THE INFLUENCE OF PRODUCT PACKAGING ON YOUNG PEOPLE S PALATABILITY RATING FOR RTDs AND OTHER ALCOHOLIC BEVERAGES

Transcription

1 Alcohol & Alcoholism Vol. 42, No. 2, pp , 2007 Advance Access publication 31 January 2007 doi: /alcalc/agl113 THE INFLUENCE OF PRODUCT PACKAGING ON YOUNG PEOPLE S PALATABILITY RATING FOR RTDs AND OTHER ALCOHOLIC BEVERAGES PETER GATES 1, JAN COPELAND 1, RICHARD J. STEVENSON 2 and PAUL DILLON 1 1 The National Drug and Alcohol Research Centre, The University of New South Wales, Australia and 2 Macquarie University, Australia (Received 14 November 2006; accepted 14 November 2006; advance access publication 31 January 2007) Abstract Aims: To investigate the influence of product packaging of ready to drink (RTD), or pre-mixed drinks, and other alcoholic and non-alcoholic beverages on the palatability ratings of adolescents and adults. Respondents were interviewed at their own schools or a campus of the University of NSW. The experiment tested palatability ratings in blind and non-blind conditions with a selection of 12- to 30-year-old Australians (140 adolescents and 210 adults) from seven private schools across NSW and students from the University of NSW and Macquarie University. Methods: Beverage palatability was measured utilizing an interval ratio scale from 1 to 7 before and after packaging was presented. Views on whether the beverage packaging was designed to appeal to the participant were also analysed. Results: In general, the alcoholic beverage packaging was thought to be designed to appeal more to adults and palatability ratings significantly increased in the non-blind conditions. In contrast, the Bacardi Breezer packaging was especially palatable to younger participants. Conclusions: Although most alcoholic beverage packaging was not of particular appeal to adolescents, some RTD beverages may be appealing more to adolescents. Unlike other alcoholic products, the difference in palatability ratings for the Bacardi Breezer, from blind to non-blind conditions, was greater for younger participants. INTRODUCTION Alcohol advertising has been shown to influence the drinking patterns and behaviours of Australian adolescents. Exposure to alcohol advertising and product packaging is associated with increased intentions to drink and is predictive of early onset of alcohol consumption (Grube and Wallack, 1994; Garfield et al., 2003; Ellickson, Collins, et al., 2005). The influence of product branding and packaging may have particular relevance when investigating ready to drink (RTD) beverages as the packaging of RTD drinks and non-alcoholic drinks such as sports and soft drinks are very similar (MacKintosh et al., 1997). The majority of research investigating alcohol advertising focuses on television advertisements. There is relatively little research relating to the influence of RTD packaging on adolescents drinking behaviour and consumption patterns (Ellickson et al., 2005). Although alcoholic beverages are allegedly not marketed towards under-age drinkers, teenage exposure to Australian alcohol advertising has been described as significant and increasing (Carroll and Cramer, 1996; Casswell, 2004). In a New Zealand-based longitudinal study of year olds (n = 500) it was found that liking a particular beer brand before reaching the age of 18 years contributed to the likelihood of being a heavy drinker (Wyllie et al., 1998). In 2002, the Australian alcohol industry spent 1.9 billion dollars on such advertisements (the majority of which placed on television) (Centre on Alcohol Marketing and Youth, 2004). However, from American sales data, an estimated three to four times this amount was spent on point-of-sale promotions and packaging (TNS Media Intelligence, 2003). Point-of-purchase marketing is increasingly important to sales and currently plays a leading role in marketing campaigns (Schlossberg, 1990; Prone, 1993). Lamb et al. (1999) suggest that the most effective means of attracting attention to a product is by *Author to whom correspondence should be addressed at: Peter Gates, The National Drug and Alcohol Research Centre, The University of New South Wales, Sydney NSW 2052, Australia Tel: ; Fax: ; p.gates@unsw.edu.au focusing attention on product brand through the use of an appropriate colour, size, language, and picture while increasing product availability. A recent UK case study on the effects of packaging found that as individual preferences become more complex and diverse, packaging becomes the major means of product branding (Hill and Tilley, 2002). Packaging also provides a more permanent impression of product brand to the consumer (Casswell, 2004). RTD alcohol is a beverage made with a spirit or wine base and a non-alcoholic drink, served in a pre-mixed package (Food Standards Australia and New Zealand, 2004). The Australian Divisions of General Practice (ADGP) has drawn attention to the similarities between the packaging of RTD beverages and other soft drinks by presenting them with the labels disguised demonstrating the difficulties in telling them apart (ADGP, 2003). In an associated news release the ADGP chair described RTDs as a serious threat to the health and well-being of the community. Further, MacKintosh et al. (1997) suggest that the RTD drinks are likely to be small and portable facilitating their illicit consumption among under-age drinkers. The increasing purchasing power of young people is beginning to effect consumer targeting strategies (Hill and Tilley, 2002). Marketing with effective brand design has been shown to elicit important purchasing cues to consumers as young as 12 years (Achenreiner and John, 2003). Currently, there is some concern that RTD beverages are using marketing strategies that parallel non-alcoholic beverage advertisement (Austin and Hust, 2005) and may appeal to a younger market (Jackson et al., 2000). The Bacardi Breezer, by way of example, is typically drunk from the bottle allowing opportunity for brand identification, and uses its brand to communicate quality by utilizing the trademark bat symbol and sophistication through the elevated price (Jackson, 2000). Given the increasing concern surrounding RTD alcohol products; it is surprising that there is little research on the effects of packaging on young consumer s palatability ratings for a range of alcoholic beverages. This paper focuses on the impact of packaging on alcoholic and non-alcoholic beverage palatability. Of interest in particular is (i) whether the 138 Ó The Author Published by Oxford University Press on behalf of the Medical Council on Alcohol. All rights reserved

2 INFLUENCE OF PRODUCT PACKAGING ON YOUNG PEOPLE S PALATABILITY RATING 139 packaging of a variety of alcohol products, including RTD alcohol, is thought to be designed to appeal to younger consumers, and (ii) the effects of a product s packaging on palatability ratings between blind and non-blind conditions and for participant s of differing age. As the intention of alcoholic product packaging is to be more appealing to adults, it is expected that as age increases the appeal of RTD packaging, and those of the other alcoholic products, will increase as measured by the relevant palatability ratings. The opposite relationship is predicted between age and palatability ratings for the non-alcoholic drinks. MATERIALS AND METHODS Participants A convenience sample of 350 participants was recruited, 70 in each of five age groups (12 13, 14 15, 16 17, 18 23, and years) spread evenly by gender. The entry criterion included being aged between 12 and 30 years, absence of medication or relevant physical or emotional disorder, residence in the Sydney metropolitan areas, and fluency in English. Adolescent participants in the first three age groups (<18 years) were recruited from government and non-government schools across Sydney. Adult participants (>18 years) were recruited by general newspaper advertisements, fliers, and word of mouth. Materials Participants completed the interview in four different sections: firstly, the screening sheet, followed by response booklet, questionnaire, and breath analysis. The screening sheet was used to ensure the participant did not have any infections or medications that may react with alcohol and were between 12 and 30 years old and had parental consent (if under 18 years). The response booklet allowed for judgements on how much the participants liked or disliked each drink, if they had ever drunk anything like it before, it s alcoholic content, and finally, if they could name the drink and the age range and gender of those who would like the drink. Experiences and attitudes about alcohol use were then collected in the questionnaire prior to a breath analysis to ensure no alcohol was swallowed during testing. Procedure Once appropriate institutional ethical approval was gained, and consent given (by participant and parent for adolescents), participants were tested at their school, while adult participants were tested at NDARC. At the completion of the interview the participants were thanked and received $AU30 (if over 18 years) or two movie tickets (if under 18 years) as a contribution to travel and related expenses. The experimental drinks were chosen to represent the most popular RTDs, alcoholic beverages, and non-alcoholic beverages in NSW as listed below. (i) Spirits (Superior Bacardi Bacardi & Co., Puerto Rico, South America, Smirnoff Vodka Smirnoff Co., USA, Jim Beam Bourbon James B. Beam Distilling Co., Kentucky, USA). (ii) RTD s (Watermelon Bacardi Breezer Bacardi & Co., Bahamas, Caribbean, Chocolate Vodka Mudshake Independent Distillers, UK, and Jim Beam Bourbon and Coke Jim Beam Brands Co., USA) (iii) Soft drinks (Raspberry Fanta Coca-Cola Co., NSW, Australia, Coca Cola Coca-Cola Co., NSW, Australia). (iv) Milk drink (Moove Chocolate Milk Dairy Farmers, NSW, Australia). (v) Tooheys New Beer Tooheys New Pty Ltd, NSW, Australia. (vi) Jacobs Creek Chardonnay Orlando Wines, Barossa Valley, South Australia. (vii) Yeos Wintermelon Tea YHS Pty Ltd, Senoko Way, Singapore (a sweet, novel beverage to control for novelty among naïve drinkers). The base spirits were made up to the same percentage of alcohol content as their respective RTD, e.g. Bacardi was prepared to give an alcohol concentration of 4.9%. The experiment consisted of two components; blind tasting, followed by non-blind tasting. Participants wore colour distorting goggles and the beverage was served without identifying clues. Blinded participants smelt, sipped, and spat a 10 ml sample of each of the set of beverages. Orange juice, a sucrose solution, and citric acid were utilized first to ensure the participants were using the palatability rating scales appropriately following that the 12 stimuli were presented randomly. After each tasting the participant was asked to rate how they liked the drink and if they tasted alcoholic content using common methodology in the preference field (e.g. Cardello et al., 2000). The second component, the non-blind testing, was identical in all respects except the participants (i) were shown the packaging of the drink and not blinded in any way and (ii) after repeating the ratings from blind tasting, were asked if they thought the packaging was designed to appeal to them. Following this, the questionnaire and breath analysis was conducted. Data analysis techniques The data in this study was analysed using SPSS (version 12.0). After confirming the data was normally distributed using descriptive statistics; the relationship between beverage palatability rating scales, thoughts on packaging, and participant gender and age (as a continuous variable) were explored. Univariate analysis of variance (ANOVA) was utilized on palatability ratings. Binary logistic regression was utilized in the case of packaging questions. In interpretation of these analyses, significant relationships were taken at P < RESULTS The 350 participants were mainly (94.3%) Australian born, with most of the sample (67.1%) having had alcohol before, initiated at a mean age of 13.6 years. A binary logistic regression was utilized on the non-blind data to determine if age or gender were significant predictors

3 140 P. GATES et al. Table 1. Binary regression depicting the relationship between thinking the beverage packaging was designed to appeal, and participant age Beverage 2 Log likelihood Odds ratio 95% Confidence interval P value b* Coke N/A N/A N/A P = 0.1 N/A Fanta P < Chocolate milk P < Wintermelon tea P < Bacardi P < * Vodka P < * Bourbon P < * Breezer N/A N/A N/A P = 0.1 N/A Mudshake N/A N/A N/A P = 1.0 N/A Bourbon and coke P < * Beer P < * Wine P < * *Negative results indicate that participants were more likely to think the packaging was designed to appeal to them as their age increased. of whether or not a participant thought a particular beverage packaging was designed to appeal to them. The unsure response on thoughts on packaging appeal was excluded (number of exclusions was dependent on the beverage, mean 10.5%, SD = 3.3) from the analysis leaving only the yes and no responses. In the case where both age and gender were significant predictors a second model with their interaction was explored. Age was found to be a significant predictor of whether participants thought a particular beverage packaging was designed to appeal for almost every beverage. The significance and results of these tests are presented in Table 1. These results support the hypothesis that younger participants would consider the packaging of non-alcoholic products to be designed to appeal to them and older participants consider the packaging of alcoholic products to be designed to appeal to them. Responses on packaging design for the Coke (P = 0.08), Bacardi Breezer (P = 0.08), and Mudshake (P = 1.0) did not vary with participants age. Thus, for these beverages the hypothesis was not supported. In the same model, the bourbon (odds ratio = 3.0, 95% CI = , P < 0.001), beer (odds ratio = 3.5, 95% CI = , P < 0.001), and Bourbon and Coke (odds ratio = 3.7, 95% CI = , P < 0.001) packaging was thought to be designed to appeal to them by significantly more male participants. In addition, Bacardi Breezer packaging was thought to be designed to appeal by significantly (odds ratio = 0.4, 95% CI = , P < 0.02) more females. The interaction between age and gender was not found to be significant for beer (P = 0.1), bourbon (P = 0.7), Bourbon and Coke (P = 0.5), or the Bacardi Breezer (P = 0.1). Thus, the results on the predictor of participant age (presented in Table 1) were not dependent on participant gender. The effect of beverage packaging on palatability ratings The difference between the mean palatability ratings between blind and non-blind conditions reflects the effect of observing a beverage s packaging. This difference is, hereafter, referred to as the mean difference. Where the mean difference is positive the ratings are higher in non-blind testing, hence packaging is making the beverage more appealing and vice versa. A univariate ANOVA on the mean difference in nonalcoholic beverages utilizing age and gender as independent variables was conducted. Of the non-alcoholic drinks, it was found that the mean difference was significantly predicted by the participants age for Coke [R 2 = 0.03, F(1,346) = 10.43, P < 0.002], Fanta [R 2 = 0.08, F(1,347) = 26.77, P < 0.001], and chocolate milk [R 2 = 0.03, F(1,347) = 10.77, P < 0.002] such that drink palatability was less appealing when presented with packaging by 0.05, 0.07, and 0.05 units, respectively, for every increase in age by 1 year. Gender was not found to predict the mean difference scores for the nonalcoholic drinks. Thus, the hypothesis that younger participants would be more particularly influenced by the packaging of non-alcoholic drinks was supported. With the exception of vodka [R 2 = 0.03, F(1,347) = 12.19, P < 0.002], beer [R 2 = 0.05, F(1,339) = 13.15, P < 0.001], and the Bacardi Breezer [R 2 = 0.02, F(1,346) = 7.69, P < 0.007], the mean difference of the alcoholic drinks was not typically predicted by participant age. For the vodka and beer, drink palatability were higher in the non-blind condition by 0.07 and 0.06 units, respectively, for every increase in age by 1 year. Thus, there was only partial support for the hypothesis that the alcoholic beverages would have a greater effect on older participant s palatability ratings. Palatability ratings for the Bacardi Breezer, however, were significantly lower in the non-blind condition 0.05 U for every increase in age by 1 year. Thus, contrary to expectancies, the Bacardi Breezer packaging was making the beverage less appealing with age. In the same model, gender was found to be a significant predictor of the mean difference for beer only [R 2 = 0.05, F(1,339) = 13.15, P < 0.001] such that male participants rated the drink as more palatable than females in the presence of the packaging. However, an ANOVA model investigating the mean difference ratings for beer including the predictors of age, gender, and their interaction showed no significant findings for age (P = 0.112), gender (P = 0.902), or their interaction (0.659). Thus, the above results were not dependent on participant gender. A univariate ANOVA was utilized to determine if a participant who thought the packaging of a beverage was designed to appeal to them tended to like that beverage more than a participant who did not have this view, regardless of their palatability rating on taste alone. The non-blind palatability ratings were analysed using yes and no responses on packaging appeal and the blind palatability ratings as independent predictor variables. For each of the beverages, the non-blind palatability rating was predicted by views on package appeal even when controlling for non-blind palatability ratings. The results of this analysis are presented in Table 2 below for each beverage. Thus, the hypothesis that a participant who thought the packaging of a beverage was designed to appeal to them tended to like the same beverage more than a participant who did not hold that view was supported. DISCUSSION This paper reports that, with the exception of some RTD preparations, the alcoholic drink packaging was thought to be designed to appeal and was associated with an increase in

4 INFLUENCE OF PRODUCT PACKAGING ON YOUNG PEOPLE S PALATABILITY RATING 141 Table 2. The relationship between non-blind palatability ratings and thinking the beverage packaging was designed to appeal controlling for blind palatability ratings Beverage R 2 F statistic P value b* Coke 0.04 F(1,328) = * Fanta 0.41 F(1,330) = * Chocolate milk 0.28 F(1,319) = * Wintermelon tea 0.51 F(1,279) = * Vodka 0.21 F(1,307) = * Bacardi 0.24 F(1,306) = * Bourbon 0.39 F(1,312) = * Mudshake 0.36 F(1,308) = * Breezer 0.29 F(1,309) = * Bourbon and coke 0.50 F(1,311) = * Wine 0.52 F(1,300) = * Beer 0.55 F(1,295) = * *Negative results indicate that the beverage non-blind palatability is found to be higher if a participant thought the packaging was designed to appeal to them, controlling for blind palatability ratings. palatability ratings for older participants. Similarly, for the majority of non-alcoholic drinks, the packaging was thought to be designed to appeal and was associated with an increase in palatability ratings for younger participants. For most beverages a participant s gender was not associated with their views on, and reactions to, beverage packaging. The appeal of packaging was positively related to palatability ratings. This observation was found to be independent of the palatability ratings from blind tasting. There were some notable exceptions to these findings amongst the group of alcoholic and non-alcoholic beverages under analysis. The Bacardi Breezer packaging was found to be appealing to the majority of participants of any age and this was especially the case if the participant was female. However, analysis on the effect of packaging on palatability ratings showed that older participants were actually less likely to find the Breezer more appealing in the presence of the packaging. This was inconsistent to the general finding that alcoholic beverages have a bitter taste. These factors suggest that this alcohol packaging is disproportionately appealing to younger participants. This finding is consistent with past research on the appeal of pictures, such as the bat symbol on the Breezer, and colours of the beverage, like the brighter colours on Breezer bottles, to adolescents (Jackson, 2000). This study focused on the effect of packaging but asked only a single question on whether the participants thought that beverage packaging was designed to appeal to them. A further limitation is that only people aged from 18 to 30 years were included in the study and the full market for alcoholic beverages, such as the spirits, which are said to be designed for an older market, were not tested. However, this age range was chosen as it represented participants of varying degrees of taste maturation, where it was considered a participant over 30 years would have a decreased sensitivity to taste (Joseph et al., 1995; Hummel et al., 1998; Duffy et al., 1999). Conclusion The present study investigated the effects of packaging for a range of non-alcoholic and alcoholic beverages. The packaging of most alcoholic products, including spirits, beer, and wine, were viewed to be designed to appeal to adults and palatability ratings increased when the packaging was presented. The opposite relationship was observed for the nonalcoholic products which appealed to adolescents. However, support was given to the concerns that some RTD alcoholic product packaging, specifically the Bacardi Breezer (and the similarly appealing Mudshake), are marketed with packaging that appeals to adolescents. Those under 18 years considered the Breezer to be designed to appeal to their age group. Further, for this product when packaging was revealed, palatability ratings increased disproportionately for adolescents in comparison to adults. REFERENCES Achenreiner, G. B. and John, D. R. (2003) The meaning of brand names to children: a developmental investigation. Journal of Consumer Psychology 13, Austin, E. W. and Hust, S. J. (2005) Targeting adolescents? The content and frequency of alcoholic and non-alcoholic beverage ads in magazine and video formats November 1999 April Journal of Health Communication 10, Australian Divisions of General Practice (2003) Ready to Drink? Alcopops and Youth Binge Drinking. Australian Divisions of General Practice Ltd. Manuka, ACT. Cardello, A., Schutz, H., Snow, C. et al. (2000). Predictions of food acceptance, consumption and satisfaction in specific eating situations. Food Quality and Preference 11, Carroll T. and Cramer, P. (1996) Advertising of Alcoholic Beverages in Australia: Expenditure and Exposure, Department of Human Services and Health, Sydney. Casswell, S. (2004) Alcohol brands in young people s lives: new developments in marketing. Alcohol and Alcoholism 39, Centre on Alcohol Marketing and Youth (2004). Youth Exposure to Alcohol Ads on Television, 2002: From 2001 to 2002, Alcohol s Adland Grew Vaster. The Center on Alcohol Marketing and Youth, Washington DC. Duffy V. B., Cain, W. S. and Ferris, A. M. (1999) Measurement of sensitivity to olfactory flavor: application in a study of aging and dentures. Chemical senses 24, Ellickson, P. L., Collins, R. L., Hambarsoomians, K. et al. (2005) Does alcohol advertising promote adolescent drinking? Results from a longitudinal assessment. Addiction 100, Food Standards Australia and New Zealand (2004) Draft Assessment Report Proposal P271 Liqueur Definition. FSANZ, Canberra. Garfield, C. F., Chung, P. J. and Rathouz, P. J. (2003) Alcohol advertising in magazines and youth readership. The Journal of the American Medical Association 289, Grube, J. W. and Wallack, L. (1994) Television beer advertising and drinking knowledge, beliefs, and intentions among school children. American Journal of Public Health 84, Hill, H. and Tilley, J. (2002) Packaging of children s breakfast cereal: manufacturers versus children. British Food Journal 104, Hummel T., Barz, S., Pauli, E. et al. (1998) Chemosensory eventrelated potentials change with age. Electroencephalography and clinical neurophysiology 108, Jackson, M., Hastings, G., Wheeler, C., et al. (2000) Marketing alcohol to young people: implications for industry regulation and research policy. Addiction 95, Jackson, N. (2000) Operation beer bottle : A Report on the Use of Bylaws in New Zealand to Control the Possession and Consumption of Intoxicating Liquor in a Public Place. May. Crown Public Health, Christchurch. Joseph C., Stevens, L., Cruz, A., et al. (1995) Taste sensitivity and aging: high incidence of decline revealed by repeated threshold measures. Chemical Senses 20,

5 142 P. GATES et al. Lamb, C. W., Hair, J. F. and McDaniel, C. (1999) Marketing, 5th edn. South Western College Publishing, Cincinnati, Ohio. Mackintosh, A. M., Hastings, G. B., Hughes, K. et al. (1997) Adolescent drinking-the role of designer drinks. Health Education 6, Prone, M. (1993) Package design has stronger ROI potential than many believe. Marketing News 13. Schlossberg, H. (1990) Effective packaging talks to consumers. Marketing News 6 7. Taylor Nelson Sofres (2003) TNS Media Intelligence. Taylor Nelson Sofres, New York. Wyllie, A., Zhang, J. F. and Casswell, S. (1998) Responses to televised alcohol advertisements associated with drinking behaviour of year-olds. Addiction 93,