At What Level Should One Cluster Standard Errors in Paired and Small-Strata Experiments?

Clément de Chaisemartin (1) , Jaime Ramirez-Cuellar (2)

1 ECON - Département d'économie (Sciences Po)
2 Microsoft - Microsoft Research [Cambridge]

In clustered and paired experiments, to estimate treatment effects, researchers often regress their outcome on the treatment and pair fixed effects, clustering standard errors at the unit-ofrandomization level. We show that even if the treatment has no effect, a 5%-level t-test based on this regression will wrongly conclude that the treatment has an effect up to 16.5% of the time, an error rate much larger than the researcher's 5% target. To achieve their targeted error rate, researchers should instead cluster standard errors at the pair level. Using simulations, we show that similar results apply to clustered experiments with small strata.

Keywords: clustered standard errors; clustering; paired experiments; stratified experiments; randomized experiments; RCT

Domains: Humanities and Social Sciences; Economics and Finance

Clément de Chaisemartin, Jaime Ramirez-Cuellar. At What Level Should One Cluster Standard Errors in Paired and Small-Strata Experiments?. 2022. ⟨hal-03873897⟩