Replication Data for: Non-ignorable Attrition in Pairwise Randomized Experiments

In pairwise randomized experiments, what if outcomes of some units are missing? One solution is to delete missing units (the unitwise deletion estimator, UDE). If attrition is nonignorable, the UDE is biased. Scholars might employ the pairwise deletion estimator (PDE) which also deletes the pairmates of missing units. This study proves that the PDE can be biased but more efficient than the UDE; the pairwise variance estimator of the PDE is unbiased in the super-population. I argue that it is easier to interpret the PDE as a causal effect than the UDE. I also propose a new variance estimator.