Objective To examine the extent (if any) that cigarette stick dimension, tipping paper design and other decorative design/branding have on Australian smokers' perceptions of those cigarettes.
Methods An internet survey of 160 young Australian adult ever-smokers who were shown computer images of three sets of cigarette sticks—five sticks of different lengths and diameters (set A), five sticks with different tipping paper design (set B) and four sticks of different decorative design (set C). Branding was a between-subjects randomised condition for set C. For each set, respondents ranked sticks on most and least attractive, highest and lowest quality and strongest and weakest taste.
Results Cigarette sticks were perceived as different on attractiveness, quality and strength of taste. Standard stick length/diameter was perceived as the most attractive and highest quality stick, with men more inclined to rate a slim stick as less attractive. A stick with a cork-patterned tipping paper and a gold band was seen as most attractive, of highest quality and strongest in taste compared to other tipping designs. Branded sticks were seen as more attractive, higher in quality and stronger tasting than non-branded designs, regardless of brand, although the effects were stronger for a prestige compared with a budget brand.
Conclusions Characteristics of the cigarette stick affect smokers' perceptions of the attributes of those cigarettes and thus are a potential means by which product differentiation can occur. A comprehensive policy to eliminate promotional aspects of cigarette design and packaging needs to include rules about stick design.
- plain packaging
- public policy
- environmental tobacco smoke
- packaging and labelling
- end game
Statistics from Altmetric.com
- plain packaging
- public policy
- environmental tobacco smoke
- packaging and labelling
- end game
Australia has legislated to reduce the effectiveness of cigarette branding by introducing plain packaging.1 ,2 The proposed legislation focuses primarily on cigarette packs but also will regulate sticks to be either plain white or white with an ‘imitation cork’ filter. No other branding, colour or design features would be allowed, except for an alphanumeric code (one designed so as not to be meaningful to a consumer).2 The law will not constrain cigarette length or diameter, at least within the ranges currently on the Australian market. This study was designed to provide evidence on the value of regulating stick characteristics. It was conducted after the government announced its intention to introduce plain packaging, but before it announced that the legislation would also regulate the appearance of individual cigarette sticks.
Cigarette design includes the selection of cigarette paper, tip ventilation and a blend of tobacco.3 Sticks can vary in length, diameter, the presence or absence of a filter, the colour and patterning of the tipping paper, the colour of the paper covering the rod and whether or not there is identifying information about the brand and/or variety on the rod. All these aspects could be used for product differentiation to facilitate the marketing of brands of cigarettes as having characteristics that are more desirable to some smokers than competing brands. Tobacco industry documents show that successful cigarette branding is defined as the integration of what the customer expects, paired with all the other aspects of the product, including name, packaging and the product itself.4
There is considerable documentation that emphasises the importance that the tobacco industry places on cigarette packaging,4 ,5 and research has investigated the impact of packaging on smokers6–9 and the role of packs as advertising.10–13 However, research on cigarette stick design is limited, with some evidence from tobacco industry documents that the industry has actively used elements of stick design to promote their products. For example, documents from Brown & Williamson Tobacco14 showed that a target female audience viewed their Superslim Capri cigarettes as “an escape from ordinary ‘fat’ cigarettes in favour of a more elegantly feminine cigarette for mature, confident women.” Following is the Capri brand statement from the same document:
Capri is the brand I wear everyday. It is an expression of my femininity. Its long, slim, delicate shape makes me feel elegant and attractive, like no other cigarette. Choosing Capri is one more way in which I show that I care about the way I present myself. It means others will notice me in the light I wish to be seen…a real lady who is uniquely feminine, elegant and stylish.
This statement reflects the importance that tobacco marketers have placed on the use of cigarette stick design as a branding tool. In Australia, all or nearly all cigarettes have the brand names on the stick, and many have other design features, such as gold bands, use of logos and patterning on the filters. Given the additional cost to print such design features on sticks, it is apparent that tobacco companies expect some commercial benefit from this practice.
Some of the evidence for the importance of aspects of stick design on consumer perception of tobacco products comes from a methodology of testing called the sensation transfer technique.15 ,16 This technique has shown that smokers' perceptions of the attributes of a cigarette (such as mouth effects and taste) can vary by changing related elements, such as branding and design. A typical test would involve a testing panel that smoked and then rated cigarettes on a range of attributes, such as mouth and throat effects, draw effort, impact, acceptability and taste level. These tests would be conducted under a number of conditions: testers would smoke masked cigarettes (brand identification markings on the stick were masked and the pack absent), cigarettes unmasked to brand (branding on the stick was visible and the pack absent) or unmasked to brand and packaging (branding on the stick was revealed and the respective cigarette pack the cigarette was from was shown). Studies by British American Tobacco using this methodology showed that some cigarettes were evaluated differently when the branding on the stick was unmasked and independent of whether the respective pack was shown.15 ,16
This study investigates the potential of using differences in cigarette sticks as a means of implying characteristics of the cigarettes. To our knowledge, there are no published experimental studies independent of the tobacco industry comprehensively examining the impact of cigarette dimension, tip colour and stick design on consumer perception of attractiveness, quality or taste. This study focuses on the perceptions of young Australian adult smokers and recent quitters. In particular, we explore whether cigarette stick physical dimensions (length and diameter), tipping paper colour and other decorative and branding elements affected ratings of attractiveness, perceived quality and strength of taste of such cigarettes. We also assessed whether branding (prestige vs budget brand) and respondent's gender affected evaluations.
One hundred and sixty Australian adult ever-smokers aged from 18 to 29 years (median 25 years) completed the study. Table 1 shows the demographic breakdown of the sample: 80% were current smokers, the rest had quit recently (median quit length was 5.2 months). An equal number of men and women were enrolled using quota sampling.
Participants were exposed to three conditions in sequence with cigarettes presented in sets on the left of the webpage. All images of cigarette sticks were computer generated and in colour, and actual size of cigarette sticks are shown. Set A consisted of five cigarette sticks of differing physical dimensions all with an imitation cork tipping paper (see figure 1: Standard length and diameter, Extra Long, Short, Standard length Slim and Extra Long Slim). The remaining two sets used only sticks of standard length and diameter. The image of the Standard stick was shown in the first (top) position as a reference (size is not as immediately salient in pictorial presentations as scale is not necessarily obvious), with the position order of the other sticks randomised below it. After completing the evaluations, respondents were then asked if they had ever smoked cigarettes sticks from this set (apart from the Standard).
Set B consisted of five sticks with differing tipping paper designs (figure 2: Plain (imitation) Cork (same as the Standard stick shown in set A), White Tip (imitation), Cork with Gold Band and Fancy Embossed Tip.
Set C was of four sticks (see figure 3: Plain cork, White tip, Branded in standard font and Current design (branded as by the company)). The latter two sticks both had a cork tip, and respondents were randomised to view either a well-known prestige (Dunhill) or budget (Horizon) brand label. In addition for set C only, the pictures of the sticks were accompanied by a picture of a prototype plain packaged pack labelled with the appropriate brand, and respondents were told all cigarettes depicted were from this brand. This was because our focal interest was the effect of branding on the stick, and we needed to establish that all sticks were of the same brand and variety.
For each set of sticks, respondents were asked to identify the stick that had most and least of three particular characteristics, with a “They are all the same” option always provided. The three questions were “Which of these cigarettes is the most attractive and which is the least attractive cigarette?” “Which of these cigarettes is of the highest quality and which do you think is of the lowest quality?” and “Which of these cigarettes do you think will have the strongest taste and which the weakest taste?”
They were also asked “Now regardless of price, which of these cigarettes would you choose to smoke?” and “Which of the cigarettes would you be least likely to choose?” Here there was no ‘equal’ option.
Overall mean ratings17 for each descriptor for each stick were then calculated by weighting as 5 points each ranking of most attractive, highest quality, strongest taste, most preferred; weighting as 1 point each ranking of the least attractive, lowest quality, weakest taste, least preferred; and scoring all other cases where the descriptor was not rated as either as 3 points (including cases where all were rated equivalent). For set A, respondents were additionally asked if they had ever smoked each of the variants shown (except the standard diameter and length, which we assumed all had used). Other data collected included demographics (age, gender, education, etc), usual brand preference and cigarette consumption.
A registered market research company (the Social Research Centre) was commissioned to administer this design via a web-based questionnaire. Respondents were drawn from a national internet panel recruited for market research purposes and who were identified as smokers in the panel database. The panel was not originally recruited with research on tobacco issues in mind. As a reward for participation in the study, respondents were awarded credits as part of a redemption scheme devised by the market research company. Ethical approval was obtained from the Cancer Council Victoria Human Research Ethics Committee.
Data analysis was conducted using SPSS V.18.0. Repeated measures analysis of variance was used to test for mean differences between sticks for each set and to identify interactions with brand or gender. Post hoc tests used Bonferroni corrections for multiple comparisons. We used a significance level of 0.05 throughout, but note that within-subject power to find effects was greater than for between-subjects effects.
Set A: stick length and diameter
Most respondents (81%) had experience with non-standard sized stick lengths/diameters: 51% with an Extra Long stick, 31% a Short stick, 60% a Standard Slim stick and 29% an Extra Long Slim stick. Purchase history of non-standard sized sticks did not significantly affect preference ratings.
There were significant differences in rankings of attractiveness (F (3.7) =40.01, p<0.001), perceived quality (F (3.8) =33.9, p<0.001) and estimated strength of taste (F (3.6) =41.39, p<0.001) (see table 2). Post hoc tests showed that the Standard stick was rated as both most attractive (all ps<0.001) and highest in quality (all ps<0.001) compared to all other stick lengths/diameters. Also, the Standard and Extra Long sticks were seen as more attractive than their slim counterparts (Standard vs Slim, p<0.001; Extra Long vs Extra Long Slim, p=0.02) and higher in quality (Standard vs Slim, p<0.001; Extra Long vs Extra Long Slim, p=0.002). The Extra Long stick was also rated higher in quality compared to the Short (p=0.005), Standard Slim (p=0.038) and Extra Long Slim sticks (p=0.002).
For estimated strength of taste, the Standard, Extra Long and Slim cigarettes were clustered as similar in rankings (all pairwise comparisons were at p>0.05), as was the Standard Slim and Extra Long Slim (p>0.05). However, both slim cigarettes were ranked weaker in taste than the other three (all ps <0.001).
Because some slim cigarettes have been marketed to women, we looked for gender effects. There was evidence of an interaction for attractiveness (F (3.4) =4.85, p=0.001), with the preferences against slims being stronger for men than for women (p=0.001), but there were no interaction effects for gender by quality or gender by taste.
It is also notable (see table 2) that most participants saw differences in attractiveness (only 13% rated all sticks the same on attractiveness), but notably fewer inferred differences in quality or strength (ie, a greater proportion of respondents rated the sticks equal on these qualities).
When asked, a majority of respondents indicated that they would prefer to smoke the Standard stick. Rankings of cigarette least preferred were mixed, with the largest group (43%) saying that they would least prefer to smoke the Extra Long Slim stick. Furthermore, stick preference was strongly associated with attractiveness (ρ (Rho) =0.65, p<0.001), moderately associated with quality (ρ=0.48, p<0.001) and hardly associated with taste (ρ=0.23, p=0.027).
Set B: stick tip design
The set of sticks varying in tipping colour and embossing were next evaluated (table 3) and were found to differ in rankings of all three: attractiveness (F (3.5) =35.94, p<0.001), perceived quality (F (3.7) =34.78, p<0.001) and perceived taste (F (3.5) =43.2, p<0.001). Post hoc tests revealed that Cork with a Gold Band was seen as most attractive (all ps<0.001), highest quality (all ps<0.001) and strongest in taste (all ps<0.001) compared to all other sticks. There was no differentiation between the Plain Cork, White Tip, and Embossed Cork Tip sticks on attractiveness scores (p>0.05). The Blue Tip was seen as least attractive and the White Tip the weakest tasting cigarette. There was little differentiation beyond the Gold Band stick in all three ratings for most attractive, highest quality and highest strength, but more variability between sticks at the other end of these scales. Again, respondents were more likely to rank the different stick designs as equivalent in quality (30% of the sample) and taste (39%) than equivalent in attractiveness (13%). There were no interactions effects with gender.
When asked to choose which stick would they most and least choose, the Cork with Gold Band was clearly most preferred, with the Blue Tip least preferred. Stick preference was strongly associated with attractiveness (ρ=0.66, p<0.001) and quality (ρ=0.67, p<0.001) and had a lower association with taste (ρ=0.29, p=0.009).
Set C: branding
The third set of stimuli focused on branding, and again, there were differences in ranking on all three dimensions: attractiveness (F (2.6) =53.71, p<0.001), quality (F (2.7) =47.30, p<0.001) and taste (F (2.7) =61.94, p<0.001). Table 4 shows that the branded sticks (Cork with Brand and Current Stick Design) were seen as more attractive, higher in quality and stronger tasting than the non-branded designs (Plain Cork and White Tip). Furthermore, post hoc tests revealed a definitive ranking from most to least attractive: the Current Stick Design, then Cork with Brand, Plain Cork and finally White Tip. Though White Tip was seen as the least attractive, equal lowest in quality with Plain Cork (p>0.05), and weakest in taste, those exposed to Horizon branding rated this less extremely than those shown Dunhill branding. Further evidence of branding interactions for attractiveness (F (1) =11.44, p=0.001), quality (F (1) =8.36, p=0.004) and taste (F (1) =4.98, p=0.027) showed that sticks with Dunhill (prestige) branding were seen as more attractive, higher in quality and stronger in taste than the Horizon (budget) branding.
Note that over a third of the sample rated the sticks as non-differentiable in quality and taste, but like in the other sets, less so for attractiveness (table 4).
Finally, over half preferred to smoke the Current Stick Design, and a majority least preferred the White Tip. Preference was not significantly associated with taste but was strongly associated with attractiveness (ρ=0.71, p<0.001) and moderately associated with quality (ρ=0.45, p<0.001).
Relationships between measures
Correlation analyses were performed on the rankings of most attractive, highest quality and strongest taste (table 5). For all three sets, there was a moderate association between sticks rated most attractive and highest quality (from ρ=0.38 to ρ=0.40), though a gender analysis for set A revealed women more strongly associating attractiveness with quality (ρ=0.60, p<0.01). Attractiveness and taste also had a low correlation (ranging from ρ=0.21 to ρ=0.27). For each set, this was attributable to only men associating attractiveness with taste (women for each set: NS). The association between quality and taste varied by set. Set A had a low association (ρ=0.22, p<0.05), set B had a moderate association (ρ=0.52, p<0.01) and set C had a strong association (ρ=0.69, p<0.01). Branding seemed to play no major role in the pattern of correlations.
The results of this study show that cigarette stick shape, colour and design are differentiable to smokers on qualities, such as attractiveness, perceived quality and estimated strength of taste. Smokers viewed the Standard stick length and diameter as the most attractive and of the highest quality, especially when embossed with a gold band and/or branding from a premium brand. Gender played a role, with men more inclined to rate slim sticks as less attractive. Additionally, the thicker ‘standard diameter’ sticks were perceived as more attractive, higher in quality and stronger in taste than their slim counterparts. These findings suggest that length and diameter can be used as a point of differentiation among cigarette sticks. The preference for the Standard stick may highlight a tendency for smokers to prefer the familiar. That said, showing the standard sized stick in a fixed reference position may have affected its rating in comparison to the other length/diameter variants, so caution should be exercised in interpreting the strong preference for it.
The results demonstrate the importance to the industry of having the name on the sticks. While this effect was stronger for the prestige brand, it was present for both brands. Unlike the brand name on the pack, branding of the stick is not necessary as sale of single cigarettes is prohibited in Australia.
There are a number of limitations to this study. The cigarette sticks were shown on the internet as computer-generated artist impressions and not photographs of real sticks. The use of pictorial images also meant that respondents were deprived of a tactile exploration of the stick before being asked to make value judgements. We are not sure what effect allowing respondents to handle real sticks would have, but note that without access to cigarette-making facilities, it is extremely difficult to produce high-quality cigarettes with design elements to order.
As respondents were drawn from an internet panel, the sample is likely to over-represent those with private internet access. It is also not a representative sample but is otherwise unselected on any aspect related to the study aims, so responses are unlikely to differ much from a representative sample. Finally, the three substudies were ordered, and it is possible responses to the earlier ones influence later ones, although we cannot postulate any likely effects. These limitations mean that caution should be taken in drawing conclusions about the magnitude of effects reported here, though the conclusion that observable aspects of cigarette stick design can affect judgements is unaffected.
The study also has a number of strengths. The use of a within-subject design to explore key comparisons means that although the sample size is relatively small, the study has considerable power. We also surveyed younger smokers and ex-smokers (aged 18–29 years), a group typically more brand conscious,18 and the group thus most likely to be affected by restrictions to cigarette design. Having recent ex-smokers does increase the generalisability of the results, although with the small sample size of ex-smokers this should be done with caution. The sample also means that caution is required in generalising the results to older smokers.
The finding that some of the rated characteristics were quite highly correlated but that the pattern of correlations differed is strong evidence that at least some of the minority preferences are likely to be real. If responding were random, there would be no correlations, and if there were stereotypic responding very high correlations. These are both possible if the respondents genuinely held no preferences. Varying correlations is evidence of varying preferences, albeit weak. As a result, we should not assume that there is no potential niche market for any of the designs we used, even those that generally rated poorly.
Research has shown that brand loyalty, once fostered, is hard to break,19 ,20 so we cannot make any strong claims about changes to stick characteristics translating into changes in smoking. However, to the extent that standardising sticks reduces brand differentiation, we might expect that some of the value the brand has acquired with smokers would be lost, and thus, there may be both a reduction in brand loyalty and a diminished sense of value in smoking, the latter of which might be expected to reduce reluctance to quit.
The findings clearly provide support for removing branding and attractive features from sticks as the Australian legislation will achieve. The Legislation does not attempt to control stick length or diameter. Our data suggest that this can be a factor affecting perceptions, and it will be interesting to see if tobacco companies increase the range of products that vary on these dimensions to increase brand differentiation in a context where plain packaging will tend to reduce it. In this case, we think that it is more the variation than the actual characteristics that may be important; while for some of the design features, there is more of a sense of something intrinsic to particular designs.
This study shows that characteristics of the cosmetic elements of the cigarette stick itself are potential means by which product differentiation can occur. As a result, standardising cigarette sticks is desirable to remove the capacity of these features to be used in the future to drive brand differentiation. As tobacco companies explore unregulated areas to further differentiate their brand, it would be surprising if stick elements were not varied to the extent that it is allowed. The Australian government have made a wise decision in greatly restricting the range of design options.
What this paper adds
This paper shows that characteristics of cigarette stick design can be used by consumers to make judgements about characteristics of the cigarettes.
Thus, they have the capacity to act as a form of promotion and should be regulated where public policy is to eliminate as far as possible promotion of tobacco products.
As a consequence, it provides evidence supporting the decision of the Australian government to regulate stick design features as part of its Plain Packaging legislation.
Funding Funded by Quit Victoria and the VicHealth Centre for Tobacco Control, Cancer Council Victoria.
Competing interests RB is a member of a Technical Advisory Committee advising the Australian Department of Health and Ageing on various aspects of the implementation of the plain packaging legislation. He did not use any information he may have gained on that committee in making decisions on the form of the study, and this study was designed and implemented completely independent of that committee.
Patient consent On-line survey.
Ethics approval Ethics approval provided by Institutional Research Review Committee. The Cancer Council Victoria Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.