School inspectorates are meant to improve student performance on the level of both the individual school and the school system. Whereas, in this context, there are no empirical findings on the efficiency of school inspectorates in Germany, international studies show that, as a rule, school inspectorates do not succeed in bringing about an improvement in performance. However, these findings are usually not very reliable due to problems with the random samples chosen for the studies. The present contribution for the first time examines for the German context which effects on student performance are empirically verifiable, using the school inspectorate Hamburg as an example. The investigation is based on trend data on Hamburg's central school leaving exam and on longitudinal data provided by the study "Students' competencies and attitudes" (German abbreviation: KESS). Possible problems with random samples are explicitly taken into account in the analyses in order to be able to come up with empirically reliable findings on the effects of school inspections on student performance.