READ: Full Ofcom decision slamming BBC and Andrew Neil

Neil made the claim while interviewing Alex Salmond on Sunday Politics during the 2017 General Election.

The statistic was challenged by a primary school teacher who had seen the interview, but at the time, the BBC claimed the data came from research completed in 2009.

READ MORE: BBC and Andrew Neil slammed for misleading claim in Salmond interview

Eventually the broadcaster backtracked and apologised for giving "incorrect statistics", then again changed the story, claiming the figures came a 2014 Scottish Government survey.

After the complainant pushed the complaint further the BBC accepted that the statistic was "not accurate".

Now, in an 18-page ruling, Ofcom has criticised the state broadcaster over its handling of the case.

Read their full report below.

*******************************

Broadcast Standards cases

In Breach

Sunday Politics

BBC1, 30 April 2017, 11:24

Introduction

Sunday Politics is a weekly current affairs programme featuring interviews and debates on political issues. This edition was broadcast during the election period for the 2017 local elections in Scotland (which ran from 22 March to 4 May 2017) and after it had been announced that there would be a UK General Election taking place on 8 June 2017.

Ofcom received a complaint that, during an interview with the Scottish National Party (“SNP”) former Scottish First Minister Alex Salmond, the presenter Andrew Neil used a “false statistic” about literacy standards in Scottish primary schools. The viewer had previouslyraised their concerns with the BBC in accordance with Ofcom’s published procedures. The BBC had upheld the complaint at the final stage of its own complaints process and published a finding *1*. However, the complainant had not been satisfied with the BBC’s response and brought their complaint to Ofcom.

During the interview, Alex Salmond said that the SNP had protected public services. The interview continued as follows:

Andrew Neil: “...If services have been so well protected, why after a decade of SNP rule do one in five Scots pupils leave primary school functionally illiterate?”

Alex Salmond: “You’ve to take these things in the round, and Nicola Sturgeon’s*2* made it a top priority to address these challenges. But let’s take another statistic: 93% of Scottish kids are now emerging from school to positive destinations. That means to further education, to apprenticeships or to work”.

Andrew Neil: “Why are one in five functionally illiterate?”

Alex Salmond: “Well you argue one statistic – I’m arguing that in the round Scottish education is putting in some substantially good performances like the 93%, a record figure, who are going on to positive destinations. You can’t have a failing education system if you’ve got that 93% and, incidentally, a record low youth unemployment in Scotland. We’ve got the second lowest unemployment rate in Europe. These pupils are being prepared by the Scottish education system”.

The interview then continued on the issue of education and other matters relating to the record of the SNP administration in Scotland.

In the original complaint to the BBC, the complainant queried the meaning and the source of the statistic, adding that it suggested that Scottish education was failing terribly, but that did not reflect the complainant’s personal experience as a school teacher in a very large primary school. The complainant had looked for the statistic in a number of Scottish education reports but had been unable to locate its source.

The complaint was made to the BBC on 1 May 2017. We took into account the BBC’s various responses to this case, including in relation to the source on which Andrew Neil based his question (“why after a decade of SNP rule do one in five Scots pupils leave primary school functionally illiterate?”):

• on 8 May 2017, in its first substantive response to the complainant, the BBC said: “A quick search online shows that the statistic you refer to comes from a 2009 study; reported here, for example: One in five Scots struggle with literacy, report reveals, The Scotsman, 4 December 2009 [and] One in five Scots children leave primary school not fully literate, The Telegraph, 4 December 2009”;

• after the complainant challenged this source, the BBC provided its second substantive response on 17 May 2017, when it said to the complainant: “First of all, we would like to apologise for giving you the incorrect statistics before. We have spoken to the programme editors who have confirmed that the statistics Andrew [Neil] was referring to in his questioning of Mr Alex Salmond was the Scottish Government’s 2014 Scottish Survey of Literacy and Numeracy [“SSLN”], not the 2009 survey…”*3*; and

• on 23 June 2017 the complainant escalated the complaint to the BBC’s Executive Complaints Unit (“ECU”). On 5 October 2017, the ECU wrote to the complainant providing the BBC’s third and final substantive response, stating that: “Having conducted our own research into the matter and consulted further with the programme, we agree that this [statement] was not accurate. The figure was drawn from the 2009 Scottish Survey for Literacy and Numeracy (which was not the most recent research into school attainment) and it was not accurate to say that this allowed the conclusion quoted in the programme. It should have been made clear that the phrase ‘functionally illiterate’ was not used in that report and that its source was the education spokeswoman of the Scottish Conservatives”.

On 28 November 2017, the BBC ECU published its finding *4* which was to uphold the complaint in this case. The BBC finding as originally published on 28 November 2017 said:

“The figure had originally been put forward by a spokesperson for the Scottish Conservatives, as being based on the 2009 [SSLN]*5*. That survey, however, contained no reference to ‘functional illiteracy’, and no data which would have justified the claim in question”.

The finding also said: the “Sunday Politics team has been reminded of the need to establish the evidential basis of claims that are quoted in its questions”.

During Ofcom’s investigation, the BBC confirmed to Ofcom in April 2018 that it had changed the published text of the ECU finding to include the following:

“The figure derived from the sum of the two lower bands for reading attainment in the 2014 [SSLN]. That survey, however, contained no reference to ‘functional illiteracy’, and no data which would have justified that form of words as a description of its findings”.

As a weekly current affairs programme, Sunday Politics is not a news broadcast, and there is therefore no requirement under the Broadcasting Code that facts discussed in the programme be presented with due accuracy. However, we considered the broadcast content raised potential issues under the following Code rule:

Rule 2.2: “Factual programmes or items or portrayals of factual matters must not materially mislead the audience”.

We therefore requested comments from the BBC on how the programme complied with this rule.

Response

The BBC’s representations

The BBC provided initial representations as well as representations from itself and from presenter Andrew Neil on Ofcom’s Preliminary View, which was to provisionally record a breach of Rule 2.2 in this case.

The BBC explained that typically before each Sunday transmission, the programme editor and producer would send a briefing to the presenter, Andrew Neil on Friday evenings. The briefing would consist of “notes on the topics to be covered in the programme and on the contributors invited to discuss them, with suggested questions”. The BBC added that there would be “exchanges between the presenter and the producer and Editor” on the day before broadcast on the proposed content of the programme and that “[v]erification of factual references is an integral part of this process”.

After a further attempt to link the statement to an official statistical source in its initial representations, the BBC now accepts that it was not based on any such source and that it was therefore not accurate. However, inaccuracy in and of itself in non-news programmes does not constitute a breach of the Broadcasting Code, and as stated above, as Sunday Politics is not a news programme, the question for Ofcom to consider under Rule 2.2 is whether the content was materially misleading to the audience so as to cause potential or actual harm or offence.

The BBC referred to Andrew Neil’s representations on Ofcom’s Preliminary View (see below), expressing the hope that having “considered the sources cited by Mr Neil” Ofcom would “conclude that, despite the phrase ‘functionally illiterate’ going beyond what was warranted by the 2014 SSLN, any inaccuracy here was not so materially misleading as to constitute a breach of Rule 2.2”.

The BBC also referred to the following statement in its initial representations: “…however characterised, the relevant statistics [from the 2014 SSLN] indisputably showed a significant decline in literacy levels amongst primary school-leavers, as the Scottish government had repeatedly acknowledged”.

In its further representations, the BBC said that this statement had arisen from an “incorrect reading of the 2014 SSLN” and had been “added at a late stage of drafting by the Head of the [BBC ECU]” as was the following statement:

“the intended effect [of the use of the term ‘functionally illiterate’] was rather to encapsulate the general situation of declining levels of literacy amongst primary schoolleavers reflected in the [2014] SSLN report”.

The BBC apologised for this error saying that the above two statements did “not represent a view held by BBC News in general or the programme-makers in particular”.

The BBC argued that rather than relating to a “single data set” as implied by the Preliminary View, an assessment of the accuracy of the programme needed to take into account a “considerably wider range of data”, as cited by Andrew Neil in his representations on the Preliminary View (see below). It added that Andrew Neil’s question to Mr Salmond (“Why, after a decade of SNP rule…?”) “related to a longer period…than is discussed in the Preliminary View”.

The BBC emphasised in its representations that neither Alex Salmond, within the programme, nor Nicola Sturgeon, speaking within the Scottish Parliament, had treated the content as materially misleading. It acknowledged that broadcasters “should not rely on others to correct their mistakes, and that the absence of challenge [by Mr Salmond] did not of itself mean viewers were not misled”. However, the BBC argued that the reactions of Mr Salmond and Ms Sturgeon were significant. It said that Alex Salmond and Nicola Sturgeon were informed about and accustomed to debating the policy area of education. It added that “…the least that can be inferred from their responses is that neither of them regarded the statement in question as materially misleading (the implausible alternative being that they did regard it as misleading but chose not to say so – indeed, in Ms Sturgeon’s case, to say otherwise)”.

The BBC cited the subsequent interview with Nicola Sturgeon in The Andrew Marr Show on 14 May 2018. In this interview, Mr Marr had said the SNP’s record on literacy was “absolutely terrible” and “…you have among 13 and 14-year olds only, less than half are now performing well in reading and writing…”, to which Nicola Sturgeon had replied:

“Firstly, let me say I’ve been very open that that’s not good enough, but just to put that into context, we have a survey that measures pupils in the second year of the secondary school, but measures them against the standard they are expected to achieve in the third year of secondary school, and we have other information that show that by the time people are in the third year more than 80% are reaching the required level”.

Andrew Marr also said: “On numeracy and literacy, there’s no question that things have got worse”, to which Nicola Sturgeon replied:

“I’m not denying that in terms of literacy and numeracy, and I’m telling you what we are doing to address that…”.

In the BBC’s view, Nicola Sturgeon’s replies to the questions posed to her in The Andrew Marr Show indicated she “accepted not only the premise that standards of literacy among Scottish children were highly unsatisfactory, but also the premise that they had worsened significantly while the SNP had been in office”. It added “In the context of a public debate in which such premises were accepted by the First Minister of Scotland, we think it difficult to maintain that Mr Neil’s question to her immediate predecessor…carried the potential for harm to viewers discerned by the Preliminary View or was so misleading as to constitute a breach of Rule 2.2”.

Andrew Neil’s representations on Ofcom’s Preliminary View

Andrew Neil also provided representations on Ofcom’s Preliminary View. While acknowledging that his statement was not based on any statistical source and that it was therefore inaccurate, Mr Neil set out the context surrounding the Scottish education debate and referred to a number of sources published before and after the broadcast, in relation to primary school leavers and secondary school pupils, to support his view that the statement was not materially misleading. He said that in December 2009, Scotland’s Literacy Commission had published a report (the “Literacy Commission report”), which had stated “A good estimate would be that 18.5% leave primary school [in Scotland] without being functionally literate”. Mr Neil added this report had “hung over Scotland’s educational debate…” and that by the time of his interview with Alex Salmond the data used in the 2009 report was 10 years old. He said that the SNP had also been in power for 10 years in 2017 and had often stated its commitment to improving Scottish educational standards. Mr Neil said it was therefore “legitimate to try to discover if things had got better in the decade of SNP power”. Mr Neil also said that his “question to Mr Salmond was framed in a provocative way to elicit a response that knocked my premise back”.

Andrew Neil said, “Evidence that Scottish literacy has improved between 2007 and 2017 is hard to come by. Evidence that Scottish literacy has got no better, and could even be worse, is clearer”. He cited “the respected independent Holyrood Magazine”, in comparing the 2012 and 2014 SSLN results, stating that “Fewer Scottish school children are good at reading and writing” and that there was “falling literacy in Scottish schools”. He added that the decline recorded in the 2014 SSLN was “not precipitous”, but it provided “scant evidence that matters were improving, even from the existing low base”. He further added that “Government politicians and policymakers in Scotland have worried, usually in private, rarely in public, that not enough progress was being made to improve the situation revealed in the 2009 Literacy Commission report – indeed they feared there were signs it was getting worse”. Mr Neil stated his belief that this was why Alex Salmond and Nicola Sturgeon did not challenge the premise of his question – that 20% of primary school leavers were “functionally illiterate” – because it was “all too credible, even if not exactly accurate, as I now realise”. He added “My specific formulation of the situation was clearly open to challenge (and subsequently was); but the underlying assumption – that literacy among young Scottish pupils is a serious problem that is not being solved – is pretty robust”.

Andrew Neil cited various more recent statistics which showed, in his view, that literacy had not been improving in Scottish schools over a number of years. For example, he stated that: “The 2014 SSLN – published two years (2015) before the interview – had already provided scant relief for those arguing literacy was improving; if anything, there was a deterioration”. He added “Several months before the interview” the OECD published (in December 2016) its latest international comparative study of schools – PISA – in which Scottish schools recorded their worst ever PISA performance with declines for scores in maths, science and reading (my emphasis)”.

Mr Neil said that in relation to primary school leavers in Primary 7:

• the 2016 SSLN report published one month after the broadcast of this programme found that “those who could write well or very well had fallen from 72% in 2012 to 65% in 2016…[The] decline in reading was less marked – but it was still down two percentage points on 2012”; and

• a “new official government study, using different methodology from SSLN, published in December 2017 found only 69%...were reaching the expected reading standard, 76% in writing. So over one in five in both cases were not”.

Mr Neil added that in relation to Secondary 2 pupils (those two years after leaving primary school):

• “those who met the expected standard in writing had fallen from 64% in [the 2012 SSLN] to less than half (49%) in [the 2016 SSLN]*6* – a dramatic fall of 15 percentage points over only four years; and

• in the 2016 SSLN those “who failed to meet the expected [Secondary 2] standard came to 16% – one in six – compared with only 7% in” the 2012 SSLN *7*. Mr Neil argued that this “suggests that something is seriously amiss in Scottish secondary schools in the early years after primary school; or that the SSLN has been underestimating the literacy problems of P7 primary school leavers”.

Andrew Neil said ‘functionally illiterate’ “does not mean illiterate in the sense of unable to read or write. It means not possessing enough basic skills to be confident of handling many normal reading and writing tasks”. He added that: “The SSLN does not use the term [‘functionally illiterate’ and] I did not attribute it to the SSLN”. He accepted that “being in the bottom two [SSLN categories] does not necessarily produce a combined ‘functionally illiterate’ category”. However, Mr Neil argued that “the SSLN’s categorisations are somewhat opaque and, for the second bottom category, not very demanding (working ‘within the level’ requires completing only 40% of the test or over)”.

Mr Neil said that “it would have been better if I had phrased my question along the lines of ‘why is literacy among Scottish school pupils still so bad – and maybe even getting worse’. But I do not accept that the formulation I used seriously misled viewers about the gravity of the situation”. He added that “evidence is strong that illiteracy in Scottish schools is still deeply embedded in the system and that, far from improving, is likely getting worse, even after 10 years of SNP government”. In conclusion, Mr Neil said that: “Questions can always be better framed in retrospect. But I refute Ofcom’s draft conclusion that my question seriously misled viewers about literacy problems in Scottish schools or misrepresented the Scottish Government’s response to them. And I do not accept that it was even close to a breach of…Rule 2.2”.

Investigation of the statistic (post-broadcast)

Ofcom also asked the BBC to explain why, having received the complaint during the Scottish local election period and ahead of the UK General Election, the statistic was not fully investigated and Mr Neil’s statement corrected at the time. The BBC said the complainant received a substantive response to their first contact within a week and a response to their second and third contacts within seven working days and two days respectively. It added that there was no correction at that time because “unfortunately, the investigation did not identify *8* the inaccuracy in the terms in which [the statistic] was described…the context was one where the claim in question had been effectively accepted, rather than disputed, by successive First Ministers”.

The BBC also said that “the ECU investigation took a good deal longer than we would have wished. It took place against the background of implementing new procedures and adjusting to new demands arising from the 2018 Charter and the assumption by Ofcom, in early April, of full regulatory authority over the BBC, and the situation was aggravated by unexpected staff absences”.

Decision

Reflecting our duties under the Communications Act 2003, there is no requirement under the Code that facts discussed in non-news programmes be presented with due accuracy. However, section Two of the Code requires that generally accepted standards are applied to provide adequate protection for members of the public from the inclusion of harmful or offensive material in programmes.

Ofcom takes account of the audience’s and the broadcaster’s right to freedom of expression set out in Article 10 of the European Convention on Human Rights. Ofcom acknowledges the importance attached to freedom of expression in the broadcasting environment, which encompasses the right to hold opinions and to receive and impart information and ideas without undue interference by public authority.

The Code does not prevent broadcasters from covering subjects which may elicit strong opinions, such as the Scottish Government’s record on education. It is crucial that broadcasters have the editorial freedom to cover such topics. However, Rule 2.2 requires broadcasters to ensure that facts are not misrepresented in a way which materially misleads the audience. This is particularly important in current affairs programmes in which audiences are likely to have high levels of trust. This is likely to be highest during election periods.

Misrepresentation

We first considered whether the question “why after a decade of SNP government, one in five Scots pupils leave primary school ‘functionally illiterate’” misrepresented a factual matter.

During the course of the BBC complaints process and of Ofcom’s investigation, the BBC pointed to a range of sources for the statement that was made during the broadcast, none of which provided a proper basis for the statistic used. In their latest representations, both the BBC and Andrew Neil have accepted that the presenter’s question was not accurate and was not based on an official statistical source.

We took into account the reference to a specific time (“after a decade”), and the repeated use of the statistic “one in five” and of the term “functionally illiterate”. In our view, these gave the overall impression that Andrew Neil was basing his allegation that “one in five [20%] Scots pupils leave primary school functionally illiterate” on a specific statistical source relating to primary school leavers, with reference to the year 2017. This approach, in our view, gave viewers the false impression that Mr Neil’s question was founded on an established fact or source, which recorded literacy levels upon leaving primary school according to a recognised benchmark in the education sector of ‘functional illiteracy’.

In light of all the above, in our view the statement misrepresented a factual matter and was misleading to viewers.

Material misleadingness

We then assessed whether this misrepresentation was materially misleading.

Under Rule 2.2, “Factual programmes or items or portrayal of factual matters must not materially mislead the audience”. Ofcom’s Guidance to Rule 2.2 states that “…it is possible that actual or potential harm and/or offence may be the result of misleading material in relation to the representation of factual issues”. The Guidance also states that it is “designed to deal with content that materially misleads the audience so as to cause harm or offence” [emphasis in original] and not with “issues of inaccuracy in non-news programmes”. This means that it is possible for a non-news programme such as Sunday Politics to include inaccurate material without breaching the Code, if the inaccuracy is not materially misleading. The Guidance also says: “Whether a programme or item is ‘materially’ misleading depends on a number of factors such as the context, the editorial approach taken in the programme, the nature of the misleading material and, above all, either what the potential effect could be or what actual harm or offence has occurred [emphasis in original]”.

Context provided by available statistics

The BBC argued that in the context of all the data given in Andrew Neil’s representations, his statement was not materially misleading. Mr Neil said that “it would have been better if I had phrased my question *9* along the lines of ‘why is literacy among Scottish school pupils still so bad – and maybe even getting worse’. But I do not accept that the formulation I used seriously misled viewers about the gravity of the situation”. He added that “evidence is strong that illiteracy in Scottish schools is still deeply embedded in the system and that, far from improving, is likely getting worse, even after 10 years of SNP government”. Mr Neil also said that “Questions can always be better framed in retrospect”.

We took into account Mr Neil’s argument that the fact his question could have been better phrased does not mean in the particular context that it was materially misleading, and that questions can always be better framed in retrospect. We also took into account that Sunday Politics is a weekly programme which the editorial team and presenter prepare in advance, as per the BBC’s explanation of the steps typically taken prior to each Sunday transmission described above. In our view, given the content and phrasing of Mr Neil’s question, ordinary viewers would have been likely to have understood that the statistic had been researched by the editorial team and was a statistic of verifiable provenance that could be relied upon. Therefore, we did not consider that ordinary viewers would have understood that the statistic Mr Neil used in his question was paraphrasing for his personal interpretation that literacy in Scottish school pupils was bad and getting worse.

We examined the sources available at the time of the broadcast which related to Scottish primary school leavers to ascertain whether Mr Neil’s statement was consistent with those indicators. Some of the statistics cited by Andrew Neil in his representations were not published at the time of the interview in April 2017, and some related to secondary school children rather than the focus of Andrew Neil’s question – namely children leaving primary school. We acknowledged that some of the statistics which were available at the time of the interview could have been used to argue that some of the metrics used to measure literacy levels at the end of primary school in Scotland had slightly declined between 2012 and 2014 (by no more than between one and four percentage points in any specific instance), or to support commentary on these literacy levels as at 2014 *10*. However, we took into account that Mr Neil’s statement was made by reference to a specific point in time (2017 – “after a decade of SNP rule”) rather than a comparison exercise of trends, for example, between 2012 and 2014. Therefore, we carefully considered whether it was materially misleading to characterise the available metrics as providing evidence that one in five Scottish primary school leavers could be described as being ‘functionally illiterate’ in 2017.

With regard to Mr Neil’s use of the expression ‘functionally illiterate’, we note that the UK’s National Literacy Trust describes ‘functional illiteracy’ by reference to adults: “Around 15 per cent, or 5.1 million adults in England, can be described as ‘functionally illiterate’. They would not pass an English GCSE and have literacy levels at or below those expected of an 11-yearold. They can understand short straightforward texts on familiar topics accurately and independently, and obtain information from everyday sources, but reading information from unfamiliar sources, or on unfamiliar topics, could cause problems” *11*. The term is not commonly used in relation to primary school children and, although it featured in the 2009 Scottish Literacy Commission Report cited by Mr Neil in his representations, it is Ofcom’s understanding that it has not been used in any official statistical sources on Scottish education since then. The 2009 Report itself only referred to “functional” literacy in its introduction and in relation to available UK data (we discuss this source further below). The Report discussed low levels of literacy and how to tackle the issue in general, among school leavers (whether from primary or secondary school) as well as adults, and preferred the use of the term “basic literacy” rather than “functional literacy” *12*.

In his representations, Andrew Neil said ‘functionally illiterate meant’ “not possessing enough basic skills to be confident of handling many normal reading and writing tasks”. In our view, this definition was Mr Neil’s own interpretation of the expression rather than a definition taken from an academic or other source.

In any event as the BBC and Mr Neil both now accept, it was inaccurate to say that “after a decade of SNP rule, one in five Scots pupils leave primary school functionally illiterate”. In order to assess whether or not this was materially misleading, we considered how the ordinary viewer would have understood Mr Neil’s question. In our view, the ordinary viewer would have understood the meaning at the time of broadcast in 2017 to be “one in five primary school leavers in Scotland are unable to carry out basic literacy skills expected of children in that age group”.

Of the information cited by Mr Neil and the BBC in their representations relevant to primary school leavers, only Scotland’s Literacy Commission Report 2009 and the 2012 and 2014 SSLNs were official sources available at the time of the interview. The 2009 report contextualised its statement that “in Scotland 18.5% leave primary school without being functionally literate” by stating that “There is no national/official measure of how many children are going through the school system without acquiring basic literacy skills but, using what information is available”, the 18.5% figure would be “a good estimate” on the basis of UK data *13*. As this was an estimate, not based on data specific to Scotland and in a report based on 2007 data which was 10 years old at the time of the interview, we considered that it was not an adequate source of information against which to assess Mr Neil’s statement in relation to educational performance in 2017 for the purpose of determining whether the statement was materially misleading. The 2012 and 2014 SSLNs were based on data specific to Scotland, and the 2014 SSLN was the most recent of the two. Therefore, we took into account the 2014 SSLN when assessing Mr Neil’s statement. We also considered the 2016 SSLN. Although not available at the time of broadcast, it was published nine days later on 9 May 2017 and we therefore considered the 2016 SSLN in assessing whether the inaccuracy was materially misleading.

• Primary 7 pupils

We considered that the 3%, 5% and 8% of primary 7 pupils who were “not yet working within the level” (i.e. the bottom level of performance) in reading, writing, and listening and talking respectively in the 2014 SSLN could be described as unable to carry out basic literacy skills expected of 11-year olds *14*. The corresponding figures from the 2016 SSLN were 3%, 6% and 8%. Pupils falling within this category would have “successfully completed” 39% or less of the items they were tested on. In our view, it would be materially misleading to use the higher categories of attainment within the 2014 and 2016 SSLNs as evidence that the pupils falling within these categories are unable to carry out basic literacy skills expected of their age group. For example, the next level of performance, “working within the level”, included pupils who had successfully completed up to 59% of the given assessment. We did not consider that the ordinary viewer would have been likely to have associated this category of performance (i.e. “working within the level”) with a lack of basic literacy skills for the relevant age group. In his representations, Mr Neil accepted that “being in the bottom two [SSLN categories] does not necessarily produce a combined ‘functionally illiterate’ category”, though he also argued that “the SSLN’s categorisations are somewhat opaque and, for the second bottom category, not very demanding (working ‘within the level’ requires completing only 40% of the test or over)”. We considered that Mr Neil’s latter comment was a matter of personal opinion rather than a widely accepted view on Scottish education statistics.

• Secondary 2 pupils

We considered that the 3%, 12% and 18% of secondary 2 pupils who were “not yet working within the level” (i.e. the bottom level of performance) in reading, writing, and listening and talking respectively in the 2014 SSLN could be described as unable to carry out basic literacy skills expected of 14-year -olds *15*. The corresponding figures from the 2016 SSLN were 2%, 16% and 15%. Pupils falling within this category would have “successfully completed” 34% or less of the items they were tested on. In our view, it would be materially misleading to use the higher categories of attainment within the 2014 and 2016 SSLNs as evidence that the pupils falling within these categories are unable to carry out basic literacy skills expected of their age group. For example, the next level of performance, “working within the level”, included pupils who had successfully completed up to 59% of the given assessment. We did not consider that viewers would have associated this category of performance with a lack of basic literacy skills for the relevant age group.

In relation to the performance of Scottish secondary school children, we also took into account the results of another source cited by Mr Neil in his representations, the 2015 Programme for International Student Assessment (PISA), published *16* in December 2016. In the PISA reading assessment undertaken by 15-year olds, students were placed in one of seven levels, from level 1b (lowest) to level 6 (highest) *17*, with level 2 as a baseline level of proficiency *18*. Level 2 proficiency is defined under the PISA classification *19* as follows: “Readers at level 2 can locate one or more pieces of information, which may need to be inferred and may need to meet several conditions. They can recognize the main idea in a text, understand relationships, or construe meaning within a limited part of the text when the information is not prominent and the reader must make low-level inferences. Tasks at this level may involve comparisons or contrasts based on a single feature in the text. Typical reflective tasks at this level require readers to make a comparison or several connections between the text and outside knowledge, by drawing on personal experience and attitudes”. We considered that viewers would not have associated level 2 ability with a lack of basic literacy skills for the relevant age group, and therefore focused on the next two levels below level 2, i.e. level 1a (the next level down) *20* and level 1b (lowest level) *21*. The 2015 PISA results for Scottish secondary pupils were that 4.4% were placed in level 1b and below and 13.5% were placed in the next level up, level 1a, a total of 17.9%.

We took into account the following with regard to the above statistics relating to the performance of Scottish secondary school children:

• in relation to the 2014 and 2016 SSLNs, in comparing Mr Neil’s reference to 20% of primary 7 pupils being “functionally illiterate” to the percentage of pupils in the 2014 and 2016 SSLN samples who were “not yet working within the level” (i.e. the bottom level of performance), we acknowledged that 18% of secondary 2 pupils were in this category for “listening and talking” in 2014 (falling to 15% in 2016) and that this was close to 20%. However, fewer secondary 2 pupils were in this category for writing and for reading (12% and 3%, respectively) in 2014 (the corresponding figures in 2016 were 16% and 2%); and

• in relation to the 2015 PISA figures for Scotland, we acknowledged that 17.9% of the Scottish pupils sampled were below Level 2, which the OECD classifies as the baseline level of proficiency in PISA and the level “at which readers begin to demonstrate the competencies that will enable them to participate effectively and productively in life as continuing students, workers and citizens” (see PISA for Development Brief 8).

However, the above sets of statistics related to Scottish secondary school pupils, and in the case of PISA, 15-year old pupils, a population which is three years older than primary school leavers. We therefore placed greater weight on the 2014 and 2016 SSLN results for primary 7 pupils, as it was pupils of this age to whom Mr Neil had referred in the interview. Significantly less than 20% of primary 7 pupils were in the bottom level of performance (3%, 5 to 6% and 8% in reading, writing, and listening and talking, respectively) in the 2014 and 2016 SSLNs.

In light of the above, we considered that the context indicated that the question posed by Mr Neil to Mr Salmond significantly misrepresented the actual situation in relation to literacy levels amongst Scottish primary school leavers in 2017 and that his statement was a significant departure from reasonable interpretations of the available data, such that it was materially misleading to the audience.

Context in the form of public debate

In their representations, the BBC and Andrew Neil also relied on the reaction of Alex Salmond and Nicola Sturgeon to the statistic as evidence that it was not materially misleading. The BBC said that, during the interview, “Alex Salmond did not seek to dispute or directly engage with the claim”. The BBC also said that following the interview Nicola Sturgeon did not “dismiss” the statistic during a Scottish parliamentary debate. Andrew Neil said that Alex Salmond and Nicola Sturgeon did not challenge the premise of his question – that 20% of primary school leavers were “functionally illiterate” – because it was “all too credible”. In our view, and as the BBC accepted in its further representations, the BBC and other broadcasters should not rely on interviewees and others to directly address misrepresentations in their programmes.

However, the BBC also argued that “…the least that can be inferred from [Mr Salmond and Ms Sturgeon’s] responses is that neither of them regarded the statement in question as materially misleading (the implausible alternative being that they did regard it as misleading but chose not to say so – indeed, in Ms Sturgeon’s case, to say otherwise (‘…I do not dismiss any of the statistics that Ruth Davidson cites…’))”. We disagreed. Although Alex Salmond did not directly dispute the statistic, and Nicola Sturgeon at first appeared to accept it, both challenged the idea that the Scottish education system was failing. In response to Mr Neil’s question, Alex Salmond said “You can’t have a failing education system if you’ve got that 93%”, a reference to “a record [number of school leavers] who are going on to positive destinations”. Nicola Sturgeon made a similar point in the Scottish Parliament after stating that Ruth Davidson, in quoting Andrew Neil’s statistic, did “a disservice to young people” *22*). In any case, irrespective of the fact that, arguably, Alex Salmond and Nicola Sturgeon only indirectly disputed the statistic in question, this did not remove the fact that Mr. Neil’s question did not reflect reality and that viewers were materially misled in this case.

The BBC also cited the First Minister’s appearance on The Andrew Marr Show on 14 May 2018, two weeks after the programme in this case. The BBC said that Mr Marr had said the SNP’s record on literacy was “absolutely terrible” and “…you have among 13 and 14-year olds only, less than half are now performing well in reading and writing…”, to which Nicola Sturgeon had replied: “Firstly, let me say I’ve been very open that that’s not good enough”. The BBC added that the First Minister had also said she did not deny that literacy had worsened. Given this context it refuted that Mr Neil’s question to Mr Salmond in the present case “was so misleading as to constitute a breach of Rule 2.2”. We disagreed. Mr Marr framed his question by reference to broadly accurate statistics, and was consistent with the reality of Scottish education at the time of the broadcast, relying on the 2016 SSLN *23*. By contrast, Andrew Neil used a statistic which both he and the BBC have accepted did not exist. Further, although we agreed with the BBC that Mr Marr characterised a “highly unsatisfactory” scale of underperformance in literacy in Scottish schools, we did not consider that this was evidence that corroborated Mr Neil’s statement that in 2017 20% of primary school leavers in Scotland were unable to carry out basic literacy skills expected of children in that age group.

Andrew Neil also quoted an article from the Holyrood Magazine as having stated that “Fewer Scottish school children are good at reading and writing” and that there was “falling literacy in Scottish schools” We considered the article, which also stated:

“Claire Telfer, Spokeswoman for the Read On, Get On coalition, led by Save the Children, said ‘It is deeply disheartening to see a standstill in reading rates for Scotland’s poorest children. 1 in 5 children from deprived backgrounds are not reading well by the end of primary school *24*’”.

In our view, the article did not contain evidence that corroborated Mr Neil’s statement. In particular, we considered that to state that 20% of Primary 7 pupils were “not reading well” was very different from stating, as Mr Neil did, that 20% of Primary 7 pupils were “functionally illiterate”. Further, the 20% statistic given in the article applied to children from deprived backgrounds, whereas Mr Neil’s question in the present case did not make such a qualification and apparently related to all Scottish Primary 7 school pupils. For all the Primary 7 pupils sampled in the 2014 SSLN, 12% were “not reading well” (i.e. 3% were “not within the level” and 9% were “within the level” and therefore not in the “performing well at the level” and “performing very well at the level” categories).

In light of the above, we considered that the wider public debate generated around the time of the programme, and the earlier 2015 Holyrood Magazine article, did not reduce the materiality of the misleading nature of Mr Neil’s statement.

Format, editorial approach and nature of the misleading material

Sunday Politics reviews the UK’s topical political issues on a weekly basis and is a traditional staple of the discursive, current affairs genre. This edition of the programme was presented by Andrew Neil, who is “one of Britain’s best-known and experienced journalists in print and broadcast” *25*, and it was broadcast at a key moment in British politics. It was the election period for Scottish local elections taking place on 4 May 2017 and it had been announced that there would be a UK General Election on 8 June 2017. Further, Andrew Neil’s question was in an interview with a prominent Scottish politician, on a topic of importance to the Scottish electorate, namely education in Scotland.

Given the timing, we considered the interview was likely to have been understood by viewers as being a significant feature of the programme. In the interview, Mr Neil challenged the Scottish Government’s record on education in primary schools, repeating a statistic twice. Therefore, in our view the statistic was a significant element of the interview and, with the emphasis placed on it, the audience was likely to have relied on it as evidence that clearly supported Mr Neil’s argument.

In his representations, Mr Neil said that the statistic was in a question to Mr Salmond which was “framed provocatively to elicit a response that knocked [the] premise back…”. We acknowledged that viewers were likely to expect Andrew Neil to use a robust and provocative interview technique in an interview with a leading politician of a prominent Scottish political party in the context of an on-going election campaign. However, we considered that, in line with the higher trust audiences are likely to have in current affairs programming, viewers would have expected to be able to rely on a statistic presented as a factual matter, and used as a prominent and important element of a political interview. This would have especially been the case given the significance of the issue of education and the presentation of the information in an authoritative manner, as part of Alex Salmond’s interview. The audience, in our view, would not have considered it acceptable, as apparently suggested by Mr Neil in his representations, for a BBC presenter to put up a false statistic and rely on the interviewee to rebut it.

In this context we considered that audience trust and, correspondingly, the potential for harm or offence due to a breach of this trust, were particularly high.

Harm or offence

It is particularly important that the content of factual and current affairs programmes can be relied on by viewers and listeners, as breaches of audience trust in those instances can result in material harm or offence.

We took into account that that the BBC and Andrew Neil accepted the absence of any actual statistical source to support Andrew Neil’s argument that “one in five Scots pupils leave primary school functionally illiterate” and that the BBC’s ECU published a finding that the 2014 SSLN “[did not justif[y] that form of words as a description of its findings” *26*. This was, at the time, the latest report of relevance to primary school leavers and it showed that 3%, 5% and 8% of them were “not yet working within the level” in reading, writing, and listening.

In our view, viewers would have expected Andrew Neil to pursue a strong line of argument in holding a senior figure within the SNP to account on its record in education. However, we also considered that the use of a statistic which did not exist gave Andrew Neil’s characterisation of underperformance in literacy among primary school leavers greater force than was justified and that the statistic exceeded by far any other contemporary characterisations of primary education attainment in Scotland. We considered that viewers would not have expected a BBC presenter to question a politician in this way.

As stated above, the requirement to ensure that facts are not misrepresented in a way which materially misleads the audience is particularly important in current affairs programmes. This is because audiences are likely to have high levels of trust, and this level of trust is likely to be highest during election periods. We considered that other factors which would have increased the likely level of audience trust in this case included the fact that a well-known and experienced political interviewer was seeking to hold a leading Scottish politician to account by citing a statistic framed in precise terms about an important and controversial issue within Scottish politics at the time, namely. educational standards in Scottish schools. In particular, the audience would have had particularly high expectations of any factual content included within a political interview on a BBC service.

Given these factors, we considered that Andrew Neil misrepresented statistics on literacy among Scottish primary school children in a way that would have had the potential to affect negatively and erroneously viewers’ understanding of educational standards in Scotland, at a time when those standards were being strongly debated during an election in that country. This was particularly relevant given that the presenter in this case was using the statistic as a basis for challenging the record on education in government of the interviewee’s party, the SNP.

For all the reasons above we considered that this content materially misled viewers so as to cause potential harm or offence.

Assessment of the actions taken by the BBC

Breaches of the Code that have resulted in the audience being materially misled have always been considered by Ofcom to be among the most serious that can be committed by a broadcaster, because they go to the heart of the relationship of trust between a broadcaster and its audience. Having reached the view that the content was materially misleading, we considered whether, and if so to what extent, the BBC’s handling of the complaint and subsequent remedial actions mitigated the potential harm to viewers.

We took into account that the BBC took some remedial action in this case in the form of its published finding that this material had breached the BBC’s Editorial Guidelines. In terms of further action, the BBC’s ECU finding stated that “The Sunday Politics team has been reminded of the need to establish the evidential basis of claims that are quoted in its questions”.

However, we had a number of concerns about the steps taken by the BBC in this case:

• in each of its three substantive responses to the complainant, the BBC erroneously referred to three different official reports as the correct source for Mr Neil’s statistic;

• the ECU finding published by the BBC on 28 November 2017 also wrongly stated that the figure quoted by Andrew Neil was based on the 2009 SSLN;

• the BBC confirmed to Ofcom (after the start of its investigation) in April 2018 that the ECU Finding had been corrected to say the following:

“The figure derived from the sum of the two lower bands for reading attainment in the 2014 [SSLN]. That survey, however, contained no reference to ‘functional illiteracy’, and no data which would have justified that form of words as a description of its findings”;

• in the BBC’s and Andrew Neil’s representations on Ofcom’s Preliminary View, it appeared that some of the basis for Mr Neil’s question was the December 2009 report by Scotland’s Literacy Commission (Ofcom is not aware of the BBC having amended the ECU Finding to reflect its final representations to Ofcom); and

• in its representations on Ofcom’s Preliminary View, the BBC and Mr Neil accepted that the statistic had no official source. The BBC said that it had provided factually incorrect statements in its initial representations to Ofcom which had arisen due to an “incorrect reading of the 2014 SSLN” and had been “added at a late stage of drafting by the Head of the [BBC ECU]”. The BBC made clear that the incorrect statements it had provided to Ofcom “does not represent a view held by BBC News in general or the programmemakers in particular”.

Ofcom is greatly concerned about the BBC’s handling of this case, where at different stages of dealing with the complaint and during Ofcom’s investigation, the BBC provided conflicting explanations on the source from which Andrew Neil’s statement was derived. Notably, the BBC was still maintaining the accuracy of Andrew Neil’s statement in its second substantive response to the complainant on 17 May 2017 and continued to provide different explanations on the source for Mr Neil’s question at all the different stages of Ofcom’s investigation. The length of time it took the BBC to admit there was no factual source to support Mr Neil’s statement is deeply unsatisfactory and we expect better standards from the BBC, both in its handling of viewer complaints and in its interactions with Ofcom.

As the BBC acknowledged, the figure cited by Andrew Neil was quoted by Ruth Davidson in the Scottish Parliament *27* three days after the broadcast. Therefore, given the apparent salience of this figure in Scottish political debate it would have better mitigated the potential harm which arose from this broadcast if the BBC had publicly corrected this misrepresentation ahead of the UK General Election that took place on 8 June 2017. By failing to establish the full facts on the basis for Andrew Neil’s statement at that time, the BBC was not able, as it should have been, to take a view as to whether it was materially misleading and to take steps to remedy or mitigate this inaccuracy. This, in our view, exacerbated the potential harm caused in this case.

Although outside of the scope of our investigation into the programme *28*, we became aware of a clip of Andrew Neil’s interview with Alex Salmond containing the content in question, which was still available on the Sunday Politics homepage until March 2018, with no clarification about the statistic in question. The BBC explained that when upholding complaints about BBC content its policy is that “appropriate action should be taken in connection with any online versions of the material in question (this normally consists of adding an explanatory note to the relevant webpage, with a link to the summary of the finding, though in some instances the material may simply be removed)”. It added that “Regrettably, the policy was not followed in this instance because of an oversight”. We agree it is regrettable that the materially misleading statistic remained online without clarification for this length of time.

We took into account the BBC’s admission of the various errors that had taken place in its handling of this complaint, including that the ECU investigation took place against the background of “implementing new procedures and adjusting to new demands” arising from Ofcom becoming the BBC’s new regulator in April 2017 and was “aggravated by unexpected staff absences”.

However, in the particular circumstances of this case, we remain concerned that the BBC did not act sooner to correct the statement. Critically, our concerns stemmed from the fact that the statement was made in a current affairs programme which dealt with a controversial subject matter at a sensitive time during an election period. For these reasons, we did not consider that the BBC’s subsequent actions sufficiently mitigated the materially misleading statement and the potential harm caused to viewers of the programme.

The BBC is rightly held to high standards by its audience. Ofcom therefore expects the BBC to take careful note of its errors in the handling of this case to ensure they do not recur. The BBC should consider how it identifies, prioritises and resolves complaints during election and referendum periods. In particular, it should consider how to identify sources of statements that are challenged by viewers and make any appropriate corrections as soon as possible when audiences have been materially misled – as in this case – before an election or referendum period ends, so as to mitigate any potential harm to viewers. Given all the above, Ofcom’s Decision is that this content was in breach of Rule 2.2 of the Code.

Breach of Rule 2.2

Notes:

1 Sunday Politics, BBC1, 30 April 2017: Finding by the Executive Complaints Unit.

2 Current Scottish First Minister

3 The complainant challenged this response on 19 May 2017 and reiterated that the BBC had failed to correct the statistic or respond to the complaint with urgency during an election period. The BBC’s replies of 2 and 13 June 2017 invited the complainant to escalate the complaint to the BBC’s Executive Complaints Unit if they were dissatisfied with the BBC’s earlier responses.

4 See footnote 1.

5 Ofcom understands there was no SSLN in 2009 and the first such survey was in 2012. The SSLN is a sample survey of Scottish pupils nearing the end of the school year in Primary 4 (age 8-9), Primary 7 (age 11-12) and Secondary 2 (age 13-14). It assesses their ability in reading, writing and listening and talking. Pupils are assigned one of four categories based on the percentage of questions they answer correctly. The categories are ‘performing very well at the level’, ‘performing well at the level’, ‘working within the level’, and ‘not yet working within the level’. For Primary 7 pupils, “not yet working within the level” meant they had “successfully completed” 39% or less of the items, and “working within the level” meant they had “successfully completed” more than 39%, but less than 60% of the “items” in the SSLN. For secondary 2 pupils the corresponding cut-off points were 34% and 60%. In the 2014 SSLN, 3% of Secondary 2 Pupils were in the bottom category and 17% in the next category up for reading, totalling 20% or one in five. For Primary 7 pupils, the corresponding figures were 3 and 9%, totalling 12% or 3 in 25 pupils. See the Scottish Government’s Statistics Publication Notice on the 2014 SSLN.

6 For secondary 2 pupils’ performance in writing, the 2016 SSLN said that 33%, 14% and 2% (a total of 49%) were “performing well, very well or beyond the level”, with 35% “working within the level”.

7 The 2016 SSLN said 16% of secondary 2 pupils were “not yet working within the level” in writing.

8 The BBC said that “New light was thrown on the matter when the Sunday Herald published an analysis by The Ferret fact-checking service published on 14 May [2017] – after the local elections but before the BBC Complaints Team [provided the BBC’s second substantive response to the complainant on 17 May 2017] – which made a strong case for regarding the claim as based on an inaccurate interpretation of the statistical data. It is a matter of regret that this analysis went unnoticed by those responsible for the 17 May message, but they would have to have been close followers of the Scottish press to have become aware of it; it went virtually unreported in the media generally and, more significantly, was not remarked upon by the Scottish government or the SNP”.

9 “…why after a decade of SNP rule do one in five Scots pupils leave primary school functionally illiterate?”.

10 For example: • At the time of the interview the results of the 2012 and 2014 SSLNs were available. These showed that the percentages of Primary 7 pupils (the final year of primary school children in Scotland): -in the lowest category of performance (“not yet working within the level”) were: (for reading) 2% and 3% respectively; (for writing) 3% and 5% respectively; and (for ‘listening and talking’) 9% and 8% respectively; and -in the next category up of performance (“working within the level”) were: (for reading) 8% and 9% respectively; (for writing) 25% and 28% respectively; and (for ‘listening and talking’) 32% and 26% respectively. (The 2014 SSLN report did not compare results for ‘listening and talking’ in 2012 and 2014 because different “assessment approaches were adopted in each year”).
• The 2012 and 2014 SSLN reports described above also provided performance percentages for Secondary 2 pupils (the second year of secondary school children in Scotland): -in the lowest category of performance these were (for reading) 3% and 3% respectively; (for writing) 7% and 12% respectively; and (for “listening and talking”) 17% and 18% respectively; -in the next category up of performance these were: (for reading) 14% and 17% respectively; (for writing) 29% and 33% respectively; and (for ‘listening and talking’) 37% and 31% respectively.
• In both age cohorts, (primary and secondary) the remaining pupils were either “performing well at the level” or “performing very well at the level”, with some instances of pupils also performing “beyond the level”. See footnote 9 for an explanation of the above performance categories and a link to the statistics.
• The Scottish Government’s report “Programme for International Student Assessment (PISA) 2015: Highlights from Scotland's results” states that a sample of pupils are assessed at the age of 15 and that in the 2015 reading assessment in Scotland “the proportion of low performers increased compared to 2012, but was lower than the OECD average”. It also showed that 17.9% of the Scottish pupils sampled were below Level 2, which is the baseline level of proficiency in PISA and the level “at which readers begin to demonstrate the competencies that will enable them to participate effectively and productively in life as continuing students, workers and citizens” (see PISA for Development Brief 8)

 11 See its webpage “Adult Literacy” as at 19 November 2017. An updated version of this webpage refers to “very poor literacy skills” instead of “functional illiteracy”.

12 Emphasis added. On page 3 of Scotland’s Literacy Commission Report 2009, functional illiteracy is described as “young people leav[ing] school … without the basic literacy skills to function in a modern society”. Where the Report refers to school leavers it is not evident that they refer to primary school rather than secondary school leavers.

13 See: A Vision for Scotland: The Report and Final Recommendations of the Literacy Commission December 2009. The report stated under the heading “UK Data” that “There is no national/official measure of how many children are going through the school system without acquiring basic literacy skills but, using what information is available, a good estimate would be that in Scotland 18.5% leave primary school without being functionally literate”, which it defined as “without the basic literacy skills to function in a modern society”.

14 Each percentage relates to different metrics for measuring various aspects of literacy (reading, writing and ‘listening and talking’), and it would therefore be incorrect to add the percentage figures to arrive at a total percentage of pupils in the bottom level of performance. As the measurements are carried out among the same sample of pupils, there would also be double or triple counting issues with this approach (for example the same pupil might be in the lower level for both reading and writing, so that by adding the reading and writing percentages that same pupil would be taken into account twice).

15 Ibid.

16 See footnote 10, final bullet point.

17 See “Table 4. An overview of reading proficiency levels as they were described in the PISA 2012 study”, as found on page 35 of the Pisa 2018 Reading Literacy Framework.

18 See footnote 10, final bullet point.

19 See footnote 17.

20 Under the PISA classification, “Readers at level 1a can locate one or more independent pieces of explicitly stated information; they can recognise the main theme or author’s purpose in a text about a familiar topic, or to make a simple connection between information in the text and common, everyday knowledge. Typically, the required information in the text is prominent and there is little, if any, competing information. The student is explicitly directed to consider relevant factors in the task and in the text”.

21 Under the PISA classification, readers at level 1b can “locate a single piece of explicitly stated information in a prominent position in a short, syntactically simple text with a familiar context and text type, such as a narrative or a simple list. Texts in level 1b tasks typically provide support to the reader, such as repetition of information, pictures or familiar symbols. There is minimal competing information. Level 1b readers can interpret texts by making simple connections between adjacent pieces of information”.

22 The BBC referred in its submissions to Ofcom to an exchange in the Scottish Parliament on 3 May 2017 between Nicola Sturgeon and Ruth Davidson, in which Ms Davidson referred to the Sunday Politics interview, noting that Mr Salmond had been asked why one in five children leaves primary school functionally illiterate. She also asked about a separate statistic. Nicola Sturgeon replied “…Although I do not dismiss any of the statistics that Ruth Davidson cites, I think that she does a disservice to young people and teachers across the country”. She then listed some achievements of the Scottish education system under the SNP administration, including “…record numbers of positive destinations, which is more young people than ever before going into employment, further education or training”.

23 Mr Marr said that “among 13 and 14-year olds only, less than half are now performing well in reading and writing, and that’s gone down from 70% in just a few years, under the SNP”. This was a reference to the results of the 2016 SSLN for Secondary 2 pupils, which were published after the Sunday Politics broadcast. They showed that 82% of Secondary 2 pupils were “performing well or very well” in reading in 2016, and that 49 % of Secondary 2 pupils were “performing well, very well or beyond the level” in writing in 2016. See Scottish Survey of Literacy and Numeracy 2016: literacy, 9 May 2017.

24 Both the 2012 and 2014 SSLNs stated that, of the sampled Primary 7 pupils from the most deprived backgrounds, 4% were ‘not working within the level’ and 15% were ‘working within the level’, in reading. See SSLN 2014 Supplementary Tables and SSLN 2012 Supplementary Tables.

25 See BBC profile of Andrew Neil

26 See footnote 1.

27 See footnote 22.

28 In this case, the complaint focused on the broadcast content, as well as referring to the fact that the misleading statement was also available in a BBC website article. We discovered the existence of the interview clip as part of our investigation. Ofcom has no enforcement powers in relation to BBC online material (such as material on the BBC websites, including written text, images, video and sound content), although under the BBC Agreement, we may give an opinion on whether the BBC has observed its editorial guidelines in its online material. Given that our investigation into the original broadcast covers broadly the same issues, we did not consider it necessary to do so.