Education Secretary David Hinds’ careful use of numbers doesn’t add up

The chair of the UK’s statistics watchdog has written to education secretary Damian Hinds for the fourth time this year, raising “serious concerns” about his department’s use of school funding statistics.

The UK Statistics Authority (UKSA) launched an investigation into the Department for Education over a minister’s claim that it was spending “record amounts” on school funding, after it emerged that the figures included billions of pounds of university and private school fees.

The figures cited by the DfE and school standards minister Nick Gibb, in defending the government’s spending on education, included the money paid out by university students on tuition fees and money that parents spent on private school fees.

Sir David Norgrove, chair of the UK Statistics Authority, wrote to Hinds this morning, chastising him for repeatedly using misleading statistics to support misleading claims.

Hetan Shah, CEO of the Royal Statistical Society, described the rebuke as “blistering”, and said it was “amazing” for Sir David to send such a letter to a minister. 

“Extraordinary that [the Uk Statistical Authority] has felt it necessary to seek the secretary of state’s reassurance that his department remains committed to the statutory code of practice for statistics and, secondary that [the DfE] will start behaving in a manner that ‘does not mislead’,” he tweeted. 

The row erupted last week after the DfE and Gibb cited figures saying that the UK was the third-highest spender on education as a proportion of economic output in the Organisation for Economic Cooperation and Development, a group of wealthy nations.

His comments came after more than 2,000 headteachers marched to Downing Street to protest at funding cuts. Headteachers also accused the government of eroding trust. Jules White, the head of Tanbridge House school in West Sussex and leader of the Worth Less? group, which has lobbied for fairer funding, said: “Parents and the wider public have a right to know the facts and the government cannot have it both ways; you cannot slash our budgets and then pretend all’s well. 

“The constant use of misinformation is placing an intolerable strain on headteachers’ relationships with the DfE. Trust is being eroded. We respectfully request the DfE to publicly set the record straight and, much more crucially, work with the chancellor to make a real-terms investment in our schools in the upcoming budget. ”

Paul Whiteman, the general secretary of the National Association of Head Teachers, said: “If trust goes, there is little left for the profession to hold on to. Failing to face up to the truth will cheat an entire generation. The chancellor must now step up and rescue education funding. It is in the national interest.”

Mary Bousted, joint leader of the NEU teaching union, branded the DfE’s use of figures as “appalling.”  

Hinds has defended the department’s use of figures today in a letter to the Statistics Authority, and pledged to work closely with it to ensure that “all departmental statistics to be both factually accurate and used in the right context”.

The shadow education secretary, Angela Rayner, said the UKSA letter represented a “humiliating rebuke” for Conservative ministers and called on the government to come clean over the crisis in school funding.

The intervention by the UKSA follows a row last week over a claim by the DfE and the schools standards minister, Nick Gibb, that the UK’s spending on education was the third highest in the world.

Complaints were made to the UKSA after it emerged in a BBC report that the OECD figures on which the claim was based included university student tuition fees as well as fees paid by private school pupils.

Hinds wrote to MPs over the weekend defending his department’s claims, but following its investigation, the UKSA ruled the figure “included a wide range of education expenditure unrelated to publicly funded schools … rather than a comparison of school spending alone. 

The result was to give a more favourable picture. Yet the context would clearly lead readers to expect that the figures referred to spending on schools

An accompanying letter by Ed Humpherson, UKSA’s director general for regulation, to the DfE’s chief statistician piled on the criticism.

“The way statistics have been presented gives a potentially misleading picture of changes in schools funding,” he wrote. “It is important that the department present statistics and data professionally and I encourage you to continue to work with communication teams to minimise the risk of misleading the public.”

The UKSA complained about a DfE tweet on school funding featuring a graph with a truncated axis which had the effect of “exaggerating” school spending figures. The information was also presented in cash terms rather than real terms.

It also criticised Gibb’s claim that in an international survey of reading abilities of nine-year-olds, England had leaped up the rankings last year after decades of falling standards, moving from 19th out of 50 countries to 8th. “This is not correct,” Norgrove pointed out. “Figures published last year show the increase was from 10th place in 2011 to 8th place in 2016.”

The UKSA also ruled on a complaint from shadow education secretary, Angela Rayner, about the DfE’s oft-repeated claim that there were now “1.9 million more young people studying in good or outstanding schools”. The authority said the figure did not give a full picture and should be set in the context of increasing pupil numbers, changes to the inspection framework and out-of-date inspections.

“I am sure you share my concerns that instances such as these do not help to promote trust and confidence in official data, and indeed risk undermining them,” said David Norgrove.

The education secretary responded to the UKSA saying his department was “looking into the precise issues raised” but he went on to largely defend the disputed claims.

Here are both letters:

The Rt Hon Damian Hinds MP
Secretary of State for Education
(via email)

8 October 2018

Dear Secretary of State

I am writing to raise with you serious concerns about the Department for Education’s presentation and use of statistics. 

The UK Statistics Authority has had cause to publicly write to the Department with concerns on four occasions in the past year. 1 

I regret that the Department does not yet appear to have resolved issues with its use of statistics. Last week, the Minister of State for School Standards wrote that, in an international survey of reading abilities of nine-year-olds, England “leapfrogged up the rankings last year, after decades of falling standards, going from 19th out of 50 countries to 8th.”2. This is not correct. Figures published last year show the increase was from 10th place in 2011 to 8th place in 2016. 

My attention has also been drawn to a recent tweet and blog issued by the Department regarding education funding. 3 As the Authority’s Director General for Regulation has noted in a letter to the Department today, figures were presented in such a way as to misrepresent changes in school funding. In the tweet, school spending figures were exaggerated by using a truncated axis, and by not adjusting for per pupil spend. In the blog about government funding of schools (which I note your Department has now updated), an international comparison of spend which included a wide range of education expenditure unrelated to publicly funded schools was used, rather than a comparison of school spending alone. The result was to give a more favourable picture. Yet the context would clearly lead readers to expect that the figures referred to spending on schools. 

The Shadow Secretary of State for Education has written to express concerns about your use of a figure that appears to show a substantial increase in the number of children in high performing schools, as judged by OFSTED. While accurate as far as it goes, this figure does not give a full picture. It should be set in the context of increasing pupil numbers, changes to the inspection framework and some inspections that are now long in the past, as an earlier letter to the Department from the Office of Statistics Regulation pointed out. 

I am sure you share my concerns that instances such as these do not help to promote trust and confidence in official data, and indeed risk undermining them.

I seek your reassurance that the Department remains committed to the principles and practices defined in the statutory Code of Practice for Statistics. In particular, I urge the Department to involve the analysts closely in the development of its communications, to ensure that data are properly presented in a way that does not mislead. 

I have asked the Authority’s Director-General for Regulation, Ed Humpherson to speak with Jonathan Slater, Permanent Secretary at the Department for Education, and to Neil McIvor, Head of Profession for Statistics at the Department for Education, about what the Department might do to improve its practice. 

I am copying this letter to the Minister of State for School Standards, to Mr Slater and Mr McIvor, and to John Pullinger, the National Statistician. 

  Yours sincerely

  Sir David Norgrove

1 Letter from Ed Humpherson to Head of Profession (March 2018) National Pupil Database Access
Letter from Sir David Norgrove to Shadow Secretary of State for Education (March 2018), School Funding
Letter from Ed Humpherson to Head of Profession (January 2018) International Reading Literacy Study
Letter from Ed Humpherson to Jonathan Slater (November 2017) Department for Education breaches of the Code of Practice

2 The Telegraph (27 September 2018): Our whizzpopping phonics revolution is transforming literacy in schools
3 Department for Education (28 September 2018): Tweet regarding school funding
Department for Education (28 September 2018): Education in the Media: Funding

 

Dear David,

Thank you for your letter. 

I appreciate you drawing your concerns to my attention and very much welcome the work by the UKSA to ensure communication of statistics across Government meets the highest standards. We are keen to work closely with the UKSA and we want all departmental statistics to be both factually accurate and used in the right context.

We are looking into the precise issues that you raise, and the Permanent Secretary will write to the UKSA shortly with a more detailed response. It may be helpful though to respond on the points of substance, including for the public including for the public
record.

Taking funding first – we need to be clear about different types of funding and spending. However, several statistics in the OECD’s 2018 report comparing expenditure in 2015 (which as you know are the latest comparative data published by the OECD) demonstrate the UK as being among the higher spenders on education at primary and secondary level, whether you look at spend as a share of GDP, spend as a share of government spending or spend per pupil. It is true to say that the OECD has ranked the UK as the third highest for total education spending – the figure which includes tertiary and private education for every country. A more direct statistic about school spending
specifically is that among G7 nations the UK government spent the highest percentage of its GDP on institutions delivering primary and secondary education.

On overall school funding, core funding is rising to £43.5bn by 2019-20. Of course, I recognise that pupil numbers are rising, we are asking schools to do more and schools are facing cost pressures. I am on record setting this out with a range of different audiences and agree that context is important.

Moving on to the survey of reading abilities, it is true to say that we have achieved our highest score in PIRLS since it first began in 2001 – in 2016 we were placed joint 8th. We agree that we could have been clearer that the improvement from 19th to 8th was between 2006 and 2016. We have put a great deal of emphasis on the teaching of phonics, introducing the phonics screening check in 2012, and since then many more six year olds are on track to be fluent readers.

Regarding the 1.9m statistic, I believe it is important to establish that the proportion of children in schools whose last Ofsted judgement was Good or Outstanding has risen from 66% in 2010 to 86% in March 2018; to make this more intelligible we tend to use the number of children rather than a percentage figure – hence we express it as 1.9m more children in Good or Outstanding schools. 

Our methodology is published at: https://www.gov.uk/government/publications/children-in-good-or-outstandingschools-august-2018. As you know, Ofsted use a range of triggers for a reinspection, such as an unexpected fall in exam performance.

Naturally we want to ensure we always present those factually accurate statements, and all others, in line with your Code of Practice for Statistics and I look forward to working with your team further on that. More widely, in the interests of making sure the public debate is well-informed, I hope that others who produce and use statistics which become regularly cited will also aspire to the highest standards of data integrity, and that the UKSA can play a role in challenging where data could easily be open to misinterpretation.

Thank you once again and please be assured of my, and my department’s, continued commitment to working with you on the integrity of statistics and informing the public debate.

---

Related

A list of official rebukes for Tory lies

Esther Mcvey forced to apologise for being conservative with the truth

It’s truly priceless that Iain Duncan Smith can accuse anyone of misrepresenting statistics with a straight face.

 

---

 

I don’t make any money from my work. I’m disabled through illness and on a very low income. But you can make a donation to help me continue to research and write free, informative, insightful and independent articles, and to provide support to others. The smallest amount is much appreciated – thank you.