The facts are clearly laid out in Measuring Student Knowledge and Skills (OECD, 1999) and The PISA2003 Assessment Framework (OECD, 2003).
Pisa is testing national samples of 15-year-olds at three-yearly intervals, beginning in 2000. The 2000 results were published in 2001 and the 2003 results are due soon. Each survey tests, not factual knowledge, but "literacy" in the areas of reading, maths and science.
One of the main aims of the surveys is to enable countries to measure their pupils' progress from survey to survey; this is more important than creating league tables of countries. To facilitate these comparisons over time, each of the three literacy areas is included in each survey, with one taking a major role and the other two a minor one. In 2000, reading was the major area, taking five of the seven hours of testing time, with maths and science having one hour each.
The 2000 results for maths and science are very tentative, being based on only 35 items for science and a similar number for maths. The results are nonetheless interesting. But they must be interpreted in terms of being measures of mathematical and scientific literacy. This places emphasis on application and use of knowledge about the content and procedures of science in authentic contexts.
It is these skills and knowledge that, according to the results, pupils in the UK are able to use more successfully than all but three other OECD countries.
In my view the favourable position of the UK in science, and to a slightly lesser extent in maths, has nothing to do with curriculum changes.
Experience shows that it is just not possible to bring about improvements that quickly. Rather it is due to a tradition of emphasising practical applications and demonstrating understanding by using knowledge.
However much some might wish for a greater emphasis on making knowledge useable and useful, we are actually well ahead of many countries in this respect. Copying and memorising "right answers" is far more common in other European countries.
It is not surprising, therefore, that pupils do well in private schools, where smaller classes and often better resources have enabled teachers to include rather more of the experiences that develop the thinking and the abilities that Pisa is assessing.
These are outcomes of education that are valued for lifelong learning. We should be welcoming the evidence of good progress towards them and continue efforts to make sure that students in all schools have opportunities of these kinds.
Professor Wynne Harlen Chair, PISA science expert group, 1998-2003