Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
We performed a power analysis based on the results of a similar experimental study (Kimbrough et al.,2014). In this study, the authors vary low and high endowment-inequality and introduce a device to avoid conflict. They find that the probability of conflict when inequality is low, is on average, 37% lower than when inequality is high (0.40 vs. 0.63, respectively). Assuming a similar effect in the probability of conflict across our treatments, we will need 80 subjects to ensure a power of at least 80% for a significance level of 5%.
Kimbrough, E. O., Sheremeta, R. M., & Shields, T. W. (2014). When parity promotes peace: Resolving conflict between asymmetric agents. Journal of Economic Behavior and Organization, 99, 96–108.