In this article the authors discuss the findings from the Programme for International Student Assessment (PISA). The initiative was implemented by the Organisation for Economic Co-operation and Development (OECD) in 2000. In a three-year cycle it was planned to compare the performance of 15-year-old students across member countries. The first cycle in 2000 focused on Reading, the second cycle in 2003 on Mathematics, and the third cycle in 2006 will concentrate on Science, with the other two areas featuring as minor components. The information gained from PISA is the basis on which the OECD develops some of the educational indicators used in its annual publication 'Education at a Glance'. In particular, indicators are derived from the data obtained in the PISA surveys that are considered to reflect the quality of educational output from each country. These output indicators, in turn, are linked to indicators of educational input such as educational spending in each country. Such bivariate correlational analyses between input and output indicators may go some way to explain differences across countries. However, more complex analyses are required to address which factors operate similarly or differently across countries to influence achievement at the student level. The complexity of such analyses stems from the fact that contextual information, for example about students' home environments, their attitudes and or about teacher and school-related factors, needs to be taken into account at the appropriate levels in order to understand what leads to differences in performance within and across countries. Finland was the highest-scoring country in the PISA 2000 data collection. In contrast, Germany performed well below the OECD average of 500. Much has been speculated about possible reasons for this difference in performance and many suggestions as to measures to be taken in Germany as a consequence. However, the article is aimed at examining the way in which student and school-level factors operate to influence student achievements in the two countries in order to base the current discussion about possibilities for improving the performance of German students on available evidence rather than on data-free claims and speculation. The authors conclude by stating that the findings in Germany underpin the strong segregation effect of the placement of students into the different school types in that country. If this observation serves as an artefact for the difference in socio-economic status, remedial efforts may be difficult to introduce. However, since about 46 per cent of the total variance in Germany is found at the student level, options exist to improve reading performance. While a lot of the current political debate in the German media focuses on the extension of school hours ('Ganztagsschule'), results of the current analysis do not provide evidence to support such a policy change. Instead, the evidence suggests that student performance in reading benefits from a strong emphasis by schools on achievement and on high expectations and standards with respect to student work.