EEF Blog: Assessing learning in the new academic year (Part 2 of 2) – how school leaders can best support pupils to regain lost learning

Having explored the different purposes for school assessment in the academic year ahead in part 1, here EEF senior associate Prof. Rob Coe looks at the role of quality assessment in helping teachers understand how best to help your pupils regain any learning lost during Covid-19 school closures.

Before the Covid-19 pandemic, one of the most high-profile topics in English education proved to be the content and aims of the school curriculum. Too often, curriculum was the subject of debate, but assessment received too little attention.

Any consideration of curriculum should be inextricably bound to understanding quality diagnostic learning assessments. If you want to use assessments to tell you whether students have learnt something, you need to be clear what it is you wanted them to learn in the first place. Clarity about the intended learning (ie, the curriculum) is crucial if we want to create or select questions for an assessment that will tell us what has been learnt.

Ideally, this includes thinking about the sequencing and anatomy of that learning. When a learning aim or task is complex, clarifying the curriculum includes identifying sub-skills or components on which the whole performance depends, specifying prerequisite knowledge and competencies that must be in place, and knowing how to combine these elements to build up the whole. If we understand this anatomy, we can create assessments that are much more diagnostically helpful. 

An example: a not-great history essay

For example, suppose we give a student some source material and a question to discuss and ask them to write an essay in secondary history.

We mark the essay and find it isn’t great, but we don’t really know why, so it may be hard to give useful feedback, activities or instruction. More useful would be if we knew that the student did not really understand the source material or its relevance; or if they understood it, but had difficulty constructing a logical and coherent argument; or they could make an argument, but had difficulty expressing those ideas in an essay format. And each of those sub-competencies could be further broken down. 

If we understand how a complex task depends on these kinds of components, then we can design an assessment task that provides much more actionable information. If we are not sure they have mastered the components, we should assess the components; once we know the components are in place, we should assess the whole. 

Support resources for schools

Support resources for schools

Tools for schools to help support home learning and maximise the impact of work set.

And so, in September and October, we may be assessing learning through targeted verbal questions, short answer quizzes, and paragraph length responses, before then assessing the whole in an extended essay.

In most cases, the kinds of assessment tasks that give the best diagnostic information are quite different from the tasks that are used in summative exams like GCSEs or SATs. If we choose or construct assessments mainly from those sources, we are unlikely to create great diagnostic, formative assessments. Given teachers will be considering addressing curriculum content in the autumn, a programme of mock GCSEs or SATs may prove ill-timed and unproductive. 

How assessment can inform teachers’ actions

For most classroom assessments, the outcome informs a choice of actions: Should we reteach that material to the whole group, or move on? What activity should this student do: Go back to relearn some prerequisite element? Practise some more similar examples? Correct minor errors? Prepare to do a parallel version of the same assessment in a week’s time?

These questions are even more salient for teachers now, given so many pupils have missed significant curriculum time, and the likelihood that differences between them will have grown wider.

Ideally, we have thought about the range of choices that are likely to be most helpful pedagogically – and that are actually feasible – when we design or select our assessment, so that it gives us the information we need to support those decisions. 

In the academic year ahead, teachers and school leaders’ decisions will be best informed by quality assessments. Effective granular assessments, like retrieval quizzes, can hold the attendant benefit of being low stakes, which can help teachers re-establish teaching the school curriculum without the potential stresses that can attend mock exam type assessments. 

Support resources to share with parents

Support resources to share with parents

Resources to share with parents on how to support home routines.

Naturally, thoughts may turn to when we should assess students. It is likely that the first weeks or months of the autumn term, we see teachers focus in on the aforementioned formative and granular assessments. Such meaningful and manageable approaches mean that students can reconnect with their teachers and their peers and teachers can re-establish curriculum and assessment.

There is an important difference between once-secure learning that has been forgotten through disuse and material that has never been learnt properly. If we assess too soon, without allowing time for familiarisation and settling in, we may not be able to separate these two. Some students may quickly regain any lost learning, whereas other students may need additional support.

It is the careful sequencing of well-discriminated, purposeful assessment that can best support pupils to regain lost learning and go onto thrive.

Key questions to consider about assessment quality and purpose

So, what questions could school leaders consider about assessment quality and purpose in the next academic year?

1. What assessment tasks will give us the best diagnostic information about the prerequisite knowledge, skills and competencies we want our students to develop through the school curriculum? Do our assessments target the anatomy of learning, not just its final product?

2. Are we clear about the kinds of choices we want information from our assessments to support? Are these choices that we actually can and do act on? Do our assessments give us the information we need to inform good choices?