Covid-19 blog:

Covid-19 could result in long-term change to educational assessment in the UK, but is the teaching profession up to it?

The Covid-19 pandemic has required unprecedented arrangements across all facets of our society. In the world of education, schools have remained open to cater for the children of key workers, but have followed colleges and universities and moved teaching into the virtual world. Schools, further and higher education establishments that had already well-developed technology platforms were clearly at some advantage and examples of cooperation and sharing of resources has reflected well on the teaching profession. Even so, not all students have access to appropriate technology and gaps in achievement will no doubt increase.

However, running tests, examinations and end point assessments has proved to be much more challenging. Education ministers were quick and decisive in cancelling all major tests and examinations in 2020. The guidance from the DfE and Ofqual has been sound and equally speedy setting out how students will be awarded general and vocational qualifications. These arrangements will protect the current tranche of candidates and ensure that they are not disadvantaged by the exceptional events of a pandemic. Indeed plans are already being developed for 2021.

But the arrangements have shown how dependent society has become on externally set and marked exams and tests used in a wide range of qualifications and settings. It also brings into stark relief the paucity of assessment generated through the professional judgment of teachers - be it in the academic or vocational fields. In short, teachers and lecturers have been relieved of a key part of their professional status. In some respects it’s easy to see why. The reliability of teacher-based assessments is at best questionable with concerns over bias and trust, worries driven by an accountability system that places test and examination results to the fore. Leaving teachers at liberty in such a system generates accusations of teachers ‘marking their own homework’. In general qualifications, the trend has been to minimise the role played by teachers in high stakes assessment through the reduction or total removal of coursework and controlled assessments. And there has been a clear rationale behind the removal or reduction in non-examination assessments such concerns over plagiarism, over-assistance from teachers and heavily rehearsed then regurgitated orals. Yet the approach taken in 2020 to run the summer series founded on the generation of centre-based assessments determined by the professional judgment of teachers appears to be a complete volt face. But let’s be clear, the moderation process being put in place for summer 2020 is most reliant on teachers’ rank ordering rather than the grades. Where the statistical evidence is strong, there will be less reliance on the grades; conversely, in small centres or in subjects with smaller entries, the statistical evidence will be weaker and there will be more reliance on the grades.

We have also seen a similar scenario in national curriculum assessments. The initial approach presented in the early 1990s was to assess pupils through Standardised Assessment Tasks (SATs) administered during normal teaching time and marked by teachers. That scenario has long gone – though use of the term ‘SATs’ has (incorrectly) lingered - after challenges by teachers over workload and payment. As a result, external assessments were introduced following the Dearing Review in the mid-1990s and more recently, the days of collecting teacher assessments at key stage 1 and 2 have been numbered.

In any assessment model, there is a tension between validity and reliability and the most recent trend has been to increase reliability through reliance on externally set and marked assessments. This has arguably reduced validity, as written examinations are limited in what they can assess. Meanwhile the teaching profession has largely gone with this move as the argument has been based on reducing teacher workload. Reducing unnecessary workload is obviously to be welcomed; but shouldn't assessment be a central facet of what the teaching profession actually does? Assessment is the fulcrum of teaching and learning, otherwise, how do we know any learning has taken place? And if we have reached a point where we cannot trust assessments undertaken by the teaching profession in high stakes contexts, is it still worthy of the title ‘profession’?

The research evidence on marking reliability in high stakes assessments suggests all forms of assessment fall far short of perfection. Yet it is also true that we lack thorough research into how teachers and others assess the work of their students and how we might improve the reliability of such assessments. We also have a generation of teachers who lack expertise in curriculum development and assessment theory and practice. They were taught and now teach in a system built around a National Curriculum and associated tests and examinations generated and marked by external bodies. And unless we can find an alternative model, this will on balance quite reasonably continue.

Some, even many, will dispute this interpretation and argue strongly that teachers are up to the job when it comes to assessment. But where is the compelling research? Where are the tried and tested alternatives? As it stands, 2020 might present such an opportunity to re-think our approach and test our assumptions. For one very unusual year, the whole examination and testing system will depend on the professional judgment of teachers to grade and rank order their students. Accountability measures such as national performance tables and internal performance management are suspended. A system to statistically standardise these outcomes is under development and students will get their results in August as usual. So if teachers can suspend their almost reflex reaction to systems of accountability and performativity, we will have some very interesting data to examine in the coming months.

This will of course be a massive challenge. Agreeing and working to a set standard within a school or college will be demanding enough, let alone rank ordering quite large groups of pupils: rank ordering a class by one teacher is one thing, doing this across large year groups is something else. In my view, centres will need to determine and exemplify what they understand as the expected standards for each grade. This should be done through the use of ‘standards files’ that contain examples of what is expected, be it for example through reference to grade descriptions in subject specifications, examples of work produced by awarding organisations, and written work, videos, the outcome of mock exams or practice tests generated in house. And someone needs to be the arbiter, the ‘chief examiner’ who has the final say in determining ‘the standard’. Then, and only then, can a process of moderation be followed. Moderation is checking that the agreed standard is being applied and if not, doing something about it. It is vital that the distinction between standardising and moderating is understood. The challenges here should not be underestimated. Even in normal times, setting standards and then moderating assessments within a school or college would be a challenge, but given concerns over social distancing for example, this raises the bar. But this is not insurmountable and centres that have well developed technology platforms and electronically held standards files will have an advantage. Technology also provides the means to share exemplified standards between institutions or across sites – an approach already adopted by many to share curriculum materials of teaching resources.

The chosen approach will of course need to be verified by the Head of Centre whose role this year is more important than ever in being accountable for ensuring that a process of evidence-based professional judgment has been implemented. In schools, colleges and places of work and study that have access to a Chartered Educational Assessor, this process will be enhanced by their knowledge and expertise. Indeed, when the government of the day backed the formation of the Chartered Institute of Educational Assessors (CIEA), it was the stated goal that every school should have, or at least have access to, a Chartered Assessor. If this were the case, we would now have in place a national system of quality assuring this year’s assessments.

So we will see this year if teachers can step up to the plate and deliver defensible professional judgments. It offers a one-off chance to test the quality of teacher-based assessments; to save millions of pounds spent every year on external examinations and, importantly, the chance to re-instate a higher level of professionalism. Conversely, it might prove otherwise. But we should think longer term. God forbid we should see anything like this in the future, but clearly it is a possibility and it may be some time before schools, colleges universities and other places of education and training get back to anything we consider ‘normal’. Indeed most buildings designated as places of learning are designed to ‘mass educate’: social distancing was never included in in the briefs developed for architects. So schooling is likely to take a different format for quite some time. We should therefore challenge our assumptions and look at contingencies.

This is not a binary choice between internal and external assessment. Tests and examinations should be part of our repertoire and we should recognise that our awarding bodies and the Standards and Testing Agency (STA) produce some excellent assessments in a well-regulated publicly accountable system. But the arrangements for 2020 present us with a rare opportunity to take stock of what it is that educational assessment is really trying to achieve and ask ourselves how best this might be done.