In the context of GCSE, AS and A level exams in England, two important events took place shortly before the holidays, on 12 December. One was Ofqual's publication of their statistics concerning the numbers of grades challenged and changed following the summer 2024 awards in England, of which more shortly; the other was the pre-appointment hearing before the Commons Education Select Committee of Sir Ian Bauckham as Ofqual's Chief Regulator.
As I watched the Select Committee proceedings on parliamentlive.tv, I was heartened by how strongly Sir Ian emphasised his commitment to ensuring that all qualifications should be fair and trustworthy. As stated in the Committee's report confirming his position, published the following day, 'Sir Ian's ambition to enhance Ofqual's reputation further to be "ever more trusted by the public and by students as the guardian of safe, fair and trustworthy qualifications" is welcome.' Indeed so.
I was therefore most surprised that, during the hearing, no one asked "How trustworthy are grades that are only 'reliable to one grade either way'?". Or "Given that 'more than one grade could well be a legitimate reflection of a student's performance', how trustworthy are the single grades shown on all GCSE, AS and A level certificates?".
As an Ofqual 'insider' – Sir Ian was appointed to Ofqual's Standards Advisory Group in March 2015, became a Director in March 2018, and took over as Chair in January 2021 – Sir Ian should be well aware of the references implied by those questions: Dame Glenys Stacey's evidence to the Select Committee on 2 September 2020 (Q1059), and the 'news item' posted to Ofqual's website on 11 August 2019, as well as former Chief Regulator Dr Jo Saxton's YouTube video of 8 August 2022 (about 9:18).
Furthermore, when Sir Ian was asked about his priorities, the delivery of reliable grades was not one of them. Which once again is somewhat surprising since Sir Ian did refer to Ofqual's statutory obligations under the Apprenticeships, Skills, Children and Learning Act 2009, Section 128 (2)(a) of which places a duty on Ofqual 'to secure that regulated qualifications give a reliable indication of knowledge, skills and understanding'. Perhaps Sir Ian considers that grades that are 'reliable to one grade either way' are reliable enough. I wonder how many students, parents and teachers agree.
Not just that. Section 128 (6) requires that 'regulated qualifications … represent value for money'. Does school expenditure of upwards of £300 M a year on exam fees, for a product that is only 'reliable to one grade either way', 'represent value for money'?
And as regards fairness, what a pity he was not asked about the fairness of the appeals process. About the barrier attributable to the up-front fee. And about the fact that a student's certificate might show a 'non-definitive' grade which cannot be discovered and corrected by a 'review of marking'. Yes, that does seem to be a paradox. But readily explained. We all know that two different examiners can give the same answer different marks, both of which are within 'marking tolerance', with the result that the script might be given, say, 65 or 67 marks. If the B/A grade boundary is 66/67, then 65 results in a certificate showing grade B; 67, grade A. If 67 is the senior examiner's 'definitive' mark, a 'review of marking' of a script marked 65 will discover no marking errors, for there are none. And so the 'non-definitive' grade B is confirmed. Is that 'fair'?
Sir Ian also, without having been asked, stated that he recognised that Ofqual is a non-ministerial department reporting not to the Secretary of State for Education but to parliament. Which is indeed the case. 'Parliament', however, is rather diffuse, but becomes real in the context of the Education Select Committee.
So here's the puzzle.
During the height of the 2020 Covid-19 crisis, recommendation 5 of a Select Committee report dated 7 July 2020 reads 'Ofqual must be completely transparent about its standardisation model and publish the model immediately to allow time for scrutiny'.
Ofqual refused, and did not publish the model until A level results day, 13 August 2020, by which time it was far too late for scrutiny. And we all know what happened next – on Monday 17 August, the results of the algorithm were scrapped.
Throughout that time, Sir Ian was a member of Ofqual's board. Did he condone Ofqual's refusal to comply with the Select Committee's instructions? And what might be the implications as regards Ofqual's governance?
As I've just mentioned, 'reviews of marking' search for 'marking errors', and the number of GCSE, AS and A level marking errors discovered and corrected can be deduced from Tables 13 and 14 of the statistics made available by Ofqual on that same day in December. That number is 221,555.
In the context of the total number of grades awarded, 6,423,785, that might appear to be trivial, a mere 3.5%.
But that is the wrong comparison.
Since a marking error can be detected only if a grade is challenged, the right comparison is against the number of challenges.
And that number was 331,330.
As a percentage of the number of challenges, the number of marking errors was about 67%, implying that 2 challenges in every 3 discover a marking error.
So how many marking errors are in the 6,092,455 grades that were not challenged? Marking errors lurking undetected simply because no one has looked for them?
To me, this raises all sorts of issues about the quality of marking, but this topic was not mentioned at Sir Ian's pre-appointment hearing either.
Maybe Sir Ian has more on his plate than the hearing addressed.
And maybe the Select Committee has more to keep its eye on in the future.