Do UK Television Alcohol Advertisements Abide by the Code of Broadcast Advertising Rules Regarding the Portrayal of Alcohol?

Aims: The aim of the study was to assess the extent to which members of the UK general public perceive television alcohol advertisements to comply with the regulatory code governing these: the Advertising Standards Authority Code of Broadcast Advertising (BCAP Code). The Code provides a general principle and 16 rules to prevent such adverts implying, condoning or encouraging immoderate, irresponsible or anti-social drinking. Methods: Quota sample of 373 adults, representative of the UK population aged 18–74 years in terms of age and gender, were recruited at a train station. Participants were shown one of seven advertisements that had been broadcast in the previous month on the two leading commercial television channels, and then completed a questionnaire with 40 statements representing the BCAP Code rules. Results: Overall, 75% of the participants rated the advertisements as breaching at least one rule from the BCAP Code. Breaches were observed for all the seven advertisements, ranging from 49 to 91% non-compliant. Rules regarding alcohol being presented as contributing to popularity or confidence, and implying that alcohol is capable of changing mood, physical condition, behaviour, or as nourishment, were seen as being breached by over 50% of participants. Conclusions: A clear majority of the UK general public perceive alcohol advertisements to breach the BCAP Code, suggesting that the current regulatory system for UK television alcohol advertisements is inadequate.

INTRODUCTION

In the UK, the total annual expenditure on alcohol advertising is around £200 million, 49% of which is spent on television advertisements (House of Commons Health Committee, 2010). Members of the alcohol and advertising industries argue that alcohol advertisements do not influence levels or patterns of consumption, but serve to promote brand loyalty (House of Commons Health Committee, 2010). This claim stands in opposition to a wealth of international evidence that has demonstrated both immediate and longer-term effects of alcohol advertising on individual and population levels of consumption (Meier, 2010). Several systematic reviews of longitudinal studies have found that the greater the volume of alcohol advertising young people are exposed to, the earlier they initiate drinking and the more frequently and heavily they drink (Booth et al., 2008; Anderson et al., 2009; Smith and Foxcroft, 2009). Further, a meta-analysis of 132 econometric studies found a small but consistent association of alcohol advertising expenditure with population levels of consumption (Gallet, 2007).

Many countries have codes that regulate the contents of these advertisements, in an attempt to ameliorate the more harmful effects of alcohol advertising. In the UK, broadcast advertising is co-regulated by the Advertising Standards Authority (ASA) and Ofcom, the television broadcast regulatory authority, with the ASA assuming responsibility for regulating advertisements on a day-to-day basis. The ASA's Broadcast Committee of Advertising Practice (BCAP), which consists of representatives from the advertising and broadcast industries, is responsible for writing and reviewing the regulations in the form of the BCAP Code.

The BCAP Code consists of general rules that apply to all advertisements, as well as rules dedicated to specific categories of products or services. The general rules require that advertisements are recognizable as such, are scheduled appropriately (e.g. so that alcoholic products are not advertised in programmes principally directed at audiences below the age of 18), do not mislead and do not cause serious or widespread offence or harm (Broadcast Committee of Advertising Practice, 2010). In addition to these general rules, the BCAP Code specifies a general principle that applies to the content of broadcast alcohol advertisements: ‘Advertisements for alcoholic drinks should not be targeted at people under 18 years of age and should not imply, condone or encourage immoderate, irresponsible or anti-social drinking’ (Broadcast Committee of Advertising Practice, 2010). This principle is embodied in 16 more specific rules exclusively governing televised alcohol advertisements (see Table 1 for description).

The ASA is responsible for investigating complaints about televised alcohol advertisements being in breach of the BCAP Code, with the complainant having no right of appeal. It is ultimately the responsibility of the broadcasters to withdraw, change or reschedule an advertisement, as a condition of their broadcast licences. Where broadcasters persistently run advertisements that breach the relevant advertising Codes (including the BCAP code), the ASA can refer them to Ofcom, which can impose sanctions including fines and withdrawal of broadcasters’ licences.

To date, there have been no adequately reported studies in the UK examining the success of the BCAP Code, in relation to whether advertisements abide by this Code. The only studies that have been conducted to date are several surveys conducted by the ASA itself, the most recent of which reported over 99% compliance of advertisements across all media with the BCAP and CAP Codes (Advertising Standards Authority, 2010). The ASA provided few details on the methods used to arrive at this figure, and the results have not been published in a peer-reviewed journal.

The figure of 99% compliance could be taken at face value as indicating that the BCAP is effective at protecting the public from irresponsible advertising. However, the figure of 99% compliance is implausible in the light of comparable international studies which have investigated the extent to which people perceive television alcohol advertisements as complying with the relevant national advertising codes.

An Australian study found that members of the general public aged 15–24 years old perceived all 12 print and television advertisements studied to be in breach of their Alcoholic Beverages Advertising Code (ABAC), most often by linking alcohol to social or sexual success, portraying alcohol as necessary for relaxation or helping them feel more confident (Jones et al., 2009). In a more recent Australian study, two research assistants performed a content analysis of 13 alcohol advertisements and found that 6 of the 13 advertisements were in breach of the ABAC, most often by associating alcohol with social, sexual or sporting success or encouraging immoderate consumption (Jones et al., 2010).

In the USA, 139 non-industry experts rated six television advertisements to be in breach of the US Beer Code, particularly by portraying beer in an irresponsible manner and by appealing to viewers under the legal drinking age (Babor et al., 2010). In a more recent US study 15 public health professionals rated between 35 and 74% of 248 beer advertisements to be in violation of the US Beer Code, particularly by associating beer with social success or appealing to viewers under 21 years of age (Babor et al., 2013b). An additional study found that 149 expert raters were less likely to report violations of the US Beer Code than 147 community raters from various vulnerable groups (Babor et al., 2013a). A similar study in Brazil found that teenagers rated many content violations of the Brazilian self-regulation code, particularly those relating to targeting young people and suggesting irresponsible and immoderate consumption (Vendrame et al., 2010).

Given this international background, there is a need to establish to what degree the general public in the UK consider alcohol advertisements to comply with the BCAP Code. To do so would be timely, in light of a recent analysis of internal marketing documents from four major alcohol producers relating to four alcohol brands in the UK that were obtained for a UK House of Commons Health Select Committee (HSC) investigation into the conduct of the alcohol industry. This analysis found that advertisements incorporated many prohibited themes, particularly relating to immoderate consumption and social and sexual success (Hastings, 2009). The author of this analysis argued that current regulations focus on content, which requires regulators to make judgments about concepts such as ‘social success’ or ‘masculinity’ which are open to multiple and contested interpretations (Hastings, 2009). Consequently, it is difficult for regulators to pin down the subtle emotional associations present in modern advertising as indicating the use of such concepts which would be in breach of the BCAP Code (Hastings et al., 2010). The present research asks the general public to be arbiters on whether these concepts are present or not, a method which overcomes many of the difficulties encountered by the current regulatory approach.

Aims

The main aim of the present research was to assess the extent to which members of the UK general public perceive television alcohol advertisements as complying with the BCAP Code.

The present study had three secondary aims:

1. To assess the extent to which there is variation between advertisements in terms of perceptions of whether television alcohol advertisements comply with the BCAP Code.

2. To identify which rules are perceived as most frequently breached by television alcohol advertisements, and those that are most frequently complied with.

3. To assess the association between the extent to which members of the UK general public perceive television alcohol advertisements as complying with the BCAP Code and demographic factors, specifically age group and gender.

METHODS

Design

This cross-sectional survey required each participant to view one of seven different advertisements, and provide his or her views on the advertisement viewed.

Participants

A total of 373 adult volunteers were recruited at Manchester Piccadilly station between 8:00 and 19:00 hours on six weekdays in March 2013, with a response rate of 39%. An aim of recruitment was to fill quotas defined by age and gender, so that the final sample was representative of the 2013 UK adult population (aged 18–74 years) in terms of age group and gender, based on projections for the 2013 UK population from the 2010 Census data (Office for National Statistics, 2011a). Participants were not eligible for the study if they had severe learning difficulties or had insufficient English.

Procedure

Potential participants were approached in the main waiting area at Manchester Piccadilly train station and were provided with verbal and written information regarding the study. Those who agreed to participate completed the questions regarding their demographic background, watched one of the seven different advertisements on a portable computer with headphones, and then completed a questionnaire regarding the advertisement they had just watched, using pen and paper. Quota sampling was operationalized through use of a grid containing details of advertisements, gender and age categories. In each period of recruitment, an aim was to obtain equal numbers in each cell of the grid.

Materials

Advertisements were sampled from 19:00 to 23:00 hours on the two leading commercial channels in the UK (ITV1 and Channel 4) from Monday 11 February to Sunday 24 February 2013. In total, eight different advertisements for seven different brands of alcoholic drink were identified.

To ensure that there was only one advertisement for each brand of alcoholic drink, the first of two advertisements for Disaronno Originale was randomly selected for inclusion in the study and the other was excluded. Thus the final sample was comprised of seven different advertisements for seven different brands of alcoholic drink, the contents of which are summarized in Table 2.

The present study used a questionnaire measure derived from previous work by Babor et al. (2008). The questionnaire used in the present study contained 40 items, with at least one item for each alcohol-specific rule in the BCAP Code (see Table 1). Some rules required multiple items as they contained multiple conditions, which were simplified into a larger number of statements with single conditions. The participants were required to rate their agreement or disagreement with each statement on a five-point Likert scale, using the following response categories: ‘strongly agree’, ‘agree’, ‘neither agree nor disagree’, ‘disagree’ and ‘strongly disagree’. Rule 19.15.2 was excluded from the questionnaire, as it was deemed unsuitable for the adult participants to assess which characters’ example was likely to be followed by, or had a strong appeal to, those under 18 years of age. Rule 19.11, which permits sales promotions provided that they do not imply, condone or encourage immoderate drinking, was deemed to be adequately represented by rule 19.2, which prohibits advertisements from featuring, implying, condoning or encouraging irresponsible or immoderate drinking, and thus was excluded from the questionnaire.

Information on participants’ demographic characteristics was collected; namely their sex, age group and ethnic group. Categories for age and ethnicity were taken from the British Household Questionnaire (Office for National Statistics, 2011b). All measures were piloted for comprehension before data collection.

Analysis

Responses recorded on five-point Likert scales were dichotomized, such that ‘strongly agree’ and ‘agree’ indicated a breach of the Code and ‘neither agree nor disagree’, ‘disagree’ and ‘strongly disagree’ indicated compliance with the Code. Responses to individual questions were grouped according to the rules, such that responding ‘strongly agree’ or ‘agree’ to one or more of the questions indicated a violation of the corresponding rule, while a violation of one or more of the rules indicated a violation of the BCAP Code. One-sample t-tests were used to assess whether proportions were significantly different from zero, and 95% confidence intervals for proportional data were calculated for each rule and for the overall BCAP Code. Z-ratios were used to assess whether rates of rule violations differed significantly between different advertisements, age groups and genders, while McNemar's test was used to test whether rates of rule violations differed significantly between different rules.

RESULTS

The demographic features of the final sample are displayed in Table 3 with regards to the number and proportion of participants in each group according to sex, age and ethnicity. The population figures for sex and age are based on the projections for the 2013 UK population based on the 2010 Census data (Office for National Statistics, 2011a), while the population figures for ethnic groupings are based on the 2011 Census data for the whole population of England and Wales (Office for National Statistics, 2013).

Overall compliance

Overall 279 (75%) of the participants perceived the advertisements to breach at least one alcohol-specific guideline in the BCAP Code [t = 33.23, df = 372, P < 0.001, 95% CI (0.70, 0.79)].

Comparisons between advertisements

The proportion of participants who perceived advertisements to breach at least one rule from the BCAP Code was significantly different from zero for all seven advertisements (all P < 0.001). Despite this, there was variation in the proportion of participants who perceived the advertisements to breach at least one alcohol-specific rule in the BCAP Code between advertisements (shown in Table 4). Rates of perceived breaches were significantly lower for the advertisements for Strongbow Pear (58% of respondents indicated at least one breach) and Disaronno Originale (49% indicated at least one breach) than the other five advertisements.

Comparisons between guidelines

The proportion of participants who perceived advertisements to breach the BCAP code was significantly different from zero for all 15 rules examined (all P < 0.05).

The proportion of participants who perceived the advertisements to breach the BCAP code varied by the specific rules (shown in Table 5). The rates of perceived breaches were highest for the rules ‘must not imply that alcohol can contribute to popularity or confidence nor imply that alcohol can enhance personal qualities’ (rule 19.3, 52% rated as breached) and ‘must not imply that alcohol has therapeutic qualities… capable of changing mood, physical condition or behaviour or as nourishment’ (rule 19.8, 51%). In contrast, the rule ‘must not link alcohol to illicit drugs’ (rule 19.9) was rated as breached by only 1.2% of respondents.

Comparisons by age group and gender

There were no significant differences in the proportion of participants in each age group and of each gender who perceived the advertisements to breach at least one alcohol-specific guideline in the BCAP Code (all P > 0.05).

DISCUSSION

Principal findings

Overall, three in four participants perceived the advertisements to breach at least one alcohol-specific rule in the BCAP Code. There was significant variation across advertisements in perceived compliance, with the proportion of perceived breaches ranging from 91% for Captain Morgan's Spiced to 49% for Disaronno Originale. Thus, even the advertisement that was seen as most compliant was perceived by nearly 50% of respondents as breaching the BCAP Code. There was also significant variation across rules, with over half of the participants perceiving breaches of rules which prohibit advertisements from implying that alcohol can contribute to popularity or confidence (19.3) or that it is capable of changing mood, physical condition or behaviour (19.8), while only 1% perceived a breach of the rule which prohibits advertisements from linking alcohol to illicit drugs (19.9). These perceptions did not significantly vary according to participant age or gender.

Strengths and limitations

The present study is the first to investigate the extent to which members of the UK general public perceive television alcohol advertisements as complying with the BCAP Code. A sample that was representative of the UK general population between the ages of 18 and 74 years in terms of age and gender was obtained, but there are several biases in terms of representativeness of the UK general population that are known, likely or at least possible.

First, we know ethnic minority participants were under-represented in the sample obtained (6%) compared with the UK population (18%). Further, it is reasonable to assume that by collecting data at a metropolitan train station during the day that people who live in other parts of the UK, people who live in rural communities, and probably low socio-economic status (SES) were under-represented in the sample obtained. In addition, no data were collected on other participant characteristics which may have affected perceived breaches, such as income, SES, education, religion, parental status, TV viewing habits and drinking habits. There are clear patterns of association between many demographic measures and alcohol consumption (e.g. Cheng and Furnham, 2013). However, no associations were observed in the present study between demographic characteristics and perceptions of breaches in the BCAP code.

In sum, the estimates presented for perceived breaches may be somewhat biased due to these known or unknown shortcomings in the representativeness of the sample to the UK general population. However, to change the overall conclusions that can be drawn from the present research, the biases in representativeness would have to be very large, and the perceptions of under-represented groups very different from those of included participants. Nevertheless, these issues warrant further detailed exploration in future studies, to quantify the extent to which different participant characteristics affect perceptions of breaches of the BCAP code.

The questionnaire was piloted prior to the study, but it was not rigorously evaluated with the target population. However, the test-retest reliability, inter-rater reliability and validity of the questionnaire are partly confirmed by previous work by Babor et al. (2008), who performed psychometric analyses of the questionnaire from which this one was derived. Further, the items were explicitly based on the rules contained within the BCAP code, to maximize relevance of this research.

A key question when interpreting this research is the extent to which the general public is able to validly recognize where advertisements breach the BCAP code. It is our contention that the public are an appropriate group, as they are the targets of the advertising. Further, it is the perceptions of the public that contribute to alcohol-related harms, not those of experts. Thus, when participants in this study indicate that the advertisement they have just seen implies that alcohol contributes to popularity, confidence or enhanced personal qualities, it is difficult to argue that their perceptions are not valid.

Relation to wider literature

Consistent with studies conducted in Australia (Saunders and Yap, 1991; Jones et al., 2010), the USA (Babor et al., 2010) and Brazil (Vendrame et al., 2010), the majority of the participants perceived the advertisements to breach at least one rule in the BCAP Code. The present study differed from those that have been conducted previously in that the proportions of perceived breaches were tested to see if they were significantly different from zero, as opposed to 0.5 (Saunders and Yap, 1991; Babor et al., 2010; Vendrame et al., 2010). This was deemed most appropriate for the present analysis as, in principle, all advertisements should be fully compliant with the BCAP Code and thus no significant proportions of participants should perceive a violation of any of the rules regarding the portrayal of alcohol for any of the advertisements analysed. Further, there was considerable variation in perceptions of breaches, which provides support to the idea that participants were more than capable of discriminating between when violations were present and where they were absent.

Irrespective of these arguments, if the comparison value is adjusted to 0.5, the overall proportions of participants who perceived the advertisements to breach at least one rule regarding the portrayal of alcohol in the BCAP Code remain significant overall, and for five of the seven advertisements individually. It should also be highlighted that the overall figure of 75% of participants perceiving a violation of BCAP code was in relation to the single advertisement they viewed. It therefore appears likely that if each participant viewed a number of advertisements, the proportion that viewed at least some advertisements as breaching the BCAP Code would be even higher.

The findings of the present study are in opposition to research conducted by the ASA, who most recently reported over 99% compliance of alcohol advertisements across all media with the BCAP and CAP Codes (Advertising Standards Authority, 2010). It is notable that since the most recent BCAP Code was implemented in 2010, the ASA have investigated nine alleged breaches by television alcohol advertisements. Of the nine alleged breaches, four were upheld, one of which was deemed to breach guideline 19.15.1, which prohibits advertisements from appealing strongly to people under 18 years, from reflecting or being associated with youth culture and from showing adolescent or juvenile behaviour, while the other three breached one or more of the rules that apply to all advertisements (Advertising Standards Authority, 2013). Thus, whatever method the ASA used to assess compliance, it produces very divergent results from those reported here.

Implications

The results of the present analysis suggest that the UK alcohol and advertising industries design advertisements that do not appear to comply with the letter or spirit, of the BCAP Code by alluding to themes such as youth culture, immoderation and social and sexual success, although they may not explicitly show them (Hastings et al., 2010). The results of the present analysis, along with the comparatively small number of ASA reported and adjudicated breaches, indicate that the co-regulation of UK television alcohol advertisements is ineffective and requires further consideration. The results of the present study suggest that the ASA are not adequately monitoring television alcohol advertisements for compliance with the BCAP Code. It further suggests that only a miniscule proportion of members of the general public who perceive advertisements as containing elements that breach the BCAP Code actually report them (Advertising Standards Authority and Committee of Advertising Practice, 2012).

Previous authors have emphasized the need for an independent review body for the regulation of television alcohol advertisements (Babor et al., 2008, 2010), with some suggesting that vulnerable groups should be involved in the review process, asserting that the best people to judge what a particular communication is saying are those in the target audience (Anderson , 2007; Hastings et al., 2010; Vendrame et al., 2010; Babor et al., 2013b). The results of the present analysis demonstrate that members of the general public are capable of using rating scales to assess the extent to which advertisements comply with the BCAP Code, and making sensible discriminations between the various rules contained with the Code.

If this is to be implemented, a decision has to be made about what is a sensible criterion regarding the proportion of the public who view a breach to have occurred for it to be decided that an advertisement is genuinely in breach. As previously noted, the present study used the criterion of a proportion of participants significantly greater than zero to indicate a breach, which is a very harsh criterion. In contrast, much previous research has used the criterion of a proportion of participants significantly >50%, which may be too lax. Consequently, a decision would have to be made about an arbitrary criterion that lies somewhere between 0 and 50% if the general public are to be used as a review body. It would clearly be important that such a review body is fully representative of the diversity of ethnicity, religion and other demographic groups.

An alternative way of viewing these results is that, given the wide discrepancy between the ASA and the general public in perceptions of alcohol advertisements, regulation based on examining content of advertisements is too fraught with complexities to produce an effective system. Thus, in line with other commentators (Hastings, 2009), regulation might more profitably focus on banning television advertising, rather than focusing on the content of such advertising, or possibly restricting the amount of advertising in conjunction with the current focus on advertising content.

Future research

Given the high proportion of perceived breaches with the present study, it is also advisable that similar studies are conducted with regards to alcohol advertisements in other media, particularly new media, of which there is increasing concern (British Medical Association, 2009).

The association between perceived breach and demographic characteristics, including religious groups, should be more fully examined in future research. Such future research might also examine the effects of examining the effects of presenting the questions relating to each guideline in different orders, to control for the order of the questions affecting the proportion of people indicating a breach for each guideline (Krosnick, 1999).

The questionnaire designed for this study should be rigorously evaluated with the target population in order to confirm or refute the reliability and validity of the findings. General public ratings could be compared with those of ‘experts’ to understand better how each group arrives at their judgments, and to evaluate the level of overlap in these judgments (Babor et al., 2013a). If the questionnaire is found to be both reliable and valid, further research should determine the feasibility of implementing such a rating procedure with members of the general public as part of an independent review body for the regulation of television alcohol advertising in the UK. ‘Expert’ groups could consist of public health practitioners, advertising practitioners or academics with expertise in human decision-making. Differences in perceptions of breaches could be compared across such groups, and with the general public, to ascertain where and why differences in views might be present.

Summary

Overall the results of the present analysis indicate that approximately three in four members of the general public perceive television alcohol advertisements to be in breach of the BCAP Code. These findings have important implications for future regulation of UK television alcohol advertisements and indicate that an independent review body would be preferable to the current co-regulatory system. If this is not implemented, then legislating against all televised alcohol advertising in the UK would avoid the many uncertainties of making judgments about content.

Conflict of interest statement

None declared.

Acknowledgements

The authors thank the staff at Manchester Piccadilly train station for their hospitality and those members of the general public who participated.

REFERENCES