The Class Size Debate: Is Small Better?

We have seen that research on the effects of class size differences needs not only to look at relations with academic outcomes, but also to provide insights into classroom processes that might explain why smaller classes differ from large classes.

In Chapter 1 I argued that research on teaching and classroom processes has tended to assume an underlying direct model, in the sense that the focus has been on the effect of teachers on pupils' attainments, while in fact teachers (and pupils) will necessarily need to adapt to the classroom context, which will include features such as the number of children in the class. But there are also contexts nested within classrooms. Before we examine links between class size differences and teacher and pupil behaviour, we first need to explore the connections between class size and one main within-class context, namely groupings within the class.

All pupils in classes are grouped in some form or another, perhaps especially at primary school level in Britain, and the group is a main context for teaching and learning. Let us take an example of a small class of 20 pupils. Here a teacher may have children seated in groups of 5 around four tables. If the teacher has given out the same worksheet to all children there will be four groupings. If later in the lesson the teacher then goes over the worksheet with the children as a class, there will be only one grouping; even though children are seated at four tables their attention and activity is focused on the teacher and the rest of the class as a whole. This example illustrates that a grouping at any moment in time may be an individual child, a pair (dyad), triad, or anything from a small group of four children to a whole class grouping.

The benefits or disadvantages of different grouping practices has aroused a good deal of comment and research in Britain and elsewhere. In Britain, 'progressive' primary education practices, including small group work, have been criticized as being ineffective (Alexander et al. 1992; Ofsted 1995b), and it has been recommended that teachers adopt whole class teaching methods, albeit

-34-