Using pooled data on instructional time and student performance by subject, our study finds evidence for the school inputs-student achievement relationship for German states. This finding is robust both to the inclusion of state fixed effects and in an extensive extreme bounds analysis. It stands in contrast to the majority of related studies. We argue that this is due to an error-in-variables problem and implied misinterpretation of existing studies that disregard the fact of learning being a cumulative process by relying on rather poor proxies for instructional time. Highschool ninth graders from the OECD Programme of International Student Assessment (PISA-E) tests' bottom percentiles bene.t most from extra-instructional time measured in cumulated form from first up to ninth grade. Besides total instructional time exposure, we identify eight further social environment and institutional variables with robust impact on student performance. In contrast to instructional time hardly any of these factors can be affected by policy in the short run.
education production function student performance school resources