Causal Inference in Rebuilding and Extending the Recondite Bridge between Finite Population Sampling and Experimental Design

TL;DR

This paper develops a unified Neymanian framework for causal inference in finite population experiments, accommodating complex treatment assignments and providing new variance estimation conditions.

Contribution

It introduces novel conditions for unbiased variance estimation under complex randomization, extending classical methods to more general experimental designs.

Findings

Derived milder conditions for unbiased variance estimation.

Provided a new justification for Neyman's conservative variance estimator.

Extended the framework to factorial treatment structures.

Abstract

This article considers causal inference for treatment contrasts from a randomized experiment using potential outcomes in a finite population setting. Adopting a Neymanian repeated sampling approach that integrates such causal inference with finite population survey sampling, an inferential framework is developed for general mechanisms of assigning experimental units to multiple treatments. This framework extends classical methods by allowing the possibility of randomization restrictions and unequal replications. Novel conditions that are "milder" than strict additivity of treatment effects, yet permit unbiased estimation of the finite population sampling variance of any treatment contrast estimator, are derived. The consequences of departures from such conditions are also studied under the criterion of minimax bias, and a new justification for using the Neymanian conservative sampling variance estimator in experiments is provided. The proposed approach can readily be extended to the case of treatments with a general factorial structure.