Official Statistics

Background information for access arrangements for GCSE, AS and A level: 2024 to 2025 academic year

Published 27 November 2025

Applies to England

Provide your feedback

We welcome your feedback on our publications. If you have any comments on this statistical release, or how to improve it to meet your needs, please complete our short survey or email our statistics team.

About this report

This annual statistical release presents data on access arrangements for GCSEs, AS and A levels, which were approved by exam boards and valid for use by students in England who were being assessed in these qualifications during each academic year. It also contains statistics on the number of requests fulfilled by exam boards for modified papers for the summer GCSE, AS and A level exam series.

Where we use the term “students being assessed” throughout this release, this includes both students sitting exams as well as those students who did not sit exams in 2020 and 2021, when exams were cancelled due to the COVID-19 pandemic, but were still assessed by their teachers and could still have access arrangements approved. Further detail can be found in the ‘policy changes over time’ section below.

Throughout this release, figures only include arrangements for students being assessed in their final year of study. For summer 2020 onwards, this applies to all students. However, arrangements for students who sat exams in their first year of study under legacy qualifications in 2015 to 2019, which were still modular, are excluded. This approach ensures consistency across the time series. Additional details can be found in the ‘2015 to 2020 reforms’ subsection below.  

Official statistics in development designation

In previous releases of these statistics (now withdrawn), figures did not give an accurate picture of the number of access arrangements in place for the exam cohort in each year. This was caused by the fact that the dataset used for those releases contained duplicates and approvals for students who were not being assessed in the exam series that the approvals were valid for.

To address these issues, we have begun to work with the exam boards to improve data quality at source, and to improve our own methodology for data cleaning and analysis. The aim of this ongoing work is to facilitate more robust reporting in future releases, particularly with the aim of reducing remaining sources of uncertainty in the data. As such, it is likely that the nature of this release will change with future iterations as we are able to make those developments.

Further details on the data, our methodology, and remaining sources of uncertainty are provided throughout the sections to follow.

In this release, we are still able to meet most principles of the Code of Practice for Statistics, and believe the release provides meaningful insights to users. However, it is important to recognise the changing nature of this and future releases. As such, this release is currently classified as ‘official statistics in development’, and will remain so until we have a stable data source and methodology in place.

We anticipate this improvement work to take at least 2 years, meaning this release may become stable from the 2028 release onwards. However, we will provide updates on likely timelines, as well as any developments made, in subsequent releases.

To help us in developing this release with user needs in mind, users are encouraged to share their feedback via the survey or email link found at the top of this page. Ofqual will also continue to engage with relevant sector bodies as part of the development of our approach to these statistics and relevant wider policy (see our regulatory report for more information on Ofqual’s wider work in this area).

Scope of the release

Figures provided within this release relate to GCSE, AS and A level qualifications taken by students in schools, colleges and other exam centres (referred to as ‘centres’ throughout this release) in England only.

Four exam boards offer GCSE, AS and A level qualifications in England:

• AQA Education (AQA)

• Cambridge OCR

• Pearson Education Ltd (Pearson)

• WJEC-CBAC Ltd (WJEC)

For access arrangements other than modified papers, all figures within this release relate to access arrangements approved for use during the academic year for students being assessed in that year. Students of all ages are included, including those resitting their previous assessments. Figures for modified papers relate to the number of requests for modified papers fulfilled by exam boards for use in the summer exam series only.

Access arrangements, such as 25% extra time, are approved once for each student and that student may then use that arrangement in any of the assessments they take during the validity period. Figures reported are therefore not affected by the number of assessments each student is taking. In contrast, an order for a modified paper is made for each individual exam paper. Therefore, an individual student may have multiple modified papers in a single exam series, and the figures will reflect this.

Figures only relate to arrangements approved for use, or the number of modified paper requests fulfilled by exam boards. Our data does not allow us to comment on the extent to which arrangements or modified papers were actually used by individual students during their assessments. Figures should not be interpreted as such.

The approval of some access arrangements is delegated to centres where exam boards judge that it is appropriate to do so (see page 115 of the Joint Council for Qualifications (JCQ) guidelines for details). The types of arrangements requiring centres to obtain approval from exam boards have changed over time. The data on access arrangements in this release only covers those types of access arrangements which have been approved by exam boards; it does not cover the arrangements where approval is delegated to centres.

Similarly, for the section on modified papers, figures relate to modified paper requests fulfilled by exam boards for use in the summer exam series only; it does not cover modified papers produced directly by centres, or those produced for use in the November exam series.

For some types of modified papers, centres only need to make one request per paper to the exam board and can then use it for multiple students that need it. Centres might also make copies on larger or coloured paper, for example. As the school or college only requested one modified paper in these instances, this will only appear as one fulfilled order in the figures reported. This means that the figures reported in this release may not necessarily reflect the total number of modified papers actually used in any given exam series. There has been some variation over time in the exam boards’ requirements for ordering modified papers. Since this data reflects the exam boards’ fulfilled orders for modified papers, the data will have been impacted by any such changes.

Policy context

Access arrangements

Access arrangements are the provisions made for students, agreed before they take an exam (or in the case of 2020 and 2021, were assessed by their teachers), to ensure that they can be validly assessed and are not unfairly disadvantaged due to a disability, temporary illness or injury. Access arrangements can be provided for any students being assessed who meet the eligibility criteria. Access arrangements granted for disabled students are provided as reasonable adjustments.

AQA, Pearson, OCR and WJEC access arrangements are set out in guidance jointly produced by exam boards through their membership organisation, the Joint Council for Qualifications (JCQ). Further detail is outlined in JCQ’s guidance on access arrangements and reasonable adjustments.

Where access arrangements are granted for students who have experienced a temporary illness or injury, or some other event outside of the student’s control, these are granted as a form of special consideration within Ofqual’s regulatory requirements. The same categories of access arrangements may be used for both reasonable adjustments and special consideration.

Any student entered for GCSE, AS or A level qualifications who meets the eligibility criteria should be granted the appropriate access arrangement. Individual students may require more than one arrangement (for example, 25% extra time and a computer reader). Once granted, an arrangement for a long-term condition will be valid for up to 26 months, whereas applications based on temporary conditions will last for one exam series only.

The term ‘special consideration’ can also refer to a post-examination adjustment to a student’s mark or grade to reflect temporary illness or injury, or some other event outside their control, which has had, or is reasonably likely to have had, a material effect on their ability to take an exam or demonstrate their level of attainment in an exam. These post-examination adjustments are distinct from the special consideration access arrangements granted prior to an exam being taken. We publish separate releases that report on post-examination special consideration requests for the summer GCSE, AS and A level series.

Modified papers

Modified papers (such as enlarged-font or braille papers) are also a form of access arrangement, and requirements for modified paper requests are also outlined by the JCQ guidance on access arrangements and reasonable adjustments.

An order for a modified paper is made for each individual exam paper. Therefore, an individual student may use multiple modified papers in a single exam series.

For some types of modified papers, centres only need to make one request per paper to the exam board and can then use it for multiple students that need it. Centres might also choose to make copies on larger or coloured paper, for example.

Policy changes over time

Outlined below are policy changes implemented over time, within the scope of this release, which have may affected some of the trends observed.

Access arrangements

From 1 September 2024, students for whom English is an additional language were able to use a computer reader or reader in exams.

Also from 1 September 2024, new categories of listening to music or white noise during exams, and timetable variation on the day of the exam were also introduced. These new categories are reported within the ‘all other remaining arrangements’ category in the main report but separately in the data tables.

From 1 September 2023, new categories of access to a mobile phone for medical purposes, remote invigilation, and timetable variation requiring overnight supervision for a student with a disability were introduced. These are reported within the ‘all other remaining arrangements’ category in the main report but are reported separately in the data tables.

Also from 1 September 2023, the extra time granted when applying for ‘bilingual translation dictionary with extra time’ was extended from 10% to 25%. These are reported under ‘bilingual translation dictionary with extra time’ in the main report and accompanying data tables. Further details of these changes can be found in the 2023 to 2024 academic year JCQ guidance document.

From 1 September 2022, the evidence requirements for 25% extra time for students with learning difficulties were changed, regarding when to use a reading speed or writing speed score as evidence to be recorded in part 2 of form 8. Further details can be found in the 2022 to 2023 academic year JCQ guidance document. The 2022 to 2023 guidance also specified that arrangements such as fidget toys, stress balls and non-electronic ear defenders or ear plugs do not require an application using the Access Arrangements Online system.

From 1 September 2021, the evidence requirements for 25% extra time for students with learning difficulties were changed. Where previously the criteria for 25% extra time was for at least one below-average standardised score of 84 or less, the revised criteria required at least 2 below-average standardised scores of 84 or less, or one below-average standardised score of 84 or less and one low-average standardised score of 85 to 89. See the relevant JCQ guidance document for full detail. However, students who were entered for examination in Autumn 2021 or June 2022 to improve upon a previous grade, and had a previously approved online application for 25% extra time, could be awarded 25% extra time based on the previous criteria of at least one below average standardised score of 84 or less. In this scenario, new applications were processed by selecting the access arrangement ‘25% extra time (2021/22 re-sits)’ category. The same process was applied the following year, with new applications processed by selecting the access arrangement ‘25% extra time (2022/23 re-sits)’ category. These re-sit arrangements are included within the ‘25% extra time’ category in the tables and charts in the main report and in the data tables.

In the 2020 to 2021 academic year, the decision to cancel summer exams and instead award teacher assessed grades (TAGs) was announced before the deadline for submitting applications for access arrangements. While centres were still encouraged to apply for access arrangements for their students as normal, the 2020 to 2021 academic year figures would likely have been higher if the summer exams had gone ahead. Applications that had expired prior to, during or after the planned dates for the summer 2020 examination series were extended until the autumn 2020 examination series.

From 1 September 2019, the 2 categories ‘practical assistant for written papers’ and ‘practical assistant for practical assessments’, were merged into a single category of ‘practical assistant’. These are reported as one combined category throughout this release, and arrangements are only counted once where students might have had separate approvals for a ‘practical assistant for written papers’ and a ‘practical assistant for practical assessments’ prior to this date.

From 1 September 2018, the coloured and enlarged paper category (previously included within the ‘all other remaining arrangements’ category of access arrangements) did not require an online application through Access Arrangements Online. The numbers reported in this category therefore significantly decreased from this point. Large-font papers are however still included in the modified papers data.

Also from 1 September 2018, the categories of ‘scribe or voice activated computer system or voice input system’ and ‘scribe or voice recognition technology’ were merged into a single ‘scribe or speech recognition technology’ category. These are reported as a combined category throughout this release, and arrangements are only counted once where students might have had separate approvals for individual ‘scribe or voice activated computer system or voice input system’ and ‘scribe or voice recognition technology’ categories prior to this date.

In September 2017, the ‘reader’ and ‘computer reader’ categories were merged in Access Arrangements Online to reduce burden for centres. To allow comparison of figures over time, these categories have been merged for all years in the report and accompanying data tables, and arrangements are only counted once where students might have had separate approvals for a ‘reader’ and a ‘computer reader’ prior to this date.

Modified papers

From the summer 2024 exam series, centres no longer needed to make a request for non-interactive electronic (PDF) question papers from WJEC. The option was still available in the system for making requests, however, and so numbers have still been reported for summer 2024. For summer 2025, the option had been removed, and so these papers no longer appear in the figures reported.

Due to the cancellation of exams in summer 2020 and 2021, there were no modified papers produced for these exam series.

2015 to 2020 GCSE, AS and A level reforms

Between 2015 and 2020, GCSEs, AS and A levels were reformed in phases by subject. Under the legacy (unreformed) qualifications, students could sit some exams in their first year of study (year 10 for GCSE, year 12 for A level) with the remaining exams being sat in their final year (year 11 for GCSE, year 13 for A level). In contrast, the reformed qualifications required all exams to be taken in the final year of study.

As a result, students studying unreformed subjects between 2015 and 2019 will have had access arrangements approved for exams taken in their first year. Most of these students would have continued using the same arrangements in their final year.  

In this release, we have chosen to report only on access arrangements approved for students taking exams in their final year of study. This decision supports:

• consistency across the entire time series, otherwise figures would be inflated in earlier years due to the duplication described above

• simpler analysis and interpretation, avoiding complications from the phased subject-by-subject reforms between 2015 and 2020

• better alignment with available data, since accounting for the modular structure of unreformed qualifications would require detailed component-level data, which is less complete and less reliable in earlier years

We are only able to exclude arrangements from each year where we are confident that the student was not being assessed in their final year of study. Where we are not confident, these arrangements may still contribute to the range of plausible values reported, as part of the area of uncertainty for those earlier years. Further details on the range can be found in the ‘methods’ section below.

Duties and regulation

If a student has a disability (as defined by the Equality Act 2010, meaning the student has a physical or mental impairment that has a substantial and long-term negative effect on their ability to do normal daily activities) they are entitled to reasonable adjustments, which awarding organisations administer as access arrangements. Awarding organisations have a duty, under the Equality Act 2010, to make reasonable adjustments to assessments for disabled students who, because of their disability, would otherwise be at a substantial disadvantage when demonstrating their skills, knowledge or understanding in an assessment. The Equality Act sets out that awarding organisations are required to take ‘such steps as it is reasonable to have to take to avoid the disadvantage’. The awarding organisations may choose to address this by taking steps such as providing modified question papers for students with visual impairments.

The qualifications covered in this release are regulated by Ofqual, which publishes General Conditions of Recognition that set out the requirements that the awarding organisations it regulates have to meet. These conditions state that “an awarding organisation must, in accordance with Equalities Law, have in place clear arrangements for making Reasonable Adjustments in relation to qualifications which it makes available.” Awarding organisations must also set out how a learner can qualify for special consideration.

Ofqual does not prescribe what arrangements awarding organisations should provide, but requires all awarding organisations to have clear, published details about who qualifies for these arrangements and what arrangements may be given. The exam boards that offer GCSEs, AS and A levels administer access arrangements collectively through JCQ. JCQ is a membership organisation comprising the 8 largest providers of qualifications in the UK, including the 4 exam boards that offer GCSE, AS and A level qualifications. JCQ provides a definition of access arrangements as pre-examination adjustments for students based on evidence of need and normal way of working. JCQ guidance on access arrangements can be found on its website. This guidance is updated annually and sets out the arrangements that centres can request of boards and the relevant eligibility and evidential criteria.

Centres will consider the particular requirements of individual students and then follow the guidelines and arrangements of the awarding organisations. Any student entered for GCSE, AS or A level qualifications who meets the eligibility criteria should be granted the appropriate access arrangement. The evidence required to support an application will vary, depending on the reasons for the particular requirement and the types of arrangement requested. For example, where a request is made for a student to have 25% extra time because of a learning difficulty without having an Education, Health and Care (EHC) Plan, the JCQ guidance requires that the impact of the student’s difficulty is evidenced using standardised tests of their speed of working.

Data sources

This section outlines the data sources used within this release. Data quality, including areas of uncertainty associated with these data sources, is covered in the next section.

Access arrangements data

Access Arrangements Online (AAO) is JCQ’s centralised system used by centres to request access arrangements. Centres can make one central application for each student through the system rather than applying to each exam board separately. Applications are made in advance of the exams taking place, and JCQ facilitates the transfer of data from AAO to Ofqual each October, following the end of the academic year.

The data Ofqual receives contains one row per arrangement per student.

Once published, figures are not usually subject to revision, unless an error is discovered.

Modified papers data

Each exam board submits data to Ofqual annually for the number of modified papers it provided to centres for use in the summer exam series. This comes from a different data source to that which is extracted from the AAO system. Data submission takes place each September, following the summer exam series.

The data Ofqual receives contains one row per type of modified paper per qualification type, separately by exam board.

Once published, figures are not usually subject to revision, unless an error is discovered.

Students being assessed data

For each exam series, exam boards supply to Ofqual details on the results that have been issued, including details on the students who received results. We refer to this as our ‘awarding data’.

We use this data to identify which access arrangements were for students being assessed, via the matching methodology described in the ‘methods’ section to follow.

We also use our awarding data to determine the total number of students being assessed in each year, which we use to calculate proportions. As this data is not complete for the 2015 to 2016 academic year, we are unable to reliably calculate proportions for that year. Missing data in that year does not prevent us from calculating the range for the number of access arrangements, but it does result in greater uncertainty, meaning the range reported is wider than in more recent years.

Centre types data  

Some figures in the report and data tables are broken down by centre (school or college) type. The centre types reported within this release combine information from the National Centre Number (NCN) register and the Get Information about Schools (GIAS) register.

GIAS centre type categories include independent school, independent special school, non-maintained special school, state-funded secondary, state-funded special school establishment and state-funded alternative provision (AP) school. NCN centre categories also include further education establishment and ‘other’. There is also ‘unknown’ category assigned to centres that could not be categorised in any of the types listed above.

Exam boards report where a student sat their exam with a centre identifier that comes from the NCN register. We use this centre identifier to determine centre type categories as they are reported in the NCN and GIAS registers.

Ofqual receives updated snapshots of the NCN register each month and the August 2025 snapshot was used in the production of these statistics. The GIAS register snapshot was also downloaded in August 2025. Centre types for each centre are based on these snapshots and are backdated for previous years.

Data quality and areas of uncertainty

There is potential for error in the data provided by exam boards via JCQ. Ofqual compares the data over time and checks for systematic issues with the data.

In addition, Ofqual is aware of the following limitations and areas of uncertainty within the data. As discussed within the ‘official statistics in development designation’ section above, we are continuing to work with exam boards and JCQ on improving the quality of data at source, and on enhancing our own methodologies for data cleaning and analysis.

Access arrangements data

The ability to extract detailed statistical information was not part of the scope for the original data recording system design in AAO. As a result, there are limitations in the data produced when used in this context.

A major source of uncertainty comes from the fact that the AAO data is not linked to actual exam entries. Additionally, centres may decide to make applications for academic years that students are not being assessed in. This can be for a number of reasons: for example, to have confirmed arrangements in place ready for mock exams, or because they might not be sure at the time of making the application whether or not the student would be sitting exams that year.

Ofqual is able to link the AAO data with our awarding dataset (also provided to Ofqual by exam boards) to try to identify just those arrangements that have been approved for students being assessed in the given year. However, there is no reliable linking field in the AAO data that would facilitate this. As such, the methodology used for this release relies on name, date of birth and centre identifier matching.

Due to data quality issues in these matching fields, particularly in the AAO data, there is a degree of uncertainty in this matching process, described more fully in the ‘methods’ section to follow. This means we are only able to report ranges of plausible values rather than absolute figures.

Challenges come from differences in spelling between the 2 datasets (for example, name variations such as “Sam” versus “Samm”, abbreviations such as “Sam” versus “Samantha”, or simple data entry errors such as “Sam” versus “San”), which can be difficult to predict or fully capture in our data cleaning process. As described in the ‘methods’ section to follow, we are exploring probabilistic matching methods for future releases as a potential way to improve this methodology.

There are also known duplicates in the source data. We are able to remove these in many cases, however any variations in data entry as above will again cause some of these duplicates to go unidentified. In addition, we have identified some cases that may be duplicates but with a degree of uncertainty. For example, this can occur where what could be the same student (based on name and date of birth) appears to be present in 2 different schools. In these cases, it could be the case that the student moved schools and therefore had 2 different applications (which is a requirement of the system); however, it could also be the case that these are 2 different students who happen to share the same name and date of birth, which can happen particularly for more common names. These cases are included within our areas of uncertainty reported, as described in the ‘methods’ section to follow.

The AAO data only contains information on applications that have been approved, and the awarding data helps identify those arrangements which are valid for students being assessed. However, neither dataset allows us to comment on the extent to which arrangements were actually used by students. Students may use their arrangements in some or all of their exams, and may use them to varying degrees within those different exams as well (for example, only using some of their 25% extra time allocation).

The AAO data also only contains information on arrangements that have been requested and approved through AAO. It does not include those that are delegated to centres. As a result, the figures do not provide a complete picture of all arrangements granted in each academic year.

Modified papers

Modified papers data is supplied by exam boards. It comes from a different source to other access arrangements and therefore is not subject to the same limitations as described above.

Similar to other access arrangements, however, the modified papers data only contains information on requests fulfilled by exam boards, not those that have been produced directly by centres. In addition, for some types of modified papers, centres may only need to make one request per paper to the exam board and can then use it for multiple students that need it. For example, they might be permitted to use a non-interactive electronic paper obtained through one request to make multiple copies on larger or coloured paper. As the centre only requested one modified paper in these instances, this will only appear as one fulfilled order in the figures reported.

In addition, like with other access arrangements, our data does not include information on whether modified papers were actually used in assessments. Students may request one or more types of modified papers (for example, one request for an 18-point bold paper and another for a 24-point bold paper) and may then choose to use one (or multiple) modified paper(s) or the standard question paper.

For the above reasons, the modified papers data does not give a complete picture of all modified papers actually used in each summer exam series.

An order for a modified paper is made for each individual exam paper. Therefore, an individual student may use multiple modified papers in a single exam series, for each of the exams taken. Figures reporting numbers of modified papers should therefore not be confused with the number of students with modified papers.

Methods

Access arrangements data cleaning and analysis

Once supplied by JCQ, Ofqual carries out significant data cleaning and processing on the data.  

First, we determine whether or not each arrangement had been approved, either automatically by the AAO system or via exam board referral. We only take arrangements forward into the analysis that have been approved, and not those that have not been approved yet, or have been rejected or withdrawn. This requires the application of a set of rules across multiple data fields, which we have obtained from and verified with the exam boards and JCQ. We also remove any duplicates, subject to the uncertainties present within the data as described above. Where students have the same approved arrangement requested for GCSE and A level, we only count these once. We also filter for arrangements requested for students in centres in England only.

Second, for each arrangement, we identify which exam series it was valid for. We do this by identifying a validity period for each approved arrangement. This requires applying another set of rules that we developed and verified in collaboration with the exam boards and JCQ. We then compare the start and end dates of validity periods against the start and end dates for each exam series, and the dates of the first and the last exam in the exam series are used as the start and end dates of each exam series. We consider an arrangement to only be valid for use in a particular exam series if the validity period covers that series entirely; if the validity period only covers part of the exam series, the approved arrangement is considered not valid for use in this exam series.

Third, we carry out data cleaning to prepare for the matching methodology, which relies on name, date of birth and centre identifier matching as there is no reliable student identifier in the AAO data. In both the AAO and awarding datasets, we clean the name fields to remove whitespace and symbols (for example, brackets).

To improve matching accuracy, we create alternative versions of names. This includes extracting a single first name when multiple names are provided. For example, from “John-James” or “John known as James”, we retain only “John”. We also extract an alternative first name when one is indicated. For example, from “John known as James” or “John [formerly James]”, we extract “James” as the alternative. A similar process is applied to surnames.

This results in the following fields:

• FirstName = the first name given before the first space and including hyphenated names

• FirstName1 = the first name given before the first space or hyphen of FirstName

• MiddleName = the first name given after the first space

• AlternativeFirstName1 = the alternative first name as described above

• Surname = full surname given including multiple and hyphenated names

• Surname1 = just the name given before the first space or hyphen of Surname

• AlternativeSurname = the alternative surname as described above

Fourth, we attempt to match each approved arrangement to a corresponding entry in our awarding dataset. This step is intended to identify just those arrangements that relate to students being assessed during the period in which their approved arrangement was valid. We attempt to find a match in all exam series, not just those the arrangement was valid for use in (to be explained below), using the above variations on names and also date of birth and a centre identifier.

We use a predefined sequence of matching keys, applied separately for each exam series. If, after all matching keys have been attempted, no corresponding entry is identified, the arrangement is flagged as unmatched. The matching keys are applied in the following order:

1. Date of birth, FirstName, (full) Surname, Centre number

2. Date of birth, FirstName1, Surname1, Centre number

3. Date of birth, MiddleName, Surname1, Centre number

4. Date of birth, AlternativeFirstName1, Surname1, Centre number

5. Date of birth, FirstName1, AlternativeSurname1, Centre number

We explored weaker matching keys as well, for example those including just the first initial of the first name. However, we decided not to include them as they could have led to falsely identifying students with approved arrangements as being assessed in an exam series and therefore inaccurately inflating the figures for some years and deflating them for others.

This is a deterministic approach to matching and is deliberately conservative to avoid introducing incorrect matches which may inflate the final figures reported. We consulted with the Office for National Statistics (ONS) Methodology Advisory Service as part of the development of this methodology, who suggested a number of potential areas to explore for future enhancements. These include, for example, probabilistic matching methods which may help to improve match rates without introducing unacceptable incorrect matches.

Fifth, we take the outcomes of the matching process above to determine the degree of certainty to which each approved arrangement may or may not be in place for a student taking exams in each academic year. This is because we know the matching methodology will not be perfect, for reasons explained above.

Each arrangement is placed into one of the following 3 categories, evaluated separately for each academic year. The same arrangement may fall into different categories across different academic years, depending on its approved validity period. If the validity period of the arrangement covers at least one series within the academic year being considered, the arrangement will be categorised as follows:

1. If a particular arrangement has a matching entry in the awarding dataset of a different academic year, but not in the academic year of interest, we can reasonably assume that the student did not take exams in the year of interest. Student names are entered more consistently in the awarding dataset across exam series compared to the AAO data, therefore the presence of the name in one year in the awarding data set but not in another should not be due to a difference in spelling. In such cases, that arrangement would be excluded from the counts for the academic year of interest.

2. If a particular arrangement has a matching entry in the awarding dataset for the academic year of interest, we can be confident that the student took exams in that year. That arrangement is therefore included in the counts for that year. Since our statistics report on each academic year, we will count such arrangements only once, even if matches are found in multiple exam series (such as both the November and June series in the same academic year).

3. If a particular arrangement has no matching entries in the awarding dataset in any academic year, we cannot be certain whether this is because the student did not take any exams, or because we did not find a match due to inconsistencies in the student’s name, date of birth, or centre identifier. This group of arrangements therefore fall into an area of uncertainty. They are added to arrangements described in category 2 above, which form the basis for the lower bound of the reported range, to form the upper bound of the reported range.

Because we include centre number for identifying unique records of arrangements in each academic year’s data set, the same arrangement may be counted multiple times if a student has moved centre and had a new application in their new centre. We anticipate this will have a small effect, as these are only potential duplicates, impacting on average fewer than 2% of cases each academic year.

A result of the above process is that the reported range for the current academic year is likely to be wider than for previous years. This is because, as we continue adding more years to the dataset, the likelihood of finding matches for previous years increases. When matches are found, some arrangements previously classified under category 3 (uncertain cases) can be reclassified into category 1 (confirmed non-assessed students), reducing uncertainty over time.

Modified papers data cleaning and analysis

Modified papers data does not come from the AAO system and therefore is separate to the process described above. For this data, only relatively simple data cleaning and summarising is needed.

Because the data provided to Ofqual by exam boards is aggregated (one row per type of modified paper per qualification type), we are unable to attempt to link this data to exam entries.

Quality assurance

Quality assurance procedures are carried out according to Ofqual’s quality framework for statistical publications, to ensure the accuracy of the data and to challenge or question it, where necessary. Publications may be deferred if the statistics are not considered fit for purpose.

Significant additional quality assurance has also been carried out on this release, due to nature of the changing methodology this year. This includes additional and enhanced quality assurance steps within Ofqual, as well as seeking views on the nature of our methodology from the exam boards and JCQ, and the ONS Methodology Advisory Service.

Confidentiality and disclosure control

To ensure confidentiality of the published accompanying data,  and to reflect an appropriate level of precision, figures have been rounded to the nearest 5. If the value is less than 5, it is represented as “fewer than 5” and 0 represents zero entries.  

Where individual rounded values have been presented in a table along with their sum total, this total may be slightly different to the sum of the individual rounded values. This is because the total has been calculated using the original unrounded values.

We also use unrounded values to derive percentages. Percentages are then rounded to one decimal place. Due to this rounding, percentages may not always add to up exactly 100%.

Related statistics

See Statistics at Ofqual for other statistical releases and interactive visualisations.

In particular, see our releases on:

• special consideration in GCSE, AS and A level

• equalities analysis

For related publications for qualifications offered in Wales, Northern Ireland and Scotland, please refer to the respective regulators’ websites:

• Qualifications Wales

• CCEA

• Scottish Qualifications Authority (SQA)

Useful links

Report and data tables accompanying this release.

Definitions of important terms used in this release.

Policies and procedures that Ofqual follow for production and release of its statistical releases.

Contact details

Head of profession: Ben Cuff

Email: data.analytics@ofqual.gov.uk