How on earth do we run 2021 admissions now?

The majority of UCAS applicants for 2021 entry will apply to university before the extended 29 January deadline, the majority will use predicted A level and Higher results as a basis for these applications.

For most UK providers, this is the majority of students that will be admitted.

Each form includes a range of information about the applicant – from simple things like where they live and study through to a personal statement, a reference from the school, and those all-important predicted grades. The dominant model of admissions sees all of the above as potential indicators of actual grades, which in a normal year would be made available to universities in mid-August. This is not, based on the first six days of data, a normal year.

At this stage admissions staff in universities don’t know exactly what further information about each prospective student will become available, don’t know when this information will be available, and don’t know how well the information we do have will predict it. Although Ofqual and the Department for Education are working hard on developing a convincing programme for producing exam replacement grades that is – to be fair – also what they did last year.

In the Commons, Gavin Williamson, hinted that “we will put our trust in teachers, not in algorithms”. We’re looking at a system of teacher assessed grades, with training and support for teachers provided to ensure fairness and consistency. There is still a lot of detail to come, but apparently national moderation will be a less visible part of the process.

Reform is coming

In England, the Secretary of State for Education has said a lot about university admissions in the last three months. His stated preference is to move the admissions system to a post-qualification system, although this would never have been for the 2021 cycle, so the default is to use a system that supposedly “limits the aspirations” of students.

The other major policy pressure on admissions is to shift away from “unconditional offers” – which are offers that do not use A level results as the determining factor. Unconditional offers are, according to Gavin Williamson, a “damaging practice”.

In summary, actual high-stakes exam results (and not coursework or anything else, thank you Mr Gove) seems to be solidifying as the only means by which students are admitted to university courses. There will be no exam results available in 2021 because there will be no exams in 2021, just as there were no exam results available in 2020.

Offered with experiential evidence that an exams-focused transition between level 3 and level 4 is not resilient, the stated policy of the Westminster Government is to take steps to make it even less resilient to future disruption.

Look to the west

In making the recommendation that A levels exams in Wales should be cancelled for 2021, the Design and Delivery Advisory Group recommended three pillars that could be used to generate grades:

• Non-exam assessments: In Wales A levels are not assessed by a single final exam, so – where possible and adapted for equity – these will continue.
• Internal assessments: These will be developed by the Welsh Joint Education Committee (WJEC) but administered at the discretion of teachers as part of the normal course of delivery. These will take place between 22 February and 23 April, with marking and moderation taking place at teacher and school level. WJEC will moderate a sample of these assessments.
• External assessments: These are set and marked by WJEC between 17 May and 29 June, and are based on material that is likely to have been covered by students during the year. Teachers and students will be aware of the topics covered in these assessments in advance.

A combination of marks from these three pillars will be used by WJEC to develop a final moderated mark.

Look to the north

In Scotland, marks will be based on teacher assessments of work during the year. The approach for Highers will likely follow the same principles as National 5 qualifications, but we are still waiting for clarification.

Here the Scottish Qualifications Authority (SQA) will produce assessment resources that can be used in schools at the discretion of teachers, with standards maintained by quality assurance at school and local authority level. SQA will feedback on samples of these completed assessments and other examples of student work (assessment evidence) during the spring, and schools will use this feedback to moderate the production of marks to be submitted to SQA. These submitted marks may be checked again by SQA before moderated marks are published.

We learned today that A level exams will be cancelled in Northern Ireland, but we don’t yet know how assessment will be conducted for 2021.

Back to the fundamentals

Over the back end of last year, we published a range of articles about how admissions happen in other countries. The received wisdom is that the UK is an outlier in relying on predicted grades – but it quickly became apparent that the UK’s reliance on a system of national pre-entry qualifications is largely unique. Teacher assessment, in the form of the grade point average, is as close to an international default as we are going to get.

In England, for whatever reason, we don’t tend to trust teachers to make these judgements – even as a basis for later moderation (things are less stark elsewhere in the UK). Which is odd as we trust them to teach, and otherwise support, our children. It is notable the criticisms last summer were of a “mutant algorithm” (that attempted to normalise attainment based on demographic norms) rather than the considerations of teachers. Robert Halfon is happy to criticise these unmoderated results of professional judgement as “grade inflation”.

If we are serious about ending “limits on aspirations” then we need to look at the practice of moderation based on the results of previous cohorts, and consider the impact that “baking in” existing inequalities has on aspiration. Students from disadvantaged backgrounds do worse at A levels, Black students do worse at A levels, disabled students do worse at A levels, boys do worse at A levels. I refuse to believe that these groups are simply less intelligent, or less likely to succeed at university.

In the summer we saw the impacts of this moderation in higher resolution than previously, and we moved instead to honour the professional judgement of teachers moderated at school level. It looks like this is also the plan for 2021. Why can’t we do that every year?

What now?

For admissions professionals, and more widely for universities, the challenge is making use of the information already available from the UCAS form – and somehow translating that into an offer that is conditional on whatever grade is finally awarded.

An approach that attracts me would be a form of “conditional unconditional” offer, a university would make an offer based on the information that is available currently – with the potential to rescind that offer should later evidences (the late summer A level grades or equivalent) demonstrate that a candidate would not be suitable for a course after all. Some providers and professional bodies may choose to develop their own aptitude tests (not knowledge-based exams) or recommend tests developed by others. This approach would massively reduce anxiety for current year 13 students, and would make planning and capacity development easier for providers. Every year.

If that sounds radical, I can assure you that it is actually very close to the system used in the US. It would, of course, currently be forbidden in England.