International studies show that the degree of teacher's professionalism and organizational characteristics highly correlate with the usage of feedback information (Visscher/Coe 2003). This fits with Luhmann’s assumption (2006) that decision premises are basic elements within organizations, on which change processes are founded. Feedback information of standardized student assessments might have an impact on these decision premises, which are persons, communication structures, and program. Connected to these decision premises, it can be assumed that there might be different approaches in using feedback information.

From the aspect of multi level structured school systems arises the question, which impact feedback information has on the relations between different levels within these systems. Especially this touches the relationship between the different levels within the individual schools and between schools and the administration. In this context, the differentiation between high and low stake tests is essential due to the possibility of constraining feedback reception. The relevance of this differentiation is reflected by comparing usage of feedback information in Norway and Germany.

The papers for this symposium use results from qualitative research projects, conducted in Norway and Germany. Interviews with actors from several levels of school systems (teachers, headmasters, administrators) are the data basis of these projects, using semi-structured interviews.

In Norway four interviews with headmasters were conducted. The German papers use results from three cooperating projects. The first comprises of 69 interviews in two states with 56 teachers, nine headmasters, and four administrators. The second project interviewed eight headmasters, 47 teachers, and four administrators. The last project had a sample of ten headmasters, 60 teachers, and four administrators. The recorded interviews were transcribed and analyzed according to qualitative content analysis. The evaluation procedure includes checking validity and interrater reliability by communicative validation.

Due to its specific settings a clear structural difference in feedback information reception between Norwegian and German headmasters was highly expected. The Norwegian paper uses school evaluation results in high stake settings and finds that the headmasters try to find a balance between accountability and development. This shows strong influences of the results on headmasters’ practice. One of the German papers looks at feedback usage by headmasters in low stake settings. Findings are that little accountability leads to little usage of feedback information, thus proving the expected outcome.

When looking at two different systems within low stake settings it is expected that a system with more reliable feedback information and coherent support strategies for schools leads to a deeper reception of feedback information, compared to another system, where usage is less intense. But the interview data does not prove this assumption, differences were not found in such a comparison.

It is expected that schools under similar conditions show no differences in internal couplings or decouplings. One project reflected on this question and found that there are no clear findings: similar conditions have no impact on how schools handle feedback information. Also within individual schools there can be different handlings of feedback information, therefore the findings are unexpected.