Matching Methods in Practice: Three Examples

By Guido W. Imbens. 2015

589 citations recorded by the source corpus.

There is a large theoretical literature on methods for estimating causal effects under unconfoundedness, exogeneity, or selection-on-observables type assumptions using matching or propensity score methods. Much of this literature is highly technical and has not made inroads into empirical practice where many researchers continue to use simple methods such as ordinary least squares regression even in settings where those methods do not have attractive properties. In this paper, I discuss some of the lessons for practice from the theoretical literature and provide detailed recommendations on what to do. I illustrate the recommendations with three detailed applications.

Subfields: Statistics and Probability

How researchers see this paper

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

In Statistics and Probability

1. M. Resa
2. Jens Ludwig
3. Shu Yang
4. Marc F. Bellemare
5. B. Mazumder
6. Patrick Sharkey
7. W. Fox
8. Dorian Carloni

Citations in the corpus

• The Selective Labels Problem: Evaluating Algorithmic Predictions in the Presence of Unobservables (2017, methodology): “(2) non-parametric methods such as propensity score matching, nearest-neighbor matching, which do not explicitly model the relationship between observed features, treatments, and outcomes [4, 15, 35, 36, 41].”
• Interpreting OLS Estimands When Treatment Effects Are Heterogeneous: Smaller Groups Get Larger Weights (2018, background): “To see this point, note that it follows from the discussion in Imbens and Wooldridge (2009) that when the conditional means of y(1) and y(0) are linear in X, we can write”
• Evaluation of subset matching methods and forms of covariate balance (2016, methodology): “This assessment is typically performed, first, by trimming the sample on the propensity score, and second, by checking balance (see, for instance, [53] and [54]).”
• What Goes Up May Not Come Down: Asymmetric Incidence of Value-Added Taxes (2020, methodology): “Our matching procedure follows Imbens (2015), and in particular that paper’s approach to analyzing the Imbens et al. (2001) lottery data (in Section 6.”
• Contract Farming and Food Security (2017, background): “[14] Imbens, Guido (2015), “Matching Methods in Practice: Three Examples, ”Journal of Human Resources 50(2):373-419.”
• Asymptotic inference of causal effects with observational studies trimmed by the estimated propensity scores (2018, background): “Unfortunately, it is common to have limited overlap in covariates between the treatment and control groups, which affects the credibility of all methods attempting to estimate causal effects for the population (King & Zeng, 2005; Imbens, 2015).”
• Aid for all: College coaching, financial aid, and post-secondary persistence in Tennessee (2016, background): “OLS estimates that hinge on extrapolation (Imbens, 2015), whereas matching will impose common support without imposing functional form assumptions about the process generating college choices.”
• The long-run effects of the 1930s HOLC “redlining” maps on place-based measures of economic opportunity and socioeconomic success (2020, methodology): “This method is a version of subclassification or stratification (Imbens 2015; Imbens and Rubin 2015), where the sample is partitioned into subclasses based on the estimated propensity score.”