring calmness, rest, absence of effort, be at ease, opportunities to take a break from regular routine. Smoking to relieve stress and tension, calming your nerves. Passive activity featured. Peace o f mind, emotional security, stability. Freedom from strangeness or alienation. Diverting the mind to imaginative activity or entertainment as an escape from reality or routine. Include cottage or nature scenes i f the message implies getting away from the fast-pace lifestyle of being in a city. Vacationing and holidays. Q46 Purity Cleanliness, absence of dust, dirt and unpleasant smells. Neatness, a lot of white visuals. Virginity, avoidance o f taboo sexuality, 'clean' thought. Q47 Romance Physical sex, kissing, fondling, caressing, holding hands, flirting, sexual attractiveness, or intimacy. Feeling sexy or sexual. Being liked or loved by the opposite sex (presuming that the sexual orientation of the people is heterosexual). Q48 Family Love Love or companionship expressed to/from parent(s), sibling(s), or grandparent(s). Being at home with other family members. Having children, concern for offspring or next generation or parents. Concern for pets. Q49 Friendship Offering or receiving friendship, sociable. Being accepted, being supportive, receiving approval, belonging (especially among same age and sex peers). Understanding, having a pleasant personality, showing kindly interest. 281 Q50 Community Relating to neighbourhoods, social organizations, and the community at large. Public spirit, group unity, making generous donations (i.e., giving to a good cause). Q51 Nationalism Being patriotic and devoted to one's own nation. Examples include depictions of a person waving a Canadian flag, being supportive of Canadians in competitions, or encouragement to buy products made in Canada. Q52 Women's Liberation Women featured in non-traditional and independent roles. Progression or social reform attained over time. Q53 Physical Activity Sports, athletics, or outdoor activities featured. Auto racing is to be considered as a physical activity. Q54 Rebellious Non-conforming, unconventional, disobedient, questioning or opposing authority, not following the advice of a parent, boss, superior. Use o f symbols such as a motorcycle. Punks, tattoos, body piercing shown. Engaging in deviant activity. Q55 Adventure/Excitement New, thrilling, or daring experience, variety, intensity o f feelings, fascination, danger, risk-taking, or exhilaration featured. Excitement that is the result o f danger or risk. Activities depicted such as skydiving, auto racing, white-water kayaking, extreme sports, riding a roller-coaster. Q56 Independence Free from outside control or influence (in control o f his/her environment and secure). Freedom from authority, liberty, spontaneity. Individualism (alone, unattached, separate, having to do with one particular person). Strong-willed, self-reliant. Autonomy, uninhibited and unbounded by rules, regulations, constraints or domination. Q57 Ruggedness Toughness, macho, determined, weathered, masculine physical features. Q58 Success Earned achievement, accomplishment, importance, careers, productivity, involvement, winning. Recipient o f a trophy or prize (e.g., winner's cheque, receiving a medal on a podium). Superiority, social recognition of ability or achievement, respect, and gaining status, attention or relative rank. 282 Q59 Wealth/Prosperity Explicit reference made to ownership, private property, possessions, elegant and formal settings, money, luxury, \"the good life\", high standard of living, or the highest quality. Being rich. Having possessions which are highly desirable and not easily attainable. Q60 Thrift Bargains, \"more value for your money\", \"more bang for your buck\", a good deal, penny pinching. Q61 Tradition Reference made to classic, legendary, long-standing (e.g., \"70 years of experience), nostalgic, or long-established customs. Q62 Modern/Technology Contemporary, improved, progressive, advanced (e.g., slightly ahead of our time, introducing the new and improved...), cutting-edge, trendy, relating to the present. Reference made to technology. Using up-to-date equipment. Q63 Humour Laughter, kidding, nonsense or outrageous, wit, funny, amusing, absurd setting (including ad copy clearly intended to be humorous or satirical). Q64 Pleasure Contentment, happiness, to have fun, enjoyment, celebration, parties, festivities, cheerfulness, hopeful, lucky, optimistic, or goodness. Physical satisfaction derived from the taste of cigarettes. Ad20 Ad19 81-PV ZI-PV Ad16 m < Ad14 Ad13 Ad12 .\u00E2\u0080\u0094 TJ < Ad10 60PV 80PV ZOPV 90PV 90PV Ad04 Ad03 Ad02 l-OPV Q01 Q02 Q03 Q04 SOD Q06 Q07 80D 60D Q10 T -o Q12 Q13 283 Appendix: Intercoder Reliability (Percentage Agreement) by Advertisement Ad001 0.98 Ad002 0.96 Ad003 0.96 Ad004 0.99 Ad005 0.98 Ad006 0.99 Ad007 0.98 Ad008 0.95 Ad009 0.97 Ad010 0.94 Ad011 0.97 Ad012 0.95 Ad013 0.94 Ad014 0.98 Ad015 0.97 Ad016 0.94 Ad017 0.99 Ad018 0.97 Ad019 0.95 Ad020 0.94 Ad021 0.96 Ad022 0.98 Ad023 0.97 Ad024 0.96 Ad025 0.95 Ad026 0.97 Ad027 0.98 Ad028 0.96 Ad029 0.97 Ad030 0.94 Ad031 0.95 Ad032 0.97 Ad033 0.98 Ad034 0.96 Ad035 0.97 Ad036 0.96 Ad037 0.96 Ad038 0.94 Ad039 0.98 Ad040 0.96 Ad041 0.98 Ad042 0.97 Ad043 0.99 Ad044 0.96 Ad045 0.96 Ad046 0.96 Ad047 0.97 Ad048 0.97 Ad049 0.97 Ad050 0.96 Ad051 0.96 Ad052 0.96 Ad053 0.96 Ad054 0.96 Ad055 0.96 Ad056 0.98 Ad057 0.96 Ad058 0.98 Ad059 0.97 Ad060 0.98 Ad061 0.97 Ad062 0.98 Ad063 0.98 Ad064 0.98 Ad065 0.97 Ad066 0.97 Ad067 0.97 Ad068 0.97 Ad069 0.97 Ad070 0.94 Ad071 0.96 Ad072 0.98 Ad073 0.97 Ad074 0.98 Ad075 0.98 Ad076 0.97 Ad077 0.99 Ad078 0.97 Ad079 0.96 Ad080 0.96 Ad081 0.97 Ad082 0.96 Ad083 0.99 Ad084 0.97 Ad085 0.97 Ad086 0.97 Ad087 0.94 Ad088 0.97 Ad089 0.97 Ad090 0.96 Ad091 0.97 Ad092 0.98 Ad093 0.97 Ad094 0.98 Ad095 0.97 Ad096 0.98 Ad097 0.98 Ad098 0.97 Ad099 0.98 Ad100 0.94 Ad101 0.96 Ad102 0.97 Ad 103 0.97 Ad104 0.97 Ad105 0.99 Ad106 0.97 Ad107 0.98 Ad108 0.97 Ad109 0.97 Ad110 0.97 Ad111 0.95 Ad112 0.98 Ad113 0.98 Ad114 0.95 Ad115 0.98 Ad116 0.97 Ad117 0.93 Ad118 0.97 Ad119 0.96 Ad120 0.97 Ad 121 0.97 Ad 122 0.98 Ad123 0.97 Ad124 0.97 Ad 125 0.95 Ad 126 0.96 Ad 127 0.97 Ad128 0.96 Ad129 0.98 Ad130 0.98 Ad131 0.97 Ad132 0.96 Ad133 0.97 Ad134 0.98 Ad 135 0.99 Ad136 0.98 Ad 137 0.97 Ad 138 0.97 Ad 139 0.97 Ad 140 0.98 Ad141 0.96 Ad142 0.98 Ad 143 0.96 Ad144 0.98 Ad 145 0.99 Ad 146 0.95 Ad 147 0.97 Ad 148 0.97 Ad 149 0.97 Ad 150 0.97 Ad151 0.94 Ad 152 0.96 Ad 153 0.98 Ad 154 0.98 Ad 155 0.95 Ad 156 0.96 Ad 157 0.95 Ad 158 0.99 Ad159 0.96 Ad 160 0.95 Ad 161 0.95 Ad 162 0.95 Ad 163 0.97 Ad 164 0.97 Ad 165 0.96 Ad166 0.97 Ad167 0.97 Ad168 0.97 Ad169 0.98 Ad170 0.97 Ad 171 0.96 Ad172 0.95 Ad 173 0.95 Ad174 0.96 Ad175 0.95 Ad176 0.94 Ad 177 0.95 Ad178 0.97 Ad179 0.96 Ad180 0.97 Ad181 0.95 Ad182 0.95 Ad183 0.98 Ad184 0.98 Ad185 0.97 Ad186 0.98 Ad 187 0.97 Ad188 0.96 Ad 189 0.97 Ad190 0.98 Ad191 0.96 Ad 192 0.98 Ad193 0.97 Ad194 0.95 Ad 195 0.99 Ad 196 0.95 Ad 197 0.96 Ad 198 0.98 Ad199 0.96 Ad200 0.95 Ad201 0.95 Ad202 0.95 Ad203 0.97 Ad204 0.94 Ad205 0.97 Ad206 0.95 Ad207 0.93 Ad208 0.95 Ad209 0.98 Ad210 0.95 Ad211 0.98 Ad212 0.96 Ad213 0.97 Ad214 0.95 Ad215 0.96 Ad216 0.97 Ad217 0.96 Ad218 0.95 Ad219 0.97 Appendix: Intercoder Reliability (Percentage Agreement) by Instrument Item Qu01 0.97 Qu18 0.98 Qu02 0.99 Qu19 0.95 Qu03 0.93 Qu20 0.98 Qu04 1.00 Qu21 0.97 Qu05 0.99 Qu22 0.99 Qu06 1.00 Qu23 0.99 Qu07 0.70 Qu24 0.98 Qu08 0.99 Qu25 0.97 Qu09 0.99 Qu26 0.99 Qu10 0.98 Qu27 0.92 Qu11 0.99 Qu28 0.93 Qu12 0.97 Qu29 0.99 Q u i 3 1.00 Qu30 0.99 Qu14 0.97 Qu31 0.96 Qu15 0.95 Qu32 0.88 Qu16 0.97 Qu33 0.98 Qu17 0.98 Qu34 1.00 Qu49 0.94 Qu35 1.00 Qu50 0.99 Qu36 1.00 Qu51 0.98 Qu37 1.00 Qu52 1.00 Qu38 0.99 Qu53 0.94 Qu39 0.96 Qu54 1.00 Qu40 0.95 Qu55 0.92 Qu41 1.00 Qu56 0.88 Qu42 0.98 Qu57 0.86 Qu42b 0.99 Qu58 0.97 Qu43 1.00 Qu59 0.99 Qu44 0.99 Qu60 1.00 Qu45 0.93 Qu61 0.98 Qu46 0.99 Qu62 0.99 Qu47 0.98 Qu63 1.00 Qu48 1.00 Qu64 0.89 Crosstabs SPONSOR * Copy Volume 286 Crosstab Count Copy Volume Total 1.00 2.00 3.00 4.00 5.00 S P O N S O R A 2 2 10 20 86 120 S 23 5 14 47 10 99 Total 25 7 24 67 96 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 89.448a 4 .000 Continuity Correction Likelihood Ratio 100.825 4 .000 Linear-by-Linear Association N of Valid Cases 219 a. 2 cells (20.0%) have expected count less than 5. The minimum expected count is 3.16. SPONSOR * Illustration Crosstab Count Illustration Total 1.00 2.00 S P O N S O R A 107 13 120 S 76 23 99 Total 183 36 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sip. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 6.071\u00C2\u00B0 1 .014 Continuity Correction 3 5.202 1 .023 Likelihood Ratio 6.075 1 .014 Fisher's Exact Test .017 .011 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 16.27. Page 1 SPONSOR * Layout 287 Crosstab Count Layout Total 1.00 2.00 3.00 9.00 S P O N S O R A 9 19 92 120 S 15 21 59 4 99 Total 24 40 151 4 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 10.8983 3 .012 Continuity Correction Likelihood Ratio 12.415 3 .006 Linear-by-Linear Association N of Valid Cases 219 a. 2 cells (25.0%) have expected count less than 5. The minimum expected count is 1.81. SPONSOR * Colour Crosstab Count Colour Total 1.00 2.00 S P O N S O R A 5 115 120 s 6 93 99 Total 11 208 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sip. (2-sided) Exact Sip. (1-sided) Pearson Chi-Square .408b 1 .523 Continuity Correction 3 .107 1 .743 Likelihood Ratio .406 1 .524 Fisher's Exact Test .550 .369 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.97. Page 2 SPONSOR * Trademark and Logo Use SPONSOR * Trademark and Logo Use Crosstabulation Count Trademark and Logo Use Total 1.00 9.00 SPONSOR A 120 120 s 71 28 99 Total 191 28 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 38.915\u00C2\u00B0 1 .000 Continuity Correction3 36.420 1 .000 Likelihood Ratio 49.512 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 12.66. SPONSOR * Package Display Crosstab Count Package Display Total 1.00 2.00 3.00 9.00 SPONSOR A 79 2 34 5 120 S 99 99 Total 79 2 34 104 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 199.7853 3 .000 Continuity Correction Likelihood Ratio 261.476 3 .000 Linear-by-Linear Association N of Valid Cases 219 a. 2 cells (25.0%) have expected count less than 5. The minimum expected count is .90. SPONSOR * Package Colour 289 Crosstab Count Package Colour 1.00 2.00 3.00 9.00 Total S P O N S O R A 29 85 1 5 120 s 58 41 99 Total 29 85 59 46 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 197.0403 3 .000 Continuity Correction Likelihood Ratio 259.816 3 .000 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 13.11. SPONSOR * Product Visibility Crosstab Count Product Visibility Total 1.00 2.00 9.00 S P O N S O R A 8 107 5 120 s 99 99 Total 8 107 104 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 199.7853 2 .000 Continuity Correction Likelihood Ratio 261.476 2 .000 Linear-by-Linear Association N of Valid Cases 219 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 3.62. Page 4 SPONSOR * Product Depiction 290 Crosstab Count Product Depiction Total 1.00 2.00 3.00 4.00 SPONSOR A 24 93 3 120 s 84 15 99 Total 24 93 84 18 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 208.9073 3 .000 Continuity Correction Likelihood Ratio 285.361 3 .000 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 8.14. SPONSOR * Demonstration of Product Use Crosstab Count Demonstration of Product Use Total 1.00 2.00 3.00 5.00 9.00 SPONSOR A 30 12 2 3 73 120 S 99 99 Total 30 12 2 3 172 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 49.370= 4 .000 Continuity Correction Likelihood Ratio 67.084 4 .000 Linear-by-Linear Association N of Valid Cases 219 a. 4 cells (40.0%) have expected count less than 5. The minimum expected count is .90. Page 5 SPONSOR * Depiction of Complementary Items or Products 291 Crosstab Count Depiction of Complementary Items or Products Total 2.00 3.00 4.00 9.00 SPONSOR A 6 3 5 106 120 S 99 99 Total 6 3 5 205 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 12.3393 3 .006 Continuity Correction Likelihood Ratio 17.630 3 .001 Linear-by-Linear Association N of Valid Cases 219 a. 6 cells (75.0%) have expected count less than 5. The minimum expected count is 1.36. SPONSOR * Identification of Partners, Co-Advertisers or Supportive Organizational Bodies Crosstab Count Identification of Partners, Co-Advertisers or Supportive Orqanizational Bodies Total 1.00 2.00 9.00 SPONSOR A 2 118 120 S 16 30 53 99 Total 16 32 171 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 63.7803 2 .000 Continuity Correction Likelihood Ratio 74.903 2 .000 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.23. Page 6 SPONSOR * Health Warning Depiction Crosstab Count Health Warning Depiction Total 1.00 2.00 SPONSOR A 120 120 s 3 96 99 Total 123 96 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 207.184\" 1 .000 Continuity Correction3 203.264 1 .000 Likelihood Ratio 273.374 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 43.40. SPONSOR * Location Crosstab Count Location Total 1.00 2.00 3.00 9.00 SPONSOR A 4 9 85 22 120 s 2 16 53 28 99 Total 6 25 138 50 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 8.834a 3 .032 Continuity Correction Likelihood Ratio 8.859 3 .031 Linear-by-Linear Association N of Valid Cases 219 a. 2 cells (25.0%) have expected count less than 5. The minimum expected count is 2.71. SPONSOR * Institutional Setting Crosstab Count Institutional Setting 1.00 2.00 3.00 9.00 Total SPONSOR A 5 2 10 103 120 S 1 41 57 99 Total 6 2 51 160 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 35.0433 3 .000 Continuity Correction Likelihood Ratio 37.300 3 .000 Linear-by-Linear Association N of Valid Cases 219 a. 4 cells (50.0%) have expected count less than 5. The minimum expected count is .90. SPONSOR * Time of Day Crosstab Count Time of Day Total 1.00 2.00 9.00 SPONSOR A 86 11 23 120 S 65 5 29 99 Total 151 16 52 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 3.885a 2 .143 Continuity Correction Likelihood Ratio 3.913 2 .141 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.23. SPONSOR * Seasonal Setting 294 Crosstab Count Seasonal Setting Total 1.00 2.00 9.00 SPONSOR A 15 80 25 120 S 8 65 26 99 Total 23 145 51 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 1.704a 2 .427 Continuity Correction Likelihood Ratio 1.722 2 .423 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 10.40. SPONSOR * Activity Featured Crosstab Count Activity Featured Total 2.00 3.00 9.00 SPONSOR A 9 79 32 120 S 47 30 22 99 Total 56 109 54 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 48.0943 2 .000 Continuity Correction Likelihood Ratio 50.940 2 .000 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 24.41. Page 9 SPONSOR * Depiction of Activity Featured Crosstab Count Depiction of Activity Featured Total 1.00 2.00 9.00 SPONSOR A 45 44 31 120 S 51 26 22 99 Total 96 70 53 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 4.5603 2 .102 Continuity Correction Likelihood Ratio 4.575 2 .102 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 23.96. SPONSOR * Number of People in Primary Focus (in Predominant Image) Crosstab Count Number of People in Primary Focus (in Predominant Image) 1.00 2.00 3.00 4.00 Total SPONSOR A 30 36 24 30 120 S 15 47 21 16 99 Total 45 83 45 46 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 8.988a 3 .029 Continuity Correction Likelihood Ratio 9.072 3 .028 Linear-by-Linear Association N of Valid Cases 219 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 20.34. SPONSOR * Sex of People Featured Crosstab Count Sex of People Featured 1.00 2.00 3.00 4.00 5.00 9.00 Total SPONSOR A 36 10 44 30 120 S 43 34 2 1 3 16 99 Total 79 44 2 1 47 46 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 55.2323 5 .000 Continuity Correction Likelihood Ratio 63.767 5 .000 Linear-by-Linear Association N of Valid Cases 219 a. 4 cells (33.3%) have expected count less than 5. The minimum expected count is .45. SPONSOR * Product Use by Males and Females Crosstab Count Product Use bv Males and Females 1.00 2.00 3.00 9.00 Total SPONSOR A 32 3 7 78 120 S 99 99 Total 32 3 7 177 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 42.872= 3 .000 Continuity Correction Likelihood Ratio 58.705 3 .000 Linear-by-Linear Association N of Valid Cases 219 a. 4 cells (50.0%) have expected count less than 5. The minimum expected count is 1.36. SPONSOR * Age Spread of People Featured 297 Crosstab Count Aqe Spread of People Featured Total 1.00 9.00 SPONSOR A 54 66 120 S 39 60 99 Total 93 126 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square .698\u00C2\u00B0 1 .404 Continuity Correction3 .487 1 .485 Likelihood Ratio .699 1 .403 Fisher's Exact Test .414 .243 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 42.04. SPONSOR * Relationship of People Featured Crosstab Count Relationship of People Featured Total 2.00 3.00 9.00 SPONSOR A 54 66 120 S 28 11 60 99 Total 82 11 126 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 17.6783 2 .000 Continuity Correction Likelihood Ratio 21.906 2 .000 Linear-by-Linear Association N of Valid Cases 219 a. 1 cells (16.7%) have expected count less than 5. The minimum expected count is 4.97. Page 12 SPONSOR * Racial and Ethnic Diversity of People Featured 298 Crosstab Count Racial and Ethnic Diversity of People Featured 1.00 2.00 3.00 9.00 Total SPONSOR A 92 28 120 S 75 1 1 22 99 Total 167 1 1 50 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 2.459a 3 .483 Continuity Correction Likelihood Ratio 3.211 3 .360 Linear-by-Linear Association N of Valid Cases 219 a. 4 cells (50.0%) have expected count less than 5. The minimum expected count is .45. SPONSOR * Non-Humans Represented Crosstab Count Non-Humans Represented 1 1.00 9.00 Total SPONSOR A 7 113 120 S 1 98 99 Total 8 211 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 3.586d 1 .058 Continuity Correction3 2.346 1 .126 Likelihood Ratio 4.112 1 .043 Fisher's Exact Test .075 .058 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 3.62. Page 13 SPONSOR * Quality Crosstab Count Quality Total .00 1.00 SPONSOR A 86 34 120 S 76 23 99 Total 162 57 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Siq. (2-sided) Exact Siq. (1-sided) Pearson Chi-Square .733\u00C2\u00B0 1 .392 Continuity Correction3 .492 1 .483 Likelihood Ratio .737 1 .391 Fisher's Exact Test .441 .242 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 25.77. SPONSOR * Taste/Flavour Crosstab Count Taste/Flavour Total .00 1.00 SPONSOR A 19 101 120 S 99 99 Total 118 101 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Siq. (2-sided) Exact Siq. (1-sided) Pearson Chi-Square 154.6466 1 .000 Continuity Correction3 151.277 1 .000 Likelihood Ratio 197.423 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 45.66. SPONSOR * Product Content Crosstab Count Product Content .00 1.00 Total SPONSOR A 119 1 120 S 99 99 Total 218 1 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square .829\u00C2\u00B0 1 .363 Continuity Correction3 .000 1 1.000 Likelihood Ratio 1.207 1 .272 Fisher's Exact Test 1.000 .548 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is .45. SPONSOR * Price/Cost Value Crosstab Count Price/Cost Value Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Price/Cost Value is a constant. SPONSOR * Availability Crosstab Count Availability Total .00 1.00 SPONSOR A 118 2 120 S 73 26 99 Total 191 28 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 29.430\u00C2\u00B0 1 .000 Continuity Correction3 27.266 1 .000 Likelihood Ratio 33.092 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 12.66. SPONSOR * Product Variations Crosstab Count Product Variations Total .00 1.00 SPONSOR A 61 59 120 S 99 99 Total 160 59 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 66.624\u00C2\u00B0 1 .000 Continuity Correction3 64.149 1 .000 Likelihood Ratio 88.886 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 26.67. SPONSOR * Popularity Crosstab Count Popularity Total .00 1.00 SPONSOR A 116 4 120 S 99 99 Total 215 4 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 3.361\" 1 .067 Continuity Correction3 1.759 1 .185 Likelihood Ratio 4.874 1 .027 Fisher's Exact Test .128 .088 Linear-by-Linear Association N of Valid Cases 219 a- Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 1.81. SPONSOR * Packaging Features Crosstab Count Packaging Features Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square .; Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Packaging Features is a constant. SPONSOR * Health Concerns and Reassurance Crosstab Count Health Concerns and Reassurance Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Health Concerns and Reassurance is a constant. SPONSOR * Guarantees Crosstab Count Guarantees Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Guarantees is a constant. SPONSOR * Competitive Advantage Crosstab Count Competitive Advantaqe Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Competitive Advantage is a constant. SPONSOR * Purchase Incentives, Contests, Premiums, and Coupons Crosstab Count Purchase Incentives, Contests, Premiums, and Coupons Total .00 1.00 SPONSOR A 120 120 S 86 13 99 Total 206 13 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 16.752c 1 .000 Continuity Correction3 14.483 1 .000 Likelihood Ratio 21.642 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.88. SPONSOR * Celebrity Depiction 305 Crosstab Count Celebrity Depiction Total .00 1.00 SPONSOR A 120 120 S 56 43 99 Total 176 43 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 64.855\u00C2\u00B0 1 .000 Continuity Correction3 62.132 1 - .000 Likelihood Ratio 81.409 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 19.44. SPONSOR * Endorsements Crosstab Count Endorsements Total .00 1.00 SPONSOR A 120 120 S 62 37 99 Total 182 37 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 53.966\" 1 .000 Continuity Correction3 51.337 1 .000 Likelihood Ratio 68.086 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 16.73. Page 20 SPONSOR * Research Crosstab Count Research Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Research is a constant. SPONSOR * Phone Number/Website Listing Crosstab Count Phone Number/Website Listinq Total .00 1.00 SPONSOR A 120 120 S 51 48 99 Total 171 48 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 74.514c 1 .000 Continuity Correction3 71.707 1 .000 Likelihood Ratio 93.177 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 21.70. SPONSOR * New Product Crosstab Count New Product Total .00 1.00 SPONSOR A 108 12 120 S 99 99 Total 207 12 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 10.474\u00C2\u00B0 1 .001 Continuity Correction3 8.632 1 .003 Likelihood Ratio 15.010 1 .000 Fisher's Exact Test .001 .001 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.42. SPONSOR * Beauty/Glamour Crosstab Count Beautv/Glamour Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Beauty/Glamour is a constant. SPONSOR * Exotic Crosstab Count Exotic Total .00 SPONSOR A 120 .120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Exotic is a constant. SPONSOR * Relaxation/Escapism Crosstab Count Relaxation/Escapism Total .00 1.00 SPONSOR A 95 25 120 S 91 8 99 Total 186 33 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Siq. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 6.8936 1 .009 Continuity Correction3 5.933 1 .015 Likelihood Ratio 7.262 1 .007 Fisher's Exact Test .013 .007 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 14.92. SPONSOR * Purity Crosstab Count Purity Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Purity is a constant. SPONSOR * Romance Crosstab Count Romance Total .00 1.00 SPONSOR A 113 7 120 S 99 99 Total 212 7 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 5.966\u00C2\u00B0 1 .015 Continuity Correction3 4.229 1 .040 Likelihood Ratio 8.612 1 .003 Fisher's Exact Test .017 .014 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 3.16. SPONSOR * Family Love Crosstab Count Familv i Love Total .00 1.00 SPONSOR A S Total 119 99 218 1 1 120 99 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square .829\u00C2\u00B0 1 .363 Continuity Correction3 .000 1 1.000 Likelihood Ratio 1.207 1 .272 Fisher's Exact Test 1.000 .548 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is .45. SPONSOR * Friendship Crosstab Count Friendship Total .00 1.00 SPONSOR A 77 43 120 S 71 28 99 Total 148 71 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 1.412\u00C2\u00B0 1 .235 Continuity Correction3 1.088 1 .297 Likelihood Ratio 1.420 1 .233 Fisher's Exact Test .249 .148 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 32.10. SPONSOR * Community Crosstab Count Community .00 1.00 Total SPONSOR A 120 120 S 97 2 99 Total 217 2 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 2.447\u00C2\u00B0 1 .118 Continuity Correction3 .723 1 .395 Likelihood Ratio 3.198 1 .074 Fisher's Exact Test .203 .203 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is .90. SPONSOR * Nationalism Crosstab Count Nationalism Total .00 1.00 SPONSOR A 120 120 S 91 8 99 Total 211 8 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 10.065\u00C2\u00B0 1 .002 Continuity Correction3 7.900 1 .005 Likelihood Ratio 13.072 1 .000 Fisher's Exact Test .001 .001 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 3.62. SPONSOR * Women's Liberation 312 Crosstab Count Women's Liberation Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Women's Liberation is a constant. SPONSOR * Physical Activity Crosstab Count Physical Activity Total .00 1.00 SPONSOR A 53 67 120 S 30 69 99 Total 83 136 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 4.430\u00C2\u00B0 1 .035 Continuity Correction3 3.860 1 .049 Likelihood Ratio 4.470 1 .034 Fisher's Exact Test .037 .024 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 37.52. Page 27 SPONSOR * Rebellious Crosstab Count Rebellious Total .00 SPONSOR A 120 120 S 99 99 Total 219 219 Chi-Square Tests Value Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Linear Association N of Valid Cases 219 a. No statistics are computed because Rebellious is a constant. SPONSOR * Adventure/Excitement Crosstab Count Adventure/Excitement Total .00 1.00 SPONSOR A 101 19 120 S 47 52 99 Total 148 71 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Siq. (1-sided) Pearson Chi-Square .1736 1 .677 Continuity Correction3 .000 1 1.000 Likelihood Ratio .178 1 .674 Fisher's Exact Test 1.000 .572 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 1.36. SPONSOR * Humour Crosstab Count Humour .00 1.00 Total SPONSOR A 120 120 S 98 1 99 Total 218 1 219 Chi-Square Tests 314 Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 1.218fc 1 .270 Continuity Correction3 .009 1 .923 Likelihood Ratio 1.593 1 .207 Fisher's Exact Test .452 .452 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 2 cells (50.0%) have expected count less than 5. The minimum expected count is .45. Page 29 SPONSOR * Pleasure Crosstab Count Pleasure Total .00 1.00 SPONSOR A 83 37 120 S 86 13 99 Total 169 50 219 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 9.648\" 1 .002 Continuity Correction3 8.670 1 .003 Likelihood Ratio 10.045 1 .002 Fisher's Exact Test .002 .001 Linear-by-Linear Association N of Valid Cases 219 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 22.60. CM CD CL 316 317 'Is s s|: < CL s o o o o o * O OT z o 0. OT 1 i 1 1 I .9 (A I - $ 1 g | S I s \" 1 ? 1 1 > 8 \u00C2\u00A7 \u00C2\u00A7 ^ 8 S S S

I * a: O CO z o a. co ! 3 8 ; 5 5 \u00C2\u00A3 i S\" 1 .5 (0 s Exact SKI. (2-sided) a~ I.M \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 Value 1 g S ^ SJ \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 S a s s 3 S Si IcOMPANY 1 E Pearson Chi-Square Continuity Corred ion\" Ukelihood Ratb Fisher's Exad Test Linear-by-Unear Association N of Valid Cases P Pearson Chi-Square Continuity Corredbn\" ' Likelihood Ratb Fisher's Exad Test Linear-by-Unear Association Nof Val id Cases a {I * J i l l s * ! 1 S3152 cc II E E .1 1 E E 4 4 2 2 a \" \" * a l l ! 5 1 1 o o \u00C2\u00BB-1 5 3 8 ! 323 >\u00E2\u0080\u00A2 z 0. s O \" 8 ! ro Q \u00E2\u0080\u00A25 a> E a. o 0) z O a. (0 \u00E2\u0080\u00A25 * Is s si s s-1 i in to \u00C2\u00A7 8 Value 1 5 ! 8 g g 8 S % 5 csi \u00E2\u0080\u00A2<* COMPANY 1 E Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Unear Association N of Valid Cases P Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction Ukelihood Ratio Llnear-by-Linear Association N of Valid Cases H I I I I i i i E E E i i I E t \u00C2\u00A3 \u00C2\u00A3 \u00C2\u00A3 2 \u00C2\u00A3 2 l i i E E E 1 \u00C2\u00A7 \u00C2\u00A7 ? S ? HI \u00C2\u00AB \u00C2\u00AB \u00C2\u00AB i i i * g * I I I ! 3 3 5 S 8 8 S S i s a c s < s O o cn c C 1 I 3 \u00E2\u0080\u00A2? s IQ 8 5 S 3 8 E Pearson Chi-Square Continuity Correction Ukelihood Ratio Unear-by-Unaar Association N of Valid Cases 5 M ! i f l f i a. R Pearson Chi-Square Continuity Correction Ukelihood Ratio Linear-by-Unear Association N of Valid Cases S t ! I l l i i i E E E I .1 1 2 2 ! 1 I 1 1 I 1 S S H I I I e \u00C2\u00AB o H i a s S I I I i s a 324 < Q. s o o _ 1 3 I n 0) LL > + O (0 z O a. co 8 8 ' \u00C2\u00A7 \u00C2\u00A7 I I O I N 5 8 Value 1 1 1 S ft e> 8 5 * 1 J ' COMPANY 1 E Pearson Cru-Sci^aro Continuity Correction Ukelihood Ratio Unear-by-Unear Association N of Valid Cases P Pearson Chi-Square Continuity Correction Likelihood Ratb Unear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction Ukelihood Ratb Linear-by-Unear Association N of Valid Cases 8 n ? S S I 1 s i E E \u00C2\u00A3 \u00C2\u00A3 I\u00E2\u0080\u0094 (-id uj | | i 1 11 . \u00C2\u00AB \u00C2\u00AB 11 It . S \u00C2\u00A3 I 1 ; s s s s a < a. 5 o o O ) c c a> co CO c o IA co a> co a o to z O a. co II \u00C2\u00A7 8 1 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 Value 2.351\" 2.432 82 s \u00C2\u00A7 3 l a 3 COMPANY 1 E Pearson Chi-Square Continutty Correct ion Livelihood Ratb Lino or-by-Linear Association N of Valid Cases 6 s i f i l l 0. R Pearson Chi-Square Continuity Corrector Likelihood Ratb Unear-by-Unear Association N of Valid Cases E E E E \u00C2\u00A3 \u00C2\u00A3 II i i s a II < 0. s o o 0) co E 325 c CO c E o TJ 0) i _ 0. c in 3 U o u. CO E *c a. c o> a. o a. E a: O to z o a. co ; 3 a 5 3 8 3 -i f 8 5 8 \u00C2\u00A7 Value 1 | \u00C2\u00A7 a | a s s a 3 COMPANY 1 E Pearson Chi-Square Continuity Correction Likelihood Ratio Linear-oy-Ltnear Association Nof Valid Cases P Pearson Chi-Square Continuity Correction Likelihood Ratio Unear-by-Linear Associatbn Nof Valkl Cases R Pearson Chi-Square Continuity Corredbn Ukelihood Ratio Linear-by-Unear Associatbn N of Valid Cases I I I I \"E Ji 4 g g g a s a i l l < 0. s o o TJ 0) CS > u < u 'Q. 0) Q * or o co z O o. (0 1 3 8 S ! (a s 3 > i l 5 5 8 \u00C2\u00A3 \u00C2\u00A7 i Value ^ a s a \u00C2\u00A7 3 8 E Pearson Cni-Square Continulry Corredbn Likelihood Ratio Linear-by-Unear Associatbn Not Valid Cases & 5 \u00C2\u00AB| 1 f l l l l i l l * 1 a. R Pearson Chi-Square Continuity Corredton Likelihood Ratio Linear-by-Unear Associatbn N of Valid Cases 11 \"8 I . i E E E I 2 II a a 327 < a. 2 o o t j CD CO CD LL fl) Q. O 0) a. \u00E2\u0080\u009E a 1 !c a . o a> a . \u00E2\u0080\u00A2 a CO 2 a . t o a> o> < \u00C2\u00AB a: O t o z o a. t o ! 3 3 ; : s s \u00E2\u0080\u00A2 8 S 81: ; ! 8 8 S J J 3 S s\\ I s w o o I i s o P 1 i 1 5 O if) tf) 5 8 8 1 \u00C2\u00A3 s s I s 3 CO 1 i l l 3 O V N I COMPANY i si ?1 1 11ill i l l f i I a 315 2 LU fl i f l i t i s| s$ III 1! e * ? a * M i l * 5 1S 31J 2 or 1 Exact Sig. (1-sided) 1 a o 5 I 5 8. I 5 $\ P 181 \u00C2\u00A3\u00C2\u00A7 ft s fl - a 8 S 2 \u00E2\u0080\u00A2tf g 8 c S 11 5 <6 it i \u00C2\u00A3 sj ?! i i hi f l i p ! \u00C2\u00B0 s- s i -a S 3 8 \u00C2\u00AB\u00E2\u0080\u00A2 -9 n \u00C2\u00A3 O J u. j z 0. S a s| J flljjj IS585I or 328 < a s o o \u00E2\u0080\u00A2o \u00C2\u00AB c in o l _ Q. 0) or in c rs E c o O CO z O a. co is s a s s I 81 Exact Siq. (1-sided) & \u00C2\u00AB \u00C2\u00AB g 1.683e .335 2.439 IIS 5 COMPANY 1 E Peareon Cr\u00C2\u00BB-Square Continuity Correction\" Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases P Pearson Chi-Square Continuity Correction\" Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases R Pearson Chi-Square Continuity Correction\" Likelihood Ratio Fisher's Exact Test Linear-by-Unear Association N of Valid Cases U S l i t S V v E E E i 1 i \u00E2\u0080\u00A2e e z \u00C2\u00A3 \u00C2\u00A3 \u00C2\u00A3 | | 1 1 III! \u00C2\u00A3 s i s t f If ? 8 S 8 l i l i > z < a \u00C2\u00A3 o o \u00E2\u0080\u00A2a % 'in L. Q o LU \u00E2\u0080\u00A2a c CO o CO * O \u00C2\u00A3.8 s a o p o l .S .\u00C2\u00BB i i ^ i i i i i ? thin el \u00C2\u00BBI i iii! i f l l i i 0. R Pearson Chi-Square Continuity Cktrrectbn* Ukelihood Ratio Fisher's Exact Test Unear -by-Linear Association N of Valid Cases 329 < a s o o o > a LL \u00C2\u00AE ID ra cc o CO z O a co 3 a 8 J S 3 81 - 91 31 1 s 1 \u00C2\u00A7 i *\u00C2\u00AB CO 8 1 ll i i i 111 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 Value j 37.411\u00C2\u00B0 34.498 47.174 89.066= 85.156 118.554 a s s E Pearson Cht-Square Continuity Correction' Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association Not Valid Cases P Pearson Chi-Square Continuity 00^*0110^ Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases R Pearson Chi-Square Continuity Correction' Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases I S ! Ill < CL 2 O a \" i * a >< 15 o * OC o co z o a. co 1 8 D 8 & I S I 8 a 8 g 2 | S | 1 % s s 3 a S 3 3 a s s 1! 3-f \" \" \" \" \" \" - *\" -8 2 ' \" c \u00C2\u00BB t t % cr | 2 S S K m a 13 1 s * t O J i l J z Si 51 p i 11 | l l h i 8 8 0. it s 8 0. f i l l I 330 ro 0_ >-Z < 331 < Q_ s o o ra * a. o co z o 0. CO 1 -5 (0 S \u00C2\u00A7 I 'fl s cn s \u00C2\u00A7 O \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 S3? s * > rVt g > o> w s 8 8 \" 2 s a g S> G > | \u00C2\u00A7 5 ? S 8 \u00C2\u00ABi J Il 1! nun I f l l i l J s i l l \" Q . i ii A, J 3 2 5 5 3 itiin or 1 3 3 I H S. 8 \u00C2\u00AB S E E 1 * ? I ? ? E l ! f >- >-. 1 | 8 S III! i s i i f I? ? fill I 332 I 333 < Q. s o o in X i 0) o x i E 0) c o QL O to z o 0. to i 3 a E als ; a I to s \u00C2\u00A7 8. 1 a .SC-OT s a 1\u00E2\u0080\u0094 i i \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 Value 1 s \u00C2\u00A7 I s 2 \u00C2\u00A3 \u00C2\u00A3 59.241e 55.913 73.824 93 % g S 3 COMPANY 1 E Pearson Chi-Square Continuity Correct on* Likelihood Ratb Fisher's Exact Test Linear-by-Unear Association N ot Valid Cases s i \u00C2\u00AB! i f l p i 0. R Pearson Chi-Square Continuity Correctbn* Likelihood Ratb Fisher's Exact Test Linear-by-Linear Associatbn Nof Val id Cases E E S \u00C2\u00A7 3 O ! l | \u00C2\u00A3 \u00C2\u00A3 I a t i t \u00C2\u00AB 3 2 \u00C2\u00BB | i i * I f f R t i l l 334 : s s; < Q. s o u o X UJ * cc o CO z o Q. CO : 3 al 5 3 8 8 J S M i ' | B fe S l a s s o a \u00E2\u0080\u00A2 o l l f l BS551 < Q. s o o o E ro O 2> 3 (0 d> m * cc o CO z o a . 13 3 5! ! 3 3 ; 3 3 5 >-z < 0. s o o * I \u00C2\u00A3 ? ~ o O W O Q . CO S 3 3 5 5\u00C2\u00BB S S 3 S 5 3 3 1 __Puritv 8 < V) a: < (0 a: 8 < CO a: i i g 1 CO t- SPON Total COUPA LU a i r H ! {li!J IP? 5 -n IS551 s i 111 ! 8 \"5 < 0-s o o * E COMPANY 1 E Pearson Chi-Square Continuity Corredbn\" Likelihcod Ratio Fisher's Exact Test Unear-by-Linear Association N or Valid Cases P Pearson Chi-Square Continuity Correction1 Ukelihood Ratio Fisher's Exact Test Linear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction\" Likelihood Ratio Fisher's Exact Test Linear-by-Unear Association Nof Valid Cases s 1 I ' 11 II t g 1 1 a a I i 3 O CN Z < 0. s o o * \u00C2\u00AB o c CO E o tC tc o CO z o o. co ! 3 S 1 .a 1 o ? 1 'a .c CO 8 -Asymp. Sig. 1 (2-sided) \u00C2\u00A7 \u00C2\u00A7 5 ! Value 1 & 11 * \ 8 E Pearson Chi-Square Continuity Corredbn* Ukelihood Ratb Fisher's Exad Tesl Unear-by-Unear Associatbn N of Valid Cases P Pearson Chi-Square Continuity Corredbn\" Likelihood Ratb Fisher's Exact Test Linear-by-Unear Assodalion N of Valid Cases l f l i p ! S 1 8 \u00C2\u00AB \u00C2\u00BB IS 315 2 337 ? as s s s s a < a. 2 o o * >. s c e = \u00C2\u00B0 E E o o * or o CO z O a. co ; 3 81 8 J S 1 J? V) s S 1 \u00E2\u0080\u00A2a c CO 1! 8 p \u00C2\u00AB 8 1 Value 1 & \u00C2\u00AB g 3 IcOMPANY 1 E Pearson Chi-Square Continuity Correction\" Likelihood Ratio Eisner's Exact Test Line ar-by-Linear Association N of Valid Cases P Pearson Chi-Square Conlinuity Correction* Likelihood Ratio Fisher's Exact Test Linear-by-Unear Association N of Valid Cases f l i p * I I P 1s i1 s i or < o. S O o (A T 3 C \u00C2\u00AB a. o CO z o 0. CO 3 \u00C2\u00A3 Exact Siq. (1-skted) | B 01 s tf> Asymp. Sig. (2-sided) \u00C2\u00A7 S I \u00C2\u00A7 \u00C2\u00A3 3 3 8 \u00C2\u00A7 \u00E2\u0080\u00A2a > 3 \u00C2\u00A7j f 3 S = 8 * 8 \u00C2\u00A7 | 3 COMPANY 1 E Pearson Crt-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Linear-by-Unear Association N of Valid Cases Q 8-1 ?l 0. R Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Linear-by-Linear Associalion N of Valid Cases 3 - 3 338 < 0 . o o c o XI (A C CU E o 5 \u00C2\u00AB or o CO z o a. co : 3 al; ! 8 s a ! .fl is 3 12 1*1 i n > .1 \u00C2\u00A3 \u00C2\u00B0 i 3 al; < a. s o o .2 S ro \" c o TO z 01 o CO z o a co : ; si; 1 s 1 \u00C2\u00A7 I B .2 CO I s CO 1. | S 3-1 r j O S 8 5 l i s Value j & \u00C2\u00AB r~ g 5.075e 3.025 6.579 S g S 3 8 8 E Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Unear-by-Unear Association N of Valid Cases P Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Unear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Unear-by-Unear Association Nof Valid Cases 8 ; 3 i 1 339 < a s o u at X I a> oc * cc o CO z O a. co : 3 sfl; S 3 > :fation s i 1 I l I s i I Pearson Chi-Squ Continuity Correc Likelihood Ratio Linear-by-Linear N of Valid Cases f i l l ! If I f I I O J J Z Pearson Chi-Sqi Continuity Correc Likelihood Ratio Llnear-by-Unear N of Valid Cases NY 1 COMPA Q. < a. s o o \u00C2\u00AB > o < \"5 u I O co. z o a. CO ! 3 3 5 \u00E2\u0080\u00A2 i 8 S 3] 1 .2 Vi s 1 s s \u00C2\u00A7 1 Asymp. Sig. (2-sided) 8 \u00C2\u00A7 8 s i s S > 1 s ^ s fc Q 11.161= 9.695 11.349 COMPANY 1 E Pearson Chi-Square Continuity Correction1 Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases P Pearson CN-Square Continuity Correction1 Likelihood Ratio Fisher's Exact Test Linear-by-Linear Associalion N of Valid Cases R Pearson Chi-Square Continuity Correction1 Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases I 8 H i I I I i i 1 >-z < Q. s o o c 0) E a> \" o X y j 2 I 3 i5 C > \u00E2\u0080\u00A2o < * or o z o a. co ! 3 S i ; i a ? a I 2 (Ii I 8 \u00E2\u0080\u00A2 1 Exact Siq. (2-sided) o \u00C2\u00A7 1 .a>~ a, a < ~~~ lO ei \u00C2\u00A3 JQ \u00C2\u00A3 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 % S 8 Value g S s S & r> s s s \u00C2\u00ABt >^ 5 3 COMPANY 1 E Pearson Chi-Square Continuity Correction1 Likdihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases P Pearson Chi-Square Conlinuity Correction1 Likdihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases R Pearson Chi-Square Continuity Correction1 Likdihood Ratio Fisher's Exad Test Linear-by-Unear Association N of Valid Cases 1 1 \u00C2\u00A7 s ? s 341 3 s i s 3 s a s a >\u00E2\u0080\u00A2 z < Q. s o o s * I in o in 0 O u 3 V) * DC o CO z O 0. (0 1 i 8 8 (9 Exact Sia. (2-sided) 8 \u00C2\u00A7 s I' 8 \u00C2\u00A7 \u00C2\u00A7 i r-. Value I 10.320* 6.272 14.168 82 ife 5 ^ S N 11) O t g a g COMPANY ' 1 E Pearson Cht-Square Continutty Correction* Likelihood Ratio Fisher's Exact Test Linear-by-Unear Association Not Valid Cases P Pearson Chi-Square Continuity Correction' Likelihood Ratio Fisher's Exact Test Unear-by-Unear Association N ot Valid Cases R Pearson Chi-Square Continuity Correction\" Likelihood Ratio Fisher's Exact Test Linear-by-Unear Associalion N of Valid Cases E E E i i g E E E ! ! 2 i i i i i i 111! ^ 1 I 1 f ? ? I ? iQ_ S_ S, f i l l >-z < Q. s o o * 2 ! % 6 o O l O ) 3 or or o to z O a. co 3 a s s s s s a 1 a \u00C2\u00A7 tf) 1 B 3 1 Exact Siq. ( \u00C2\u00A7 8 ? Value 1 s a a rvi d rr> 3 6 a l l l i j l ! 111112 i s i l \u00C2\u00AB!. f i l l j 1S1 i 51 s a ! I S 1 f l l i l ! HI HI 5 8 Ul 0. a: ti a 6 -o 343 < 0. s o u * o o c u o t E CO \" D O S * OC o CO z o a. CO 8 3 3 5 9 8 s ? 3 1 8 in ? SJ 8 < \u00C2\u00AB < CO < CO o at Q cc Q 3S \u00C2\u00A3 1 \u00C2\u00A3 1 CO t-CO (-8 a a: 1 CO ! a I 3 (0 1! s B 5 S 8 8 8 5 > 1.227= .010 1.803 93 COMPANY 1 E Pearson Chi-Square Continutty Correction* Likelihood Ratio Fisher's Exact Test LJnear-by-Unear Association Nof Valid Cases P Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Llnear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Llnear-by-Unear Association N ot Valid Cases 1 \u00C2\u00A7 o\" c * & 8. I i i i BBS 1.1 ; E I 1 1 t | 111 i | 5 I 3\" *? \u00C2\u00A3 \"C a? a? \u00C2\u00BB 0 p o g \"8 8 8 35 1 ? ? i O CN Z Q. s o o * c o '\u00E2\u0080\u00A25 (0 OC o co z O CL CO 3 a S 3 ; 5 3 8 3 3 1 s s \u00C2\u00A30 1! 1 fi 1 \u00C2\u00A7 O CO 8 \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 Value \u00C2\u00A7 \u00C2\u00A7 \u00C2\u00A7 s .5 s linn III151 m P Pearson Chi-Square (Continuity Correction* Likelihood Ratio Ftsher's Exact Test Unear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction* Likelihood Ratb Fisher's Exact Test Linear-by-Unear Associatbn N of Valid Cases 11 S 1 \u00C2\u00A3 \u00C2\u00A3 a l l ! \u00E2\u0080\u00A2 ? S \" f ?g ~ s 8 a | t i n 344 18 S \u00C2\u00AB < 0. S o o (A n o CC o CO z o 0. CO ; s si; 8 5 ! Exact Sic. (1-sided) ' 8 i I y w 1 I \u00C2\u00A3 q 811 8 ? S Value 1 ^ 8 R \u00C2\u00A3 S 5 % 3 COMPANY \u00E2\u0080\u00A2 1 E Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association Not Valid Cases P Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher'a Exact Test Linear-by-Unear Association N of Valid Cases R Pearson Chi-Square Continuity Correction* Likelihood Ratio Fisher's Exact Test Unear-by-Unear Association N of Valid Cases K 8 S \u00C2\u00A3 E E E S 1 1 I S i ! i \u00C2\u00A3 K i \u00C2\u00A3 S i 1 ! ! 5 a m i \" ! \u00E2\u0080\u00A2 ! 4 S I * f - - ? f i l l O o o \u00C2\u00BB-< CL s o o o E X \u00C2\u00AB or o CO z o 0. V) is s a . (1-sided) 3 Exact Sic Exact Siq. (2-sided) | =_ a \u00C2\u00A7 8 *5 1 \u00C2\u00A7 \u00C2\u00A7 a s 3 s > \u00E2\u0080\u00A2e an* (Social ion .& E i \u00E2\u0080\u009E 1 on* ssocial Ion COMPANY 8- P B & f IfIs*5 1S 3151 h l i l i Q. R Pearson Cra-Squai Continuity Correct* Ukelihood Ratio Fisher's Exact Tes Linear-by-Unear A N of Valid Cases I s I I I 8 345 WARNING CIGARETTES LEAVE YOU BREATHLESS Tobacco use causes cr ippl ing, often fatal lung d iseases such as emphysema. Htallii Canada IT'S LOUD. IT'S FAST. IT'S HARD. QUESTIONS? C D CD EXTRERIE SPORTS seriES o J3-to I sponsoied by: i f "@en . "Thesis/Dissertation"@en . "2003-11"@en . "10.14288/1.0076851"@en . "eng"@en . "Interdisciplinary Studies"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Tobacco marketing and public policy in Canada, 1960-2002 : the role of sponsorship"@en . "Text"@en . "http://hdl.handle.net/2429/14984"@en .