Delaying 2021 exams is a good idea – here’s why

12 Sep 2020, 5:00

Pushing exams back a few weeks next summer is the best option. Arguments for more major exam reform fall into a series of traps, writes Jon Coles

It’s clear that the Covid-19 impact on 2021’s exam groups will be greater than on 2020’s. The debate about next summer’s GCSEs and A levels is already well underway, with some of the ideas proposed better than others.

One of the issues is whether 2021 exams should be delayed to later in the summer, in order that pupils get much-needed time to catch up. Although Ofqual published the outcome of its consultation on 2021 GCSEs and A-levels last week, the document didn’t include any decision about delaying 2021 exams.

It’s my view that a delay of a few weeks is a good idea. This won’t solve all the problems, but it avoids some pitfalls of other suggestions.

First, it is generally mistaken to think educational problems can be solved through the assessment system. Pupils have missed time in school; on average cohorts will know and understand less than they normally would; gaps between rich and poor will be wider. These are real-world problems. You could design an assessment system which tries not to illuminate the facts, but that would do nothing to solve the problems and nothing to create equity.

Major reform of the exam system is not a solution for this year

Second, major reform of the exam system is not a solution for this year. Introducing such change initially depresses raw achievement and widens gaps – because test familiarity has such a big effect and because the most advantaged have the most capacity to prepare for change. This is true always and everywhere.

Third, making rushed changes to exam questions is anyway a terrible idea. Questions are developed, tested and researched over years before being included in exams. As a result they usually differentiate well between candidates. Introducing untested items into exams carries an extremely high risk of unintended consequences.

Fourth, trying to run an exam system without exams is not a good idea. Whatever the case for reforming exams, this year’s experience does not strengthen it.

The fifth and final argument is that we should just run with teacher assessment again. But this typically produces wider deprivation gaps than exams. While Ofqual should have provided training for teachers this year, it takes a huge leap of faith – unsupported by evidence – to think training or “moderation” would fully solve that problem.

So, what should we do?

Ofqual’s relatively modest proposed changes are actually quite sensible. Those who say the changes don’t go far enough generally make one or more of the five mistakes above.

Next, it is sensible to delay exams as much as possible – but that’s at most by three to four weeks. The marking and awarding process has, at best, two weeks’ slack within it. We could additionally delay results from their usual August slots by perhaps a fortnight. Beyond these changes, problems multiply fast – end of term, getting scripts marked in teachers’ summer holidays, impact on destination institutions, etc. Three weeks isn’t enough to save the world, but I’ll take it.

Government should do everything it can to make a normal exam series workable. One obvious change is that the “tier 2 local lockdown” guidance should prioritise year 11 and year 13, so they attend at all times – even where there is a rota system. This could be achieved with minimal health impact by removing the requirement for 100 per cent attendance of vulnerable and key worker children in a rota system, since the rationale which existed for that in national lockdown has disappeared.

Finally, we will need plans B and C in case full exam series are not possible. These can use assessments closely similar to those in a normal series. There are viable options for a cut-down exam series and for differently timed interim exams using properly pre-tested assessment items.

We can’t solve this year’s problems with the exam system, but we can avoid making them worse. Meanwhile, solutions to the educational problems lie where they always lie – in the classroom, in the hard work of teachers and pupils and in supporting families.

More from this theme


Key stage 2 SATS: 59% achieve ‘expected standard’ as maths and writing plunge

Results drop to lowest levels since 2016 after Covid disruption, but silver lining as reading improves

Freddie Whittaker
Exams, Schools

‘Shambles’: Heads wait hours for SATs results as Capita portal crashes

Leaders waiting since 7.30am say delays 'completely unacceptable', as national statistics website also goes down

Freddie Whittaker

Scaled scores for 2022 key stage 2 SATs announced

DfE publishes marks pupils need to meet the 'expected' standard

Freddie Whittaker

Ex-DfE adviser and ESFA boss appointed in Ofqual board shake-up

New board members to bring 'invaluable insight', says regulator chair

Samantha Booth

Private schools over-inflated GCSEs during Covid, analysis suggests

But academics say it's 'still not clear' why private schools issued more generous grades than expected

Freddie Whittaker

AQA staff vote in favour of strike action over pay

Unison considering 'next steps' after 71% vote for industrial action it claims could disrupt exam results

Freddie Whittaker

Your thoughts

Leave a Reply

Your email address will not be published.