Ending the use of predicted grades in UK admissions could harm disadvantaged students’ hopes of winning a place in a selective university rather than aid them, it has been warned.
The Westminster government is consulting on moving to a system whereby students apply to university after they get their actual grades, amid concern that the use of predicted marks to assess applications and make offers is imprecise and disproportionately harms applicants from the most disadvantaged backgrounds.
But a published by the 黑料吃瓜网 Policy Institute on 18 March claims that, according to available data, “these two supposed problems” are “very likely false”.
Mark Corver, co-founder of consultancy dataHE and a former director of analysis and research at admissions service Ucas, says although it was true that forecasted grades tended to be higher than the actual grades achieved, they are “better seen as a reliable estimate of the highest grades an applicant might realistically get through intrinsically uncertain exams”.
This was because in about one in four cases students do hit at least their predicted grades and there is a very high probability that they will get results that only just miss the forecast.
“If you had to choose a single statistic of potential to underpin good matching of university offers then this would probably be it,” he says, adding that the pure uncertainty of exams means that making accurate predictions for individual students was inherently difficult.
“Perhaps clarifying the nature of predicted grades by expanding the current single value to a likely upper and lower level of attainment would help this be more widely understood,” he adds.
Dr Corver also addresses the argument that students from disadvantaged backgrounds are more likely to have predicted grades that understate their final results.
He says the data suggest that some of the most under-represented groups in selective universities, such as black students and those from low participation neighbourhoods, actually have a larger overvaluation with predicted grades than the average.
Using a simplified model where the most selective universities admitted the top 30 per cent of students, he shows that using exam results would lower the entry chances of applicants from the most disadvantaged areas by 5 per cent and by about 20 per cent for black students.
Dr Corver also adds that analyses suggesting pockets of top-performing students from under-represented groups are much more likely to get underpredicted grades miss the fact that this is “simply reflecting that there are more [disadvantaged] students with lower predicted points”.
“The belief that predicted grades harm equality is not supported by the data,” he concludes. “The pattern is mixed across under-represented groups, but overall predicted grades are probably more an aid than a hindrance.”
Elsewhere in the Hepi report, Mary Curnock Cook, who was chief executive of Ucas during a previous attempt to look at post-qualification admissions, says that “paradoxically” predicted grades were “highly predictable in their unreliableness”.
She suggests that rather than the risks of moving wholesale to using exam results, it might be better to either support teachers in making better predictions or use banded grade forecasts.
But in another chapter, James Turner, chief executive of the Sutton Trust, reiterates that its research had highlighted it was “bright low-income youngsters who are most likely to have their grades underpredicted, potentially contributing to an ‘under-match’ of students in higher education and suppressing social mobility”.