Research, consultation, planning, communications

This study assessed adherence of brand alcohol advertising to the Advertising Standards Authority and Broadcasting Standards Authority rules on advertising on television which came into force in February 1992.  One hundred and forty brand alcohol advertisements which first screened between 1st February 1992 and 30th June 1993 were assessed in terms of the interpretations of the Codes made by five groups.   These five groups were:

  1. (1)the anti-alcohol advertising lobby;

  2. (2)public health professionals;

  3. (3)"general viewers”;

  4. (4)the liquor industry and broadcasters; and

  5. (5)the Advertising Standards Complaints Board (ASCB) and Broadcasting Standards Authority (BSA).


The assessment was commissioned by the Alcohol Advisory Council to help it consider its position on the self-regulation of televised brand alcohol advertising.  The objectives were to:

  1. Bulletdefine a rigorous and standardised method to characterise and describe breaches of the Codes as understood by each group, and quantify the level of adherence to the Codes resulting from these breaches;

  2. Bulletuse this method to evaluate for each group adherence to the rules and principles of the Codes of all brand liquor advertising between February 1992 and June 1993, inclusive;

  3. Bulletprovide definite conclusions and recommendations on the extent to which the Codes and BSA principles are adhered to, can be adhered to, and/or can be improved; and

  4. Bulletprovide a basis upon which to assess the viability of a self-regulation regime for alcohol advertising.


These objectives were met by the development of a computer-based expert system.  The basic idea of expert systems is that "expertise" can be modelled in a way which is reproducible by turning it into a set of rules, procedures and an information base.  Thus a non-expert can be guided to make decisions in a way that an expert would.  Expert systems have been developed for a wide range of uses in medicine, engineering, business, geology, tax analysis, and the law.


In this case, the "expertise" to be modelled was how people decided whether an advertisement adhered to the Codes.  This was defined through the analysis of verbatim transcripts of a series of interviews and discussion groups with the various interested parties.  Separate systems were developed for each of these groups.


The results of the assessment produced by the expert system followed a predictable and credible pattern, from least to most breaches assessed in the order:

  1. Bulletalcohol industry/media - one advertisement in breach (1%);

  2. Bulletthe ASCB/BSA - two (1%);

  3. Bulletthe general viewer - 21 (15%);

  4. Bulletpublic health professionals - 68 (49%); and

  5. Bulletmembers of the anti-alcohol lobby - 103 (74%).  


Of the 21 advertisements assessed by the “general viewer” as in breach of the Codes, seven were assessed as breaching three or more provisions of the Codes.


An analysis of the discourse of the general viewers canvassed in this study and the way they evaluated advertisements in the interview stage of the study shows that their opinion is more closely aligned with the interpretation of the ASCB/BSA than the other groups in finding fewer breaches of the Codes and identifying breaches on similar rules.  It may be that a relatively modest shift in emphasis and interpretation would be required to achieve a much closer fit between the assessments of these two groups as the divergences of opinion occur only on a few of the rules.


The report concluded that the regime may have needed some refinement but that it was capable of addressing the interests and concerns of many people.  It was suggested that ultimately, if there is clarity in the rules, reasonable consensus on interpretations, and  consistency in the decisions on complaints then it appeared that a voluntary adherence regime would be workable and credible for many people.


The report recommended a review of the outcomes of this study with representatives of the various interest groups as a basis for clarifying the wording and interpretation of the rules.  It also recommended that sponsorship advertising was looked at in the same way as this brand advertising study.


Using expert systems in this way is valuable in processing significant amounts of information in a systematic and consistent way, without the influence of individual biases affecting particular assessments or the effects of fatigue on the part of the evaluator.  In this case, the 140th advertisement was assessed as rigorously as the first.