CMPO Viewpoint

By:  Simon Burgess

There is a paradox here. The key research finding is that teacher effectiveness is hugely important for pupil attainment – much more so than class size, IT in the classroom, or any of the other policies that politicians typically reach for. Perhaps unusually, this evidence has made its way into the policy debate fairly quickly and is now widely accepted. Teacher effectiveness is about pupil attainment: a highly effective teacher is one whose pupils make strong academic progress. Teachers do many things for their pupils but policy should surely be focussed first on attainment. In this country, our work 

showed that putting effective teachers in front of a pupil instead of ineffective teachers for all subjects just for the two GCSE years wipes out half of the GCSE points gap between kids from disadvantaged families and the rest. So the prize for a great policy is huge.

You might think that this would mean that teachers are at the centre of the policy debate, with the parties and think tanks producing detailed discussion of the best policies to raise teacher effectiveness. With two main exceptions discussed below, there has been little new policy making about teacher effectiveness.

In fact, this relative neglect is probably justified. It is proving hard for researchers to progress from that initial, crucial, finding about teacher effectiveness. We researchers have not yet solved the questions of how to improve teacher effectiveness. It should be said that this is not through want of trying: this is a major focus of research effort here and in the US. The research agenda lying before myself and other researchers in this field includes: what do effective teachers do? Can this be learned? Or is it about selection? Can we measure effective teaching? What should the teacher “professional journey” be, and what contract forms support that? The difficulties are in obtaining robustly causal estimates, and of running large scale random control trials in schools. Hopefully we will have some of that evidence in time to inform policy soon.

The two areas of policy affecting teachers are the demand for only qualified teachers in schools; and the devolved introduction of performance pay for teachers.

 Unqualified teachers? or continuously re-qualified teachers?

The Conservatives have allowed unqualified teachers, and UKIP support this. Labour say that they would require that all teachers hold or are working towards Qualified Teacher Status (QTS). To me, it seems unlikely that this is a first order issue for pupil attainment. While there is no specific evidence relating robust estimates of effectiveness with QTS status, in general the bulk of the international evidence shows that there is little relationship between the individual’s own academic career and her/his effectiveness as a teacher; nor any consistent gaps in effectiveness by route of entry into teaching.

So in itself it is probably not a big deal. However, it may be a politically astute way of bringing in, as a package, another item in the Labour manifesto. This is the pledge to require teachers to “keep their skills and knowledge up to date” throughout their careers “as a condition of remaining in the classroom”. This is potentially a very important proposal, albeit trailed previously, and if pursued vigorously could make a significant difference to average teacher effectiveness. The key questions will be about implementation, and while there is scope for this to be game-changing, there is also scope for it to simply provide more paperwork for no real gain.

Performance pay for teachers

Since September 2014 all state-funded schools in England, academies and non-academies, are required to have a performance pay system for teachers. Not “allowed to”, but “required to”. The international evidence on the likely effects of this is genuinely mixed. Those on both sides of the argument can point to high quality studies by leading researchers that find substantial positive effects, or no effects.

But in a sense the potentially more impactful element of the policy is that schools are left to determine the nature of the scheme themselves. Designing performance pay schemes is complex with many factors to decide, and many opportunities to make crazy choices. In fact some, maybe many, schools have adopted policy templates produced by others – their LAs, headteacher unions, and so on. But still, the scope for some disastrous outcomes is not negligible.

At CMPO we have investigated performance pay in the public sector for teachers, doctors, and bureaucrats in JobCentres,  and in Customs and Excise (as was), also evidence reviews here.

The two common findings are: firstly, incentives work, sometimes powerfully; but secondly that that powerful effect can be mis-directed if the scheme is badly designed. Having 30,000 schools write their own performance pay schemes is certainly in line with a less prescriptive, more autonomous schools system; it is also playing with fire.