Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
To estimate the minimum detectable effect (MDE) we simulate the model a thousand times using the randomized sample and historical data of exit. In 2018, there was approximately a 9% rate of child exit from residence. This is the rate of exit we use for the control group. For the treatment group, we increment the rate of exit to estimate de MDE.
For a power of .88 we can detect an impact of 6% (Increasing exit from 9% to 9.54%).