This article is a response to four commenters in a symposium on our paper, Kosuke Imai, Gary King, and Clayton Nall. "The Essential Role of Pair Matching in Cluster-Randomized Experiments, with Application to the Mexican Universal Health Insurance Evaluation," [with discussion] Statistical Science, Statistical Science, Vol. 24, No. 1 (2009): Pp.65-72, copy at http://gking.harvard.edu/files/abs/cluster-abs.shtml. (Abstract: HTML | Article: PDF)Abstract
We are grateful to our four discussants for their agreement with and contributions to the central points in our article Imai, King, and Nall (2009). As Zhang and Small (2009) write, "[our article] present[s] convincing evidence that the matched pair design, when accompanied with good inference methods, is more powerful than the unmatched pair design and should be used routinely." And, as they put it, Hill and Scott (2009) ``do not take issue with [our article's] provocative assertion that one should pair-match in cluster randomized trials `whenever feasible'.'' Whether denominated in terms of research dollars saved, or knowledge learned for the same expenditure, the advantages in any one research project of switching standard experimental protocols from complete randomization to matched pair designs (along with the accompanying new statistical methods) can be considerable. In the two sections to follow, we address our discussants' points regarding ways to pair clusters (Section 2) and the costs and benefits of design- and model-based estimation (Section 3). But first we offer a sense of how many experiments across fields of inquiry can be improved in the ways we discuss in our article.
Also see related research.