Content-Type: text/html Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking Influence of Cigarette Promotion on Juvenile Susceptibility to Smoking: A Path Analysis Hye-ryeon Lee The University of Arizona Kristie A. Taylor The University of Arizona & Stacey Nofziger Kansas State University All inquiries and correspondence regarding the paper should be directed to: Hye-ryeon Lee, Ph.D. Assistant Research Scientist Arizona Cancer Center The University of Arizona 2302 E. Speedway, Suite 204 Tucson, AZ 85719 Tel: (520) 318-7100 Fax: (520) 318-7104 E-mail: [log in to unmask] Influence of Cigarette Promotion on Juvenile Susceptibility to Smoking: A Path Analysis Hye-ryeon Lee (The University of Arizona) , Kristie A. Taylor (The University of Arizona) & Stacey Nofziger (Kansas State University) Abstract As restrictions on advertising tobacco products increase, tobacco industry resources have shifted from traditional advertising to the distribution of promotional products. Using data collected from middle and high school students in Tucson, Arizona, this paper investigates the link between exposure to tobacco promotion and juvenile susceptibility to smoking. Results show that, while exposure to tobacco promotion is correlated with other risk factors of smoking uptake, tobacco promotion has an independent influence on susce0tibility to smoking. Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking Influence of Cigarette Promotion on Juvenile Susceptibility to Smoking: A Path Analysis The link between tobacco advertising and youth smoking behavior has drawn increasing attention in recent years. Evidence from internal tobacco company documents released during litigation indicate that youth have been intentionally targeted by tobacco advertising (e.g., RJ Reynolds, 1973). In 1996, an estimated 5.1 billion dollars was spent by the tobacco industry to advertise and promote cigarette consumption. This figure is significantly lower than the $6 billion expenditure recorded in 1993 but still represents a 4% increase from 1995 (Federal Trade Commission 1998). Although the commercial marketing of cigarettes is supposed to be targeting adults--who are the legal tobacco consumers (Gostin, Arno & Brandt, 1997)--a great deal of research has demonstrated that cigarette advertising and promotion is associated with juvenile smoking (Pierce, Gilpin, Burns et al., 1991; USDHHS, 1994). Advertising and promotion also has significant costs; Pierce, Gilpin and Choi (1999) estimated that between 1988 and 1998 there were 7.9 million new adolescent tobacco experimenters who were drawn to smoking because of tobacco advertising and promotion. Most studies of advertising and smoking clearly demonstrate that juveniles of young ages are aware of cigarette advertisements and are able to match specific advertisements or logos with the correct cigarette brands (Altman, Levine, Coeytaux, Slade & Jaffe, 1996; Schooler, Feighery & Flora 1996; Evans, Farkas, Gilpin, Berry & Pierce, 1995; Unger, Johnson & Rohrbach, 1995; Botvin, Goldberg, Botvin & Dusenbury, 1993; Aitken, Eadie, Hastings & Haywood, 1991; Klitzner, Gruenewaod & Bamberger, 1991; Aitken & Eadie 1990; Aitken, Leathar, O'Hagan & Squair, 1987). Juveniles who are more aware and appreciative of tobacco advertising are also more likely to smoke. For example, smokers are better at identifying the brand associated with photographs of cigarette advertisements (Aitken & Eadie, 1990: 404). Furthermore, the most popular advertisements among smoking and non-smoking teenagers are for the brands that most teenager smokers smoke (Arnett & Terhanian, 1998). In addition to differences between smokers and nonsmokers, studies have also found significant differences in the recognition and appreciation of smoking advertisements among groups of juveniles with different degrees of susceptibility to future smoking. Unger et al. (1995) and Feighery, Borzekowski, Schooler and Flora (1998) found that juveniles who do not smoke but indicate they may in the future were more likely to report liking cigarette advertisements than nonusers who indicated that they would definitely not smoke in the future. As restrictions on advertising tobacco products increase, tobacco industry resources have shifted from traditional advertising to the distribution of tobacco product promotions. In fact, from 1995 to 1996, the total expenditures to newspapers and magazines decreased a total of 10.8 million dollars, a 25% drop in this type of advertising. In contrast, 42% of the money spent on promotion and advertising in 1996 was invested in tobacco promotions. This represented a total of $2.15 billion dollars (Federal Trade Commission, 1998). Such promotions include free samples, coupons, free clothing such as t-shirts or hats with the cigarette brand name or logo, or free smoking related items such as lighters. Over a period of one year, approximately 4 million people received 14 million promotional items from the Philip Morris Marlboro Adventure Team campaign alone (Altman et al., 1996 : 1590). This type of promotion reaches a significant portion of juveniles. Coeytaux, Altman and Slade (1995) found that 35% of minors in their sample had participated in a promotional campaign by owning a tobacco promotion item, collecting coupons or having a catalog. Schooler et al. (1996) found that 16.5% of 7th graders had received a mailing from a cigarette company and 26.5% owned a promotional item such as a t-shirt, lighter or sunglasses. Since most previous work on the relationship of tobacco advertising and juvenile smoking has focused on juveniles' ability to recognize and identify cigarette advertisements, the effect of exposure to tobacco product promotions needs more scrutiny. The consistent link between smoking status (or susceptibility) and tobacco advertisement recognition (or appreciation of promotions) indicates only a correlation between smoking and advertising, without indicating the causal direction. The majority of researchers assume that increased exposure to the messages and images in cigarette advertisements or promotions leads juveniles to have relatively more positive images and ideas about smoking, and these positive views then lead juveniles to smoking. In short, exposure to cigarette advertisements and promotions makes juveniles more susceptible to smoking. However, the majority of studies do not provide a clear causal link between advertising or promotion awareness and juvenile smoking (McDonald, 1993). There are two significant alternative explanations of the correlation between advertising and smoking. The first alternative is that juveniles who are interested in smoking become more attuned to and interested in tobacco advertisements and promotions. For instance, juveniles who are exposed to cigarettes through their own smoking or that of peers and family members, would logically notice advertising for cigarettes (McDonald, 1993: 283). In addition, juveniles who smoke would be more likely to participate in promotional campaigns that require the saving of cigarette labels or proof-of-purchases in order to receive free products. In one of the few longitudinal studies of promotions and smoking, Biener and Siegel (2000) found that youths with high receptivity to tobacco advertising (who both owned a promotional item and named a tobacco brand that attracted their attention) were more likely than those with medium or low receptivity to progress from no smoking or experimentation to established smoking. This suggests that promotions do in fact lead to smoking. A second alternative is that exposure to tobacco promotion may be associated with general problem behaviors such as rebelliousness and risk-taking. Hence, it may not be the marketing per se, but personal characteristics that are accountable for differential exposure to both marketing and smoking. Most studies that examine the link between tobacco promotion or advertising and smoking, including Biener and Siegel (2000), have generally failed to control for the complex of factors such as general risk behaviors, peer or family smoking, or individual views about smoking, so these alternative explanations have not been clearly ruled out (McDonald, 1993: 284). The current study fills several important gaps in the literature on cigarette promotion and juvenile smoking. First of all, this study shifts the focus from simple recognition of advertising to actual exposure to and participation in cigarette promotions and the subsequent effect of exposure and participation on smoking behavior. Second, the possibility that exposure may be associated with personal characteristics is explored by investigating the relative impacts of tobacco promotion and other key risk factors. Third, to eliminate the possibility that the causal order is reversed and that smoking itself might generate exposure to and participation in cigarette promotion, the data analysis is conducted with students who are not currently using tobacco, with susceptibility to smoking as the dependent variable. A path model is used to explore the causal logic suggested by the literature and to explore the possible links between tobacco promotions and smoking. Methods The data used for this study was collected through a self-administered survey conducted at 63 middle and high schools in Tucson, Arizona as part of the evaluation of the Full Court Press project (FCP) funded by the Robert Wood Johnson Foundation. As shown in table 1, a total of 12,290 surveys were conducted in two waves with students in 7th through 12th grades. The first wave in 1996 surveyed 8,222 students in 7th, 8th, 11th and 12th grades, and the second wave in 1998 surveyed 4,068 students in 9th and 10th grades. ------------------------------------------------------------------ Table 1 about here ----------------------------------------------------------------- For this analysis, data from the two waves were merged and the sample was limited to nonsmokers (those who said they had) "never smoked a cigarette, not even a few puffs" ) and experimenters (those who said that they have smoked "one cigarette 'just to try' but have not smoked one in the last month"). These allows us to isolate the effect of promotions and advertising on students who are not regular smokers. The final sample was 8,248. Of these, 65.1% were surveyed in 1996 and 34.9% were surveyed in 1998. Overall, 26.3% were in 7th & 8th grades, 34.4% were in 9th & 10th grades, and 37.0% were in 11th & 12th grades. Whites were 40.9% of the sample, Hispanics were 38.3%, and females were 50.8%. The dependent measure for this study, Susceptibility to Smoking, reflects whether the youth have consciously decided to not smoke or whether they are open to the possibility of smoking. Research conducted by Pierce, Farkas, Evans et al. (1993) established that a lack of resolve to not smoke is a good predictor for future smoking. As shown in table 2, the susceptibility index was created by adding three Likert-type items. The main independent measures, Exposure to Tobacco Promotion and Participation in Promotion were measured by a total of four questions. Actual question wordings and descriptive statistics for the variables, Conflict with Family and Parental, Peer and Reference Group Norms, are provided in table 2. ------------------------------------------------------------------ Table 2 about here ----------------------------------------------------------------- Rebelliousness was measured by the index of 11 Likert-type items shown in table 3. The index is a modified scale that includes a subset of the rebelliousness scale developed by Smith and Fogg (1979). ----------------------------------------------------------------- Table 3 about here ----------------------------------------------------------------- Personal attitudes and beliefs regarding smoking were measured by asking 11 questions that deal with potential benefits of smoking. Based on the results of the factor analysis shown in table 4, two indices were created: one for Perceived Benefits of Smoking and one for Positive Social Image of Smoking. ------------------------------------------------------------------ Table 4 about here ----------------------------------------------------------------- We use a path model to examine the direct and indirect effects of participation in and exposure to promotion. We examine the direct effects of promotion on susceptibility to smoking, then we examine the indirect effects of promotion via "positive social image" and "benefits." Personal characteristics, rebelliousness, and peer, reference group and parental norms are included. All data are standardized as z-scores and calculations of indirect and total effects are based on significant unstandardized regression coefficients. Findings and Discussion The data show that tobacco promotion reaches youth quite widely. As shown in table 2, 18.2% of the students reported buying items or receiving free products given out by a tobacco company. Coupons, proof-of-purchase seals, or bar codes to get gifts or other were saved by 9.5% of the students, and 4.2% of students already use tobacco promotional items. Another 24.0% report that they would be willing to use promotional items if they had the opportunity. To compare the impact of promotion to the impact of personal characteristics, we use the path model in figure 1. We first regress perceived benefits of smoking, positive social image, participation in promotion, exposure to promotion, and the personal characteristics on susceptibility, and we then regress participation in promotion, exposure to promotion and the personal characteristics on both perceived benefits of smoking and positive social image (as shown in figure 1)[1]. Indirect effects are calculated by multiplying all significant coefficients along each path, and total effects are calculated by adding direct and indirect effects. ------------------------------------------------------------------ Figure 1 about here ----------------------------------------------------------------- The findings about personal characteristics, positive social image and social benefits are consistent with previous studies. Positive social image has the largest association with susceptibility. Respondents who thought that smoking enhances socially valuable traits such as toughness or sexiness were more likely to report higher levels of susceptibility (see table 5). Rebelliousness had the second strongest association with susceptibility. Peer norms and reference group norms are both positively associated with susceptibility; respondents whose friends and reference groups did not discourage smoking reported higher levels of susceptibility. Females were more susceptible than males, and students in higher grades were more susceptible than students in lower grades. Family conflict was also positively associated with susceptibility. Parental norms were negatively associated with susceptibility, suggesting that students whose parents thought it was important to not smoke were actually more likely to be susceptible to smoking. This is consistent with the finding that rebelliousness is related to susceptibility. ------------------------------------------------------------------ Table 5 about here ----------------------------------------------------------------- What does the data suggest about the relationship between promotion and susceptibility? Net of other factors in the model, participation in promotion has a stronger association with smoking susceptibility than exposure to promotion. As shown in table 5, participation in promotion is positively related to susceptibility, while exposure to promotion not significant in any of the models. Furthermore, participation in promotion is the third largest factor in the model overall. The previously overlooked factor of promotion participation is a far stronger predictor of susceptibility than simple exposure to promotion. In addition, participation in promotion has a significant effect on susceptibility via positive social image. Given that participation often involves wearing or displaying tobacco-related items such as hats, key-chains or lighters, this finding is not surprising. There is also a smaller effect of promotion via perceived benefits of smoking benefits. Promotion has a significant impact on susceptibility to smoking, even when personal characteristics are accounted for. Participation in promotion is a stronger predictor of smoking susceptibility than exposure to promotion. Furthermore, the model suggests that this impact is independent of smoking behavior, since the analysis was limited to nonsmokers. Conclusions We show that promotion is a significant predictor of susceptibility to smoking among nonsmokers even after accounting for impact of school level, rebelliousness, parental, peer, and reference group norms, perceived benefits and positive social images of smoking. When the effect of participation in promotion on susceptibility is accounted for, exposure to promotion is not significant. The analysis suggests that while exposure to tobacco promotion is correlated with other risk factors of smoking uptake, tobacco promotion has an independent influence on susceptibility to smoking. Findings from the study support concerns about the negative effect of tobacco promotion on youth. Regardless of whether these tobacco promotions are intended for adults or minors, the reality is that these promotion efforts reach youth and clearly serve as additional an risk factor for smoking uptake among youth. References Aitken PP, Eadie DR, Hastings GB, & Haywood AJ. (1991). Predisposing effects of cigarette advertising on children's intentions to smoke when older. British Journal of Addiction, 86:383-390. Aitken PP, & Eadie DR. (1990). Reinforcing effects of cigarette advertising on under-age smoking. British Journal of Addiction, 85:399-412. Aitken PP, Leathar DS, O'Hagan FJ, & Squair SI. (1987). Children's awareness of cigarette advertisements and brand imagery. British Journal of Addiction, 82:615-622. Altman DG, Levine DW, Coeytaux R, Slade J, & Jaffe R (1996). Tobacco promotion and susceptibility to tobacco use among adolescent aged 12 through 17 years in a nationally representative sample. American Journal of Public Health, 86(11):1590-1593. Arnett JJ, & Terhanian G. (1998). Adolescents' responses to cigarette advertisements: Links between exposure, liking and the appeal of smoking. Tobacco Control, 7:129-133. Bierner L, & Siegel M. (2000). Tobacco marketing and adolescent smoking: More support for a causal inference. American Journal of Public Health, 90(3):407-411. Botvin GJ, Goldberg CJ, Botvin EM, & Dusenbury L. (1993). Smoking behavior of adolescents exposed to cigarette advertising. Public Health Reports, 108(2):217-24. Coeytaux RR, Altman DG, Slade J. (1995). Tobacco promotions in the hands of youth. Tobacco Control, 4:253-257. Evans, N., Farkas A, Gilpin E, Berry C & Pierce JP (1995). Influence of tobacco marketing and exposure to smokers on adolescent susceptibility to smoking. Journal of the National Cancer Institute, 87(20):1538- 1545. Feighery E, Borzekowski DL, Schooler C, & Flora J. (1998). Seeing, wanting owning: The relationship between receptivity to tobacco marketing and smoking susceptibility in young people. Tobacco Control, 7:123-128. Gostin LO, Arno PS, & Brandt AM. (1997). FDA regulation of tobacco advertising and youth smoking. Journal of the American Medical Association, 277(5):410-418. Johnston L (1996). Smoking rates climb among American teenagers, who find smoking increasingly acceptable and seriously underestimate the risks. Ann Arbor: The University of Michigan news and Information Service. Klitzner M, Gruenewald PJ, & Bamberger E. (1991). Cigarette advertising and adolescent experimentation with smoking. British Journal of Addiction, 86:287-298. McDonald C. (1993). Children, smoking and advertising: What does the research really tell us? International Journal of Advertising, 12:279-287. Pierce JP, Gilpin E, Burns DM, Whalen E, Rosbrook B, Shopland D, Johnson M. (1991). Does tobacco advertising target young people to start smoking? Journal of the American Medical Association, 266(22):3154-3158. Pierce JP, Gilpin E, & Choi WS. (1999). Sharing the blame: Smoking experimentation and future smoking-attributable mortality due to Joe Camel and Marlboro advertising and promotions. Tobacco Control, 8:37-44. Pierce JP, Farkas A, Evans N, Berry, C, Choi WS, Rosbrook B, Johnson M, & Bal DG (1993). Tobacco use in California 1992: A Focus on preventing uptake in adolescents. Sacramento: CA. California Department of Health Services. RJ Reynolds. 2 February 1973. Confidential research planning memorandum. Schooler C, Feighery E, & Flora JA (1996). Seventh graders= self-reported exposure to cigarette marketing and its relationship to their smoking behavior. American Journal of Public Health, 86(9):1216-1221. Smith GM, & Fogg CP. (1979). Psychological antecedents of teen-age drug use. Research in Community and Mental Health, 1:87-102. SPSS [computer program]. Version 9.0.0 for Windows. Chicago, IL: SPSS Inc; 1998. Unger JB, Johnson A, & Rohrbach, LA. (1995). Recognition and liking of tobacco and alcohol advertisements among adolescents: Relationships with susceptibility to substance use. Preventive Medicine, 24:461-466. U.S. Department of Health and Human Services (1994). Preventing tobacco use among young people: A report of the Surgeon General. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office of Smoking and Health. Washington DC: U.S. Government Printing Office. Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking Table 1 Description of Survey Waves Year Grades Number of Schools Number of Respondents Number of Nonsmokers 1996 7th, 8th, 11th, 12th 62 8,222 5,398 1998 9th, 10th 16 4,068 2850 Total 7th-12th 63* 12,290 8248 * The total number of schools is not cumulative, as 15 of the 16 schools included in the survey in wave 2 were also included in wave 1. Table 2 Description of Key Measures Indices/Items N Mean/% SD Susceptibility* (Cronbach=s Alpha=.86) If one of your best friends offered you a cigarette, do you 8,191 1.34 .61 think you might smoke it? At any time during the next year do you think you might 8,174 1.43 .69 smoke a cigarette? Do you think you might try a cigarette soon? 8,185 1.31 .57 Family Conflict# (Correlation Coefficient r=.51) I have a lot of arguments with my family. 8,196 2.12 .86 My family looks for things to nag me about. 8,170 2.12 .90 Parental Norm+ (Correlation Coefficient r=.54) How important does your mother (or stepmother) think 7,727 1.42 .81 staying off cigarettes is? How important does your father (or stepfather) think 7,391 1.60 .93 staying off cigarettes is? Peer Norm+ How important does your best friends think staying 7,751 1.90 1.12 off cigarettes is? Reference Group Norm+ How important do most people your own age 7,805 2.49 .93 think staying off cigarettes is? Exposure to Promotion (Cronbach=s Alpha=.23) Have you received mail addressed to you personally 8,248 13.4% from tobacco companies (such as surveys coupons, free gifts, or catalog)? Have you ever bought or received for free any product 8,248 18.2% which was given out by a tobacco company (such as clothing, hats, bags, posters, or lighters)? Participation in Promotion (Correlation Coefficient r=.33) Have you ever saved coupons, proof-of-purchase seals, 8,248 9.5% or bar codes from cigarette packs to get gifts or promotional items? Would you ever use tobacco promotional items? 8,248 4.2% using now 24.0% yes *Scores range between 1 and 4 where 1 means definitely not and 4 means definitely yes. #Scores range between 1 and 4 where 1 means strongly disagree and 4 means strongly agree. +Scores range between 1 and 4 where 1 means very important and 4 means not important at all. Table 3 Item Descriptions for the Rebelliousness Scale: Means and Standard Deviations* (N=7,708) Mean SD 1. I feel guilty when I break a rule.** 2.04 .71 2. When rules and regulations get in the way, 2.48 .72 I sometimes ignore them. 3. If I don't like an order I have been given, 2.34 .79 I may not do it, or may do only a part of it. 4. If I don't like something I'm told to do, 2.27 .79 I often put it off or just don't do it at all. 5. Sometimes I enjoy seeing how much I can get away with. 2.17 .86 6. When I make a decision, I usually go by 1.96 .75 what my parents taught me.** 7. I sometimes get myself into trouble at school. 2.01 .87 8. When I'm told to do something by a teacher, I do it.** 1.80 .65 9. I get a kick out of doing things every now and then 2.55 .88 that are a little risky or dangerous. 10. If anyone upsets me I usually try to get revenge. 2.13 .82 11. I don't mind lying to keep my friends out of 2.35 .88 trouble with the authorities. Cronbach=s Alpha=.80 * Scores range from 1 to 4 where 1 means strongly disagree and 4 means strongly agree with the statement. Items 1 through 8 are a subset of Smith and Fogg (1979). ** These items are coded reversely. Table 4 Factor Analysis of Perceived Benefits of Smoking (n=7,689) Smoking_ Factor 1 Benefits Factor 2 Positive Social Image helps people relax. .84 helps reduce stress. .84 helps people feel more comfortable in social situations. .81 helps people keep their weight down. .57 helps people forget their worries. .75 cheers people up when in a bad mood. .76 makes people feel more self-confident and sure of themselves. .69 makes people look more grown up. .83 makes people look tough. .85 makes people look sexy. .83 is enjoyable. .61 Variances Eigenvalues 53.82% 5.92 12.52% 1.37 Cronbach's Alpha .90 .84 Factor scores are from Varimax rotation. Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking Table 5 Direct and Indirect Effects of Selected Variables on Susceptibility, Ranked by Magnitude* ** Positive Social Image--Direct Effect 0.323 Rebelliousness--Direct Effect 0.170 Indirect Effect Through Image 0.050 Indirect Effect Through Benefits 0.011 Total 0.231 Participation in Promotion--Direct Effect 0.160 Indirect Effect Through Image 0.030 Indirect Effect Through Benefits 0.007 Total 0.197 Peer Norm--Direct Effect 0.087 Indirect Effect Through Image 0.031 Indirect Effect Through Benefits 0.005 Total 0.123 Gender--Direct Effect 0.078 Indirect Effect Through Image n/s Indirect Effect Through Benefits n/s Total 0.078 ** Perceived Social Benefits--Direct Effect 0.054 Reference Group Norm--Direct Effect 0.035 Indirect Effect Through Image n/s Indirect Effect Through Benefits 0.003 Total 0.038 Family Conflict--Direct Effect n/s Indirect Effect Through Image 0.015 Indirect Effect Through Benefits 0.002 Total 0.017 Parental Norm--Direct Effect -0.037 Indirect Effect Through Image n/s Indirect Effect Through Benefits n/s Total -0.037 Exposure to Promotion--Direct Effect n/s Indirect Effect Through Image n/s Indirect Effect Through Benefits n/s Total n/s Grade Level--Direct Effect -0.125 Indirect Effect Through Image -0.024 Indirect Effect Through Benefits n/s Total -0.149 * Only significant coefficients are used in these calculations. Non-significant coefficients are indicated by n/s. ** Neither positive social image or perceived social benefits have indirect effects. Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking Figure 1. Path Analysis of Participation in Promotion and Exposure to Promotion on Susceptibility [--- ??? Graphic Goes Here ---] Note: Solid lines represent significant effects. Dotted lines are insignificant (and unreported) effects. Influence of cigarette promotion on juvenile susceptibility to smoking Influence of cigarette promotion on juvenile susceptibility to smoking ACKNOWLEDGMENTS Funding for this project was provided by a grant from the Robert Wood Johnson Foundation. Authors would like to thank Drs. David Buller, Laurie Chassin, David MacKinnon, and Jennie Kronenfeld for their feedback on earlier versions of this paper. Authors acknowledge William Peterson for his contribution to the preparation of this paper as well. [1] To control for survey year and experimenter status, two dummy variables were initially included in all regressions. However, omitting these dummy variables did not effect the magnitude, sign or significance level of any other variables. To simplify presentation, all results presented here omit these two control variables. All analyses were done in SPSS for Windows 9.0.