Effective School Improvement in Mathematics

This article addresses the evaluation of the Mathematics Improvement Programme. The results show that the programme improved the learning results of pupils in grade 3 with half a standard deviation and reduced the percentage of struggling pupils to less than 1%. Fifteen percent of the variance in pupil results is to be explained at the school level. About a quarter of this 15% can be explained by differences between the experimental and the comparison group. All of this condition variance is explained by the experimental variables. Five out of 10 implementation features contribute significantly to differences in pupil results.