For many teachers, large classes present problems which adversely affect their practice and their pupils’ learning. This is what our surveys show. But researchers and commentators often have a different view. For them the class size debate can be summed up with the question: does class size affect pupil attainment?

As we show in our new open access book, ‘Rethinking Class Size: The Complex Story of Impact on Teaching and Learning’, published by UCL Press this week, researchers (contrary to a practitioner view) commonly find that the statistical association between class size and attainment is not marked and so conclude that class size does not matter much. This has led some to even suggest that we could raise class sizes, and instead invest savings in professional development for teachers. Currently, in the wake of the Covid pandemic and teacher absences, there are reports of some schools being forced to create supersized classes of 60 pupils.

The view that class size is not important is probably the predominant view among researchers and policy makers, and so they may be relatively relaxed about increases in class size. We therefore need – more than ever – good quality evidence on class size effects, but in our view much research is limited and leads to misleading conclusions.

We identify three problems. One issue is that the exclusive concern with the association between class size and academic attainment in first language and mathematics is limited.  It may be that teachers in large classes prioritise these subjects (and this might help account for the relatively small difference in pupil academic outcomes, compared to smaller classes) but this raises a question about the possible way large classes affect other pupil ‘outcomes’, e.g., creativity, independent and critical thinking and motivation to learn. There is very little research that addresses this question.

Second, class size is not an ‘intervention’ like the distinct pedagogical approaches with which it is commonly compared in meta-analyses of research findings. It is not something one adds to the classroom like a reading scheme, but rather one aspect of the classroom context to which teachers and pupils have to adapt.

The third and most important problem with the usual research approach is that it does not take into account ways in which class size affects classroom processes, particularly teaching. At the heart of the claim that class size does not matter is the assumption that teaching is essentially about conveying information to students. If teaching was just about lecturing, then class size is much less important. Good teachers can control and lecture a very large class of 60. Indeed, in higher education we routinely see massive classes of more than 100, listening to a lecture. No doubt, some people would argue that this kind of teacher directed, didactic style of teaching is preferable anyway.

But this is a very narrow view of teaching, especially at primary school. In ‘Rethinking Class Size’ we present results from our own 20-year research programme involving extensive classroom observations, national questionnaire surveys and detailed case studies – probably the largest dedicated study on class size worldwide. We show that large classes: make differentiated teaching and individual support more difficult; mean reduced knowledge about individual pupils; make classroom management more demanding; reduce the amount of educationally valuable activities like practical and investigative tasks; increase the demands of marking, report writing, planning and preparation; and increase teacher stress.

Large classes are particularly demanding given the diverse pupil intake found in many UK schools, as well as policies of inclusion, which mean more individual support is often needed. Worryingly, we found in our research that it is low attaining pupils and those with SEND who are most disadvantaged in large classes, for example in terms of classroom engagement.

Our central conclusion is that the effect of class size on academic attainment is not a direct one, as is assumed in much research, but interconnects in complex ways with classroom processes like the balance of whole class, group and individual teaching, classroom management, relations between pupils, classroom tasks, and administrative activities like marking.

To understand the effect of class size we have as researchers to do more than produce a statistical correlation with academic attainment because without information on what is going on in classrooms we cannot reliably interpret the correlation. In the book we develop a distinctive social pedagogical framework to account for the complex contextual, dynamic and relational factors at work.

Obviously at a time of national emergency, schools have to adapt to staff and pupil numbers as best they can, and large classes may in the short term be necessary. Teachers will also find inventive ways of teaching large groups. But we should not assume this is a long-term solution. We need to be aware of the implications for teaching and the breadth of learning.

Peter Blatchford is Emeritus Professor in Psychology and Education at the UCL Institute of Education. Anthony Russell has had an international career in teaching, teacher training, curriculum development and education research.