This summer, around six million grades will be awarded in England of which about 1.5 million will be mistaken, but they cannot be discovered or corrected. Dennis Sherwood says it’s time to reform Ofqual’s appeals process


On April 18, Ofqual published its Official Student Guide (Ofqual, 2023). Under the heading, What to do if you think there is a mistake in your results, it states: “If you think there has been a mistake in the marking of your exams or assessments, you should talk to your school or college. Your school or college can ask the awarding organisation to check if there were any errors in how your exam or assessment was marked.”

It sounds reassuring: any student who fears that there has been a mistake can get the marking checked.

Phew! That’s a relief! And reference to any of the exam boards’ websites will confirm that a “review of marking” will discover, and subsequently correct, any “genuine marking errors or unreasonable marking” (to quote AQA).

It is easy to see that a “marking error” could result in a candidate being awarded a mistaken grade.

But is it possible for the marking to be error-free, yet the resulting grade to still be mistaken? If this were possible, then it could happen that a student is awarded a mistaken grade, but for this not to be discovered because there were no “marking errors”.

In which case, the appeals process is failing to do what it is supposed to do – to enable students “who think there is a mistake in (their) results”, to quote Ofqual once more, to get that mistake corrected.

Yes, I know it sounds crazy to suggest that error-free marking can result in “mistaken” grades. But actually it isn’t crazy at all.

Suppose an examiner assigns six marks to an answer to an A level geography examination question. Suppose further that a different, equally qualified examiner were to mark the same answer. The second examiner might give the same mark, but the mark might be five, or perhaps seven.

These marks are all compliant with the mark scheme – there are no “marking errors”. They are simply the consequence of something we – and Ofqual – know very well: different examiners can legitimately have different academic opinions. So “it is possible for two examiners to give different but appropriate marks to the same answer” to quote a 2016 Ofqual blog about marking reviews and appeals (Ofqual, 2016a).

That geography script might therefore be given a total mark of 64, or, say, 66, depending on who did the marking. If grade B is defined as “all marks from 61 to 69 inclusive” then no matter – the candidate is awarded a grade B.

But, and you can see where I am heading, if the B/A grade boundary is 65/66, then the candidate’s certificate shows either grade B or grade A, depending on the lottery of who marked the script – quite an important difference when it comes to things like university admissions.

One of those two grades must be mistaken. But which one?

Ofqual resolves this dilemma by designating the grade attributable to the mark of a senior examiner as “definitive” or “true” (Ofqual, 2016b).

So suppose, in this case, that the senior examiner’s mark would have been 66, corresponding to the “definitive” grade A. Accordingly, if the candidate’s grade is based on 64, then the grade B on the certificate must be “non-definitive” or “false” – or, in plain English, “mistaken” or “wrong”.

Yet that mark, 64, is fully valid and there are no “marking errors”. So if the candidate, who needs that A for a university place, were to request a “review of marking”, no “marking errors” would be discovered.

The originally awarded “non-definitive” or ”false” or ”mistaken” or ”wrong” grade will be confirmed, even though a senior examiner, whose grade is “definitive” and “true” would have awarded grade A.

Of course, if this were a very rare possibility, the impact would be modest. But it is not rare at all.

This summer, around 1.5 million “mistaken” grades will be awarded, which cannot be discovered or corrected. How do we know this?

In the mid-2010s, Ofqual carried out a substantial research project to find out just how rare – or not – this is (Ofqual, 2018). Entire subject cohorts in 14 subjects were marked twice: once by an “ordinary” examiner and again by a “senior” examiner. Every script therefore had two marks and two grades.

You might expect that the vast majority of those grade-pairs would be the same, and that any discrepancies would be random across the subjects. No.

Across all subjects, for about one script in every four (HEPI, 2019), the grade resulting from an ordinary examiner’s mark was different from the senior examiner’s “definitive” grade, with wide variability by subject: about four in every 100 were different in maths; about 44 in every 100 in history.

So if applied to this summer’s exam cohort, about six million grades will be awarded in England of which around 1.5 million will be “mistaken”. But since there are no “marking errors”, these 1.5 million “mistakes” cannot be discovered or corrected.

To put that in context: the number of “mistaken” grades is greater than the number of candidates sitting exams (around 1.2 million at most).

On average, we could argue that this will impact every student in the country. You as a teacher or a school will certainly have very many students affected.

As I have said, this is not rare at all. What it is, is hugely unfair. So I say to you simply: Ofqual’s appeals process should be repealed and replaced by a process that delivers fairness.

Even better – shouldn’t Ofqual deliver reliable grades in the first place, so that an appeals process to correct “mistakes” were hardly ever needed?

  • Dennis Sherwood is an independent consultant and author of Missing the Mark: Why so many school exam grades are wrong, and how to get results we can trust (Canbury Press, 2022). Visit https://bit.ly/3NHO5Xk


Further information & resources