MatchIt: Nonparametric Preprocessing for Parametric Causal Inference

MatchIt implements the suggestions of Ho, Imai, King and Stuart (2007) for improving parametric statistical models by preprocessing data with nonparametric matching methods. It provides a wide range of sophisticated matching methods, which makes it possible to greatly reduce the dependence of causal inferences on hard-to-justify, but commonly made, statistical modeling assumptions. The tool is designed as a preprocessing step applied before the parametric analysis is run.

The software easily fits into existing research practices because, after preprocessing data with MatchIt, researchers can use whatever parametric model they would have used without it, yet produce inferences with substantially more robustness and less sensitivity to modeling assumptions. MatchIt is an R program and also works seamlessly with Zelig, making matching-based causal inference accessible within a familiar statistical workflow.