Politicians like them, parents like them, teachers like them: smaller class sizes are surely a no-brainer when it comes to improving learning, right?

It may seem blindingly obvious that reducing the number of pupils in a class will improve the quality of classroom teaching. But, overall, the evidence suggests that it actually has a relatively small impact on student progress, which is disappointing given the high costs of employing more teachers.

So why don’t pupils benefit from having more individual attention from their teacher?

It is only when class size is reduced to under 20 pupils that we observe noticeable learning gains - equivalent to around three extra months’ progress during an academic year. This appears to be the tipping point after which a genuine change in classroom behaviour occurs: teachers are able to adapt their practice to provide more specific and effective feedback to students or to organise more effective interaction in groups, for example. Smaller reductions, say from 30 to 25 pupils, have little noticeable impact on teacher-learner interaction, though they may help to reduce overall teacher workload and stress.

But are there particular circumstances where a smaller class is needed?

The key is to think about which teaching approaches are best for the learners at a particular point in their development - whether it is whole-class teaching, small group work or targeted one-to-one tuition. A preoccupation with class size merely distracts schools from focusing on the single most powerful factor impacting on pupil progress: the quality of classroom teaching.

A preoccupation with class size merely distracts schools from the quality of classroom teaching

Indeed, some commentators argue that the world’s best performing education systems are those that prioritise the professional development of teachers over the flawed (but popular) policy of reducing class sizes.

Are smaller classes more important for younger pupils?

The strongest evidence of impact relates to younger primary school pupils in the US and the UK. The learning gains from smaller classes of fewer than 20 pupils persist for several years, particularly from the first year of school. The studies also indicate larger effects for lower achievers and disadvantaged pupils. There is limited evidence for secondary pupils, but class size tends to vary less here.

What do I need to know?

The research suggests that reducing class sizes can be effective when teachers are given guidance on what strategies to consider when teaching smaller groups of children. In particular, teachers need to harness the opportunity for more feedback to identify the next learning steps for pupils, or to develop more effective group collaboration for learning. An alternative approach is to work more intensively with smaller groups of children with particular teaching goals, either in mixed or same-attainment groups.

Lee Elliot Major is chief executive of the Sutton Trust and Steve Higgins is a professor of education at Durham University. Together, they authored the teaching and learning toolkit, now the Education Endowment Foundation toolkit