The essential role of pair matching in cluster-randomized experiments, with application to the Mexican Universal Health Insurance Evaluation

By Kosuke Imai, Gary King, Clayton Nall. 2009

298 citations recorded by the source corpus.

A basic feature of many field experiments is that investigators are only able to randomize clusters of individuals — such as households, communities, firms, medical practices, schools, or classrooms — even when the individual is the unit of interest. To recoup some of the resulting eciency loss, many studies pair similar clusters and randomize treatment within pairs. Other studies (including almost all published political science field experiments) avoid pairing, in part because some prominent methodological articles claim to have identified serious problems with this “matched-pair cluster-randomized” design. We prove that all such claims about problems with this design are unfounded. We then show that the estimator for matched-pair designs favored in the literature is appropriate only in situations where matching is not needed. To address this problem without modeling assumptions, we generalize Neyman’s (1923) approach and propose a simple new estimator with much improved statistical properties. We also introduce methods to cope with individual-level noncompliance, which most existing approaches incorrectly assume away. We show that from the perspective of, among other things, bias, eciency, or power, pairing should be used in cluster-randomized experiments whenever feasible; failing to do so is equivalent to discarding a considerable fraction of one’s data. We develop these techniques in the context of a randomized evaluation we are conducting of the Mexican Universal Health Insurance Program.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/0444a1b897addb972316e9d13cfeef0e94287615

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Unbiased Estimation of the Average Treatment Effect in Cluster-Randomized Experiments (2015, background): “First, as Imai, King and Nall (2009) suggest, this estimator is not location invariant.”
Adjusting treatment effect estimates by post‐stratification in randomized experiments (2013, background): “(2011) or Imai et al. (2008). Regardless, under the PATE approach, the variances of all the estimators increase, but the substance of this paper’s findings remains.”
Misunderstandings between experimentalists and observationalists about causal inference (2008, background): “) However, despite this gain in efficiency, a blocked experiment has fewer degrees of freedom and so can have lower power in small samples; simulations indicate that this is not an issue except in very small data sets (Imai et al., 2007), and so blocking is almost always preferable when feasible.”