No, private schools do not result in two years extra progress

Henry Stewart's picture

On Friday all the major British newspapers ran remarkably similar headlines on the story about research from Durham University, comparing private and state school GCSE performance: 

Guardian: "Private school gives pupils a boost worth two extra years of education, research shows"
Telegraph: "Private school pupils 'two years ahead' of state educated peers by the age of 16" 
Times: "Pupils who go private get ahead by two years"
Daily Mail: "Private pupils are two years ahead of state rivals by the age of 16"

The problem with these headlines is that the statistic is wrong. Durham's own figures give no basis to the claim of private school students being two years ahead.

The False Durham Claim

The Durham University report, available on the Independent Schools Council web site, did make that claim. To be specific it stated:

"The difference of 0.64 GCSE grades ... equates to a gain of about two years’ normal progress and suggests that attending an independent school is associated with the equivalent of two additional years of schooling by the age of 16".

This is not true. One GCSE grade equals one level of progress, in education data language. As anybody with the most basic knowledge of English secondary education knows, the DfE expects students to make three levels of progress over the five years from age 11 to age 16 (and the majority do make that progress, for instance 65% in Maths and 70% in English in 2014). That means that a difference of 0.6 grades is equivalent to around one year of schooling, not two.

If it is true that private schools result in one extra years progress, that is still a story. But it is half the level that was so universally reported. And shouldn't a glaring error in the headline finding have caused journalists to question the quality of the research overall? 

Although 3 levels of progress is the expected measure, those with a stronger starting point (who will be over-represted in private schools) advance more. In Maths, children achieving 5b at age 11 saw an average 3.6 levels of progress by age 16, and those on 5a advanced an average 4.2 levels. Was this differentail fully taken account of in the research?

Repeating claims of private school sector without question

The reason the headlines were so similar is that they were taken directly from the press release of the Independent School Council.

The Independent Schools Council is, of course, not a neutral source. It exists to promote the interests of private schools and to argue (as reported in several of the articles) that private education is “worth paying for”.

So why did all major newspaper accept their statement without question and basicly reprint their press release?

If a university produces research on energy funded by the oil industry or a report on healthy food funded by sugar companies, it may have some substance but you would expect journalists to show a healthy scepticism and to seek alternative views. So why was no such balance, or even basic fact checking, sought here?

What we have here is a piece of research funded by a group with a strong vested interest and with a fundamental error in the headline finding. I do not know whether the detailed analysis is flawed but, for me, these two facts leave the whole study open to serious question.

Of course you would expect private schooling brings some advantage. Take the most advantaged children in the country, with few from difficult backgrounds or with special needs, and then spend more than two-and-a-half times the funding of state schools, and you would expect to achieve some advantage. 

However the OECD found that (as my colleague Janet Downs points out here), after adjusting for social and economic background, English private schools did less well than state schools. And one must ask how the Durham researchers could compare, for instance, how disadvantaged students perform in both systems when so few attend private schools.

Education journalists: please engage your critical faculties and do your job

But this is a story of education journalists failing to do their job. Faced with a press release from an organisation with a clear and obvious vested interest they failed to check the most basic of facts and repeated the claims without question.

Journalists, please remember: just because a press release (from the DfE, a government minister or the Independent Schools Council or any other body with a vested interest) makes a claim, it does not mean that it is true. It is your job to check it before publishing it. Please engage your critical faculties.

Or ask us: we're always happy to help check any dubious data claims. Simply email me at :)


Data note

The statement that private schools receive over 2.5 times as much as state schools is based on the Telegraph estimate of an average annual fee in 2014 of £12,582 for private day schools (£28,500 for boarding) and TES figures showing the median local authority 2015 funding per pupil is £4,529 (TES, 29 January 2016). Even before the private school increase for 2015, this suggests private school per pupil funding is 177% more than that for state schools.

Figures on % of students making 3 levels of progress in English and Maths are taken from the standard school Raise Online report.

Figures on average levels of progress in Maths for 5b and 5a students are taken from the DfE's "2014 KS2-4 National Subject Transition Matrices

Share on Twitter Share on Facebook

Be notified by email of each new post.


Henry Stewart's picture
Sun, 28/02/2016 - 12:35

I should note that there was one exception, one publicaiton that did not simply reprint the ISC press release claims. This was Schools Week. Freddie Whittaker covered the story with the much more measured headline "Troubling’ variation in state and private school grades needs investigation, say subject experts". The Two year claim was included but in the last paragraph and within a quote from the ISC General Secretary - thus making clear it was opinion and not fact.

It is good to see a decent journalistic example alive in one publicaton.

Janet Downs's picture
Mon, 29/02/2016 - 09:27

The Headmasters' and Headmistresses Conference (HMC) which represents 'leading  independent schools' also pushed out a press release.  It contained a quote from HMC chair, Chris King, who said:  “It is now proven beyond reasonable doubt that independent schools add significant value to children’s education even when their socio economic background is discounted.'

I left a comment (posting as ex-secondary modern teacher) under the article which said the Durham report actually said ''...the jury is still out with regard to the true extent to which attendance at independent schools in England enhance the academic achievements of students when student and school-level differences are accounted for.'  This was held up for moderation.  

When I checked this morning the note about moderation had gone but my comment was not published.

I also left a comment on another page noting that Durham university had not cited the OECD finding (2010) that UK state schools outperformed UK private schools on the reading scale once socio-economic background was taken into account.    That comment has not been published either.

Perhaps it's because these two comments undermine the 'two years' ahead claim.

Janet Downs's picture
Mon, 29/02/2016 - 09:59

The Durham researchers made it clear in their report (p42) that there was 'significant limitation'  which meant they were 'unable to give a confident and precise estimate of the causal effect of attending an independent school'.   The advised that the 0.64 grade estimate of independent school advantage should be 'viewed with caution'.

This was ignored by the ISC.  Their press release confidently stated:  'Independent schooling accounts for a 0.64 of a GCSE grade increase.'

Janet Downs's picture
Mon, 29/02/2016 - 10:11

I can't find in the report whether the Durham researchers looked at GCSE results from ALL indpendent schools or only those who are members of groups such as the HMC for which ISC acts as an umbrella group.     Membership of these groups is not open to all but only those of a particular 'quality'.  If the research was limited to only these groups, then this would affect the results.   There are over 1000 independent schools which are not members of these groups.   If these were included then it would likely reduce the alleged independent schools advantage.

Nigel Ford's picture
Mon, 29/02/2016 - 13:27

Well done Janet for your tenacity.

Michele -Lowe's picture
Mon, 29/02/2016 - 15:50

Regarding this discussion, BBC R4's Start the Week this morning were discussing life chances and the nature/nurture debate.  And the contributors were not pulling their punches.  

Add new comment

Already a member? Click here to log in before you comment. Or register with us.