11622-22 Duncan v The Sun on Sunday

    • Date complaint received

      26th January 2023

    • Outcome

      Breach - sanction: action as offered by publication

    • Code provisions

      1 Accuracy

Decision of the Complaints Committee – 11622-22 Duncan v The Sun on Sunday

Summary of Complaint

1. Stephen Duncan complained to the Independent Press Standards Organisation that The Sun on Sunday breached Clause 1 (Accuracy) of the Editors’ Code of Practice in an article headlined “POLL BOOST FOR UNION SINCE DEATH”, published on 18 September 2022.

2. The complainant was one of a number of individuals who raised concerns about the article; he was selected as IPSO’s lead complainant.

3. The article reported that “support” for Scottish independence had “plummeted” following the death of Queen Elizabeth II. It said that an exclusive poll by the publication, 10 days after the monarch’s death, had “found a seven per cent drop in those wanting a breakaway.” It stated that “[j]ust 42 per cent of Scots would vote for independence if there was a referendum tomorrow, compared to one poll last month that found 49 per cent were in favour of a vote”.

4. The article was accompanied by three infographics which detailed the findings of the polls and which read, respectively: “42% would vote for independence, down from 49% last month”; “28% of Scots think the Queen’s death has strengthened the Union”; “55% are not in favour of a second independence referendum”

5. A substantially similar version of the article appeared online, headlined “UNION SUPPORT BOOST Support for Scottish independence plummets since Queen’s death”.

6. The complainant said the article was inaccurate and misleading in breach of Clause 1 (Accuracy). He said that, in claiming that there had been a “seven per cent” reduction in support for Scottish independence, the publication had compared figures from two separate polls by two separate organisations: Panelbase and Deltapoll, respectively, and that these polls were not comparable.

7. While the complainant accepted that a comparison of the two polls suggested a reduction in support for independence, he said that it was inaccurate and misleading to report that support had “plummeted” since the Queen’s death. He said that when the publication had used the first of these polls – conducted by Panelbase in August 2022 – as a basis for the comparison it had done so by excluding from the dataset the respondents who were “undecided” on the matter as well as those deemed “unlikely to vote” and had then calculated the level of support as a split between the remaining respondents. He said that this was how the publication had arrived at a figure of “49 per cent” of people supporting independence.

8. In contrast, for the second poll – conducted by Deltapoll in September 2022 – the complainant said that the figure of “42 per cent” in support of independence cited by the publication was a proportion of the entire dataset that included respondents who were “undecided” and who “would not vote” in a hypothetical referendum on the matter. The complainant said that the article misleadingly presented the polls as a “like-for-like” comparison without making this distinction sufficiently clear.

9. The complainant said that there were further reasons why the polls were not comparable: they had been completed by separate organisations and followed different methodologies. This included, for example, omitting certain cohorts: the Deltapoll survey did not include respondents who were 16-17 years-old; a segment of the Scottish population normally represented as they would, as per the 2014 Independence Referendum, be entitled to vote on the matter.

10. The publication did not accept a breach of the Editors’ Code. While it accepted that the “7 per cent” figure had been calculated as the complainant described, it did not accept that the article had distorted the findings of the two polls or that it contained any significant inaccuracies. The publication said it was entitled to compare polls on the same subjects from different polling companies, when they used similar questions, even if the precise polling methodology varied, and that it was common practice for newspapers to do so and as such readers would not be misled.

11. Further, it said that regardless of which variables were taken into account, a comparison of the two polls showed a clear fall in support for Scottish independence following the death of the Queen varying from the “7 per cent” reported (“don’t knows” excluded from the first poll and included within the second) to 4 per cent (including “don’t knows” in both) and 2 per cent (excluding “don’t know” in both). The publication said that a difference between the figures was insignificant and inconsequential; each comparison represented a fall in support for Scottish independence and to such an extent that would be sufficient to result in a ‘No’ vote, were a further referendum held on the matter. In addition, the newspaper said that “plummets” was clearly a subjective characterisation of this fall, and recognisable as such.

12. Notwithstanding this, at the beginning of IPSO's investigation, the publication offered, in an effort to resolve the matter, to remove the online article and publish the following clarifications on page two in print in its Corrections and Clarifications column and on the Corrections and Clarifications page of its website:

“A 18th September article said that support for Scottish independence had fallen 7% since the Queen's death. This figure was based on two polls whose methodologies were not strictly comparable. In fact, the figures suggest a drop of between 2% and 4%, depending upon assumed variables in the polling.”

“A 17th September article, now removed, said that support for Scottish independence had fallen 7% since the Queen's death. This figure was based on two polls whose methodologies were not strictly comparable. In fact, the figures suggest a drop of between 2% and 4%, depending upon assumed variables in the polling.

13. The complainant did not accept this offer of resolution, as the proposed wording did not adequately address his concerns. He said that a correction should be published as prominently as the original article, which had been published on page 17 and appeared on the home page of the website. The complainant also said that the wording of the correction should include an apology.

Relevant Code 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.

iii) A fair opportunity to reply to significant inaccuracies should be given, when reasonably called for.

iv) The Press, while free to editorialise and campaign, must distinguish clearly between comment, conjecture and fact.

Findings of the Committee

14. The Committee recognised the important and continual role that opinion polls play in the media landscape: providing publications and readers alike with insight into the public’s view on a particular issue at a specific point in time. The newspaper was entitled to compare the findings of two polls, but it was obliged to take care not to mislead readers in doing so.

15. The Committee considered whether the article was inaccurate to report that support for Scottish independence had “plummeted” by “7 per cent” (meaning seven percentage points) following Queen Elizabeth’s death. This figure had been calculated by comparing the findings of two separate polls which asked respondents whether Scotland should be an independent country; the first of which was conducted before the Queen’s death and the second following it. The “49 per cent” figure – the outcome of the first poll – was a percentage only of the respondents who had answered “Yes” or “No” to the statement, and not of the wider constituency of respondents who had also said they were “undecided” and/or were deemed to be “unlikely” to vote. In contrast, the “42 per cent” figure was the percentage of those in support of Scottish independence from a dataset which did include such respondents. The publication accepted that a like-for-like comparison of the figures produced, depending on how each was calculated, a lower percentage in support for independence than claimed in the article:  between two and four percentage points. In the view of the Committee, failing to make clear that the publication had applied different methodologies in calculating each of the figures in the comparison, and where that resulted in a greater difference between the two figures than might have otherwise been the case, constituted a failure to take care over the accuracy of the article, raising a breach of Clause 1 (i).

16. The reporting of the comparison without making this clear, particularly when this interpretation formed the basis of the claim that support for Scottish independence had “plummeted” by such a margin following the Queen’s death, was significantly misleading and as such required correction under Clause 1 (ii).

17. The Committee next considered whether the remedial action taken by the newspaper was sufficient to meet the terms of Clause 1 (ii). The Committee noted the complainant’s request for an apology. While there may be some circumstances in which an apology may be appropriate, the nature of this breach was not one of those occasions. It related to a general point of fact and did not personally affect a specific person or people. Upon receipt of the complaint, the publication had offered to publish a clarification, in print and online, and to remove the online version of the article. The proposed clarification clearly set out both the original inaccuracy and the true position. It was offered promptly and its location, the publication’s established Corrections & Clarifications column – online and in print, on page two – was sufficiently prominent. The clarification was therefore sufficient to meet the terms of Clause 1 (ii). There was no further breach of Clause 1 (ii).

18. While the Committee noted the complainant’s concerns about the omission of certain cohorts from the second poll, it noted that “16-17-year-olds” were not a separate, individual cohort within the sample of the first poll; rather, this group was incorporated into a much larger age group “16-34”.  It was therefore unclear as to the precise impact that this particular age group, which had been allowed to vote in the Scottish Independence in 2014, had upon the poll’s findings and by extension the accuracy of its comparison to the second poll. Taken in this context, and where readers would understand that a comparison between two different polls, undertaken by two separate organisations, at separate times, under different conditions and employing different methodologies, would have their own limitations, the Committee did not consider that this rendered the article inaccurate or misleading. There was no breach of Clause 1 on this point.


19. The complaint was partially upheld under Clause 1.

Remedial Action Required

20. The clarifications which were offered clearly put the correct position on record and were offered promptly and with due prominence. In relation to the in-print clarification, the Committee noted that this appeared in the newspaper’s Corrections and Clarifications column on page 10, rather than on page two as originally agreed. While the Committee recorded its concerns that the newspaper had published this clarification in a different location and had failed to communicate this decision to IPSO prior its publication, it did not consider that further action was required. The published clarification had identified the original inaccuracy and set out the correct position clearly. Further, this clarification had appeared in the newspaper’s regular and recognisable Corrections and Clarifications column and was considered sufficiently prominent given the prominence of the original article (page 17). In the circumstances, no further action was required.

Date complaint received: 19/09/2022

Date complaint concluded by IPSO: 15/12/2022