Learning from exam results crisis: the way students’ work is assessed needs to change

This year’s school results in England have highlighted how a narrow focus on so-called academic standards can hide much greater issues of justice in how we assess student performance.

The coronavirus pandemic has laid bare underlying problems – but these issues have existed for a long time.

This is an opportunity to go back to some basic principles and devise a system which recognises the impact of assessment on individual young people and their future lives and wellbeing.

Grade inflation

Firstly, while the basic idea that assessment should be robust and credible is right, we have been wrong to accept successive government policies that equate standards solely with avoiding so-called grade inflation, where increased grades are assumed to mean lower standards.

The obsession with grade inflation assumes that past systems have always been right, and that they provide a “gold standard” by which to judge current and future generations of students. It is incompatible with any sense that social progress can be achieved through education.

Secondly, the preoccupation with standards is deeply flawed because it is based on “norm-based assessment”, where results depend on where a student comes in a ranking. This was what the algorithm developed by Ofqual was based on. An individual student’s result is determined by how well other students do, rather than simply the quality of their work.

Under norm-based assessment, the same piece of work could get an A one year and then a C the next, depending on its rank among other students’ work. Even if you do believe that the purpose of assessment is to rank students – which I would dispute – then you can never attach a grade to that ranking because it has no constant meaning. The norm-based underpinnings of this year’s algorithm meant that it was always fatally flawed.

Another approach is to use “criteria-based assessment”. All pieces of work are judged against set criteria, such as knowledge of key facts or strength of argument. These criteria then link to a grade band and any work which meets the criteria gets that grade.

This has two advantages. It ensures that assessment results genuinely relate to the individual student’s performance. Secondly, it ensures a stability in standards. What an A stands for remains largely the same regardless of how many students achieve that standard in a given year.

Ingrained inequalities

We must also recognise that assessment is not neutral. Ingrained inequalities in society and our education system are magnified in our assessment approaches. The assumptions which shaped the algorithm used to adjust marks benefited a certain type of student: those who studied in smaller classes and took less mainstream subjects, such as Latin. In other words, already advantaged students were given further advantages.

In addition, part of overcoming privilege is to break the link between prestige and actual achievement. Doing well in your results should mean that you have a stronger foundation for future learning or work and that you have more choices. Far too many students are taught to use their school results to attend the most prestigious university they can – when another university may offer advantages they had not considered.

Convincing students that the more prestigious their university, the better it must be, actually reduces the range of choice which their good results should have given them. And it is based on false premises, as research is clear that there is no link between prestige and quality of education.

Finally, we should ask whether we have lost the educational function of assessment – and, for that matter, school itself. To see the final outcome of years of schooling in terms of simply abstract grades undermines the personal and intellectual growth that should be the focus of education. Are we so focused on the tests that we have lost sight of the education?

As a society we have allowed the worst of all worlds to develop in terms of how and why we assess students’ work. We have encouraged young people to define their entire self-worth and future prospects on a narrow assessment of their abilities. Then, to make it so much worse, we have broken the essential link between what an individual student has actually done and the mark they receive.

Fixing this is the biggest lesson to learn from this year’s crisis. We must get the fundamental principles of robust and just assessment in place and then find a system that can deliver that.

Jan McArthur receives funding from Economic and Social Research Council, the Office for Students and Research England (grant reference: ES/M010082/1) and National Research Foundation, South Africa (grant reference: 105856).