The government has hired experts to come up with a way of preventing grammars, and other secondaries with high ability intakes, from having an unfair advantage in new school league tables, TES has learned.
A ministerial adviser says the Department for Education is attempting to tackle the “bonus” that he says academically selective schools enjoy under the current “flawed” system.
He revealed that the DfE has commissioned outside help on alternatives because officials do not believe that grammars can be as “brilliant” as existing measures suggest.
But grammar school heads deny they are unfairly advantaged and say they will oppose any “contrived” attempt to play down their success with students.
The DfE concern centres on the method used to calculate how much schools “add value” to their pupils in terms of academic achievement.
This will assume much greater importance from 2016 when a new accountability system will see a valued added measure used to set crucial floor targets that can lead to the closure of under-performing schools.
Tim Leunig, an aide to Liberal Democrat schools minister, David Laws, revealed government unease about the existing “grammar school value added bonus” earlier this term.
“We know at the moment that the value added methodology is not fair in the technical statistical sense,” he told an Association of School and College Leaders (ASCL) conference.
“It is somewhat generous to grammar schools and other people with particularly high performing intakes. We have hired Simon Burgess from Bristol [University] to work out how we can do this better.”
Professor Burgess - an economist, who has been working on the project with Dave Thomson, a statistical consultant at the Fischer Family Trust - told TES he hoped their final report would be published before Christmas.
“We were asked to produce a system that was fair to all schools and that is what we have been trying to do,” he said.
Mr Leunig described an earlier draft as “very clear” and “very straightforward”. “It does get rid of that sense that you are more likely to do well in value added if you have kids with particular key stage two starting points,” he said. “And that’s fair.”
But Barry Sindall, chief executive of the Grammar School Heads Association, does not accept that selective schools have been unfairly boosted by the current system.
The former head of Colyton Grammar, Devon, said the association “would have no quarrel” with the new plans, providing they were based on comparisons between students with the same prior attainment.
However he warned: “If they are trying to come up with a contrived system that actually plays down grammar school performance then that is something clearly we would have a view on.”
The main performance measure in the new secondary accountability system will show whether a school’s pupils have performed better than expected at GCSE level, when their test results at the end of primary school are taken into account.
Known as “Progress 8”, it will look at eight GCSEs including English and maths and other academic ‘English Baccalaureate’ subjects and will include an optional slot for three “high value vocational” qualifications.
The measure will be calculated by comparing results with those achieved by pupils with same prior attainment who took their GCSEs three years previously. Professor Burgess’s report will recommend exactly how that is done.
Mr Leunig said he had initially been “worried” that Progress 8 would be too tough for “top end grammar schools”.
“Would we be saying that they had to average an A* for all their kids so that if one of them was ill on the day they kind of got a guaranteed fail?” he told heads.
“So I got our number crunchers to crunch the numbers and they came back and they said ‘No, no every grammar school – all of them but three - is in the top quarter by value added’ and I thought this cannot be right, it is just implausible.
“I mean if [grammars] really are that good we should expand them left right and centre if they are that brilliant.
“So either we should be doing that or we should say ‘What’s wrong with our VA (value added) methodology?’. And then I dug round and discovered that our VA methodology was flawed and therefore we had Simon do something better.”
Professor Burgess said: “We are looking at the statistical pro and cons of different measures and the best way to model student progress.”
Malcolm Trobe, ASCL deputy general secretary, said: “They are concerned that there is a fair system in place and that is right and proper.
“More able students make faster progress so the methodology has to have something built into it to allow for that differential rate of progress. It won’t be easy. But mathematically it is possible to do, I am sure.”