Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
In this pilot, we work with a small sample. To detect the minimum detectable effect (MDE) associated with our sample size, we use a cluster power calculation for the parents' sample. Each school represents a cluster. We fix the statistical level at 5%, and the statistical power at 80%. In our analysis, we work with standardized outcomes, estimating all the effects in units of standard deviations. Given that, we consider a variance of one for the outcomes of interest. Considering an intracluster correlation of 0,05, we will be able to detect an effect of 0,46, which is a large effect.