Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
Assuming alpha= 10% (significance level, given our sample size), with a power set to 80%, on a binary outcome (e.g., has increased or not consumption), 12 p.p. difference is the minimum detectable effect (MDE). With continuous outcomes (e.g. mean income growth expectations), and a standard deviation of outcomes of 1, the MDE amounts to 0.25.