Getting the science straight: the schools minister’s suggestion that private schools convey little academic advantage does not stand up to scrutiny

Francis Green. 

A recent report from the Sutton Trust reveals that positions of public influence are still disproportionately cornered by the privately educated, with little progress since their previous report. So the Johnson – Hunt (Eton – Charterhouse) contest to be prime minister is but the tip of an iceberg. It is curious, then, to find Schools Minister Nick Gibb and genetic psychologist Robert Plomin seemingly agreeing on an ungrounded assertion: that there is little difference in the academic outcomes of state and private schooling in Britain, and that private is assuredly not worth the money.

The science is not on their side.

Professor Plomin asserts that: “Even though schools have little effect on individual differences in school achievement, some parents will still decide to pay huge amounts of money to send their children to private schools… I hope it will help parents who cannot afford to pay for private schooling … to know that it doesn’t make much of a difference in children’s school achievement. Expensive schooling cannot survive a cost-benefit analysis on the basis of school achievement itself.”

The root problem with this assertion is that it lumps together the arguments and evidence about selective state schools with the argument and evidence about private schools. This conflation makes no recognition of the enormous resources gap between Britain’s private and state schools – something like three to one and rising. It beggars belief that such a large resources gap would have no effect on the academic outcomes (let alone the other outcomes) of education.

Plomin’s assertion about private schools rests on just one scientific paper. The paper focuses on one measure of educational outcome, the average of Maths, Reading and Science scores at GCSE. Like others, the authors find that it is important to control for socio-economic background and prior achievement when comparing the outcomes for children in different school types.

Their contribution is to add a summary measure of genetic influences on years of education. A ‘polygenic score’ is derived from a large number of genetic variants each of which have been found in another survey to have a small correlation with years of education. This additional variable explains some more of the variation in children’s achievement. After all controls are included, there is a statistically significant, if modest, average effect of private schooling on their measure of GCSE outcomes. It is not stated whether the size of this average effect is modified by controlling for the polygenic score after socioeconomic background and prior achievement had already been controlled for.

Yet there are at least 16 other studies by economists and sociologists which have concluded that private schools convey both significant educational advantages – often broader than just a few core subjects – and subsequent benefits in adult life. Few of these studies are acknowledged or recognised by Plomin. None are as good as a randomised controlled trial, but collectively they tell a plausible story because there is a clear mechanism for private schools to be more effective – primarily, their vast resource superiority. The overall conclusion is that private schooling has modest but significant cumulative effects on academic outcomes at every stage of schooling, and substantial long-term benefits. The privileges gained by the privately educated cannot easily be explained away by genetic factors, as suggested by Professor Plomin.

One can only assume that Plomin and co-authors must have thought these other studies are flawed because they do not include polygenic scores as control variables. Yet most are based on high-quality data with good controls for prior cognitive ability and socio-economic background. They can’t be ignored.

Let us not, therefore, pre-judge the evaluation of private schools’ effects. Paying huge fees may not turn out to be cost-effective if one wants to take a pecuniary attitude to it all; but that is by no means certain. Our latest study, presented at a recent international conference of education economists, finds a 17 percentage point wage premium for privately educated alumni once they reach the age of 25. If replicated at a later age, that is the sort of long-term advantage that can begin to warrant the ‘investment’ that many rich parents choose to make.

One can perhaps understand why a schools minister might like to talk up state schools, especially if his party has been in power for some time. Yet the criticism that private schools are a waste of money leads to a very conservative (small “c”) position: if you want the system to change, one has only to properly inform parents, then wait and watch while they desert the private sector. This approach has been used to justify a lack of reforming action at more than one point in history over the last century.

For critics like myself, David Kynaston (my co-author on Engines of Privilege and very many others,  the science is in support of what private school leaders and their customers believe to be the case: the schools are, on the whole though with plenty of exceptions, educationally effective. One might ask, how could you sensibly expect it to be otherwise, with such an enormous resource advantage?

Our normative critique is that the system is unfair (most people agree with this value judgement), inefficient, and leads – as the Sutton Trust amply demonstrates – to a democratic deficit in this country. Most reformers would seriously like to see something better for the new generation. To get there, however, we must start by recognising the facts for what they are, not what we might wish them to be.

Photo: Charterhouse by Chris Hartford via Creative Commons.