Exam chiefs have had to withdraw 10 years of data after admitting they got figures for extra exam time wildly wrong.
In an extraordinary blunder, Ofqual, the exams regulator, overestimated the proportion of children getting extra time in exams for special needs by more than 50 per cent.
In addition, figures suggesting private schools might be ‘gaming the system’ for their pupils have also been shown to be false.
The mistake was caused by including in the data children who were not taking exams that year, as well as duplicate applications from the same candidates.
The revelation is significant because the supposed enormous rise in children being granted extra time for special needs has become a political headache for the Government.
Bridget Phillipson, the Education Secretary, has been battling how to mend the country’s overloaded and ‘broken’ special needs system.
Tom Bramley, executive director of research and analysis at Ofqual, said: ‘We are correcting the record as soon as possible.
‘This issue is limited to our access arrangements dataset and our other statistics are not affected.’
Exam chiefs have admitted they overestimated the proportion of children getting extra time in exams for special needs by more than 50 per cent (file picture)
The latest figures showed an incredible 30 per cent of children got extra time in their A-level and GCSE exams, following a massive rise since 2016.
However, today, the regulator admitted the real figure is likely to be less than 19 per cent.
The mistake was discovered after Mrs Phillipson asked Ofqual to look into why there was a large disparity between the percentage of children who get extra time at private schools, at 42 per cent, compared with those at state schools, at 27 per cent.
Today, Ofqual admitted that the real figures for private and state schools will likely be broadly in line with their respective special needs populations.
In a blog post, Mr Bramley added that the proportion of students receiving special exam arrangements – including extra time – was ‘much more closely aligned than previously reported to the proportion of students with special educational needs reported in the school population.’
Pupils with special educational needs and disabilities (SEND) can apply for extra time in exams – with 25 per cent extra time being the most common.
It is known as an ‘access arrangement’ – and others in this category include use of a reader or scribe.
Julie Robinson, chief executive of the Independent Schools Council, said of the revelation: ‘Ofqual is supposed to be the trusted source for exam statistics and as a result of these significant errors, independent schools have wrongly seen their results undermined and their integrity questioned.
‘We are pleased that the investigation instigated by the Department for Education will lead to a correction of the record and we hope an apology will be forthcoming.’

It comes as Bridget Phillipson, the Education Secretary (pictured), has been battling how to mend the country’s overloaded and ‘broken’ special needs system
A DfE spokesman said: ‘Ofqual has acted appropriately in firstly reviewing the data and then identifying and addressing the issue as quickly and transparently as possible.
‘We welcome Ofqual’s commitment to improving data quality and transparency, and we will continue to work closely with them to ensure high standards across the system.’
Paul Whiteman, general secretary at school leaders’ union NAHT, said: ‘While it is regrettable that problems have been identified with these statistics, Ofqual has done the right thing in withdrawing them and working to correct and republish them.
‘Having correct datasets on access arrangements is vital to identify trends and changes across the system.
‘School leaders have certainly seen more requests for access arrangements in recent years in the aftermath of Covid, increased reports of anxiety and mental ill-health, and significant rises in numbers of children with additional needs.’
Ofqual is due to publish revised statistics in late 2025, covering the academic years from 2020/21 onwards.