Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference

TLDR

Published studies often present only a few causal estimates from many model runs, raising doubts about their accuracy and the statistical properties of the chosen specification, while matching methods offer a promising but frequently misinterpreted alternative. The authors propose a unified approach that lets researchers preprocess data with matching and then apply their preferred parametric techniques, thereby avoiding misinterpretations. The method involves preprocessing data with matching using provided software, followed by application of the researcher’s chosen parametric techniques. The procedure yields more accurate and less model‑dependent causal inferences from parametric models.

Abstract

Although published works rarely include causal estimates from more than a few model specifications, authors usually choose the presented estimates from numerous trial runs readers never see. Given the often large variation in estimates across choices of control variables, functional forms, and other modeling assumptions, how can researchers ensure that the few estimates presented are accurate or representative? How do readers know that publications are not merely demonstrations that it is possible to find a specification that fits the author's favorite hypothesis? And how do we evaluate or even define statistical properties like unbiasedness or mean squared error when no unique model or estimator even exists? Matching methods, which offer the promise of causal inference with fewer assumptions, constitute one possible way forward, but crucial results in this fast-growing methodological literature are often grossly misinterpreted. We explain how to avoid these misinterpretations and propose a unified approach that makes it possible for researchers to preprocess data with matching (such as with the easy-to-use software we offer) and then to apply the best parametric techniques they would have used anyway. This procedure makes parametric models produce more accurate and considerably less model-dependent causal inferences.