How Michael Gove manipulated education statistics

The Education Secretary's misleading claim that the UK has plummeted down the international league tables.

One of Michael Gove's favourite arguments for his school reforms is that Britain has plummeted down the international education league tables. In June 2011 he told Policy Exchange that the UK had fallen from "4th to 16th place in science; from 7th to 25th place in literacy; and from 8th to 28th in maths" between 2000 and 2009 in the OECD's Programme for International Student Assessment (PISA).

But how reliable are the statistics? In this week's NS, Peter Wilby draws attention to a story that deserves more than attention than it has so far received (no national paper has reported on it). Last month, in response to a letter from David Miliband, Andrew Dilnot, the chair of the UK Statistics Authority, expressed "concern" about the Department for Education's unqualified use of the figures. He noted that the OECD's 2009 report for the UK included the following "important caveat":

Trend comparisons, which are a feature of the PISA 2009 reporting are not reported here because for the United Kingdom it is only possible to compare 2006 and 2009 data. As the PISA 2000 and PISA 2003 samples did not meet the PISA response-rate standards, no trend comparisons are possible for these years.

In other words, Gove should not be comparing results from 2000 with those from 2009. Dilnot wrote: "While I understand that some users of these data would like to make comparisons between the first PISA study in 2000 and the most recent in 2009, the weaknesses relating to the response-rate standard in earlier studies should not be ignored."

He concluded: "These uncertainties and weaknesses are not just a technical footnote; they are themselves an important part of the evidence, and affect interpretation and meaning. League tables and the presentation of international rankings can be statistically problematic, and require clear and careful commentary alongside them."

The statistics chief also noted a review published by the Institute for Education which concluded that "problems with identifying change over time" meant the apparent decline in secondary school pupils' performance should not be treated as a "statistically robust result". The Department for Education is yet to respond.

This isn't the only recent instance of the coalition playing fast and loose with statistics. David Cameron is fond of boasting that "one million" new private sector jobs have been created since the coalition came to power, but, as I've noted before, what he doesn't mention is that 196,000 of these were simply reclassified from the public sector.

After complaining for years about Gordon Brown's manipulation of economic statistics, the coalition came to power promising a new era of transparency. But Gove and Cameron's behaviour suggests it's not prepared to practise what it preached.

Education Secretary Michael Gove was criticised for ignoring "weaknesses" in the statistics. Photoraph: Getty Images.

George Eaton is political editor of the New Statesman.

Photo: Getty Images
Show Hide image

Why are boundary changes bad for Labour?

New boundaries, a smaller House of Commons and the shift to individual electoral registration all tilt the electoral battlefield further towards the Conservatives. Why?

The government has confirmed it will push ahead with plans to reduce the House of Commons to 600 seats from 650.  Why is that such bad news for the Labour Party? 

The damage is twofold. The switch to individual electoral registration will hurt Labour more than its rivals. . Constituency boundaries in Britain are drawn on registered electors, not by population - the average seat has around 70,000 voters but a population of 90,000, although there are significant variations within that. On the whole, at present, Labour MPs tend to have seats with fewer voters than their Conservative counterparts. These changes were halted by the Liberal Democrats in the coalition years but are now back on course.

The new, 600-member constituencies will all but eliminate those variations on mainland Britain, although the Isle of Wight, and the Scottish island constituencies will remain special cases. The net effect will be to reduce the number of Labour seats - and to make the remaining seats more marginal. (Of the 50 seats that would have been eradicated had the 2013 review taken place, 35 were held by Labour, including deputy leader Tom Watson's seat of West Bromwich East.)

Why will Labour seats become more marginal? For the most part, as seats expand, they will take on increasing numbers of suburban and rural voters, who tend to vote Conservative. The city of Leicester is a good example: currently the city sends three Labour MPs to Westminster, each with large majorities. Under boundary changes, all three could become more marginal as they take on more wards from the surrounding county. Liz Kendall's Leicester West seat is likely to have a particularly large influx of Tory voters, turning the seat - a Labour stronghold since 1945 - into a marginal. 

The pattern is fairly consistent throughout the United Kingdom - Labour safe seats either vanishing or becoming marginal or even Tory seats. On Merseyside, three seats - Frank Field's Birkenhead, a Labour seat since 1950, and two marginal Labour held seats, Wirral South and Wirral West - will become two: a safe Labour seat, and a safe Conservative seat on the Wirral. Lillian Greenwood, the Shadow Transport Secretary, would see her Nottingham seat take more of the Nottinghamshire countryside, becoming a Conservative-held marginal. 

The traffic - at least in the 2013 review - was not entirely one-way. Jane Ellison, the Tory MP for Battersea, would find herself fighting a seat with a notional Labour majority of just under 3,000, as opposed to her current majority of close to 8,000. 

But the net effect of the boundary review and the shrinking of the size of the House of Commons would be to the advantage of the Conservatives. If the 2015 election had been held using the 2013 boundaries, the Tories would have a majority of 22 – and Labour would have just 216 seats against 232 now.

It may be, however, that Labour dodges a bullet – because while the boundary changes would have given the Conservatives a bigger majority, they would have significantly fewer MPs – down to 311 from 330, a loss of 19 members of Parliament. Although the whips are attempting to steady the nerves of backbenchers about the potential loss of their seats, that the number of Conservative MPs who face involuntary retirement due to boundary changes is bigger than the party’s parliamentary majority may force a U-Turn.

That said, Labour’s relatively weak electoral showing may calm jittery Tory MPs. Two months into Ed Miliband’s leadership, Labour averaged 39 per cent in the polls. They got 31 per cent of the vote in 2015. Two months into Tony Blair’s leadership, Labour were on 53 per cent of the vote. They got 43 per cent of the vote. A month and a half into Jeremy Corbyn’s leadership, Labour is on 31 per cent of the vote.  A Blair-style drop of ten points would see the Tories net 388 seats under the new boundaries, with Labour on 131. A smaller Miliband-style drop would give the Conservatives 364, and leave Labour with 153 MPs.  

On Labour’s current trajectory, Tory MPs who lose out due to boundary changes may feel comfortable in their chances of picking up a seat elsewhere. 

Stephen Bush is editor of the Staggers, the New Statesman’s political blog. He usually writes about politics.