Stability of school effects in Dutch secondary education

This chapter reports the results of an investigation into the stability across both years and subjects of school effects in Dutch secondary education. What distinguishes the present study from previous ones dealing with the stability of school effects is the fact that two types of instability have been investigated simultaneously. Not only the instability across years and subjects has been established, but also their interaction. This interaction effect should be interpreted as follows: a school may produce outstanding results with respect to a certain subject one year, while the next year the same school may reveal rather poor results for the same subject. The following specific research questions were addressed: (1) What percentage of the total variance in student achievement per subject can be attributed to differences between schools and to what extent are these effects stable across years? (2) To what extent are school effects stable across subjects? (3) To what extent does the instability across years interact with the instability across subjects? The school effects per subject were found to be fairly stable across years, but schools appeared to produce remarkably divergent results across subjects. A substantial interaction effect of instability across years and subjects was detected as well. The findings largely corroborate the conclusions of recent studies stressing the important role of departments in secondary schools. The general differences between schools with respect to student achievement urned out to be very modest, making up no more than 4 % of the total variance in student achievement. Stability of School Effects in Theory and Research Much research in the field of school effectiveness has been inspired by a strong ambition to direct educational policy (Ralph & Fennessey, 1983). Many authors hav