A study at York University, based on new government figures, shows there is no advantage for able pupils in attending a grammar school.
The findings, presented at a conference organised by the Campaign for State Education, also show that pupils who have to go to secondary moderns because they are not selected for grammar schools get lower results than they would if they were in comprehensives.
The research comes at an explosive time as parents across the country are trying to secure ballots on the future of the remaining 163 grammar schools. Ministers have introduced legislation to allow local parents to decide whether they want to keep the 11-plus.
Professor David Jesson, of York's centre for performance evaluation and resource management, looked at the 60,000 brightest pupils in grammar schools and comprehensives who took GCSEs in 1998.
He compared the GCSE scores of comprehensive pupils with those of grammar school pupils who achieved the same standard in national curriculum tests in English, maths and science at 14. He found that the able and very able did better at comprehensives than at grammar schools while the extremely able did just as well.
Professor Jesson attacked the argument advanced by pro-grammar school campaigners that grammar schools raise the standards for all pupils, as well as those who are selected.
He compared the GCSE scores of pupils in secondary modern schools with those of pupils in comprehensive schools. Those in secondary moderns did considerably less well than pupils of similar ability in comprehensives.
His study also examined two similar local education authorities, one comprehensive and one selective. In the comprehensive authority 52 per cent of pupils got five or more A*-C grades compared with 48 per cent in the selective one.
Professor Jesson said: "Some hundreds of children in the selective authority are not achieving what they would have achieved in the comprehensive authority.
"Opening all schools to all pupils would improve results by an average of three percentage points for the proportion of five A*-C passes and an additional two to three higher grades for every pupil in the system - equivalent to three years' improvement at current rates of progress."
Grammar school heads questioned the value of using only the two years before GCSE to measure pupils' progress. But Professor Jesson, who has yet to carry out detailed comparisons of pupils' progress between the ages of 11 and 14, said that an initial glance at the figures suggested that the picture was likely to be similar.
Earlier, Dr John Marks, of the Educational Research Trust, and a member of the right-wing think tank, the Centre for Policy Studies, said that comprehensive schooling had replaced selection by ability with selection by class and house prices. "Grammar schools are the surest way for talented children to get out of ghettos," he said.
He called on the Government to double the number of grammar schools by setting up new schools in London and other big cities.
Sixth-formers will shun tough new five-subject A-level courses unless they know universities will be looking for breadth as well as depth when they are offering a place, headteachers warned yesterday. John Dunford, general secretary of the Secondary Heads Association, said students and teachers had been left confused and worried by changes to A-levels next year. Ministers want sixth-formers to study up to five subjects in the new half-size AS-levels, and top up three subjects to the full A-level standard.Reuse content