ASA Adjudication on Procter & Gamble UK

Ad

A TV ad for Fairy washing up liquid showed a woman's hands engaged in various tasks: gardening; carrying shopping; washing; scrubbing a surface; cycling and peeling potatoes. The voice-over stated "After what your hands go through every day, don't you think they need just a little bit more glamour? Fairy clean and care aloe vera and cucumber helps look after while you do the washing up". On-screen text stated "Tested in panel of 256 Glamour readers". The voice-over continued "That's why Glamour readers recommend Fairy Clean and Care. Sink into Fairy."

Issue

A viewer challenged whether the claim "That's why Glamour readers recommend Fairy Clean and Care" was justified, because he believed the basis for the claim "Tested in panel of 256 Glamour readers" was insufficiently robust.

BCAP TV Code

Response

Proctor and Gamble UK Ltd (P & G) said that to ascertain whether a panel size was robust was not simply a matter of considering the numbers involved in the panel size but also the manner in which the panellists were recruited. They said their panellists had been recruited in a wholly random manner and selection was not biased in favour of readers who were or were not users of Fairy liquid. They said Glamour held the names of five thousand readers who had consented to take part in surveys such as this one. They said that the process for recruitment on this occasion was to e-mail all those on that list to ask if they would be interested in participating; they said the first four hundred readers to sign up were then sent a bottle of Fairy Clean and Care, and that 256 went on to complete the survey. They said they believed results from a panel of this size, provided the panel has been selected randomly, would produce a statistically significant result. They said their Head of Consumer and Market Knowledge had calculated that the survey returns showed a 99% confidence level that the results demonstrated by the panel were correct. They said the percentage of all readers of Glamour magazine who would recommend Fairy Clean and Care was extremely high and would fall within the range of 80.4% to 91.6%.

P&G said that, going forward, they would be happy to include the further specifics of the claim and state "That's why 86% of Glamour Readers recommend Fairy Clean and Care" in addition to the on-screen text stating the size of the sample. They provided a copy of the Fairy consumer test questionnaire sent to Glamour readers and a summary the results.

Clearcast said, because the 400 respondents to the first call for participation were selected on a first come first serve basis from Glamour's pool of readers who had signed up to complete such surveys and were not pre-selected for any favourable disposition to Fairy, then the results were statistically significant, i.e. to a 99% degree of certainty. They said that, from experience of dealing with other similar claims, they were of the view that the numbers fell comfortably within the acceptable parameters of statistical significance.

Assessment

Upheld

The ASA understood that Glamour magazine had a monthly circulation figure of around 46,000. We considered that, in order to extrapolate to the larger group for the claim "readers recommend", P&G would need to show in this instance that the sample was selected randomly from all Glamour readers such that it was representative and that the results showed a high positive. In such an instance, we considered an ad should avoid giving the misleading impression that the results indicated 100% of readers would be likely to recommend the product if that was not the case, for example by including the percentage, "86% of Glamour readers recommend" as well as the sample size.

However, we noted we had not seen evidence to show that Glamour's pool of 5000 readers who had agreed to be available to take part in surveys was demographically representative of Glamour's readership as a whole, nor whether the sample of 256 respondents in the Fairy survey was likely to be so, as both groups were self-selecting. We also noted that some questions in the survey asked respondents to "please indicate your agreement with the following statements: Strongly Agree/ Slightly Agree/ Slightly Disagree/ Strongly Disagree" and that the statements which followed were positive statements about Fairy Clean and Care. We considered that, by asking respondents to "indicate your agreement" rather than "indicate your views" the format could predispose respondents to circle the "agree" rather than the "disagree" options. For these reasons, we considered that the panelists could not, with confidence, have been said to have been recruited in a wholly random manner and that the survey design could also have influenced the results. We concluded the claim was likely to mislead on that basis.

The ad breached CAP (Broadcast) TV Advertising Standards Code rules 5.1.1, 5.2.1 (Misleading advertising) 5.4.2 (Superimposed text) and 5.4.6 (Comparative advertising).

Action

The ad must not appear again in its current form.

Adjudication of the ASA Council (Broadcast)