- Difference-in-Differences website (with Bret Zeldow and Laura Hatfield).
- Foundations of design-based causal inference. Standard expositions often compress formal derivations; these guides show every step. The results concern the Difference-in-Means estimator and two estimands: the sample average treatment effect (SATE), which averages individual effects over the study units, and the population average treatment effect (PATE), which averages individual effects over the superpopulation from which the study draws its units.
