You won't want to miss out on the world-class speakers at TNW Conference this year 🎟 Book your 2 for 1 tickets now! This offer ends on April 22 →

This article was published on August 13, 2020

Algorithm downgrades nearly 40% of English exam results, hitting poorer students hardest

Unions want the government to replace the grades with teachers' predicted results

Algorithm downgrades nearly 40% of English exam results, hitting poorer students hardest Image by: Kobe Michael from Pexels

An algorithm used to determine school-leaving grades in England has downgraded 39% of the results predicted by teachers — with disadvantaged students suffering the biggest drop.

Exam regulator Ofqual adopted the system when exams across the country were cancelled due to the COVID-19 pandemic. Instead, teachers were instructed to submit their predicted grades for each student. The algorithm then moderated their assessments by comparing them to their school’s historic performance and each pupil’s past results.

In total, the algorithm downgraded about 280,000 A-level results, with poorer students more likely to receive lower grades.

Ofqual’s figures revealed pupils at fee-paying schools received double the improvement in the top A* and A grades compared to students at state-funded comprehensives.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!

[Read: Algorithm that determines school exam results risks ‘baking in inequality’]

The watchdog admitted poorer students would be hit the hardest.

“There was a tendency for some more generosity to be there in the predictions for students from lower socioeconomic status backgrounds,” said Michelle Meadows, the watchdog’s deputy chief regulator.

“So you see a very small effect where there is some more – I don’t want to use the term downgrading, I think that’s the wrong term here – but there is a small effect of a greater difference between the end calculated grades and the centre assessment grades.”

Labour MP Wes Streeting said the system had exacerbated educational inequalities.

“As someone from a poor family in a poor postcode I dread to think what would have happened to my grades if I’d been victim to this flawed system,” he tweeted.

Calls to ditch the algorithm

A similar system used in Scotland was abandoned this week after its calculations provoked a public outcry. A quarter of all pupils had received downgraded results, and students in deprived areas had their pass rates dropped by more than twice the rate of those in the country’s richest regions.

In response, the Scottish government withdrew the results and replaced them with the teachers’ predictions. Unions are now calling for ministers in England to follow Scotland’s lead.

“The government needs to accept it has got this badly wrong, stop trying to pull a rabbit out of the hat and keep things simple by using teacher predictions – as happened in Scotland,” said Jo Grady, general secretary of the University and College Union (UCU).

Education Secretary Gavin Williamson has dismissed the idea, but says pupils can resit their exams in the fall, or swap their grades for those they got in mock exams earlier this year. However, schools will need to provide evidence of a “valid” mock result to appeal against the grades.

“We still don’t know how long appeals are going to take, how much they will cost, and what the grounds will be,” Laura McInerney, a former teacher and education blogger, told the Financial Times. “Until you answer those questions, headteachers won’t be able to give proper guidance to their pupils about what to do.”

The uncertainty will only add to the stress of students — particularly those from disadvantaged backgrounds.

So you like our media brand Neural? You should join our Neural event track at TNW2020, where you’ll hear how artificial intelligence is transforming industries and businesses. 

Get the TNW newsletter

Get the most important tech news in your inbox each week.

Also tagged with