In German schools of all federal states students participate in comparison tests in the 3rd and 8th grades (VERA) that are administrated by the Institute for Educational Quality Improvement (IQB) in Berlin. In Hamburg, these tests were complemented by additional tests in the 2nd, 5th, 7th, and 9th grades and were renamed KERMIT (Kompetenzen ermitteln; Lücken et al. (2014)). These tests were embedded in a coherent test system to assess the competence development. In primary schools in the tests comprise the domains German reading comprehension and orthography and mathematics, in secondary school English listening, English reading comprehension, and natural sciences were added. Teachers will get an elaborate feedback of the competence-distribution and -development of their classes. The main purpose of KERMIT is to support schools in their data based school development. The results should trigger a reflection process in schools to identify sources of competence differences, to set targets and to implement new teaching procedures, resulting in a permanent quality cycle in school. The results of the KERMIT-tests are also given to representatives of Hamburg’s administration. They are used to monitor the educational system of Hamburg as well as to enable supervising authorities to support their schools. Therefore, KERMIT seems to be fully implemented as a support system for school development. An online questionnaire with about 1000 participants was introduced to assess the actual imple-mentation of KERMIT in schools. First, this questionnaire aims at analyzing how teachers and princi-pals evaluate KERMIT and second, it targets the question, if the quality cycle is really an integrative part of the school development according to a framework of educational application of comparison tests suggested by Groß Ophoff et al. (2007). Another aspect is the experience with the supervising authority regarding the data-based school development. The answers reveal that teachers were capable of understanding the KERMIT-results, but the reflection was often limited to the individual teacher. A collective reflection with colleagues was significantly less common. Mediation analyses show otherwise that collective reflection within teams is crucial to increase the actual implementa-tion of new teaching procedures. In addition, the evaluation of the supervising authorities’ support with the data-based school development was heterogeneous, ranging from ‘no help at all’ to ‘great support’. Against the background of these results we discuss approaches to foster the collective reflection of data in general as well as the role of supervising authorities to increase the impact of data-based school development.

Lücken, M., Thonke, F., Pohlmann, B., Hofmann, H., Golecki, R., Rosendahl, J., Benzing, M., & Po-erschke, J. (2014). KERMIT – Kompetenzen ermitteln. In: D. Fickermann & N. Maritzen (Hrsg). Grundlagen für eine daten- und theoriegestützte Schulentwicklung. Konzeption und Anspruch des Hamburger Instituts für Bildungsmonitoring und Qualitätsentwicklung. S. 127-154. Münster: Waxmann. Groß Ophoff, J., Hosenfeld, I. & Koch, U. (2007). Formen der Ergebnisrezeption und damit verbun-dene Schul- und Unterrichtsentwicklung. Empirische Pädagogik, 21(4), 411-427.