The 2026 National Reference Test (NRT) results are out today. They show a long-term trend of slightly improved attainment in GCSE maths, but not in English language. So, to ensure that standards are maintained, we told exam boards to take the results of the NRT into account when setting some grade boundaries in GCSE maths. ## What is the National Reference Test? Every February and March, a sample of year 11 students at more than 330 schools in England sit a short test featuring similar-style questions to those set in GCSE English language and maths exams. The tests have been running since 2017, when the current GCSE specifications were first examined. That year serves as the benchmark for making comparisons over time. As it is a consistent test sat each year, it is a highly reliable measure of whether overall attainment in these subjects is going up, down, or staying the same. Where there are genuine changes in the level of attainment over time, the NRT enables us to see it. This extra evidence helps us to maintain standards and ensure that outcomes in GCSE English language and maths reflect any trends in student attainment. ## What has the NRT shown and what has Ofqual done? ### Maths The NRT has shown a small rise in attainment in maths at grade 7 in recent years. The trend is now clear and sustained against the 2017 benchmark. It is highly likely that this improvement is also present across the higher grade range. That change is a genuine improvement in what students know, understand and can do. On this basis we instructed exam boards to reflect this evidence of improvement in their setting of grade boundaries, in the upper grade range. It means those grade boundaries are lower than they otherwise would have been. ### English language In English language, it’s a different picture. Results at grade 7 are not statistically significantly different to 2017. At grade 4, there is an indication that attainment has fallen since 2017. However there has not yet been a consistent enough trend over time to justify any action this year, given the cautious way in which we use evidence from the NRT. As usual we will be looking closely at this again next year, in light of the 2027 NRT results. ## What does this mean for students and schools? The impact on today’s GCSE maths results is a small increase in the number of students achieving higher grades. At grade 4 results are stable as attainment is stable. Results in English language are stable, with any variation being normal fluctuations we see each year in the proportions of students achieving each grade. The main takeaway is that the NRT has enabled us to ensure that standards are maintained and rising attainment in GCSE maths has been reflected in students’ results. You can find more detail about the NRT results in our 2026 NRT annual statement. And for all the data, there’s the National Foundation for Educational Research’s full report. **Michelle Meadows, Chief Standards Adviser, Ofqual** * * *

More than 880,000 A and AS Level results have been issued today to students in England, with another 221,000 results for vocational and technical qualifications. These results open doors for students to move on to the next stage in their lives, whether that be university, employment or apprenticeships. And they serve as recognition of – and reward for – the effort and quality of work displayed by those students. I offer my congratulations to everyone receiving results today. ## Here are some key details about today’s A Level and level 3 results * More than 1.1 million results have been issued across all level 3 qualifications. * A Level outcomes are stable, with small changes only, largely driven by subject entry patterns. * We have seen an increase in students taking T Levels as these qualifications, still relatively new, become more established and grow to cover more subjects. ## What are the trends behind the results? This year’s results picture is one of stability. Small changes in grade distributions year-on-year are normal. Looking across A Level subjects with large enough entries to compare meaningfully, most show only small movements up or down. Some rising, some falling, typically by less than a percentage point either way. This kind of variation happens every year and largely reflects the different mix of students taking each subject. The standard of work required to gain a particular grade remains constant. There are noteworthy patterns and trends in the popularity of subjects at A Level. Maths was the subject studied by more students than any other, accounting for about one in 8 of the A Level results released today. Maths A Level has increased in size as a proportion of entries in recent years, and that trend has continued this year. Further maths A Level too has again seen strong growth this year. After maths, the most commonly-taken A Level subjects were psychology and biology – the same as in 2025 – and chemistry and economics complete this year’s top 5. The popularity of maths at A Level means outcomes have a significant effect on national trends. The very small rise in the proportion of students achieving A* or A at A Level overall (from 28.2% in 2025 to 28.3% this year) has been largely driven by a small increase in those grades in maths (from 41.6% in 2025 to 42.1% this year). The standard of work required to achieve any grade in maths this year was the same as previous years, so this rise reflects a slightly stronger cohort of students. While subject level changes such as this are important, the overall national picture for A Level results in 2026 is one of stability. ## What do I need to know about AS and A Level results? * 835,950 A Levels have been awarded to 320,505 students, from 2,835 different schools and colleges, in 57 different subjects. * This is an increase of 2.7% on 2025; the population of 18-year-olds has increased by 4.3% on 2025, according to ONS projections. * About two-thirds of A Level students took 3 subjects. * 7,245 18-year-olds achieved an A* grade in every subject they took – and 25 of those took 5 or more A Levels. * The regions with the highest percentage of A Level results at A or A* were London and the South East. * 48,835 AS Level results have been issued, across 35 subjects. ## What do I need to know about level 3 vocational and technical qualification results? * 19,665 results for T Level have been awarded, up from 11,980 last year. * 18.3% of students taking the occupational specialism in the second year of their T Level course achieved a distinction. * 221,095 VTQ results have been awarded overall, with 195,190 of those for applied generals. * The most widely taken subject sector areas were business, administration and law; social sciences; and leisure, travel and tourism. ## Where to find more details about A Level and level 3 results Ofqual has published a statistical report on this year’s results. This is the place to go if you want more in-depth analysis of the results and outcomes for A Levels, AS, applied generals, tech levels and technical qualifications within T Levels. We have also published interactive visualisations of level 3 results on Ofqual Analytics. This gives detailed breakdowns of outcomes according to types of school or college, by sex, for subject combinations, and more. This year is the first time you will be able to see results and subject uptake broken down geographically at local authority level. Results day each August represents the culmination of hard work and commitment by students, teachers and many others. Congratulations to all students collecting their results this week. **Amanda Swann, Executive Director for Delivery, Ofqual**

When we think about qualifications – GCSEs, A Levels, T Levels, and the many others offered in England – it’s tempting to focus on what they contain: the subjects, the assessments, the grades. But before any of that, we need to ask a question. Do we really understand what the qualification is for? New research from Ofqual, titled Designing qualifications that will be fit for their purposes, sets out a detailed framework for what we call 'purpose analysis'. This is the idea that getting clarity on a qualification's purposes should be the essential first step in designing it. Skipping or short-changing this step is one of the main reasons for failure. ## More complicated than it looks Designing a qualification might seem relatively straightforward: decide what it's for, work out what it should assess, and build it accordingly. In practice, it's far more complex. Qualifications are asked to serve many different purposes simultaneously, and those purposes can conflict with each other. The report identifies 3 distinct perspectives that shape what a qualification needs to do. The **expertise** perspective focuses on the body of knowledge, skills and understanding that learners should acquire – the learning itself. The **engagement** perspective considers how the qualification motivates learners and their teachers, and whether it promotes effective teaching and learning. The **information** perspective looks at what qualification results tell us, and how accurately they support decisions about individuals, institutions, and the wider education system. All 3 perspectives matter, but they often pull in different directions. A qualification designed primarily to motivate disengaged post-16 learners, for instance, may look very different from one designed largely to provide reliable signals for university admissions. When these tensions aren't confronted head-on – if designers try to serve all purposes equally without making explicit choices – the result can be a qualification that serves none of them well. ## Learning from the past The report draws on a rich seam of history to illustrate what goes wrong when purpose analysis is neglected. The General National Vocational Qualification (GNVQ), for example, was expected simultaneously to motivate disaffected learners, expand routes into higher education, prepare students for work, achieve parity of esteem with A Levels, and more. This was a range of demands so broad and diverse, the aims were never likely to be achieved satisfactorily. This is not the only example of a qualification designed with insufficient attention to its driving purposes, subsequently failing. ## Four questions every designer should ask So, what does good purpose analysis look like? Our report suggests designers need to work through 4 key considerations. * First, who is the qualification for – what are the characteristics and circumstances of the learners it will serve? * Second, what purposes should it serve, and from which perspectives? * Third, how does it fit within the wider qualification landscape, and does it maintain the structural, cognitive, and ethical integrity of the system as a whole? * Fourth, what practical constraints – financial, human, physical, political – will shape what's achievable? Crucially, this analysis should result in an explicit, prioritised profile of purposes. Not a list in which everything is equally important, but a clear-eyed account of what the qualification most needs to achieve, and what trade-offs are acceptable. There is no such thing as a perfect qualification. ## Why this matters now This research is relevant right now. England's qualification system is undergoing significant reform, with changes affecting qualifications at a range of levels. The frameworks explored in this report, developed at Ofqual during the past decade and used in the drafting of new regulations, should help ensure the ongoing reform efforts are built on solid foundations. The report is deliberately theoretical in places, offering conceptual tools rather than simple answers. But its practical message is clear: qualification design is always a matter of compromise, and the best way to navigate that is to be explicit about purposes from the very start. **By Paul Newton, Research Chair, Ofqual**

Every year, millions of students across England take regulated qualifications – GCSEs, A Levels, T Levels, assessments for apprenticeships and many more. Behind each qualification are rules that exist for one fundamental purpose: to make sure every student's result is valid and trustworthy. Qualifications only have value if people can trust them; that trust depends on awarding organisations following the rules. When they do not, we act. Today, we have issued the first Chief Regulator's Rebuke – a new enforcement tool I introduced last October as part of a significant update to Ofqual's Supporting compliance and taking regulatory action policy. The rebuke has been issued to WJEC-CBAC Ltd (WJEC), following serious failings involving 4 of its Eduqas GCSE, AS and A level qualifications between 2019 and 2025. I want to take this opportunity to explain the rebuke – a public censure – and why it matters for students. ## Why we updated our approach Good regulation should be proportionate, flexible and effective. Our updated policy reflects our experience of using our enforcement powers over years, and our ambition to be a modern, agile regulator. We want awarding organisations to get things right in the first place. The vast majority of the time, they do. Where concerns arise, we expect them to resolve issues early. But when things go seriously wrong, we need the right tools to respond appropriately. ## What the Chief Regulator's Rebuke is for Our formal enforcement options include financial penalties, special conditions and, in serious cases, withdrawal of recognition. These remain important tools. But there was a gap: cases serious enough to warrant a public outcome, but where the circumstances did not merit a fine. The Chief Regulator's Rebuke fills that gap. It is a non-statutory instrument that allows me to hold an awarding organisation publicly to account for serious failings, in a transparent and proportionate way. This non-statutory option reflects our commitment to working collaboratively with awarding organisations – something they told us mattered to them during our consultation process. ## The WJEC case WJEC admitted failing to collect and monitor the declaration forms required to demonstrate that schools and colleges had complied with subject content requirements for its Eduqas GCSE drama, GCE drama and theatre, GCSE geography and GCSE computer science qualifications. These are not trivial requirements – they exist to protect the integrity of qualifications and give students, schools and employers confidence in the results. There is no evidence of direct adverse effects on students, and I am grateful that WJEC co-operated fully and openly and accepted the rebuke. But the failings were serious, spanning multiple subjects and years. They had the potential to undermine public confidence in regulated qualifications. That is why a public outcome was right. ## **Protecting students: the bigger picture** Whether we are issuing a rebuke, imposing a fine or working informally with an awarding organisation to resolve a concern, everything we do is ultimately to serve students, with qualifications that people can value and trust. **Sir Ian Bauckham CBE, Chief Regulator**

Our new Apprenticeship Assessment Regulatory Framework sets out the final conditions, requirements and guidance for reformed apprenticeship assessment. These reforms are designed to accommodate the Department for Work and Pensions’ new principles of flexibility in the way that apprenticeships are assessed. This framework provides that flexibility while ensuring that standards for assessment will be upheld. Together, our regulatory framework and Skills England’s new style assessment plans will form a foundation for more flexible approaches to assessing apprentices, while ensuring that apprenticeships remain a trusted and valued route to the skills and knowledge needed for the job. ## What this means for awarding organisations offering apprenticeship assessments Please familiarise yourselves with the new conditions, requirements and guidance. The new framework embeds more flexible assessment principles from the Department of Work and Pensions. This allows greater flexibility in how you design and deliver assessments, but with that comes greater responsibility. Specifically, you must: * produce and maintain an assessment strategy for each apprenticeship assessment you offer, bringing these assessments in line with other Ofqual-regulated qualifications * ensure your assessments reflect the relevant occupational standard and assessment plan published by Skills England Ofqual’s new regulatory framework will apply to new apprenticeship assessments when updated assessment plans come into effect. ## What this means for colleges or training providers Your primary relationship for apprenticeship assessment remains with your awarding organisations. You should work with your awarding organisations to understand the apprenticeship assessment arrangements and read the specifications your awarding organisations will publish for each apprenticeship assessment. Under the new regulatory framework, providers may have a greater role in delivering and marking elements of assessment. This will be decided and overseen by the awarding organisation, not by Ofqual directly. If you have questions about your specific responsibilities, please contact your awarding organisations. ## What this means if you are an employer Employers play an important role in confirming that apprentices are ready for assessment and that assessments reflect the needs of your industry. You do not need to take any immediate action but you should be aware that, as updated assessment plans come into effect, the way assessments are structured may change. These reforms introduce greater flexibility in assessment design and reduce duplication but our framework will ensure that apprenticeship assessment remains a trusted measure of occupational competence. ## What do apprentices need to do? New apprenticeship assessment arrangements are being rolled out gradually. This might mean that you are assessed during your apprenticeship rather than at the end. Your training provider will confirm your assessment arrangements. You do not need to take any action unless your training provider or employer tells you otherwise. These reforms are designed to offer flexibility in how apprenticeships are assessed, while ensuring that standards are upheld and that they remain valuable for employers and employees. * * * **Catherine Large OBE, Executive Director of Policy, Ofqual**

Following the Curriculum and Assessment Review, the government is introducing 3 new types of qualification for post-16 learners: V Levels, Foundation Certificates and Occupational Certificates. These qualifications will join A levels and T Levels to enable young people to take different pathways toward employment and further and higher education, using high-quality qualifications. Ofqual has launched a consultation on the criteria that awarding organisations must meet to be recognised to deliver these new qualifications. This blog explains what we're proposing and why. ## Why new recognition criteria? Delivering national qualifications is a demanding, complex endeavour. It requires mature systems and processes, sophisticated expertise, and the capability to deliver at scale. Awarding organisations must have policies and procedures that secure quality and fairness, as well as the capability to manage risk and ensure resilience – whether that's responding to cyber security threats or managing the logistical demands of running national assessments. Ofqual's regulatory experience has shown the importance of these requirements. We've seen what happens when things go wrong, and we know the impact on students. Our approach is designed to protect learners taking these new qualifications. Our proposed new criteria set the bar high for awarding organisations, because students deserve qualifications they can trust. Only organisations with proven capacity, capability and governance will be able to offer these national qualifications. ## A phased approach This consultation concerns the first tranche of the new vocational and technical qualifications, covering a small number of subjects for first teach in September 2027. The timetable is tight, and awarding organisations must be ready to deliver from day one. We recognise that our proposals may mean relatively few organisations would secure recognition for the first tranche. Awarding organisations that do not yet have the necessary capabilities will have limited time to develop them before 2027. However, all awarding organisations will have the opportunity during the first phase to understand our requirements and to develop the expertise, systems, processes and governance they will need to apply for recognition in later tranches. We will review our approach before the second tranche (for qualifications with a first teach date in 2028) and will consult again. This gives the sector time to prepare and ensures our approach remains proportionate as the qualifications landscape develops. ## What are we proposing? The proposed criteria focus on 4 areas: * * * * **Prior recognition:** Applicants must already be recognised by Ofqual, so we can have confidence their experience was developed within our regulatory framework. * **Expertise:** Awarding organisations must demonstrate proven capability in delivering qualifications at the required scale and complexity. * **Systems and processes:** Organisations must have reliable, established systems in place at the time of application – not just plans to develop them. * **Governance:** Organisations must show they have arrangements in place to make effective and consistent decisions throughout development, delivery and awarding. * * * ## What happens next? We want to hear from awarding organisations and anyone with an interest in these new qualifications. Your views will help us ensure our criteria are effective and proportionate. We believe that these proposals will enable the awarding organisations we regulate to successfully introduce a high-quality suite of new vocational and technical qualifications that will give post-16 students a clear, credible and flexible route into higher education and employment. The consultation is open until 22 April. You can respond through our online consultation platform. * * * **Catherine Large OBE, Executive Director of Policy, Ofqual**

Artificial intelligence (AI) tools are now part of daily life for most young people. Students use them to brainstorm, research, summarise and write. However, AI should not be used to generate coursework without proper disclosure. This is cheating, and the consequences, which can include losing a qualification entirely, are serious. Despite this, some students may be about to submit their coursework without ever having a direct conversation with their teacher about what is or isn't acceptable. This blog explains why that conversation is important, and the resources we’ve developed to help you have it. But beyond the rules, there's a more important reason to address this with your students: when AI does the work, they miss out on the learning. ## Real learning builds real knowledge and skills The knowledge and skills students develop through completing their own coursework, finding their own material, working through a problem and thinking independently will serve them well in higher education and employment. In an AI enabled workplace, being knowledgeable, and skillful enough to apply that knowledge critically to what the AI is doing, will add critical value. When a student submits AI-generated work as their own, they aren't finding a shortcut to success. They're bypassing the very process that builds the capabilities they'll need later. As a teacher, spotting and addressing this isn't just about enforcing rules. It's about protecting students’ long-term prospects. ## Starting the conversation Many teachers are already thinking about how to respond. Fifty-five per cent plan to ask students to explain their work if they have concerns, and 40% intend to speak to students about AI rules before submission. That's encouraging. If you're not yet sure how to approach this, you're not alone. You don't need to have all the answers before starting the conversation. Frame it around what students stand to gain from doing their own work, and what they stand to lose if they don't. Most students respond better to understanding the 'why' than simply being told the rules. ## Resources to help you To support you in having these conversations, we've developed 2 classroom resources: * * * * A lesson plan designed to help you or your colleagues lead a discussion with your students about AI and coursework. It walks through how to open the conversation and explain why it matters, and covers the real consequences of malpractice, from losing grades to affecting future applications. The plan includes suggested scripts, discussion questions and tips for handling grey areas, so you can hold the conversation with confidence. * A senior leadership team briefing pack to help school and college leaders develop a whole-school approach to AI in assessed work. Consistent messaging across your institution makes it easier for every teacher to hold the line, and ensures students hear a clear, unified message. * * * The pack supports leaders in equipping their staff and ensuring students understand the rules before it's too late. Both resources are available to download now. ## Your role makes a difference You're well placed to shape how your students approach this. A single honest conversation can make a lasting impression. Real learning builds genuine skills. That's worth protecting. * * * **Amanda Swann, Executive Director of Delivery, Ofqual**

Ofqual has published new research exploring artificial intelligence’s potential role in marking, as well as its current capabilities and constraints. Rather than providing definitive answers, the published working paper offers an overview of the topic, shares our current thinking, highlights key challenges, and aims to stimulate further debate across the sector. This blog offers a taster of the questions we’re grappling with as we consider the future of AI in marking. The working paper draws on a series of discussions with technology experts, awarding organisations and academics, who helped us to test and challenge our thinking about the potential applications and limitations of AI in marking. These conversations have informed our current perspective and highlighted the complex questions that still need to be addressed. ## **Ofqual's current position on using AI in marking** Our conclusion is clear: AI is promising for quality assurance and marker training, but for the moment it’s nowhere near ready to take over high stakes marking. Our position remains that the use of AI as the sole mechanism for awarding marks does not comply with our current regulations. Such use does not meet our requirements for a human based judgement to be used in marking decisions. In addition, the potential for bias, inaccuracies and a lack of transparency in how marks are awarded could introduce unfairness into the system. This would be unacceptable in the marking process. ## Student expectations and trust High stakes qualifications represent a social contract between students, educators and society. Students invest years of effort into their education with legitimate expectations about how their work will be assessed, and how their qualifications will be valued and accepted by others. Currently, those expectations include having their exams marked by trained human examiners who understand both the subject matter and the assessment criteria, and who apply their academic judgement fairly and consistently. Qualifications only hold value when students, parents, teachers and employers have confidence in their fairness, transparency and legitimacy. Any change to marking, particularly involving automation, must be carefully considered for its impact on this fundamental trust. ## **The challenges posed by AI in marking** Current AI technologies, especially large language models and deep neural networks, have advanced capabilities but lack true semantic understanding and the capacity for human-like judgement. These systems often function as ‘black boxes’, making it difficult even for experts to explain how specific marks are generated. This lack of transparency is a major challenge for maintaining trust in qualifications. We must be able to understand how marks are determined. If explanations and evidence for why certain marks were offered are unreliable, or not even available, the fundamental contract of assessment breaks down. It is a basic principle that underpins public trust and perceptions of fairness that students can challenge exam marking and secure a review. However, when marks are generated by AI, explanations and evidence for why certain marks were awarded may be unreliable, unavailable, or difficult to interpret, making it harder to uphold the transparency and accountability essential for trust in qualifications. There is much more work to do to understand and agree what this would mean where exams have been marked by AI rather than by a human marker. We explore these issues in more detail in our research paper. Additionally, AI systems can perpetuate or amplify biases present in their training data, and empirical studies have shown that performance can vary across different demographic groups. Ensuring that criteria are applied consistently and fairly, regardless of candidate or marker identity, is a core principle of valid assessment. Transparency and fairness are not just ethical ideals; they are essential for upholding the integrity and public confidence in qualifications. ## The case for using AI in marking There are promising applications for AI in marking, particularly in quality assurance of human marking and in training new markers. These uses could potentially improve marking consistency while keeping human expertise and judgement at the centre of the process. But the vital caveats we have outlined remain. In addition, the performance of AI in marking varies significantly across different contexts. For example, there are substantial differences in what is needed to mark a maths exam compared with an English exam. Current evidence does not support an overall or general case for using AI in marking. There needs to be context-specific evidence to make the case for using such tools. ## The future of AI in marking Our priorities are clear: ensuring fairness for students, protecting security, maintaining the validity of qualifications, and preserving public confidence, while enabling responsible innovation. As AI technology develops, we will continue gathering evidence and fostering constructive discussion about its role in assessment. The trust that everyone in education places in the fairness and transparency of qualifications remains at the heart of our regulatory approach. **Jo Handford, Associate Director Strategic Projects and Innovation, Ofqual** **Joanna Williamson, Research Fellow, Ofqual**

As teachers, you know better than anyone that GCSEs, AS and A levels are high-stakes qualifications that shape young people's futures. Every decision about how these exams are delivered must start with one question: is this fair for students? That's why we're consulting on a carefully balanced approach to on-screen assessment. One that protects the integrity of qualifications you and your students work so hard for, while enabling measured innovation where the evidence supports it. ## A balanced approach We're proposing a controlled approach that enables innovation while protecting what matters most: standards, fairness, and your ability to prepare students effectively. Here's what this means in practice: * Pen and paper remain central. We're not proposing that traditional exams disappear. Handwriting isn't going anywhere. Most GCSEs, AS and A levels will continue to be assessed the way they always have been. * Your school or college retains autonomy and choice. In the vast majority of cases schools and colleges have a choice of specifications across exam boards. You can choose the specification that best meets your needs, factoring in the mode of assessment to your choice. * Limited, manageable pace of change. Each exam board would be allowed to introduce up to 2 new on-screen specifications, making up a maximum of 3% of the total specifications, on top of the small number already available today. * Restrictions on high-volume subjects. The most widely taken subjects (those with over 100,000 entries nationally) won't be moving to on-screen assessment for the foreseeable future. * Clear guardrails. We recognise the importance of safeguarding the security, accessibility and fairness of new qualifications. Any new on-screen assessment must pass rigorous accreditation by Ofqual. ## Protecting fairness for all students We know you're concerned about fairness – and so are we. Our proposals include specific protections: * No student-owned devices. Students won't use their own laptops for exams. This prevents unfairness arising from differences in device quality or access. Schools will provide the devices, ensuring consistency. * Strong accessibility standards. Platforms must be easy to use for students. * Consistent standards. Standards will be maintained across specifications that are on paper and those on screen. ## Built on solid evidence These proposals are informed by extensive research, including: * studies on how students perform differently on paper versus on screen * analysis of the practical challenges schools and colleges face * international evidence on digital assessment * direct feedback from teachers, school leaders, and exams officers We've also been clear about the risks. On-screen assessment introduces new challenges around cyber-security, technical failures during exams, and maintaining standards across different modes. Our controlled approach ensures these risks are carefully managed. ## We're listening to your concerns Our research tells us that while there's interest in greater use of technology in assessment, teachers have legitimate concerns. You've told us about: * concerns about fairness for students who don't have equal access to technology * variable IT infrastructure in schools and colleges * differences in students' and staff digital skills and confidence * the operational demands of running on-screen exams alongside paper-based assessments * space requirements, including larger desks and suitable exam venues * uncertainty about whether certain subjects are suited to on-screen assessment We've listened. Our proposals reflect these realities. ## What won't change Your professional expertise remains at the heart of preparing students for qualifications. The knowledge and skills being assessed won't change. The rigour and value of qualifications won't be compromised. And traditional pen and paper assessment will remain the primary mode for most exams. ## We need your voice This consultation runs for 12 weeks, and we particularly want to hear from teachers and school leaders. You understand the realities of exam delivery. You know your students. You see what works and what doesn't. Your responses will help us ensure that any changes serve students' interests and are deliverable in schools and colleges. ## Looking ahead If these proposals are taken forward, detailed rules and guidance will follow in 2026. We'll keep the 2-specification limit under review, learning from how on-screen assessments work in practice. This is about evolution, not revolution. It's about protecting the qualifications system that serves your students well, while carefully enabling innovation where the evidence supports it. It’s worth remembering that this is optional.  In the vast majority of cases you can choose the specification that suits your preferences. You will be able to read the full consultation and share your views from 9:30am. Your expertise and experience are essential to getting this right. Together, we can ensure that any changes maintain the standards and fairness that make England's qualifications trusted and valued. **Jo Handford** **Associate Director Strategic Projects and Innovation, Ofqual**

Back in July we withdrew our official statistics on access arrangements for GCSEs, AS and A levels covering 2014 to 2024, after identifying significant issues with how data had been recorded and reported. Today we are pleased to be following that up by publishing revised statistics, alongside new research and a regulatory report that confirms what we expect from exam boards. The report sets out several specific areas where we believe improvements can be made. Here's a summary of: * what the new stats show * what our research found * what we expect from exam boards ## What the official statistics show As we explained in July, previous figures did not give an accurate picture of the number of students receiving access arrangements within a specific year’s exam cohort. This was not because arrangements were being granted inappropriately, but rather due to limitations of the data source we used for our reporting. These included counting students who held valid arrangements but didn't sit exams in the given year, and not being able to filter out duplicate applications. The new statistics – for the 2024 to 2025 academic year and going back to 2015 to 2016 – use an improved methodology that focuses only on arrangements granted to students who were actually assessed in each year. They show that 25% extra time was granted for 16.6% to 25.5% of all students being assessed in 2024 to 2025, compared with 14.7% to 20.6% in 2023 to 2024. This means the proportion of exam students with an approval for 25% extra time is lower than the previous data suggested, and broadly in line with the rate of special educational needs (SEN) in the student population (as reported by the Department for Education). We have given ranges rather than absolute numbers to reflect the complexity of the data. The upper end of the ranges will decrease when these figures are updated next year, as new data becomes available. You can read more about our improved methodology in the regulatory report. ## What our research found Like everyone else, we had seen for some time that demand for access arrangements had been rising, and with it pressure on exam boards and schools and colleges to keep on meeting that demand. We wanted to better understand this, and to identify anything that could be done to improve the system. So our research, data and policy experts went to work, analysing the available evidence and data. Our work focused on the granting of 25% extra time in exams, as this is the most common exam-board approved arrangement, as well as the concept of time in assessment. Our research found limited evidence of students experiencing time pressure (also known as “speededness”) in GCSE, AS and A level exams, although this varied across subjects and student groups. While these exams are not designed to assess students' ability to work under time pressure, the interaction between exam tasks, duration and individual students can create unintended speededness for some students. We heard a range of views about the current system, including concerns from schools and colleges about the increasing workload and costs involved in identifying appropriate arrangements, evidencing student need, and administering access arrangements. There was, however, support for the core principle that access arrangements should be based on a student’s individual needs. The system generally meets this purpose, but we have identified some aspects which could be improved. ## What we expect exam boards to do, and what we will do Ofqual already has rules in place for what exam boards need to do in relation to access arrangements. They must also comply with equalities law. But as well as this, and as a result of our findings, we have told the exam boards that we expect them to: * review and improve the access arrangements system, focusing on the clarity and accessibility of the information they provide – both to support schools and colleges, and to build a wider understanding about access arrangements, including among students and parents * ensure that their requirements allow students to receive appropriate access arrangements for their needs while balancing school and college workload * improve data about exam board-approved access arrangements * strengthen how they monitor that the system is functioning as it should We will monitor the exam boards’ progress in these areas as part of our normal regulatory approach. Also, as we respond to the government’s proposals following the Curriculum and Assessment Review, we will consider our requirements around assessment duration for GCSEs, AS and A levels. We will also take this into account in setting our requirements for the proposed new post-16 qualifications, such as V levels. We will confirm our regulatory expectations for exam boards following that work. ## An access arrangements system we can all be confident in The research and analysis we have done shows our commitment to ensuring the access arrangements system functions effectively and is reported on accurately. It has taken time, but that’s because it’s complex. It’s also really important. And the improvements we’ve asked exam boards to make will ensure students continue to receive the support they need to demonstrate their knowledge, skills and understanding, while minimising burden on the schools and colleges managing these vital arrangements.

Every year, over 11 million qualifications are awarded in England – from GCSEs and A levels to vocational qualifications in healthcare, construction, childcare and many other sectors. These qualifications are the foundation of opportunity. They recognise students’ achievements, help them progress, and give employers confidence in the people they're hiring. The system works well. But when things go wrong – when awarding organisations fall short of the standards we all expect of them – the consequences matter for the students taking qualifications and for everyone who relies on qualifications being fair and accurate. In some cases, failure to meet the high standards we expect of our qualifications can lead to stress for students and delays to their progression. We need a regulatory approach that is clear, proportionate and effective – one that enables awarding organisations to take swift action to deal with problems but also enables us to act decisively when our high standards are not met. Today, we're publishing an updated policy that does exactly that. ## An agile enforcement policy for regulating exams and assessment Ofqual has published an updated compliance policy for awarding organisations, now called Supporting Compliance and Taking Regulatory Action. Ofqual's previous policy was last updated in 2012. We received extensive feedback through consultations in 2019 and 2025, which directly shaped this revision. The updated policy reflects how we regulate today, with a focus on early engagement with awarding organisations, proportionality, and transparency. ## Key updates There are 3 main updates: 1. A new regulatory tool called the Chief Regulator's Rebuke The Chief Regulator’s Rebuke is a new regulatory tool that fills an important gap in our enforcement options. We would propose to issue a rebuke when we have determined that an awarding organisation is in breach of our rules, but the matter may not be serious enough to warrant a financial penalty. 2. A new appeals process for awarding organisations Awarding organisations will have a right of appeal if they disagree with a determination of non-compliance or the issuing of a Chief Regulator's Rebuke. Appeals are to be heard by Ofqual's Enforcement Panel, providing a clear and fair route to challenge regulatory decisions. This adds an important layer of procedural fairness to our regulatory approach. 3. Introducing a more efficient process when taking regulatory action We have introduced more flexibility in how we issue Notices of Intention that reflects regulatory practice in other sectors. This will reduce the burden on awarding organisations and allow for the process to move swiftly to resolution. ## Implications for awarding organisations This policy does not introduce any new requirements for awarding organisations. Instead, it emphasises our commitment to working flexibly by using non-statutory tools – such as setting requirements or accepting undertakings – before considering formal enforcement action. This proactive and proportionate approach aims to resolve issues efficiently while maintaining the regulatory standards that protect learners and the value of qualifications. We have also clarified how and when we publish enforcement decisions, improving transparency while ensuring fairness. For example, as with our existing approaches, we will give awarding organisations a fair opportunity to consult with us and challenge decisions before they become public. These changes reflect Ofqual’s stewardship approach to regulation –  ensuring that qualifications remain high quality, and fair for all. **Amanda Swann** **Executive Director for General Qualifications, Ofqual**

More than 5.6 million GCSE results have been issued to students in England today, along with more than 368,000 results for level 1 and 2 vocational and technical qualifications taken alongside them. These results enable young people to progress to further study or an apprenticeship. Well done to all students receiving their results today, and to all the teachers and others who have supported them every step of the way. This year’s picture is one of stability, with outcomes showing only minimal variation from previous years. This falls within the boundaries of natural year-on-year fluctuations. ## Key points That said, here are some key facts and figures from today's GCSE and level 1 and 2 results: * Overall GCSE results are similar to 2024. Outcomes at grade 7 and above are 21.8% compared with 21.6% in 2024, and outcomes at grade 4 and above are 67.1% compared with 67.4% in 2024. * 5,661,970 GCSE results were issued in 133 different qualifications. * There were more entries from post-16 students this summer for GCSE English language and maths. However, within both the 16-year-old and post-16 groups, results are stable. * Fewer students sat GCSE biology, chemistry and physics but there was an increase in entries for combined science. There was also a decline in entries for history compared with 2024, but a small increase in geography. * Overall, 4.1% (14,920) of the 368,065 vocational and technical qualification grades issued are at the top grade available. Of those, 355,590 were Technical Awards and 12,475 were Technical Certificates. * The most widely taken vocational and technical subjects were Technical Awards in Leisure, Travel and Tourism; and Health, Public Services and Care. * The geographical region with the highest percentage of GCSE grade 7s and above was London, and the north-east had the fewest. * Boys’ results improved slightly while girls’ results fell marginally. ## What are the trends behind the results? **Stable levels of performance   ** Stability is the watchword with the 2025 GCSE results. The national picture at grade 4 shows just a 0.3 percentage point reduction from 2024, and just a 0.1 percentage point increase compared to 2019 (see Figure 1). _Figure 1: GCSE outcomes across all subjects, for all ages, at grade 4/C and above, England only_ While grades overall are stable, there are still small variations reflecting differences in student performance. However, the standard of work that students must achieve to gain a grade is held consistent year on year, and these results maintain that standard. The 2025 National Reference Test results, also published today, demonstrate the continuing stability of English language and maths performance outcomes for 16-year-olds. **Post-16 GCSE candidates  ** Where there is a more notable change is the 14% increase in post-16-year-olds getting GCSE results today. This is due to the greater number of post-16 students taking English language and maths resits. The higher number of resit candidates is reflected in the changes to English and maths results at and around grade 4 for all candidates combined, compared to last year. In fact, performance within the 16-year-old age group and within the 17+ age group is broadly similar year on year. **Girls and boys  ** The gap between boys’ and girls’ GCSE attainment has narrowed, due to a combination of boys performing marginally better and girls performing marginally less well than in 2024. These are, again, only small changes. **Regional disparities  ** As in any year and at any level of qualification, today’s results reflect the regional disparities that exist across England. Within that wider picture, though, it is worth noting that the gap between London and the rest of the country has narrowed slightly at GCSE grade 4. **Vocational and technical results   ** This is just the second year of Tech Award results since these qualifications were strengthened by changes to their design and structure. Tech Awards are typically the size of a GCSE and cover subjects such as sport, hospitality and catering, and health and social care. ## Where to find more details If you’d like to dig into the data, take a look at our new statistical report on this year’s results and our interactive outcomes charts. As normal, the Joint Council for Qualifications (JCQ) has published results in England, overall and by subject. These results are also shown in our interactive charts. Today is the culmination of a lot of hard work, not just by the students involved but also their teachers and those who have supported them along the way. Congratulations to you all. **Amanda Swann  ** **Executive Director for General Qualifications, Ofqual**

More than 850,000 A and AS level results have been issued to students in England today, along with just over 250,000 results for vocational and technical performance-table qualifications taken alongside them. These are results that enable young people to progress along a range of pathways – to university, apprenticeships or employment. They are the reward for the effort and quality of work that those students have displayed, and I would like to offer congratulations to everyone receiving their results today. Across the range of level 3 qualifications, a total of more than 1.1 million results have been issued this summer. The highest number of those are for A levels, with 814,335 being taken by 313,190 students in 137 different subjects. A further 50,475 AS level results have been issued today as well. **Here are some other key facts about today's level 3 results:** * A total of 6,845 18-year-old students received A* grades in all the A levels they took this summer – 35 of them took 5 or more A levels * About two-thirds of A level students took 3 A levels * The subjects taken by the highest number of A level students were maths, psychology and biology * Applied generals were the vocational and technical performance-table qualifications with most results issued and the most widely taken subject sector areas for applied generals were business, administration and law, social sciences, and science and mathematics * The number of students expected to complete T Levels this summer increased by more than 60% on last year, to 11,909 * The geographical region with the highest percentage of A level results at A or A* was London ## **What are the trends behind the results?** This year the number of 18-year-olds is up on 2024, and there are some small but noticeable patterns in the type of qualifications they have chosen. The number of A levels being taken this year has actually fallen slightly since 2024, by about 2,500, and the trends in the choices of subject have continued. STEM subjects such as mathematics, further maths and physics have risen in popularity, even in a slightly smaller cohort, whereas some arts and humanities subjects have declined. There has been a small increase in pass rates. We can’t say for sure why this is the case, but there are some factors that could be part of the story. Based on prior attainment, such as these students’ GCSE grades, and relative to the national population of 18-year-olds, this year’s cohort of A level students seems to be stronger than last year’s. Moreover, this year’s A level students made their post-16 choices in 2023, which was the first year GCSE grading returned to normal after the pandemic. That could mean those students had a more accurate picture of their strengths to guide their choices. We can’t be certain but it may be that some students who might have taken A levels in previous years instead opted for some of the level 3 vocational options, such as applied general qualifications (for example BTECs) or T Levels. So today’s results may reflect changes in the choices of students 2 years ago, and the types of qualifications they wanted to study to take their next step. It is important to emphasise that while each year’s cohort of students will be slightly different, the standard of work needed to get a particular grade remains comparable with previous years. This means any change in numbers achieving a grade is entirely due to a change in student performance, not a change in the standard expected. You can find out more about this process here. ## **Where to find more details** Today, for the first time, we are publishing a statistical report on this year’s results. If you want more in-depth analysis of outcomes for A levels, AS, applied generals, tech levels and technical qualifications within T Levels, this report will be a great place to find it. We have also published interactive visualisations of outcomes by centre type, variability in school and college A level results, A level outcomes in England, A level grade combinations, and an interactive map of England showing A level results in different subjects by grade and county. As normal, the Joint Council for Qualifications (JCQ) has published results in England for AS and A level, overall and by subject. These results are also shown in our interactive visualisation. The annual results period is the culmination of a lot of hard work, not just by the students involved but also their teachers and those who have supported them along the way. Congratulations to you all. **Amanda Swann** **Executive Director for General Qualifications, Ofqual**

I want to explain why we have withdrawn our official statistics on access arrangements today, and what we plan to do next. We know that these statistics are important for understanding how the exam system works, and have been the subject of some discussion. That is why we are keen to share what we have found and start the process of setting the record straight as soon as possible. Let me start by saying that this problem is about data collection and reporting, not the access arrangements system itself. No student has been or will be affected. Here’s what you need to know. ## **What are access arrangements?** Access arrangements are based on students’ needs. They fall into 2 categories: “reasonable adjustments” for disabled students and arrangements for those with temporary injuries or other access needs. Schools and colleges apply to exam boards for access arrangements to help ensure that those students can sit their exams without being unfairly disadvantaged. The most common form of exam board-approved access arrangement is the granting of 25% extra time in exams. ## **What happened?** In 2023 we started considering trends in access arrangements data and began research into how the arrangements were used and perceived in the education sector. As part of this work we requested more detailed data from exam boards, which we received at the end of 2024. We have always recognised that there are limitations to these figures. Our analysis has now revealed that they have led to a significant overstatement of approved access arrangements in place for the exam cohort in each year. The issue was not with how arrangements were granted, but with how the data was recorded and reported. For example: * The reporting included students with valid access arrangements but who didn't sit exams that year. * Duplicate applications for the same student were not filtered out. * Reporting was based on data that was not linked to individual student exam entry records, making validation difficult. Initial findings from our analysis suggest that the proportion of students receiving access arrangements – including 25% extra time – is more closely aligned than previously reported to the proportion of students with special educational needs (SEN) reported in the school population. ## **Important points to remember about access arrangements** Our analysis gives a more accurate picture of access arrangements in place for students actually taking an examination in any given academic year. In doing so, it has not revealed any evidence of misuse or systemic failure. This has been a technical issue with reporting rather than how the system has been used. The approach to access arrangements remains unchanged. Schools and colleges should continue to apply for access arrangements as they always have, following the same JCQ guidance. ## **What’s next?** We will publish revised statistics in late 2025, covering the academic years from 2020 to 2021 onwards. These will be designated as “official statistics in development” and will include confidence ranges to reflect the complexity of the data. They won’t be perfect, but they will give a much more accurate picture than previously. We will be working with exam boards to improve the quality of the data they give us in future so that the quality of the overall statistics continues to improve. We also plan to publish a report explaining our review of evidence (including the new data analysis), as well as research into perceptions of 25% extra time in exams. This long-term research, as well as our work to improve the official statistics, shows our commitment to improving the understanding and transparency of this important part of the qualifications system. **Tom Bramley Executive Director, Research and Analysis**,**Ofqual**

As we close in on the ninth summer of awarding GCSE grades 9 to 1, we can say they aren’t quite so ‘new’ any more. And yet we still see thousands of visits every month to our explainers on the 9 to 1 grading scale from when they were first introduced. We know why this is. Every year sees a new cohort of students and parents keen to understand the scale and relate it to the A* to G scale they have heard about from previous generations. GCSE exams have just finished and attention is turning to results. So with 8 years’ experience of the grades bedding in, we have published refreshed information on our website to answer the most common questions. ## Why did GCSE grades change to 9 to 1? The change was made when GCSEs were reformed so that: * it is immediately clear whether students studied the old pre-reform GCSEs or the updated versions of them * there is more differentiation at the top of the scale, where grades 7, 8 and 9 together are equivalent to grades A and A* on the old scale See GCSE 9 to 1 grade scale explained for more information.