Last December, damning headlines followed when England was placed 18th out of 41 countries in maths, 12th in reading and 11th in science in rankings of tests taken in 2003 by 250,000 15-year-olds.
Now an analysis of these results has shown that most of the English schools which took part in the study were high-achieving.
The performance, in a study by the Paris-based Organisation for Economic Co-operation and Development, followed a stronger showing in 2000, when England was placed eighth in maths and reading and fourth in science out of 40 countries.
But ministers were able to shrug off some of the negative press last year by pointing out that the 2003 results in the Programme for International Student Assessment (Pisa) tests were unofficial. England had not been included in the official results by the OECD because too few English schools - only 64 per cent of those surveyed - took part in the tests for the findings to be statistically robust.
However, this week academics and senior civil servants were presented with an analysis, commissioned by the Department for Education and Skills, which cast new light on the figures. They were also told that the DfES is so concerned about schools' reluctance to take part in Pisa that it is may pay them Pounds 500 each to administer next year's tests.
On average, English schools taking part were high achievers, based on an analysis of GCSE and key stage 3 results. They had only half as many of England's lowest-achievers at GCSE and KS3 as the national average, the analysis by Professor John Micklewright of Southampton university showed.
When Professor Micklewright's team made a statistical adjustment to take account of this bias, England's score in the 2003 reading tests dropped from 506 to 499. This would have pushed England down a place or two in the international rankings, the conference was told. No figures were presented for the other subjects, but the effect is believed to be the same.
The analysis, which was presented in draft form, is to be published next year.
However, Andreas Schleicher, chief spokesman for Pisa, said it was not possible to say from comparing England's 2000 and 2003 results that standards had declined over this period. One of those present at the meeting said: "To know that standards have gone down over time, you have to do a longitudinal study which tests the same schools over time. Pisa does not do that so Mr Schleicher is right."
However, another said: "England's Pisa results have dropped quite sharply from 2000 to 2003, but our domestic test results keep going up every year.
What is going on here?"
A DFES spokesman said: "As the OECD has made clear, it is not possible to make meaningful comparisons with other countries based on this data. The most comprehensive measure of our pupils' performance against the national curriculum is our system of key stage national tests and GCSEs."