Professor Peter Tymms says the leap in primary standards under Labour has been a mirage. And more and more experts agree. Warwick Mansell reports
The idealism that once drove Peter Tymms to spend three years teaching in a remote part of sub-Saharan Africa is still with him, he says. Maybe it is just as well.
For the 56-year-old son of a missionary - one of Britain's most prominent education academics - has needed a strong sense of right and wrong in recent months.
Professor Tymms, professor of education at Durham university and director of its Curriculum, Evaluation and Management Centre (CEM), has been at the centre of a bitter row with the Government over one of its biggest education success stories.
Last year he wrote to the Statistics Commission, an independent body set up by the Treasury, with research findings that called into question Labour's achievements in primary schools.
His 17-page paper considered the results of 11 studies of primary performance from 1995, and claimed that dramatic improvements recorded in key stage 2 reading and maths test scores in the late 1990s overstated the rise.
Professor Tymms found that on other tests of reading and maths, pupils'
improvements were not nearly so dramatic. Results of these independent tests were combined with the national curriculum test data to produce the set of results shown in the graph.
The commission rejected some of Professor Tymms's claims, but backed his central argument.
The Department for Education and Skills was enraged. Sir David Normington, the permanent secretary, wrote to the commission and urged it to back down.
Ruth Kelly, the Education Secretary, told the BBC's Today programme that the commission's findings were not being accepted.
But the commission has refused to capitulate.
"I'm just amazed that Ruth Kelly is arguing that black is white," he said.
"There's no argument. It's hard to find anyone who is genuinely independent saying that the Government is right.
"Standards of reading have hardly moved (since the mid-1990s). That's been shown to the be case using lots of data. Only the official test scores say there has been a massive rise in standards."
Peter Tymms was born in Johannesburg, the son of an Anglican missionary who was working in a poor white mining community. At the age of four, his parents returned to England, settling in the north-east.
He attended a primary school but was moved to prep schools and then Giggleswick, the north Yorkshire public school, after his family "discovered I had picked up a local Geordie accent".
As a teenager, he was taught English by Russell Harty, later of TV chat show fame.
He excelled in science, eventually reading natural sciences at Christ's college, Cambridge, before gaining a PGCE at Newcastle and heading to Zambia for more than three years' teaching science at a boarding school 200 miles from the nearest paved road.
There, he met his wife, Beena, a teacher from India. Returning to England, he taught chemistry at three secondaries and an FE college in the north-east. The couple have two children.
His break into education research came in the mid-1980s, when he enrolled on a master's course led by Carol Fitz-Gibbon, who was beginning to analyse results in secondaries.
He worked for her, specialising in primary education and developing independent tests now used by thousands of pupils. Two years ago when she retired, he took over the CEM centre.
The centre carries out "value-added", or progress, tests, for schools to ascertain how they can help pupils improve.
Its work is increasingly international: Professor Tymms recently returned from a trip to New Zealand, Hong Kong, Singapore and China. Each country runs or is interested in the tests. In China, he is helping the Government develop its system for monitoring school standards.
Professor Tymms has also spent the past year heading a commission for the Liberal Democrats looking at the future of education in Newcastle.
His varied professional interests give him a powerful base from which to criticise the Government.
He bats back suggestions that he might be old Labour, but said he cared for society's underdogs, and was "extremely disillusioned" with the Government.
Fellow academics describe him as more measured in his criticisms than Professor Fitz-Gibbon.
David Jesson, of York University, said: "He's less intentionally confrontational. Though we have not always seen eye to eye professionally, I've got a lot of time for Peter."
Some murmur that the Durham team has sometimes been too bold in its claims.
But even critics concede that Professor Tymms has succeeded in promoting the centre's work.
Margaret Brown, professor of maths education at King's college, London, and Dr Dylan William, one of Britain's leading testing experts now working in the United States, said he was unusual in his mastery of the quantitative aspects of education policy.
Professor Fitz-Gibbon said: "Peter reads factual and scientific books and papers voraciously. He is an excellent scientist and he gets things right."
Trevor Millum, secretary of the National Association for the Teaching of English, who worked with him for three years in Zambia, said Professor Tymms was a man of integrity reluctant ever to tell a lie.
It is a steadfastness that appears to be serving him well. As one observer said at a recent conference: "Peter has been plugging away on this issue for years. It's bearing fruit and the fact the Statistics Commission is supporting what he says shows he's been right to stick at it."