28913-20 Garrity v The Times

Decision of the Complaints Committee 28913-20 Garrity v The Times

Summary of Complaint

1. John Garrity complained to the Independent Press Standards Organisation that The Times breached Clause 1 of the Editors’ Code of Practice in an article headlined “Twelfth poll in a row shows majority back independence”, published on 6 November 2020.

2. The article reported on the results of a poll surveying Scottish adults about how they would vote in a Scottish independence referendum. Both the headline and the article reported that a “majority” of Scottish adults said they would vote Yes in an independence referendum, with the article then going on to state that the poll “placed support for leaving the UK at 54 per cent when undecided voters were excluded.” The article and headline also said that the poll was the “[t]welth poll in a row” to show a majority backing for independence.

3. The article also appeared online in substantially the same format.

4. The complainant said that the article was inaccurate in breach of Clause 1. He said that it was inaccurate to state that a “majority” of Scottish adults backed independence: 47% of poll respondents had said they would vote yes, 40% had said they would vote no, and 13% said they did not know. Where less than 50% had said they would vote yes in an independence referendum, the complainant said that it was inaccurate for the publication to state that a “majority” of Scottish adults had said they would vote yes. He then said that it was inaccurate to extrapolate the polls findings to the Scottish population where Don’t Knows had been excluded. This was because, even when “Don’t knows” were excluded, it remained the case that only 47% of respondents had voted yes. In order to achieve the 54% support, a new data set had to be created, including only those who gave a decisive yes or no answer, which was not reflective of the Scottish voter population, and therefore the complainant said that the 54% figure could not be applied to the Scottish population. He also said that only 3 polls since 2015 showed a majority of support (meaning over 50%) for Scottish independence, and that it was inaccurate for the publication to state that twelve polls “in a row” showed that a “majority back independence.”

5. The publication said it did not accept that the Code had been breached. It said that it did not agree with the complainant’s narrow definition of the term “majority”, as the term can also mean “the greater number of.” It said it was not inaccurate to state that a greater number of poll respondents had, in both the reported poll and the 11 prior polls referenced in the article, said that they would vote yes in a Scottish independence referendum, and provided 11 previous polls in which more respondents had said that they would vote Yes in a Scottish independence referendum.  It accepted that there was some debate over how publications should approach the matter of respondents who answer “Don’t Know” in response to polls, but said that it was accepted consensus amongst polling companies and publications to omit Don’t Knows and to make clear when they have been omitted, as doing so renders the findings of a poll comparable to referendum votes, where only votes cast in favour or against are counted; as such, the findings of the poll were representative of the Scottish voting population, where voters who had expressed no opinion would not vote and therefore would not be counted. It said that in this instance, where the article made clear the “Don’t Knows” had been excluded, it was not inaccurate in the way suggested by the complainant. It also said that, while it understood the complainant had concerns with the methodology of the poll, the publication’s responsibility was to accurately report on the results of the poll, which the publication said it had done; it was not responsible for the methodology by which the poll was conducted.

6. The complainant reiterated his position that it was inaccurate to use the term “majority” when referring to less than 50% of respondents. The complainant also said that, whether or not it was the consensus to not include “Don’t Knows” in the reporting of polls in both the polling and newspaper industry, it was incumbent upon the publication to ensure that the articles it published were accurate. He suggested that the publication could have done so in this instance by making clear exactly what percentage of voters voted Yes, No, and Don’t Know, or by qualifying the 54% figure with a statement such as “of those who expressed a voting intention.” He said that, without any such qualification, the article was inaccurate.

Relevant Clause Provisions

Clause 1 (Accuracy)

i) The Press must take care not to publish inaccurate, misleading or distorted information or images, including headlines not supported by the text.

ii) A significant inaccuracy, misleading statement or distortion must be corrected, promptly and with due prominence, and — where appropriate — an apology published. In cases involving IPSO, due prominence should be as required by the regulator.

Findings of the Committee

7. Clause 1 (i) requires that publications take care not to publish headlines which are not supported by the text of the article. It does not require a headline to give the full context of the story in question, but the article must support the headline. The Committee found that the claim in the headline and opening paragraph - that a “majority” of Scottish adults were in favour of independence - was supported by the text of the article. The article set out that support for Scottish independence was polled at being 54% “when undecided voters were excluded.” As such, the headline claim that a “majority” of Scottish adults supported independence was not an inaccurate summary of the poll results, where the body of the article set out the method by which this conclusion was reached and so informed readers that the responses of some participants had been excluded and was therefore not reflective of the full population sample. The Committee also found that it was not inaccurate to state that a “majority” of respondents had backed independence, where majority can simply mean “the greater number of” and where it was not in dispute that a greater number of respondents in the reported poll had said they would vote yes in an Independence referendum. A greater number of respondents in the reported poll had said that they would vote yes in a Scottish independence referendum, and so the article was not inaccurate and there was no breach of Clause 1 on this point.

8. The second disputed claim in the headline was that the reported poll was the twelfth poll to show “majority” support for Scottish independence. The complainant pointed out that only three of these polls had found over 50% of respondents supported independence. The Committee noted that the article had not mentioned that only when undecided voters were excluded from the data sets did support surpass 50% in the cases of all 11 previous polls. In this instance, where there was no dispute that in each case support for independence had been found to outweigh opposition, the Committee concluded that the headline was not misleading or inaccurate; there was no breach of Clause 1. That said, it emphasised the importance when reporting on polls of clearly stating when undecided respondents were excluded from the results.

Conclusion

9. The complaint was not upheld.

Remedial action required

10. N/A

Date complaint received: 07/11/2020

Date complaint concluded by IPSO: 01/04/2021