Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
student outcomes: MDE = 0.19 (assumptions: 30 students/cluster, 140 clusters (schools) per arm, pairwise comparison of treatment arms, units=standard deviations, sd=1, power=80%, size=5%, ICC=0.3)
school-level outcomes (management practices): MDE = 0.27 (assumptions: 1 observation/school, 140 clusters (schools) per arm, pairwise comparison of treatment arms, units=standard deviations, residual sd=0.8, power=80%, size=5%)