NBER

Jaime Ramirez-Cuellar

University of California at Santa Barbara
Santa Barbara, CA 93106

E-Mail: EmailAddress: hidden: you can email any NBER-related person as first underscore last at nber dot org
Institutional Affiliation: University of California at Santa Barbara

NBER Working Papers and Publications

July 2020At What Level Should One Cluster Standard Errors in Paired Experiments, and in Stratified Experiments with Small Strata?
with Clément de Chaisemartin: w27609
In paired experiments, units are matched into pairs, and one unit of each pair is randomly assigned to treatment. To estimate the treatment effect, researchers often regress their outcome on a treatment indicator and pair fixed effects, clustering standard errors at the unit-of-randomization level. We show that the variance estimator in this regression may be severely downward biased: under constant treatment effect, its expectation equals 1/2 of the true variance. Instead, we show that researchers should cluster their standard errors at the pair level. Using simulations, we show that those results extend to stratified experiments with few units per strata.

National Bureau of Economic Research
1050 Massachusetts Ave.
Cambridge, MA 02138
617-868-3900
info@nber.org

Twitter RSS

View Full Site: One timeAlways