This article is more than 3 years old

DfE’s announcement on A level grades is another headache for university admissions teams

David Kernohan predicts an interesting admissions cycle following the Department for Education's announcement on A level grades
This article is more than 3 years old

David Kernohan is Deputy Editor of Wonkhe

As of today we have no way of knowing how this year’s A level results will compare to last year’s – or any year’s.

While new proposals from DfE and Ofqual look like a sound way to take account of a variable learning experience and curriculum coverage, and the idea of trusting the teachers that have worked so diligently over the past 12 months rather than the algorithms that did not, has an obvious appeal this route leads to a monumental headache for admissions teams.

As yet, there are no clear signs that DfE have considered the impact on the higher education sector short of thinking about allowing space between a week-earlier results day and a potentially modified confirmation date for a more comprehensive appeals process.

Previously on examnishambles…

After an entire cohort of Covid-bedevilled English A level students reacted angrily at the way a place- and characteristics-based Ofsted algorithm mauled their Centre Assessed Grades, DfE decreed that those same CAGs would become the awarded grades for 2020 – sending UCAS and the higher education sector’s admissions teams scurrying back to the 2020 cycle drawing board. Admissions rose, particularly to high tariff courses, as higher-than-expected grades meant more applicants met the terms of their conditional offers. And everyone forgot about vocational qualifications.

Gavin Williamson, who somehow remained Secretary of State for Education during this period, immediately decreed that in-person exams would go ahead as normal during 2021, but it turns out life comes at you pretty fast, and in early January it emerged (as we were still digesting being back in a nationwide lockdown and schools being closed, despite being told that neither of these things would happen) that 2021 A level exams would not be happening. Ofqual and DfE then released a consultation, which prompted more than 100,000 responses – a sample size far larger than the voter surveys Number 10 would otherwise be looking at to determine policy.

Fifty-two percent of those responses came from current schoolchildren.

Now read on

The consultation hedged a populist “teachers not algorithms” headline with a series of suggestions around the use of externally set papers sat under exam conditions that felt like an attempt to have one’s cake while it sat rotting in a lorry stuck on a Kentish airfield.

But the alternative – teacher grades, based on collected evidence which may be but does not have to be derived from externally developed questions – does leave some other questions unanswered. Though the idea of assessments being made by the educational professional closest to the candidate in question makes an instinctual sense – and indeed, seems to function perfectly well in many other parts of the world – it prompts two other perceptions. One is of favouritism (or its absence). The other is of comparability between HE applicants, schools, and cohorts.

On the first count – appeals will be available in 2021 to all students. These start at centre (school, college, MAT) level and focus on whether the process of grade determination has been followed correctly. If further satisfaction is required, this is escalated to the exam board – who would use the evidence gathered by the teacher to make their own assessment of the grade.

All this appeals action would be expedited for candidates where a university place rests in the balance – UCAS will be having words with providers on extending the confirmation date to allow for this. And – for the disappointed – an autumn 2021 exam series would allow for at least January 2022 university entry.

Admissions vectors

All of this makes for a very interesting 2021 UCAS cycle – as, far from the original plan of tying this year’s grade profile to last year’s, we are now in a situation where we have very little control over how many people get an A* at chemistry A level. Offers based on predicted grades are, canonically, a means of inspiring applicants to work to obtain those grades or above even though actual evidence suggests most underperform on the day. For 2021, this assumption may not hold true.

On the popularly understood evidence of last year, teachers were more generous in their grading than the exams would have been. However, it should be added that the exam results would have passed through an algorithm (very similar to the PM’s “mutant” one) to keep the overall grade pattern within their decade-old boundaries. It is entirely likely that the uncompensated 2020 exam grades would have looked a lot like the CAGs.

Extremely good for actually recognising attainment (and all the other stuff education researchers go on about) but presenting a problem for admissions teams. Generally we know a certain proportion of a cohort will get a given grade, and we can use that in setting our offer thresholds. Here, we have no way of knowing if 200,000 people will get an A* in A level chemistry – or just 40.

So one of the big unanswered questions is whether universities would be supported in taking on far more (or conversely far fewer) undergraduates in 2021 than would happen in a predictable year. Pressure will be put on capacity for the lucky ones (just in time for Wave 4 of Covid-19) and on finances for those who manage less well. Then we have entry to formally capped courses like medicine, and informally capped courses like Broadcast Media (they’d have 57 students to a microphone), to consider.

Open the box

Teachers can take account of various aspects of pupil performance in determining A level and A level-like grades (professional vocational stuff will be assessed by “Covid-safe” exams instead). This could include in-class assessments, written and practical work, mock exams, those exam board questions (that will work just like in-class assessments and be marked by the teacher) and general classroom performance. This choice exists so teachers can assess based on what has been taught.

The external questions are the controversial ones – 51 per cent of parents and carers like the idea, as did 69 per cent of teachers. However just 26 per cent of pupils were keen – likely a huge factor in the decision to go along with what we are told is a majority preference for optional papers.

All of this evidence needs to be carefully collated and preserved against future appeals – there’s no information yet as to whether these appeals will be at a financial cost to the student or school. There will be guidance to teachers from exam boards about applying grades, and this will be bolstered by process decisions made within school. There will also be some sampling of this evidence by exam boards.

Students get these grades on 10 August, only very slightly earlier than a regular year. Pupils will know about what evidence a teacher is likely to use by early April, with the optional external questions available around Easter. Grades are submitted on 12 June, with external quality assurance done during July.

A proposal

While currently willed together with glitter, hope, and fancy string, A level results still remain the only factor by which admission to most courses can be offered by higher education providers. With teachers moving to the outer reaches of the filing cabinet to determine these grades, should we not be encouraging universities to take advantage of Zoom interviews, bespoke assessments, portfolios, references, and UCAS personal statements to make offers? And – given that most of these rely on information that is known already – is there a sound reason why these should not be unconditional? And could some of these offers only become conditional if accepted, so providers would have some idea as to how many students would be arriving in September and thus how many more they could pick up in August?

And – to be frank – couldn’t we do this every year?


Gavin Williamson’s letter to Ofqual has been released, setting out some more detail and confirming A level results day as 10 August. One notable new aspect is that we learn that exam boards will be responsible for setting out a list of potential sources of evidence for student performance, and how these sources should be used. We also get detail as to what materials exam boards will provide – grade descriptors for at least alternate grades, sets of questions, mark schemes, example answers – to help teachers ensure grades are comparable to previous years.

5 responses to “DfE’s announcement on A level grades is another headache for university admissions teams

  1. Thank you… really enjoyed reading it, especially the last line: “couldn’t we do this every year?”. Indeed. So here’s a mind-game. Imagine a world without school exam grades… how would things work?

    1. Well the irony was not lost on me that this morning my institution, in line with many others, agreed to accept US UG applicants on the basis of high school grades only since SAT Subject Tests have been scrapped. And last week we had a request from a UK school to accept their own internal GCSE equivalent awards instead of the real thing. Personally I would rather see the UK retain the broad curriculum at Level 2 but do something far more creative at Level 3 – retaining a broader basket of assessments since this is much more aligned to how HE works now. ALs as they are look old hat (obviously a gold standard hat). As for the numbers game – the question is have the offers we’ve held back accounted for enough of that and who has lost out by the selective courses having to turn down so many? Our WP metrics are looking very good but there are other segments I am worried about… My head still hurts!

  2. It would seem that Lecturers may have to do impossible work in the first year of courses to get the new intake of students to the same place, especially if students have missed some study and their grades are inflated to compensate. This can also be found in level 3 courses after inflated GCSE grades.

  3. Yes, we could do this every year, and, I guess the world would remain a sane place for all concerned. What a thought.

  4. Quite like your suggestion that we should think about – and explore meaningfully – what is involved in ‘doing this every year’ rather than seeing the last two cycles as a departure from a ‘fit for purpose’ norm – see various contributions by Dennis elsewhere on the issues -particularly but not only in humanities subjects – with the reliability of AL grading.

    Very interested in the point made by Katherine that HEIs will happily accept US educated applicants this year based on high school grades given the scrapping of SAT Subject Tests. Whilst the numbers involved are obviously very different from those progressing with UK quals this example does concede the principle that school/centre based assessment can be trusted as a basis for admission; if this is sen as good enough in US and other overseas contexts why should not this be explored more deeply at the same time as the HE and pre-HE sectors respond to the DfE PQA consultation thereby enriching substantially this otherwise highly limited ‘quick fix’ procedural proposition which by itself simply ‘bakes in’ an alleged AL high stakes examination gold standard.

Leave a Reply