Class size and teaching: width and quality both matter
By Blog Editor, IOE Digital, on 3 February 2017
David Aaronovitch is a good journalist and there is much to admire in an article he wrote for the Times newspaper last week (‘Teachers must get out of their ideological rut’, January 26, 2017). I suspect however that he has never taught a class of children. The line he takes – following the well publicised view of the head of OECD educational policy Andeas Schleicher – is that class size doesn’t matter. David remembers the words of his grandmother: ’feel the quality not the width’.
With my colleague Tony Russell, I am working my way through the carefully collected views of hundreds of primary school teachers, headteachers, Teaching Assistants and pupils, along with careful classroom observations and case studies (part of a large scale study we conducted at UCL Institute of Education), and what stands out are the many ways that class size does indeed matter. We have found that having fewer children in the class tends to mean more individual attention, a more active role for pupils in class, better relationships between pupils, easier classroom management, more individualisation for pupils with SEN, and more classroom engagement amongst lower attainers in secondary schools.
Of course teachers might be misguided and could be accused of being self serving. But there is also the possibility that they are – as the most privileged participants in classroom life – actually closer to the reality of what goes on than those who disparage concerns about rising class sizes.
The important thing to recognise, it seems to me, is that the effects of class size are not simple. It’s a mistake to look for a simple direct correlation between class size and say pupils’ attainment levels, and then, when one finds the association is not strong, conclude that class size is unimportant. There are so many other factors that might be affecting the association including the kinds of pupils in the class, and teacher’s preferred methods of teaching. What emerges from the work we are doing is that class size is a main feature of the classroom environment but that it exerts its influence in complex and little understood ways not always shown up in tests for simple statistical associations. Contrary to what is said in the article, class size also impacts on teachers’ job satisfaction; large classes can be extremely stressful for teachers.
I am in agreement with a main point David makes – that is, the importance of critical thinking and reasoning over rote learning. But the mistake I think is to assume that there is a choice between these things and class size. It is an interesting fact that in a number of countries and regions in East Asia, e.g., in Shanghai, Hong Kong, Taiwan and Macau, despite high scores on PISA tests, there have been government and state led initiatives to reduce the number of pupils because of a wish to change predominant teaching approaches and bring about more critical thinking and collaboration between pupils. There is an understanding there that class size reduction and changes to teaching are best seen as working together, not in opposition. Width and quality both matter.
The worrying thing is that despite the ever continuing debate over class size and the strong views expressed, we in fact have very little dedicated research on the topic on which to base informed judgements.
2 Responses to “Class size and teaching: width and quality both matter”
The man who invented the PISA tests | brentwoodteachingandlearning wrote on 26 February 2017:
[…] Schleicher speaking at the Global Education Summit Response to the Aaranovitch’s article from the IOE […]
The research that I recall most clearly may now be overtaken. But, when statutory class size limitations at KS1 were being considered in 1997, that which existed suggested that class sizes mattered – but only when such large reductions were made that they could not be afforded in the public sector.
The statutory limitation was introduced nonetheless, at a smaller level, to fulfil a political commitment. There is little evidence that this, on its own, made much difference. That said, measurements were made against movements in educational standards of achievement, not the other factors to which you rightly refer.
Is your work covering all the key stages? I imagine there may be a differentiated effect in the various age groups.