Standards of achievement could be raised by two GCSE grades in every subject if classroom teachers applied assessment techniques which identify individual pupils' learning needs.
Properly used, the strategy could push British schools' current middling performance in mathematics into the top five in international comparisons, with even average children performing as well as those currently in the top 35 per cent. Low achievers would make the biggest strides.
This is the conclusion of researchers Professor Paul Black and Dr Dylan Wiliam of King's College, London. They have spent a year collating more than 600 studies of assessment world-wide, ranging from nursery school to undergraduate level.
"Current school testing systems contribute little to improving pupil performance. The emphasis on competitive testing and achieving national targets may actually be counter-productive because it reinforces low-achieving pupils' sense of failure," said Professor Black, who was the chairman of the Government's Task Group on Assessment and Testing which devised the reporting structure for the national curriculum.
"Since 1988 national policy has moved on from using assessment as a tool to promote a competitive market in education to the more mature position of target-setting for individual pupils."
The report argues that consistent use of "formative assessment", in which teachers use regular tests to identify children's learning problems and offer high-quality feedback to help them overcome them, is the most effective way to improve achievement.
But Professor Black believes that the Government needs to go one stage further and focus on what goes on inside the classroom. Too many existing classroom practices allow pupils to "get by", he suggests.
For instance, in question-and-answer sessions there is often no time to think through an answer, so pupils simply guess what they think the teacher expects to hear. Time to think is much more likely to enable children of all abilities to achieve.
The researchers are also concerned about the effects of constant grading of pupils. In an Israeli research project 11-year-olds were given either written feedback on their tests, a graded mark, or both. The performance of the group who received only comments increased by one third, while that of the other two groups declined.
Professor Black and Dr Wiliam say that formative assessment is not just another "magic bullet" solution. But they acknowledge that further research would be needed to establish the most effective methods of delivery in British schools.
Areas to be studied would include diagnostic testing, feedback, self-assessment by pupils and opportunities for group discussion, all of which have been shown to improve pupils' performance.
Professor Black and Dr Wiliam will explain their findings today at a seminar at the London headquarters of the Nuffield Foundation that will be attended by senior inspectors, academics and representatives of the school standards and effectiveness unit, the Teacher Training Agency, the Qualifications and Curriculum Authority and teacher union representatives.
They will suggest that the first step towards improving formative assessment in Britain would be to set up a small group of schools from different environments - urban, rural, large and small and with different intakes - to work out effective ways of making use of the research findings.
Research Focus, page 22