Education Committee
Oral evidence: The Work of the Office of Qualifications and Examinations Regulation, HC 1608
Tuesday 13 January 2026
Ordered by the House of Commons to be published on 13 January 2026.
Members present: Helen Hayes (Chair); Jess Asato; Sureena Brackenridge; Darren Paffey; Rebecca Paul; Manuela Perteghella; Mark Sewards; Peter Swallow; Chris Vince; Caroline Voaden.
Questions 1 to 49
Witnesses
I: Sir Ian Bauckham CBE, Chief Regulator, Ofqual; Michael Hanton, Deputy Chief Regulator, Ofqual.
Witnesses: Sir Ian Bauckham and Michael Hanton.
Chair: Welcome to our first public Education Committee session of 2026; I wish everybody a happy new year. This is an accountability session with Ofqual, and we have the chief regulator and the deputy chief regulator of Ofqual here to give evidence. Welcome to you. Would you like to make any opening remarks before we begin our questioning?
Sir Ian Bauckham: I am very pleased to be here and have the opportunity to talk about Ofqual's work over the last year. A happy new year to all members of the Committee is reciprocated, and I am very much looking forward to engaging in the questions.
Chair: Michael Hanton, do you want to introduce yourself?
Michael Hanton: Good morning everyone and thank you. I am deputy chief regulator and am pleased to be with you this morning.
Q1 Chair: Thank you. I will begin our questioning this morning. Sir Ian, it has now been two years since you first took up the role of chief regulator on an interim basis and a year since you were permanently appointed to the role. Can you set out for us what your priorities have been during that time, what you have delivered against your priorities, and what is still to be done?
Sir Ian Bauckham: As you rightly say, it has been two years since I took up this role. First and foremost, looking back across the two years, our responsibility was to safely deliver qualifications that form part of the portfolio of qualifications in this country.
In 2025, and a similar situation pertained in 2024, we successfully delivered about 7.5 million regulated qualifications across the summer series and just under 12 million overall across the year. In addition, 1.3 million students successfully took GCSEs and A-levels.
This year, and when I say this year I mean the most recent series, we evinced some notable improvements on the previous year and we had a fall in errors in assessment materials in 2025 in comparison to 2024. I should say that no level of error in high-stakes assessment material is acceptable as far as I am concerned but, when the trajectory is in the right direction, I regard it as progress.
We overcame some significant threats to the qualifications system. One example is the sale of hoax exam papers online, which had begun to emerge as a problem in 2024. Owing to concerted action across the qualifications industry, including by us, we saw a marked reduction in that phenomenon in 2025, which I am pleased about because it misleads and effectively defrauds students.
In 2024 and 2025, we maintained qualification standards, by which I mean that the level or standard expected for any particular grade at GCSE, A-level, or any other qualification is maintained constant so it becomes no easier and no harder to achieve that grade. That is in terms of delivery.
There has also been significant progress in terms of reform, particularly over the last year. The Curriculum and Assessment Review issued its final report and the Government responded setting out their policy intentions for reform. It covers a range of things to do with curriculum and assessment, as the name implies, but certainly includes qualification reform. The result is that we have a clear trajectory of work for qualifications reform set out, which we are beginning to engage with now as an organisation.
I want to say something about our enforcement powers. When we were established by Acts of Parliament in 2009, we were given a range of powers to use by way of enforcement to ensure that awarding organisations offering qualifications in England complied with the high standards we expected of them in order to safeguard the interests of learners. Where awarding organisations fall short of those standards, it is important that we signal that clearly and send a message both to the offending awarding organisations and to others about the standards that we expect. We are using those enforcement powers proportionately but without fear or favour. We have imposed 26 enforcement actions over the 12 months, January to December 2025.
More widely, I hope we have made a strong contribution to public debate on qualifications in a range of areas. I have been very clear: it is appropriate and right that Ofqual contributes to public debates from the viewpoint of expertise and experience in qualifications. To that end, we have maintained a strong presence in social media and in the wider media about relevant issues.
We have published a range of high-quality research pieces that stand up to the best quality academic research globally on qualifications and assessment. In fact, we have a further piece of research that we will be publishing later this week on the use of artificial intelligence in marking, so we are keeping up that work. We have re-focused the debate on access arrangements for GCSEs and A-levels and there may be an opportunity to talk about this in more depth later. I became aware that the data that Ofqual had historically published on this did not speak appropriately to the legitimate public questions that there were in this area. We have re-focused and are doing better in that space.
A personal priority for me is to make sure that I continue to get into providers: schools, colleges, and others who prepare students for qualifications. Over the last 12 months, and similarly in the previous 12 months, I visited all regions of England at least once and went into all types of provider preparing people for qualifications, to understand what qualifications feel like and how effectively they work from the point of view of those providers. They included schools, FE colleges, apprenticeship providers, independent training providers, and some other providers including two visits to prison education units.
Finally, in the last couple of months we have published an important policy consultation on on-screen assessment in high stakes qualifications, which I am sure will be the subject of further discussion later this morning. I hope that gives a rounded overview of the things we have been doing.
Q2 Chris Vince: Thank you, Sir Ian and Michael, for joining us today. Ofqual was an observer member of the Curriculum and Assessment Review Panel and you told us that this meant you could not be a signature to the recommendations. Could I ask what your views are on the Review’s recommendations and the Government’s response?
Sir Ian Bauckham: I was chair and observer at the Curriculum and Assessment Review, so I was able to witness the deliberations of the panel and, where appropriate, offer expert advice or advice from the perspective of the qualifications regulator on areas where qualifications reform was in debate.
In terms of the life cycle of those recommendations, they were made by the Curriculum and Assessment Review and the Government then published their own response to it, which set out the recommendations they are going to move ahead with being all recommendations as far as qualifications specifically are concerned.
If I may, I will comment on the recommendations in two broad categories. First, what we call general qualifications being GCSEs and A-levels. Secondly, vocational and technical qualifications.
On GCSEs and A-levels, the recommendations are principally about undertaking a review of GCSE and A-level subject content appropriately, with a particular focus on GCSE subject content with a range of objectives in mind. That is a sensible thing to do at this stage. It has been a number of years since those GCSEs were first reformed, so taking stock of the content that is delivered in those GCSEs feels like a sensible thing to do.
Most notably and relevantly for us as the qualifications regulator, an ambition is set out to reduce the overall examination time for a typical student taking between eight and nine GCSEs by two and a half to three hours in total on average. That is the way the Government’s response to the review has framed the ambition. From my perspective, that is sensible and rational. I am bound to say it is also achievable without jeopardising the high standard of reliability that we expect from GCSE grades in particular. I can say a little more about that if useful but it is a principal focus for us in the reform process of GCSEs.
I will finish my remarks there on general qualifications and move on to the vocational and technical side of the Review. There some recommendations are a little more far-reaching.
First, they concern the establishment of a new V-level qualification which is designed to sit within what will be the suite of advanced V-level and technical levels for students aged 16 to 19 at level 3. It potentially represents a significant step forward for qualifications in that space. Up to now, what we have generally referred to as AGQs, which have been the qualifications taken by those at level 3 not taking A-levels or T-levels, have not been informed by nationally set content. Therefore, they have been more difficult to accredit and quality assure.
The establishment of nationally set content to sit behind those qualifications gives us the opportunity to quality assure proposals from awarding organisations more rigorously. It also gives us the opportunity to establish a national approach to standard setting and to grading for V-levels, which is a welcome development in a field that has been more diffuse and diverse in terms of standard setting and grading. In saying all that, the policy intention is not to force V-levels into a mould dictated either by T-levels or A-levels but to create and then maintain a distinctiveness for V-levels.
If I may say, Chair, I noted in your own review, which you published in September 2025, that you recommended the maintenance of the middle routes that, at the time, you referred to as AGQ. The V-level reform speaks to that priority.
A couple of things on that side of the Curriculum and Assessment Review. The establishment of the level 1 stepping stone qualification for English and Maths to address the needs of students who have not achieved a grade 4 at GCSE in English and Maths by the end of year 11. This, I should say, is after 11 years of teaching and learning in English and Maths. The establishment of that qualification as a stepping stone qualification provides a way to identify shortfalls in learning that will have emerged over the 11-year period during which those students have been studying English and Maths, and address them before students step back on to the GCSE track and aim to achieve the valued and important grade 4 at GCSE in English and Maths. I very much welcome that step forward.
Finally, on the vocational and technical side, the establishment of foundational and occupational certificates at level 2 for those students who still need to work at level 2 post-16 is particularly welcome because the landscape of qualifications and provision for those students in particular post-16—so these are students achieving on average below grade 4 across the board at GCSE—has hitherto been sub-optimal. Devoting resource and energy to the intentional design of high quality qualifications for students in that situation is also welcome. So, broadly speaking, I welcome the recommendations.
Chair: Thank you very much. We would certainly agree with that final point very strongly.
Q3 Chris Vince: There are two points I want to make. You touched on stepping stone qualifications. I have to declare an interest: I used to be a maths teacher so that is very interesting to me. It will be interesting to hear your thoughts on the timescale and how it is going to develop. Also, I just wondered if there was anything that was not in the Curriculum and Assessment Review that you would have liked to have seen as well?
Sir Ian Bauckham: Can you just repeat the point about maths because I did not quite catch it?
Chris Vince: It was just with regard to the stepping stone qualification you mentioned. I always mention in this place that I am a maths teacher; it is my thing. It was just the timescale for that and how you saw it developing.
Sir Ian Bauckham: These qualifications take time to develop properly. The first step in all these qualifications is the development of content and key parameters to do with qualification design. The Department for Education has committed to run a consultation on the stepping stone qualifications in the spring of this year. That was set out in the Government response and we are working intensively with the Department for Education to provide expert advice on that. In terms of the earliest date on which the qualification could arrive in schools for teaching, are we able to pinpoint that date yet, Michael?
Michael Hanton: That date has not been confirmed yet. There is also the interplay with revisions to GCSEs in English and Maths, making sure that is a coherent suite of qualifications across reforms to the GCSE and then the new qualification.
Sir Ian Bauckham: In terms of your wider question about whether I thought there was anything significant missing, I do not think there was. The Curriculum and Assessment Review I witnessed was a very rigorous and thorough overview of the whole curriculum, assessment, and qualifications landscape in England. Every aspect of that was examined critically, in an unbiased way, using the evidence that was available and the headline recommendations the Review arrived at are broadly sensible.
If I may make one final comment on the English and Maths qualifications, Chair? I noticed that you expressed strong views on the treadmill of resits in your own reports. I agree because there is no point in a student who has spent 11 years studying English and Maths, and only achieved a grade 1 or 2, repeatedly sitting an examination in the hope that, by luck, one day they might scrape enough marks to pass. That is because achieving a grade 4 at GCSE in English and Maths is not a matter of luck but a matter of being able to demonstrate the skills, knowledge, and understanding required.
You said in your report that it was important that the Government also look at the root causes of students making so little progress that they only achieve a grade 1 or grade 2 and try to address that much earlier on. I would say that goes right back to the first stages of primary education because we know there are still too many children leaving primary school not having reached the expected standard in English and Maths. So it tracks right back through primary and secondary to try to ensure that we do not have students who have these gaps in their learning.
As somebody who has picked up a low achieving GCSE group in year 10 and year 11, it is absolutely alarming to discover how significant and basic those gaps can be: failure to understand basic place value in Maths, for example. It is really important that is addressed early as well as having the right qualifications for what hopefully will become a small minority who still do not quite get there by the age of 16.
Q4 Peter Swallow: One of the findings of the Curriculum and Assessment Review was that too much time was spent on exams. The Government are committed to working with Ofqual to reduce GCSE exam time by two and a half to three hours on average. It would be great to understand what that process is going to look like and, in terms of outcome, are we looking at fewer exams, shorter exams, or both?
Sir Ian Bauckham: The answer is probably going to be a combination of the two, but probably shorter exams in most cases. Most GCSEs are assessed through two exams, some are assessed through three; so there is an opportunity to consider whether two or three is the right number of exams in each case. Exam lengths vary, interestingly, even in the current dispensation between exam boards offering GCSEs in the same subject.
One of the exercises we did in the lead-up to offering advice to the Committee in this regard was to look across the subjects offered by a range of exam boards and try to answer a simple question. The reality of this is unlikely, so it is a hypothetical question, but if a school said, “Reducing examination time is our number one priority above all other priorities in choosing an exam board specification therefore we are going to choose the shortest specification in each subject regardless of all other factors” then, if we take the most common subject combinations—eight or nine GCSE subjects—it would already, in the current dispensation, be possible to reduce on average the assessment time by about an hour and a quarter. Without doing any additional reform, even in the current dispensation, that much saving is within the bounds of achievability because all these qualifications have been accredited. They are all good quality qualifications.
You might legitimately ask why they vary in lengths of time. That is because, at the last point of reform a little over a decade ago now, assessment time was not considered a priority. It is now considered a priority so, when we go through the reform process this time round, we can be more rigorous, questioning, and demanding with exam boards that are proposing qualifications, with a view to pushing down the overall assessment time but, importantly, without compromising the high level of grade reliability that we have.
There is a correlation between three things here. First, the volume of content to be taught. So how much content lies behind the GCSE. Secondly, the reliability of the exam grade. In other words, how precise an indication does it give of the mastery of that content?
Thirdly, the assessment time itself. We think there is sufficient margin to make those reductions in overall assessment time, which are relatively modest in the grand scheme of things but sufficient to be noticeable for students to make those reductions while respecting the high level of reliability that we have for grades without making overly bold assumptions about content reduction. Content reduction is a matter that sits with the Department for Education.
Q5 Peter Swallow: It is helpful that you set out that clarification because some people have a perception that this is about lowering standards and the quality of the assessment. Of course, you can have an assessment as long as you want on the same spec; it does not inherently make it a better exam just because it is longer and it was helpful for you to set that out.
You talked about your evidence base for this and how you have come to the conclusion that this is a feasible and achievable goal. Will you commit to publishing the modelling that you submitted to the review which demonstrated that this cut is achievable?
Sir Ian Bauckham: Yes, I have described some evidence that we offered to the review already in terms of the analysis of the existing landscape of specifications and the relative length of examinations. The evidence we provided was for the purposes of ministerial policymaking, so we would not routinely publish that information at an early stage. However, we will be consulting on any proposals to do with reforming GCSEs and the evidence associated with those proposals will be published at that point.
Peter Swallow: You will commit at least to publishing further information in the future?
Sir Ian Bauckham: Yes. We will publish further information at the point we consult on the proposals but we do not normally publish information or advice provided for ministerial policymaking.
Q6 Manuela Perteghella: Sir Ian, you have already discussed a significant period of reform for post-16 qualifications. What is Ofqual's role in developing and rolling out the new qualifications? Can you update us on the progress you have made so far? Do you have a say in how inclusive these qualifications should be for all students' abilities and needs?
Sir Ian Bauckham: Yes, I will give some headlines in answer to that response and I may ask my colleague to supplement my answer. I will focus mostly on the qualifications in the vocational and technical space to answer that question because the GCSE reforms are more straightforward, reduction in assessment time notwithstanding.
In terms of V-levels, the occupational and foundation certificates, and the stepping stone qualification in English and Maths, progress is being made on both. The first mover, if you like, in designing policy for qualifications is the Department for Education because there are key policy decisions that have to be taken about the purpose of qualifications and the content that sits behind them. The lead agency, at this point, in all three of those strands is the Department for Education as it makes those decisions.
The consultation on V-levels closed yesterday, so the Department will be considering the outcomes of the consultation before finalising its policy. That does not mean that we are doing nothing at this time; we are absolutely providing detailed advice to the Department for Education on all three of those strands. On content, although decisions about content sit with the Department and ultimately Ministers, there are consequential impacts to decisions that are made about content on qualifications. One thing that we do during the process is offer advice to Ministers in the Department for Education on what those impacts might be so that those decisions are taken with eyes open.
Michael, can you offer some further supplementary information to support my answer?
Michael Hanton: I would think about our role in three key ways. First, we are really committed to making sure that the awarding organisations delivering these qualifications are fit for purpose and that they have the expertise and the right experience to deliver those qualifications. That is a strand of our work.
Secondly, we think about qualification design. What is the purpose of these qualifications? Questions like the grading approach; making sure these qualifications are designed in a way that mitigates any risks or threats to them so they are going to be valid and fit for the purposes they have. Thirdly, thinking about the quality of these new and important qualifications. We will be looking at the quality of these qualifications upfront, before they are taken by students.
You mentioned inclusivity. One thing the Curriculum and Assessment Review said, which again we fully support, is that it is essential to think about accessibility of qualifications upfront. From the design of the qualification; thinking about how we design these qualifications so that they are accessible to the students they are intended for.
Sir Ian has talked about the different qualifications at level 2 and level 3, and making sure they are targeted at the students who are going to take those qualifications. We have extensive guidance, that awarding organisations are required to follow, to think about how they design assessments in the right way. Accessibility is part of our upfront quality review as well.
Q7 Manuela Perteghella: Have you already begun consulting on the V-levels and occupational certificates?
Michael Hanton: We will be consulting in the spring. We will have our first policy consultation setting out all the issues I have just talked about.
Q8 Manuela Perteghella: In the spring, thank you. Are you confident that Ofqual has the resources to be able to focus on this work without negatively affecting its work in other areas? Will any other work be deprioritised?
Sir Ian Bauckham: Thank you for that question about resourcing; I will just say a couple of words about resourcing overall. We have a settlement in the current spending review that, in common with all Government Departments, is not without challenge but, none the less, we have been through a rigorous overview of our core work and I can confirm that we are able to continue with all aspects of our core work and discharge them effectively.
The new work associated with qualifications reform is complementarily funded by the Department for Education. What I can say is, providing funding is forthcoming as discussions have indicated it will be, that we will have sufficient resource to deliver the reforms without compromising any other core aspects of our work.
Q9 Chris Vince: I want to talk briefly about AI and technology in terms of assessment. Obviously, there is real opportunity here but I know that the Curriculum and Assessment Review highlighted the risk of using generative AI, particularly in coursework. I am also aware that Ofqual was conducting some research to understand the implications of AI for qualifications and assessment. I just wondered if you could touch upon where you are with that research and if you have been able to look at the concerns that the Curriculum and Assessment Review highlighted in terms of AI in coursework.
Sir Ian Bauckham: Thank you for the question about artificial intelligence. There are three key points that I want to make about artificial intelligence. First, the use of AI upstream from the point students take qualifications within the qualifications industry. This is the positive bit. There are lots of opportunities for exam boards and awarding organisations to use AI to support quality assurance, reduce costs, and to improve efficiency. We are very positive about all that work, which we encourage exam boards to do because it basically makes qualifications more secure, less error prone, and can help support the quality assurance of things like marking and the quality of set papers.
Secondly, artificial intelligence used for the marking of scripts. As I alluded to earlier, we will be publishing a specific piece of research later this week on this. In short, it maintains the position that we have taken hitherto, which is that we do not believe AI is yet ready to undertake sole marking of student work. That is not to say it has no role in marking; it can be used to quality assure or detect unexpected patterns in examiners' marking and so on, but not for the sole marking of student work. That is party because AI still makes mistakes, it still hallucinates, and partly to do with transparency and the importance of transparency. If I am taking a high-stakes qualification, it is right that I can ask why the examiner awarded me mark X or mark Y. It is difficult to do that when the marking has been done by a complex series of algorithms inside an impenetrable black box.
Thirdly, the importance of challengeability. It is a principle of natural justice in qualifications that if I think the grade I have been awarded is wrong and does not match my recollection of my performance, I should have the right to challenge it and ask for a different human being to take a view of my work and determine whether they agree with the determination the original marker made.
If you think about what that looks like in the event that AI has done the marking instead of a human at that primary level, it feels quite complex and difficult. What would happen if the AI said, “I did not make a mistake” when a human said, “It looks to me as if the AI has got this wrong and this candidate should have been awarded a mark for this question?” What does that mean for all the other candidates who were marked by the same complex set of algorithms? Are they all wrong, even if they have not asked for a review? There are questions there about transparency and challengeability. That is the second broad area to do with AI.
The third broad area to do with AI is malpractice. This may be what you were primarily alluding to in your question and I can make a few comments on that. This is a particular challenge for qualifications that have extended writing coursework; it is less of a challenge for qualifications that have other types of coursework. Coursework that involves producing a physical artefact, whether it is an item that has been produced in design technology, a piece of physical artwork, performances, or whatever it is, is less open to AI malpractice. You have subjects like PE, Dance, and Drama where there are particular performances of various sorts that are recorded which are also less open to AI cheating.
Extended writing is open to AI cheating. Fortunately, across the GCSE and A-level suites, we have relatively little extended writing as non-examined assessment. There are several places where it does exist: the two I would highlight in particular are English and History A-level, where about 20% of the qualification in each case is extended writing coursework. I would not be transparent if I did not say that I am concerned about it.
As I alluded to in my opening remarks, I visited a number of colleges and schools up and down the country, spoke to teachers, and a number raised this with me. In response, we made sure the advice we gave to the Curriculum and Assessment Review on making decisions that result in qualifications where this kind of vulnerability existed was taken on board. The Curriculum and Assessment Review reached a sensible place taking it into account.
I have signalled my anxiety to the Department for Education, particularly about History and English A-levels, because part of the driver for extended writing coursework lies in the way the content itself is designed. As I said earlier, that is a matter for the Department for Education.
I have also commissioned a significant piece of work internally to Ofqual, as the regulator, to ascertain the extent to which this risk is being realised at the moment and what the options might be in any reform of those qualifications for closing down that risk and assuring ourselves of the integrity of these qualifications. I share what I read to be the premise of your question; there is a cause for anxiety here.
Q10 Chris Vince: My concern comes from the fact that there is a danger that we could decide to model a qualification based on avoiding malpractice and not on doing what we think is best for the qualification. We might feel that, in some subjects, those extended pieces of writing and coursework are the best way to assess something but we might not choose to do so because of malpractice. That is ultimately the danger, is it not?
Sir Ian Bauckham: Yes, that is absolutely true. There are some ways to not eliminate but reduce the risk of malpractice and maintain writing in coursework or non-examined assessment. They include ensuring that candidates are required to present the same information in multiple different formats. That is still achievable with AI but it starts to make it more difficult to use AI. We can ask candidates to complete extended disclosures, which are countersigned by teachers that really probe the extent to which the candidate may have accessed AI.
It is possible to tell an untruth in such a disclosure but the more detailed and probing it is, the more difficult it gets to tell an untruth. It is possible to build in compulsory checkpoints through the completion of the work, which the teacher signs off as a way of verifying that the work is the student's own. None of those are foolproof but they are alternative methods to removing it altogether, which can reduce the risk.
Q11 Chris Vince: I declare an interest as the chair of the All-Party Parliamentary Group for Schools, Learning and Assessment and the All-Party Parliamentary Group for Classics.
On the multifarious uses of AI, I note that Ofqual has recently looked at the grade boundaries for MFL and recognised that students might historically have been graded too harshly for those subjects. Are you looking to use AI across other subjects to see whether there might be anomalies? Can AI look at a particular cohort of students receiving lower grades in particular subjects compared to others, spot anomalies, and then address them? I know that one of the challenges in dealing with the issue has been building an evidence base around it.
Sir Ian Bauckham: You are right to reference the work that we did in German and French GCSEs, which was begun in 2019 and, in fact, completed post the pandemic. It was to answer the challenge of whether it was unduly harder to achieve higher grades in GCSE French and German than in other subjects. We did not use AI in doing that work; we used traditional data analysis and standards analysis methodologies.
The conclusion we reached was that it was slightly harder to achieve a particular high grade in French and German than it was in Spanish, so we committed to make adjustments to the standard of work expected to get those high grades in German and French to bring that standard in line with Spanish because it felt like a natural comparability within that suite of GCSEs.
Would we use AI to expedite the work in other subjects? I do not see any reason why we would not. I have to tell the Committee that we have not done so hitherto. We would be at the beginning of that journey but it is something we can certainly look at.
Q12 Chair: It is pretty common practice now for students to use AI to mark their practice work. They run an essay through AI software to find out how they are doing. What advice would you give to students about that practice and the accuracy of the picture they are getting about their own work as they prepare for public exams?
Sir Ian Bauckham: It is very good if students are undertaking practice work and seeking feedback on it. All I would say about AI, because there are multiple versions of AI available commercially and lots of different ways in which it can be used, is that I would encourage them to seek corroboration from an experienced teacher as well as any feedback they might get from AI.
As I said earlier, I visited a number of schools and colleges recently and I remember particularly a visit to Reading College in Berkshire that had done some quite advanced work in the use of AI to support students, mark their work, and give them feedback. I was impressed with what I saw. These were students principally taking English GCSE again in year 12 and year 13.
The teacher had a large class and had developed an AI application that was able to support the teacher with the marking of work and feedback to students in order to speed up the turnaround so that students could progress more rapidly. It was good quality feedback but it was very much designed and supported by an experienced teacher. I see no problem with students using the resources available to them online, but I would say, “Make sure you are on the right lines by checking in occasionally with your teacher as well. Do not attempt to replace your teacher with AI because that might lead you into difficult waters.”
Peter Swallow: Hear, hear!
Q13 Darren Paffey: Ofqual is currently consulting on proposals to use more on-screen assessments. I was wondering if you could say a little more about your expectations around the timeline for that and when we might see it implemented and rolled out?
Sir Ian Bauckham: We have done a lot of research on the question of on-screen assessments over recent years. One significant piece of research in particular highlighted some potential benefits to the on-screen assessment of GCSEs and A-levels, but quite a significant range of risks as well.
The situation at the moment is that there is no formal regulatory method by which we could prevent an awarding organisation bringing an on-screen assessed qualification to market. Our appraisal of the various risks and challenges involved in on-screen assessment led me to conclude that we should, in fact, have a method to control the entry of any on-screen assessment into the market and therefore delivery in schools.
The current consultation is first and foremost about a proposal to give ourselves the power to manage the entry of any on-screen assessment into the market. We have taken a relatively conservative approach, so the first principle is no on-screen assessments unless they meet certain criteria or fall into certain categories. It is a locking down and a safeguarding of the system as it is now because of all the risks and potential difficulties associated with on-screen assessment. It is then a very judicious potential opening up in a number of areas.
First, we have taken off the table, in line with a steer from the Secretary of State, all general qualifications with current entries of over 100,000. At GCSE, that takes out English, Maths, Combined Science, Physics, Chemistry, Biology, French, Spanish, History, Geography and Business Studies. If you look at the most common combinations of GCSEs that students do, they are by far the most common qualifications that are taken, so it is a very judicious proposal with regard to GCSE. At A-level, we have one qualification which exceeds 100,000 in entry and that is Maths, so that is taken off the table as well.
The second control that we have put in place is limiting the number of specifications that any exam board may propose to a maximum of two per exam board. We have four exam boards. Should all those materialise, it is a maximum of eight. In terms of what that looks like as a proportion of the total number of GCSEs and A-levels offered, we have 270 GCSEs and A-levels so that maximum rate is basically a very small percentage.
The third control is that any such proposal would not be automatically approved. Even if proposed, those specifications developed by exam boards would need to go through a rigorous accreditation process from us where we would make a determination about whether they were fair and could be introduced without, for example, exacerbating existing inequalities gaps, were deliverable in schools, of good quality, and all the usual things about being a reliable indicator of students' knowledge, skills, and understanding of the specifications.
In short, research tells us that while there might be some advantages there are a range of risks. I am happy to say more about the risks if that is useful. Therefore, we need to control entry into the market here and we currently have no methodology for doing that. This is, first and foremost, about giving ourselves the power to have a methodology to control entry into the market and then it is about potentially, and this is a genuine policy consultation, opening up some limited avenues whereby some qualifications could come to market with all those conditions.
Q14 Darren Paffey: They seem very reasonable controls. When do you expect to have those controls within your power—this year, next year? Do you have any reason to believe that, of the few qualifications left below the limits and outside those caps, anyone is racing to get there before you can stop them?
Sir Ian Bauckham: On the question of racing, it will not have escaped the attention of those of you who vigilantly watch the education media that there has been significant coverage of some exam boards' enthusiasm in this space. It is an enthusiasm that we seek to temper for all the reasons I have outlined. Yes, there is enthusiasm among the awarding body community but my job as regulator is not to do what awarding bodies and exam boards want me to do. Quite the opposite. My job is to safeguard the examination system in the interests of students, society, and the economy in this country. Yes, there is racing.
How long is it likely to take? The current consultation is a policy consultation. It is only the first stage in the consultative process and closes on 5 March this year. After that, we will consider the results of that consultation. I detect, from observing the public debate on this issue, that there will be a significant response to that consultation.
If we decide to move ahead—“if” is an important word in that sentence—then the next stage will be to consult on the technical changes to our rules, which have the force of statute, to do the things we propose to do either as consulted or in an amended form. Again, that would need to be a full three-month consultation. Exam boards would then need to develop any specifications they wanted to submit for accreditation. We would have to go through the accreditation process which is rarely a rapid process because, when exam boards submit specifications for accreditation, they very rarely meet the required standard first time round. They nearly always need to have more than one go at it.
Then, of course, those specifications in the normal run of things would need to be in schools a year before first teaching to allow teachers to prepare for teaching them. You then have the two years of teaching the course leading up to the examination. When you put all that together, the first exams we are likely to see, in the event that we go ahead with the proposals as consulted now, is about 2030.
That is a very long answer to a simple question, apologies, but hopefully it is helpful to have the background.
Q15 Darren Paffey: A final question on this. You will be aware, in the wider debate, that yesterday the Education Secretary made announcements around screen time for younger children and how evidence is suggesting that it impedes quite a bit of learning. How do you square that with evidence for later learners, so adolescents, and the timescale you are proposing for when exams might go online with the opportunities that, as has been identified in the discussion, SEND children might find on-screen assessments better? What evidence are you leaning on in order to make those decisions for the good of as many learners and examinees as possible?
Sir Ian Bauckham: There are lots of competing areas of evidence and opinions, and you are quite right to allude to that. We have good quality research in this area, bespoke pieces of research. We have an academic research team at Ofqual who make it their business to ensure that they are abreast of all research developments in this area. Our proposals are underpinned by solid academic research.
I will put my cards on the table and say I share what is emerging: the beginnings of a consensual view that excessive screen time for children is potentially damaging. I am mindful that proposing there could be, albeit in the very controlled way that I have described, some assessments at GCSE or A-level on-screen might be seen to be supporting more screen time. I do not think that is the case for the reasons I have described. The volume that we are proposing as a maximum is so restricted that I do not think it will have any impact on those older children.
I share the concern about young children spending too much time on screens and on forms of on-screen applications including, particularly damagingly, social media, where children expose themselves to the judgment of potentially millions of other people; the human brain is not designed to cope with that, probably at any age but certainly not as a young child.
Q16 Jess Asato: Following on from the previous questions, what assessment have you made of schools' capacity to implement on-screen assessments and the level of investment that will be required?
Sir Ian Bauckham: The concern that underpins the premise of your question relates to capacity in the school estate, which has certainly been a matter of discussion with the schools and colleges I have visited over the last year.
If I can once again allude to the visit to the sixth form college in Hampshire that I made recently, I asked the exam secretary what it would it be like from their perspective—as someone organising the public exams for a large number of young people taking A-levels and other qualifications at 18 or 19—if any significant proportion was done on-screen. I detected a certain blanching and then, as is typical, a series of very practical concerns such as, “Where do all the laptops come from? How do we manage in an exam hall? What about extension leads? What if it does not work on the day of the exam?” It is sometimes easy for policymakers sitting above all this to ignore practical issues that become very real indeed when you talk to the real people in the grassroots institutions who have to deliver these qualifications. That absolutely underpins why we are taking this in hand.
First, my view is that—should it happen at all—we cannot allow on-screen specifications to saturate the market in an uncontrolled way, whereby we do not have the tools to manage the volume. Secondly, the extent to which we open the sluice gates should, in the first instance, be very minimal indeed, to make sure that anything that is allowed to enter the market is small scale and therefore sufficiently manageable for schools and colleges.
Obviously, the provision of infrastructure in schools and colleges is a matter for the Department for Education, and the Secretary of State's steer letter that accompanied our consultation—where the Government makes their own comments on their ambitions for IT and their take on the IT infrastructure of schools—has been published. But we have to deal with the world as it is, and for teachers, exam officers, headteachers and principals, this feels challenging. That is why, as I say, we are taking a very controlled approach.
Q17 Jess Asato: On the back of that, do you think there is any risk that schools might consider dropping some subjects if they are not sufficiently supported to offer on-screen assessments?
Sir Ian Bauckham: The risk of that happening is low. In many cases, not all cases, there is more than one alternative. We have a market of awarding organisations and four exam boards. I deem the likelihood of all qualifications in a particular subject moving on-screen across all exam boards as not impossible, but very low. One of the complexities, but also strengths, of a market-based system for qualifications is that there is choice and diversity within qualifications. So those choices will still be available to schools.
Q18 Jess Asato: It is not impossible, and the risk may be low; but you could potentially end up with a situation where in one subject, all the exam boards go on-screen. There will be some schools who do not have the funding to cover that.
Sir Ian Bauckham: The subject where that is most likely to happen is computer studies or computer science, where in fact we already have some on-screen assessments. Beyond that, you are quite right to say that it is not a theoretical impossibility; but once again, I come back to the way in which the awarding organisation market operates because it is principally about securing market share.
Qualifications are their business; if by moving on-screen there is a prospect of losing schools and therefore losing market share, that acts as a strong disincentive for exam boards to make such a decision. They do not want an exodus of schools taking their qualifications. That is not a hard control, but it is a market operation control.
Q19 Jess Asato: I have two more questions. One is around students with SEND who clearly, as we have heard, would benefit from having on-screen assessments; obviously that would be hugely welcome. But we also know that there is quite a lot of evidence that, for some students, handwriting helps them both to retain and interrogate information in a way that perhaps works better for their memory skills. Do you see a situation where there might be some students who have the reverse argument, which is, “You are jeopardising my learning by making this on-screen because the way I have learned is with paper and pen and that actually helps me?” Can you see that argument developing in the future?
Sir Ian Bauckham: Yes, to a certain extent. Chair, if I can pick up the Member's question in reverse order, taking the handwriting point first; I strongly agree with the premise that underpins that question. Handwriting is cognitively important during the learning process. We may not yet have rock-solid evidence for that, but the emerging evidence is that learning to handwrite does support cognitive development. Ensuring that we do not create incentives in the system that reduce the emphasis on handwriting at earlier stages of learning is important. Again, that is why we are proposing to control on-screen assessments and only allow them in very limited circumstances.
Coming back to children with special educational needs and disabilities, again the evidence is not overwhelming but there is some evidence—as our research indicates—that some children with some types of SEND may benefit from the opportunities that on-screen assessment offers. It is of course already the case that under equalities legislation children with a specific SEND can access arrangements for qualifications on-screen to mitigate a learning need.
Q20 Jess Asato: Finally, you have already said that subjects with the highest numbers will not be able to be taken on-screen, but what is the likelihood of this changing in the future, given that it is a market?
Sir Ian Bauckham: Well, it will not change up to now because while it is a market, it is a regulated market; how the market behaves is controlled by Ofqual because we are the regulator. If the results of the policy consultation do not make us revise our view and we implement that cap, it will exist until we choose to revoke it.
What I want is to watch both how the wider state of education and the resourcing of schools develops, and what the research is telling us over the next two, three, four, and five years. I want to see how the development and accreditation of any actual on-screen specifications is going, how successful that is, and what kind of quality results, before I would propose any revision of that.
Q21 Caroline Voaden: Ofqual is proposing that exam boards must use substantially different questions in on-screen and paper-based specifications. Could you explain why that is? How will you ensure that this is done fairly and that the qualifications will be comparable?
Sir Ian Bauckham: I may invite my colleague to come in on this, but let me just give you a quick headline.
It is already the case that we have different specifications and therefore different exams, in the same subject between exam boards, and they are comparable in standards. So the methodology for ensuring that we have comparable standards across different question papers in the same subject already exists. The rationale for having different questions in anything that goes from paper to on-screen is largely to do with the way in which mode effects play out. Mode effects are the impact that the method of assessment has on the quality of the student response. The quality of the response is in part shaped by whether they handwrite it or whether they type it.
If you imagine a situation where you have an on-screen specification in, let us say—give me a subject, Michael?
Michael Hanton: Business.
Sir Ian Bauckham: It will not be a business GCSE because that is out of the equation, but just for the sake of the argument let us pretend that there is an online business exam and a handwritten business exam, and the questions are exactly the same.
What could happen is that with the online exam, the mode effects lead either to higher marks or lower marks. Because of those mode effects we would have to set standards differently to compensate. If the questions have been the same but the standards are different, we think that would jeopardise public confidence.
I am aware I have not explained this very clearly, so I am going to ask my colleague to have another go.
Michael Hanton: Thank you, and apologies; I was thinking of A-level business. That was my mistake.
The key point is essentially that we want to ensure that standards are comparable across all specifications. As Sir Ian said, we go through a process with exam boards every year to ensure that exams are no easier or harder with one exam board compared to another. That is a core principle of the regulatory framework. It is easiest to achieve when they are independent specifications, where examiners are looking at the quality of work as it has been presented through the mode of assessment that students have taken. That is then aggregated and looked at by examiners alongside the range of other evidence that they have. This is essentially the most effective way to maintain standards.
Alongside that, we are also interested in making sure that exam boards are making the best opportunities of digital. We do not want them necessarily to be constrained by taking exactly the same questions and simply replicating them on-screen, but to think about how items are presented on-screen, making the best use of the digital technology that they have. So there are those two arguments.
The final control—again Sir Ian has mentioned this—is what we will be looking at in our accreditation process. We will be looking at how they are going to ensure standards are maintained across the different forms of assessment on paper and digitally, and how they are going to ensure that the questions are of an appropriate difficulty. We will be scrutinising that upfront as part of our approval process.
Q22 Caroline Voaden: Your own research highlights that on-screen assessments could give students from wealthier backgrounds an unfair advantage over others. How are you going to monitor the impact of this? If students are negatively affected, what changes could you make?
Sir Ian Bauckham: Once again, the risk of that happening is one of the reasons we need to take this in hand and prevent an uncontrolled entry of on-screen assessments into the market. The headline principle is that we need to have the tools to manage this so that we can control it.
One of our core principles—I am absolutely firmly committed to it—is that no qualification can be an experiment as far as young people are concerned. We will only allow regulated qualifications to be used in schools and colleges if we have reached a very high bar of confidence that they will not, by design, unfairly discriminate against students from less advantaged backgrounds. We will do that process during the accreditation; we will check that there is nothing in the design intention or implementation of the qualification that would inadvertently cause specific harm to students from socio-economically less advantaged backgrounds. There will be no experiments here, and that is a very high bar.
Q23 Caroline Voaden: It is probably fair to say, though, that schools in richer areas, possibly private schools, have more income and will be more likely to offer online assessment because they will have the necessary IT equipment.
Sir Ian Bauckham: Yes, but they will have very little opportunity to do it precisely because we allow so little of it to exist in the system.
Q24 Caroline Voaden: If I am gauging this right it seems that you are really reluctant; you are really not keen on the idea of on-screen assessments at all. So why do you not just say it is not going to happen?
Sir Ian Bauckham: As I said, the first thing we needed was to make sure we had the tools to control entry to the market, so that is what we did first. We are mindful of the significant risks but as has been alluded to in other questions, there are also some potential benefits.
Technology is moving fast. It is a plausible argument to say that many aspects of life are increasingly on-screen, despite all the concerns we hear about that. We thought it was right to consult on a proposal that allowed exam boards to show whether they could develop a qualification that could be delivered fairly for students—albeit in relatively small entry subjects—and we would evaluate that on its own merits. We felt it was appropriate to open the door, but only that wide.
Q25 Manuela Perteghella: I want to talk a little about exam boards. Ofqual has issued a number of significant fines to awarding bodies this year for breaching regulations regarding exams. Why do these problems occur, and what more can be done to prevent future breaches?
Sir Ian Bauckham: That is a very pertinent question; once again, Chair, I may ask my colleague to come in on the back of my opening remarks.
The regulatory powers given to us by Parliament have been used to the full. On arriving as chief regulator I inherited something of a backlog in enforcement cases, and I have been very keen to clear that backlog. Our powers were given to us by Parliament to send clear strong messages to exam boards and to the sector that they will be sanctioned where there are breaches that are potentially damaging for the integrity of the qualification or for students.
I am particularly keen that this is done in a timely way, in other words as quickly as can safely be achieved. The adage that justice delayed is justice denied applies in this circumstance; if a student feels that an exam board has done something that has rocked the student’s confidence in the exam that they have worked hard to gain, then something must be seen to be done quickly about that.
We have used our formal legal enforcement powers on 26 occasions in the last year, from January to December 2025, which is significant. Not all of those involved fines—for example, we also have the power to impose legally binding special conditions on exam boards as well, to make them do things that we think they might not otherwise do.
Why do these things happen? There is a range of reasons: they happen by accident, they happen because exam boards sometimes take their eye off the ball and are not sufficiently attentive to risk management, and they happen because we are operating in a commercial market and exam boards have an eye to what competitors are doing for the same market share.
What can be done to stop these things happening in the future? One of the things we always do when we identify non-compliance and move towards using enforcement action is to make sure that the exam board has fully understood the problem, carried out an appropriate root cause analysis of why it happened in the first place, and put in remediation to prevent it happening again. We take a very dim view of exam boards that repeatedly make the same breach of our requirements. That is a very serious offence. If it is a different breach, that is also serious but less serious than repeat offences in the same area. We have the power to escalate the scale of our penalties where exam boards are not learning from past non-compliances and are making the same mistake a second time.
Q26 Manuela Perteghella: Can an exam board or an awarding body that is making the same mistakes year after year be disqualified?
Sir Ian Bauckham: Yes. The nuclear option is to withdraw recognition. Formal recognition gives the power to an exam board to offer regulated qualifications; if we withdraw recognition, that means that the qualifications that exam board does can no longer be funded in state-funded institutions in England. So that is the nuclear option. We would always consider the impact it would have on students taking those qualifications at the time, but ultimately it is a sanction that is legally at our disposal.
Q27 Manuela Perteghella: Many of these breaches occurred several years ago—2016, 2017, 2019—so is the delay in Ofqual issuing fines because of this backlog?
Sir Ian Bauckham: Yes.
Q28 Manuela Perteghella: How has the backlog formed?
Sir Ian Bauckham: The backlog probably formed for a couple of reasons. It was partly because of covid; during the covid years, little enforcement activity took place. That was a situation I inherited as chief regulator. It was also partly because—I hope I can say this—the priority has not always been to speedily bring enforcement cases to a conclusion.
One of my personal priorities is to ensure that where there is non-compliance it is speedily brought to enforcement, for the reasons that I alluded to earlier. My view is that justice delayed is justice denied. If I took an exam and somehow the exam board messed it up and I have been caused stress and anxiety as a result, I do not want to wait for three years before I see something done about it, I want something done about it in the current academic year. That is the view I take as somebody who has worked with young people in schools for many years.
Q29 Manuela Perteghella: Lastly, it was reported in the summer that you were scrutinising Pearson's approach to A-level maths, following concerns about this year's replacement papers. What updates can you give us about the results of this scrutiny?
Sir Ian Bauckham: Pearson is the largest provider of maths A-level in England, and maths A-level is the largest A-level in the country, with a little over 100,000 entries. It has three components, papers 1, 2 and 3; following paper 1, there was speculation on social media about the potential predictability or apparent familiarity of the questions. As a result, Pearson took the decision to withdraw the paper 2 they had originally lined up and replace it with a different paper.
When that paper was sat there was further negative commentary from students, particularly on social media, about the range of topics that had been covered. We analysed that and reached the conclusion that when papers 1, 2 and 3 were taken together—paper 3 had not been affected by any of this—the examination gave an accurate indication of the skills, knowledge and understanding of the content, so we deemed the grades to be reliable.
However, I was concerned about the anxiety that had been caused to students and the commentary I had read online, so I asked my team to look hard at this and try to make a decision about whether Pearson had in any sense been non-compliant. We have not quite finished that piece of work but when we do it will of course be published in the normal way.
Q30 Darren Paffey: The exam boards that you regulate obviously exist in the context of the real world and its threats. One of those threats is the increasing number of cyber attacks on public sector bodies. The National Audit Office has deemed this to be a threat which is severe and advancing quickly, so I just wonder how confident you are that the cyber security arrangements across all the bodies that you regulate are sufficiently robust?
Sir Ian Bauckham: Thank you for that question. With your permission, Chair, I am going to invite Michael Hanton to take that.
Michael Hanton: Thank you. The first thing is to recognise the point you have made: that 2025 demonstrated that no industry or sector is immune to cyber threats; unfortunately that is absolutely true for the exam industry as well. In light of that, we are focused on the threat of cyber issues and attacks on the exam industry.
We are very grateful for the work and the support we have had from the National Cyber Security Centre; we have worked closely with it and with the exam boards too. The exam boards that are offering high stakes qualifications have ISO 27001, which is the international standard for cyber security. That is an important step but we continue to look at all aspects of how exams are delivered, including looking at third party suppliers and the different interdependencies of what is quite a complex system.
In that context it is also worth noting that there are 6,000 schools, colleges and training providers delivering exams each summer; so many other organisations are involved in secure delivery. In light of that, for the past two years we have run our cyber awareness campaign, which is a month long campaign targeted particularly at schools and colleges. We are making sure they have access to the resources that are specifically for them, to help them think about their own individual cyber resilience preparedness and encourage them to check over their arrangements. From our point of view it is an end to end endeavour; this is a topic that we will continue to work on and focus on.
Q31 Darren Paffey: You have mentioned the cyber awareness campaign. Are there any other things that you are doing to your processes—updating, changing, or transforming them—so that you can be sure you are maintaining the confidentiality and the security of exam papers and that you, as well as those you regulate, have a system that is fit for purpose? What is changing to achieve that?
Michael Hanton: One of the threats that arises in a cyber context is the confidentiality of assessment materials, and that is obviously very important. There are a range of controls in place across our regulatory approach, but essentially the heart of it is to ensure that the fewest number of people possible are aware of the content of assessment materials before they are sat.
There are organisations that need to be involved in that process; part of the work we are doing at the moment is to ensure that they have the right security arrangements in place: asking exam boards to look at their third party suppliers, and essentially looking across that broader system of the different stages of exam delivery. That goes, again, all the way through to schools and colleges; there are clear regulations around the storage of exam papers once they come into schools and colleges.
There are requirements on things such as the opening of packages, which must be done with two people present, and ensuring that the right papers are opened at the right time so that there are no vulnerabilities right up to the last point before students are taking materials. It is an issue we take very seriously, again, looking end-to-end.
Q32 Darren Paffey: Presumably this is part of your deliberations over online, on-screen assessments as well, as to how—with even greater vulnerabilities in such cases—you might replicate the same monitoring of first sight within a digital system.
Sir Ian Bauckham: Yes, absolutely. As Michael has said, there are already vulnerabilities in the system and as part of the process of accrediting any on-screen proposals we would want to make sure that there were no material additions to such vulnerabilities.
Of course, on-screen assessments do not have to mean online assessments; the two things are different. It is possible to have a pre-loaded assessment that is not accessed via the internet. But yes, securing the security of assessment materials is absolutely fundamental to our objective of maintaining public confidence and trust in what is a highly trusted qualification system already. We are not going to take any risks in that space.
Q33 Jess Asato: You have updated your compliance policies for awarding organisations this year to introduce a new Chief Regulators Rebuke, as well as new principles for awarding bodies to follow. What prompted you to introduce these measures?
Sir Ian Bauckham: That is an interesting question and one that has produced some debate, as you might imagine. We felt that from time to time it was appropriate for us to review our policy on compliance, sanctions and enforcement. As I said earlier, we have a range of powers, and we must be clear with awarding bodies on how those powers are used and who the decision-makers are. We have an enforcement committee that hears contested cases and makes a decision on any settlement cases where a settlement is proposed. I do not personally do that—a group of appointed board members makes those decisions—but we just wanted to clarify the routes to enforcement, the appeals mechanism, and so on.
The additional sanction that we introduced was, as you say, a Chief Regulators Rebuke. We felt that there was space for a sanction where things such as a special condition were not appropriate, where it did not meet the bar for a fine, or where the process of getting to a fine would involve a disproportionate and unjustifiable use of public resources, but where public signalling of non-compliance was none the less an important thing to do.
As the name suggests, introducing a rebuke allows us to make a formal statement rebuking an awarding organisation for doing something that was non-compliant and potentially not in the interests of students taking the qualifications, and making that public. It is an addition to our regulatory enforcement toolkit.
Q34 Jess Asato: Have you issued any rebukes since the tool was introduced?
Sir Ian Bauckham: Not yet.
Q35 Jess Asato: So you do not know whether it actually does anything? It sounds a little like telling someone off; they might not care.
Sir Ian Bauckham: Interestingly, we consulted on it and we got very sharp feedback from awarding organisations that they would find it a very difficult thing to receive indeed. It would give chief executives cause to have difficult conversations with their boards, and in the case of awarding organisations that are commercial entities with shares, it could have an impact on share price. So they did not like it very much, which gave me the assurance that it might be effective.
Q36 Jess Asato: We will wait to see you do the rebuke.
I have one final question. The new principles state that awarding bodies must act with honesty and integrity. Do you have any reason to believe that is not already the case?
Sir Ian Bauckham: You are alluding to our principles condition. We have published a series of conditions under the principles condition, which basically encapsulates the key qualities that we expect from exam boards in the way they work, including honesty, integrity, transparency and co-operation with the regulator.
As we have already discussed, there are breaches, and where there are breaches it could be that exam boards may not have acted with the level of integrity that we expect. But what the principles were really intended to do was to extract from the general conditions of regulation the high level principles that effectively inform the conditions that follow.
There is nothing new in principles such as honesty and integrity; they are already effectively implicit in all the conditions. But one of the reasons we felt it would be useful to extract and headline those principles is to support awarding organisations in dealing with novel situations, which may not quite fit the exact wording of any of the conditions now. They could go up a level and refer to those principles and think, “Is this course of action at least in line with the principles, even if we cannot quite make it match the other conditions?”
Secondly, one of the things we have been doing over the last couple of years is supporting governance in awarding organisations. Governance bodies in awarding organisations are often not familiar with the granularity of the conditions, but we felt it would support governors in holding their own awarding organisations to account if we were able, in relatively straightforward language, to extract high level principles and equip those in governance roles to provide a challenge to their executives.
Q37 Sureena Brackenridge: In your opening address, you referred to inaccurate data on access arrangements that will lead to legitimate public questions; let us consider that in a little more detail.
It has been claimed that Ofqual knew about the problem with the statistics on access arrangements for at least two years before any action was taken. Why did it take so long for investigations to begin?
Sir Ian Bauckham: What Ofqual has historically published, Chair, on access arrangements is the data that is provided to Ofqual by exam boards of access arrangements that are in force in the year in question. When an access arrangement is granted in a school or college, it is granted for a period of 26 months. The rationale for that is clear: if you get an access arrangement at the beginning of May in year 11, it will expire at the end of June in year 13, and for all sorts of workload reasons, you then have an access arrangement that covers, in a traditional setting, both GCSE and A level.
But what that means is that you have access arrangements that are technically in force, as I say, from the beginning of year 11 through to the end of year 13, including in year 12 when currently the vast majority of students do not take examinations. So while the data that we were publishing was published in good faith, with data received from exam boards on the access arrangements in force at the time that the publication related to, what I realised with a degree of clarity was that the data was not answering the basic question, “What proportion of students taking their GCSEs and A levels in 2025 did so with the support of an access arrangement?” The public legitimately want an answer to that question.
The follow-up questions are, “Is it rising, and if so, why is it rising? Is it the right proportion, and if so, why is it the right proportion?” You cannot address any of those questions unless you have the clearest possible answer to the original question, “What proportion of students in this year had access arrangements?” We were not answering that question, and potentially we risked misinforming the public debate.
So we took the decision to withdraw those statistics, not because they misrepresented what exam boards had told us—they did not do that—but they were not accurately answering the questions the public had. The work we have been doing since then is to try to ascertain exactly what proportion of students have access arrangements in a particular exam year in question.
Let us take 2025 as an example; now that might sound like an easy thing to do but unfortunately it is not. Access arrangements are historically done by exam boards working together, through a system that was originally designed some years ago by the Department for Education. It is a different system to the exam entry system. The exam entry system in schools and colleges normally happens through the school's own management information system, so it is very clear and accurate; it is benchmarked against the actual names of students on roll.
But then you have this access arrangement system that happens separately, and you get mismatches in the data between the access arrangements database, the management information system and exam entries, which may involve different spellings of names, nicknames being used, erroneous dates of birth and so on. Reconciling the two is actually quite difficult and laborious and we cannot always do it with accuracy, but I asked my team at Ofqual to do a significant piece of very granular work to try to get the data from the exam boards and work out what proportion actually had access arrangements in the exam series 2025.
We have done that to a certain extent, but unfortunately we do not have a precise figure. We have a range, and the reason why we have a range is because of these data inconsistencies between, on the one hand, the accurate exam entry database and on the other, the access arrangements database. In the light of that, we are doing two things; we are continuing to try to get more accurate data, and we have also told exam boards that we expect them to reconsider the systems that they use for access arrangements in order to ensure those are better reconciled with the entry database so that we have more accurate data and we can speak to the legitimate question that the public has on this.
Now the new figures that we published for 2025 show that we still have significant access arrangements in the system. They also show that access arrangements are rising year on year; modestly, but rising none the less. They are significantly lower, as you would expect, than the historic figures because of the problems I described associated with the older figures.
Importantly—this is also a topic for public debate as well—the figures are still higher in independent schools than they are in state-funded schools, although I would add as a rider that identifications of special educational needs and disabilities are also higher in independent schools than in state-funded schools.
I am sorry, that is a very long answer to the question but I hope it made sense.
Q38 Sureena Brackenridge: I will pick apart a few things, starting with the independent schools point that you raised. The statistics are still proportionally much higher in independent schools; let us take for example the numbers of children who have access arrangements for 25% extra time. Nearly a third of children in independent schools have those arrangements compared to less than 25% in state schools. The Independent Schools Council has called for an apology because of the inaccurate data on access arrangements. Have you apologised and if not, are you prepared to do so?
Sir Ian Bauckham: No, I have not apologised. What I have done is to explain clearly why we published the data that we originally published and historically had been publishing. It was data received in good faith from exam boards about the total quantum of access arrangements in place, and that is always what it was. When we drill down into it and try to answer that legitimate public question we arrive at a lower figure, but there is still a differential.
Q39 Sureena Brackenridge: Yes, it is still significantly higher. In terms of the assessments for this year, how confident are you that exam boards are going to meet the number of actions that Ofqual has set out, and how will you monitor their compliance?
Sir Ian Bauckham: It is going to take a little while for exam boards to redesign the access arrangements system. That will not be possible to achieve for summer 2026. For all sorts of reasons, schools often apply for access arrangements either during year 10 or earlier on in year 11. If you are in 11-16 years old schools—about half of our secondary schools are 11-16 years old schools—there is no particular incentive for you to leave the application until year 11 because what the student does after year 11 is not your business. Many of those schools will choose to apply for access arrangements in year 10, so those access arrangements will already be in place. Effecting visible change in this area is going to take a little while; it will not be in summer 2026.
We have an expectation that exam boards will comply with equalities law, and many of these access arrangements—those that are under the category of reasonable adjustments—are provided under exam boards' direct responsibilities under equalities law. JCQ, which is effectively the trade body for the exam boards, administers the system and has reviewed the way in which access arrangements are granted; it has not found a specific incident of bad practice.
Exam boards are accountable, not just under our regulation but under equalities law, for granting access arrangements in cases where evidence is provided that points to the need for an access arrangement. Where that happens in greater numbers in independent schools than it does in state schools, I do not know what that points to; as I have said already, there are higher proportions of identified SEND in independent schools than state schools. It may be partly a reflection of that, or it may be evidence of other kinds of needs. I do not know the answer, but exam boards are under an obligation under equalities law to do this fairly, and would ultimately be answerable under equalities legislation if they did not.
Q40 Mark Sewards: During the pandemic we obviously saw widespread grade inflation, as teachers awarded grades. I was one of those teachers; I was trying to make sure I did it in line with expectations.
We know that the inquiry into covid-19 has taken evidence from a range of witnesses about these arrangements into exams. Once that module of the inquiry is reported back, what steps will Ofqual take to engage with those findings, and what lessons do you think you will learn so that you can be informed in response to any future pandemic?
Sir Ian Bauckham: Regarding the question about the covid pandemic and the inquiry, Module 8—the schools and children module—has now finished hearing but has not yet reported. It is clearly a statutory inquiry and I have given written evidence that is in the public domain and is in considerable detail. Once the module results and the findings are published we will, of course, engage with them seriously and take any steps that we have not already taken. But this provides an opportunity for me to say briefly that, as per my evidence, there are two principal areas of learning that we take from the pandemic experience.
The first is that an attempt to award qualifications and grade them in a way that is separate or divorced from the actual performance level of the work of individual students is unlikely to command public confidence. That is effectively what happened in 2020. The arrangements that were originally put in place in 2020 were done largely under direction from the Secretary of State for Education at the time, but they overly separated the actual performance of individual students from the grades they got. That did not hold public confidence, and a principal learning I take from that experience is that any such approach in the future is equally unlikely to hold public confidence.
Q41 Mark Sewards: Will they have to do exams in a future pandemic?
Sir Ian Bauckham: At the beginning of the pandemic our advice to the Government was not to cancel exams. In the event they were still cancelled. Our preference will always be that students should take exams if it is at all humanly possible to do so because that provides the only robust, fair way to award qualifications based on student performance.
If it is impossible to take exams—this moves on to the second area of learning that I wanted to talk about—there needs to have been pre-scenario planned ways of dealing with qualification awards in the absence of exams. They are likely to be imperfect; as you have already alluded to, we saw grade inflation in 2021. That was an imperfect system but it did at least hold the link between student work and the grade received.
There are two broad areas where we have been active in this. First, is to impose a resilience condition, C2.6, in our statutory framework. That requires exam boards to ensure that schools abide by our advice and retain evidence of student work within their examination course that we can have recourse to should we need evidence for awarding qualifications and holding that link between grade and work.
Secondly, as I alluded to earlier, is detailed scenario planning for the cancelling of exams at a range of different points in the school year and in a range of different ways, with an eye to whether there is a likelihood that exams might be able to restart or happen in some shape or form. No actual scenario is going to directly match any planned scenario, but at least having a range of planned scenarios enables us to look at what possible recourses there are in the event that exams are cancelled.
I need to underline, Chair, that without exams, the awarding of qualifications will be an imperfect exercise. But as I say, holding that link between individual student performance and grade is something that we have learned from the pandemic. It is also helpful for me to underline—which I did in writing in my evidence—that we are sorry for the anxiety and stress that was caused to students in the summer of 2020. I do not want to make any bones about that, but the background and how we got to that point is documented in detail in the written evidence. As I say, once Module 8 reports we will look at the findings with interest, and if there is any further action for us to take we will of course take it.
Q42 Mark Sewards: If somebody was interested, where might they find these detailed scenario planning documents?
Sir Ian Bauckham: We have not published those yet; it is work that we do internally.
Q43 Caroline Voaden: I would like to ask you about the new British Sign Language GCSE, which was set to be rolled out in September last year. Ofqual did not start consulting on the assessment regulations until April and there has been huge disappointment in the deaf community about the length of time it is taking to bring this GCSE into existence. What is the reason for the delay?
Sir Ian Bauckham: By way of response, Chair, I will just give a headline if I may, and then invite Michael Hanton to give some more detail in answer to that important question.
We are absolutely committed to playing our part in bringing the British Sign Language qualification to schools, ensuring that we have done all the policy work that we need to do to ensure that the subject content that the DFE has developed can be developed into a specification by exam boards.
Michael, can you give a bit more detail on timelines, please?
Michael Hanton: Perhaps the place to start is to say that in November 2025 we confirmed all our regulatory requirements, so in essence everything is now in place for exam boards to develop a GCSE in British Sign Language. The point I would reflect on is that it is obviously vital that a GCSE in British Sign Language stands up alongside other GCSEs and that it is a rigorous, high-quality qualification. Alongside that, we were very committed to working closely with the deaf community and making sure that we were drawing on expertise in terms of how to develop this qualification.
There are a couple of interesting, quite unique features about BSL as a language; there is no written form, for example, and there are regional variants. So there are particular challenges in terms of designing the qualification and ensuring that it can operate effectively.
It did take some time to work with the deaf community to make sure we had a qualification that reflected the genuine needs of the deaf community, including looking at content around the culture of BSL as a language and its place in the deaf community. It took some time, but I am pleased that we were able to complete it and everything is now in place ready for the next stage.
Q44 Caroline Voaden: There was an article in Schools Week that said that AQA have said it does not want to offer it, and there are three other exam boards that have not said yet whether they will. So there are concerns that there may be no exam board that wants to offer it. What steps can you take if that is the case?
Michael Hanton: As you say, four exam boards are currently recognised to offer GCSEs in England; they are making their own individual decisions about what qualifications they offer and obviously there are new qualifications as we have talked about this morning from the Curriculum and Assessment Review. We do not have the power to force an exam board to offer a particular qualification, that is not something that we—
Q45 Caroline Voaden: Can you rebuke them if they do not do it.
Michael Hanton: Gosh!
Sir Ian Bauckham: We can only rebuke them, Chair, for regulatory non-compliance, and unfortunately it would not be regulatory non-compliance, it would be effectively a commercial and market decision.
Michael Hanton: I would just say—I know there has been some commentary on this in the press as well as among the deaf community—that there is the opportunity for an awarding organisation to become recognised to offer a GCSE in this particular area. That is a route that is open and we would welcome conversations about that.
Q46 Caroline Voaden: Could it be an organisation that just does BSL GCSE and nothing else?
Michael Hanton: That is right.
Sir Ian Bauckham: There is no legal obstacle to a new awarding organisation that is not currently recognised to deliver GCSEs coming forward and asking to be recognised to deliver a GCSE specifically for British Sign Language. That could happen.
Q47 Caroline Voaden: Is it quite a lengthy process for them to get recognised?
Sir Ian Bauckham: This is a “piece of string” answer: it depends how strong they are and how ready they are for recognition. As Michael has alluded to, GCSEs are high prestige qualifications and we would want to make sure that British Sign Language GCSE stood muster against all other GCSEs, that it was of the same quality and the same standard. There would be a bar that the awarding organisation would have to meet, but it is not impossible that it could be met.
Q48 Jess Asato: In 2022, our country offered a welcome to people fleeing the war in Ukraine, and since then 27,000 children have been displaced, together with 7,000 from the wider diaspora. A huge number have been identified as wanting to take Ukrainian GCSE, but instead many parents feel that their children are being forced to take Russian instead, which is clearly a very unsatisfactory situation. Why has the progress on achieving a Ukrainian GCSE been so slow, despite so many calling for one, including the Children's Commissioner and the Education Ministers last year in June 2025?
Sir Ian Bauckham: This is essentially a parallel answer to the one about BSL: we do not have the power to make an exam board offer a GCSE or an A-level or any other qualification. Once the subject content is developed, which is the DFE's responsibility, an exam board needs to step forward and develop a specification. I know that Ministers have met with the exam boards and encouraged them to consider developing a specific Ukrainian GCSE, but as yet no exam board has picked that challenge up and run with it. That is essentially the reason why there is no regulated GCSE in Ukrainian.
Q49 Chair: Thank you very much. Is there anything else that you would like to bring to the attention of the Committee this morning?
Sir Ian Bauckham: I do not think so. It has been a wide-ranging and, from our point of view, helpful discussion. Thank you for your questions. I hope our answers have provided clarity and assurance for the Members who have posed them.
Chair: Thank you very much for coming to give your evidence to us today. We look forward to seeing you in a future session. That brings our session for today to a close.