Accurately predicting A-levels ‘near-impossible’, says research

The experts at UCL analysed the GCSE results of 238,898 pupils to see whether they could accurately predict their A-level results

Accurately predicting A-levels using students’ past academic results is a “near-impossible” task, academics from UCL have concluded after studying the records of nearly 250,000 past candidates.

The system of awarding A-levels this year needs to be overhauled to reduce inaccuracies that can lead to unfair disadvantages for some students, the UCL researchers added.

The experts at UCL analysed the GCSE results of 238,898 pupils to see whether they could accurately predict their A-level results. Even after removing any opportunity for bias – and accounting for pupils’ gender, ethnicity and socioeconomic status – they could only predict one in four pupils’ best three A-levels accurately.

The team found 23% of comprehensive school pupils were under-predicted by two or more grades compared to just 11% of grammar and private school pupils.

Prof Lindsey Macmillan, from the UCL Institute of Education, said the study raised questions over the system of awarding A-levels this year. Instead of sitting exams this summer, teachers assigned pupils calculated grades, which were then moderated by exam boards with reference to the school’s overall historic performance. “If you’re a straight-A student at a grammar or private school, you’re more likely to continue that to A-levels. But this research is telling us there’s a lot more movement around the grades between the two exam levels for comprehensive students,” she explained.

The UCL findings come on the same day that new joint research from the University of Birmingham and University of Nottingham indicated stark discrepancies in A-level students’ experiences of schooling this year.

This research raises the question of why we use predicted grades at such a crucial part of our education system. This isn’t teachers’ fault – it’s a near-impossible task
Prof Lindsey Macmillan (UCL Institute of Education)

Prof Kalwant Bhopal, at the University of Birmingham, and Dr Martin Myers, at the University of Nottingham, surveyed more than 500 17- and 18-year-olds between April and July and interviewed a further 53 of them.

According to the survey, just 21% of students are happy exams were cancelled and nearly half (46%) would have preferred to sit their exams. Whilst 81% of pupils from fee-paying independent schools were satisfied with how their school managed the crisis, only 67% of pupils in state comprehensive schools were satisfied. Black students were significantly less likely to report feeling satisfied with their schooling during lockdown (67%) than white pupils (82%).

Following interviews with the students, Prof Bhopal and Dr Myers said many reported having limited access to mental health support and resources to support the completion of work that would contribute towards their final grades. Many students also felt they had not received enough information from teachers on how grades would be awarded.

Although particularly pertinent in this year’s admissions cycle, the researchers at UCL said their study indicates that moving towards a post-qualification applications and admissions (PQA) system would help remove potential inequalities in future years, because students from disadvantaged backgrounds – who are statistically more likely to be under predicted – might be encouraged to apply to higher tariff universities.

A survey of 128 school, college and university leaders in the UK by the University and College Union today found that more than 60% thought current university admissions systems are not fit for purpose and 80% thought a PQA system was worth exploring.

A-level results: comprehensive pupils more likely to be underpredicted, says research
Michelle Donelan has written to vice-chancellors to ask them to be flexible in admissions decisions this year


Earlier this morning, the minister for universities, Michelle Donelan, wrote to every vice-chancellor in England to ask them to be flexible in admissions decisions this year.

“You will also be in the best position to take a holistic view, using a wider range of evidence, of an individual’s capability – working with their school or college. If as a result of this, you have reason to believe that [a student] will be able to benefit from moving on to your institution, I would encourage you to allow them to do so,” Donelan said.

Donelan’s letter follows an apology from Nicola Sturgeon yesterday [10 August] for the approach of the Scottish Qualifications Authority (SQA), which downgraded about 125,000 estimates from teacher assessments.

“Despite our best intentions I do acknowledge that we did not get this right and I am sorry for that,” she said.

Scottish education secretary John Swinney is today [11 August] expected to announce how the government will resolve the problem.

Read more: A-level and GCSE results: improving schools can appeal grades

Leave a Reply

Send an Invite...

Would you like to share this event with your friends and colleagues?

Would you like to share this report with your friends and colleagues?

You may enter up to three email addresses below to share this report