The publication of the OECD report on the PISA 2000 study induced a public outcry in Germany. On average, German students participating in this standardized test performed considerably below the OECD average and substantially worse than those of other European countries, like Finland or Ireland. However, the results presented by the report consist mainly of country averages which do not take into account any other covariates of individual student achievement. This paper provides a comprehensive econometric analysis of the association of the individual-level reading test scores of German students with individual and family background information and with characteristics of the school and class of the 15 to 16 year old respondents in Germany to the survey. The results of several quantile regression analyses demonstrate that many popular explanations, like too much regulation of schools or the substantial share of non-citizens among the participating students, are by no means supported by the data. Rather results point towards a considerable impact of schools aiming at a more homogenous body of students in terms of their educational achievement.
student achievement school quality quantile regression