August always means an anxious wait for results days, but this year pupils will be feeling particularly apprehensive. England’s exams regulator, Ofqual, has said that national results will be lower than last year’s and are expected to be similar to those before Covid. Some reports estimate that around 50,000 A-level students will therefore miss out on getting the A* and A grades they could have expected if they took their exams last year. They will also face intense competition for top university places given the record numbers of international students applying too.

Readjusting after the grade inflation of the pandemic was always going to be painful. In 2019, 25.5 per cent of A-level results were grades A or A*; in 2021 this skyrocketed to 45 per cent, and then fell to 36.4 percent last year. If, as predicted, this falls another 11 percent or so, then there will be a lot of disappointed pupils who may feel penalised simply for not being born 12 months earlier. In their defence, they were also adversely affected by the pandemic: this cohort will have spent their GCSE years in and out of lockdowns, and will have only sat their first ever public exams in May and June as their exams were cancelled in 2021.

Dr Jo Saxton, the chief regulator of Ofqual, has hailed this ‘return to normal’ as ‘an important milestone’. But what does ‘normal’ even mean anymore? By constantly moving the grade boundaries goalposts, we undermine the objectivity of the exams process, and what should be an objective metric has instead become a political football. How am I supposed to tell my students that exams are a fair, standardised, reliable indicator of their achievements, when the standard they need to achieve varies so wildly from year to year?

Goodhart’s Law states that ‘when a measure becomes a target, it ceases to be a good measure. Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.’ This is exactly what has happened with exam results. Quangos continue to set quotas which are often heavy-handed and ruthlessly statistical, designed to continually adjust the bell curve on the basis of political decisions and party optics. If a government wants to show schools are improving, then it has to accept a degree of grade inflation, but too much and the credibility of those results is questioned. Therefore, rather than accepting a degree of natural fluctuation from year to year, we have this incredibly complicated system of standardisation and intervention in order to mould results into a pre-ordained statistical shape. Behind this subtle tweaking is a lot of subjectivity.

What is more important: keeping grades consistent, or recognising how knowledgeable a cohort are?

Of course, we need exam regulators like Ofqual to ensure compatibility and comparability between exam boards. I have taught three different exam boards for English Literature A-level, all of which have startlingly different grade boundaries. For OCR, grade boundaries for A*s are often over 90 per cent; for Edexcel, they are closer to 75 per cent. The text choices and questions for Edexcel are not necessarily any more difficult, but I often have to adjust the generosity of my marking so that the result fits the grade rather than the other way around. As the grade boundaries fluctuate so much from year to year, it becomes increasingly more difficult to predict.

Some people may wonder why exam boards don’t just award grades on the basis of norm referencing: judging pupils’ performance in relation to the rest of their cohort and then fixing the percentages accordingly. For example, the top 5 per cent achieve A*s, then the next 10 per cent achieve As and so on. The problem is that such a system would encourage excellence, but also exclusivity: you have to accept a certain number of students always ‘failing’, and there is no way to track improvements or progress in schools. It also raises the question of what is more important: keeping the grade profile consistent, or giving grades consummate with how knowledgeable a particular cohort actually are. Instead, GCSEs and A-levels combine norm-referencing with criterion-referencing: measuring students’ performance against certain standards, which allows governments to, once again, turn measures into targets. For example, the government has said that it wants 90 per cent of Year 6 students to meet the expected standard in reading, writing and maths, and the average grade in GCSE English Language and Maths to increase to a 5.

There are no easy answers or quick fixes, and, whatever happens next week, young people deserve praise for their achievements in the face of years of disruption. Rather than nostalgically reminiscing about when ‘times were tougher’ and lamenting rampant grade inflation, we need to consider how we can depoliticise exam results so that they can truly reflect performance rather than perpetuate a particular narrative.