Exams are horrible – but name me a better system

Iwas never much good at school. But aged 16, when I moved back to Britain from the US and threw myself into the A-level system, I finally saw the light. And it was called: exams. For the well-prepared but unexceptional student (like me) and the arrogant genius alike, they are a good chance to be judged not on personality or subjective factors like “attitude” – a favourite of Americans – but on actual performance.

If only the Government, now embroiled in a ludicrous and tragic imbroglio over an exam-replacement system dependent on a bizarrely regressive algorithm, had thought a bit harder about the virtues of exams.

Instead of teachers’ predictions, the poorly vetted algorithm – briefly the object of furious protests before a Government U-turn last week – assigned grades based on the distribution normally seen in schools over the past three years.

Small schools and small subjects lacked enough of this so-called predictive data and, for them, teachers’ more lenient predictions were used and marks were higher.

Many of those corseted into the oppressive algorithm, who attended big state schools and took big subjects like English and maths, found themselves marked down by up to two grades and had their hopes for the future scuppered.

According to the experts, this ghastly algorithm and its explosive results could have been avoided. The exam body Ofqual refused the scrutiny offered from the Royal Statistical Society back in March, for instance. Gavin Williamson, the Education Secretary, did not ask tough questions in July when he had the chance. Ofqual refused to be transparent about the algorithm before it was applied.

But the best way to have avoided this mess would have been to have held actual exams, as they managed in Germany.

Britain’s handling of Covid-19 has seen a bewildering array of confusing and arbitrary measures and approaches. At first we were the most blasé – lethally so – country in Europe. Then we became paranoiacs; pointless one-way systems everywhere; a peculiar reluctance to let nail salons and other low-density, easily virus-proofed outlets reopen, and painfully mixed messages over schools.

If we had set our mind to it, we could easily have allowed school-leavers and those taking GCSEs to actually sit their exams, with lockdown repurposed as a blissfully elongated revision period. There is nothing more socially distanced than an exam; I sat my A-levels in a vast hall, metres from anyone. They might have had temperature checks on the way in. Or, with real ingenuity, marquees open at the sides could have been erected, as they have for some primary schools, with desks set therein. This would have been the Smart Covid approach. Prof Guy Nason, chair in statistics at Imperial College London, has been one figure to express surprise that no in-person assessment was attempted, as it was in Germany to no discernible harm.

Instead we kept to our old ways, brittle and illogical. I don’t think for a minute that the Government doesn’t care about educational opportunity. But I think its culture of data-worship has led to some shoddy, arrogant thinking about the superiority of number-crunching over qualitative human effort. Throughout modern life, the idea seems to have snuck in that the more you can erase actual human beings from the picture, the better. When in doubt, use an algorithm.

The stock of exams has been falling for years. They’re seen as unfair, pressurising, impersonal, anxiety-inducing. They’re also seen as sexist. In 2013, outrage followed the pronouncement by Mike Nicholson, then-head of admissions at Oxford, that boys do better than girls at exams because they’re better at taking risks. But that outrage was mixed with a kind of agreement: that exams, many multiple choice, are themselves sexist.

In 2020, the data showed something else, proving that – as with many claims about gender, a large pinch of salt is needed. The British Council found that when it came to exams, especially in languages, girls were more likely to perform well, but boys were more often successfully re-marked up because they were more forward about demanding it.

The reality is that – whether you’re an “exam person” or not – hard work pays off. Having failed to be teachers’ pet, and with wildly oscillating coursework, I found they offered me a fair chance. I revised like mad, to the mockery of my too-cool-forschool peers. I learnt all the past papers, memorised every historic multiple choice answer, and – despite terrible nerves on the day – found that my efforts paid off.

The absolute carnage the nation’s exam-takers have been plunged into could have been avoided. With less disdain for the old-school, analogue concept of the exam, a little creative thinking about how they could actually be sat, and a more critical stance towards the promise of algorithms, we’d be in far better shape. And so would the self-esteem of our youngsters.

The culture of data worship has led to some shoddy, arrogant thinking