BBC Newsnight last night took the PISA tables as the new gospel. The tables show which pupils got a teaching that best taught to the PISA test. That's all. The PISA science test will only say a very limited amount about how 'good' someone is at 'science'. If we think science should include 'the ability to do experiments' say, we would need to know if the PISA science test included that. If it didn't include that, then it wasn't any good for testing that. So one country might be very good at 'experimental science' and may well have spent a lot of time on it but this won't necessarily show in the test.
That principle applies across the board with any testing procedure. The test only tests what it tests. That's all it tells you. One style of teaching usually suits a test better than another. Usually that style of teaching is the teaching that most resembles the test. It's a 'self-serving' system. We have to ask if we want that style of teaching, that kind of education.
As for the 'reading' test, again - we weren't told what was tested: reading out loud? (ie nothing to do with 'reading for meaning'?) 'comprehension'? (that usually only includes 'retrieval' and 'inference' and not 'interpretation') ...and so on.
These issues are fundamental to the question of what kind of education we want and for what purpose. In general terms, if the outcome of the PISA tables is that schools end up spending even more time teaching in a way that suits PISA tests, anything that is not testable gets squeezed out of the curriculum. You know, like stuff to do with ethics, say. Or inventing things. Or coming up with feasible alternatives. Or teamwork. Or compassion. Or planning...