Matching using estimated propensity scores: relating theory to practice.

By Donald B. Rubin, Neal Thomas. 1996

How researchers see this paper

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

Citations in the corpus

FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference (2017, methodology): “(R2) represents the desire for estimating the most granular causal estimates possible while (R3) addresses the need for matching covariates to be associated with the potential outcomes (Rubin and Thomas, 1996; Brookhart et al., 2006).”
Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference (2007, unknown): “We also know from theoretical and simulation results that, in a wide range of scenarios, using matched samples can result in substantial bias and variance reduction, compared with using random samples of the same size (Rubin and Thomas 1992, 1996).”
Propensity score techniques and the assessment of measured covariate balance to test causal associations in psychological research (2010, methodology): “…to discuss the benefits of assessing the balance between treatment and comparison groups and selecting a PS method before the treatment effect is estimated (Rubin, 2001; Schafer & Kang, 2008; Rubin & Thomas, 1996; Ho et al., 2007a; Imai et al., 2008; Imai & van Dyk, 2004; Hansen & Kleyman, 2008).”
Optimal Full Matching and Related Designs via Network Flows (2006, background): “…where covariance adjustment may not be appropriate, a theme with echoes in later work of Rubin (1976) and his coauthors (Rosenbaum and Rubin 1985b; Rubin and Thomas 1996); related methodological literature promotes the use of matching in combination with covariance adjustment in order to…”
Causal Inference With General Treatment Regimes (2004, background): “Indeed, there is a large literature on the advantage of adjusting for the estimated propensity score rather than the true propensity score in both observational studies and randomized experiments (e.g., Rosenbaum 1987; Rubin and Thomas 1992, 1996; Hill, Rubin, and Thomas 1999).”
Linking Smallholders to Markets: Determinants and Impacts of Farmer Collective Action in Kenya (2012, background): “However, Rubin and Thomas (1996) argue that probit models for PSM should not be “trimmed” in the name of parsimony.”
Propensity score adjustment as a weighting scheme for volunteer panel web surveys (2006, background): “It seems to be a reasonable practice to include all available covariates from a given dataset, as Rubin and Thomas (1996) suggest.”
Affinely invariant matching methods with discriminant mixtures of proportional ellipsoidally symmetric distributions (2006, background): “The Rubin and Thomas [18] results predict a gain of efficiency for differences in covariate means by a factor of approximately two, and Hill et al.”
Impact of participation in extra-curricular activities during college on graduate employability: an empirical study of graduates of Taiwanese business schools (2014, methodology): “To satisfy the assumption of ignorable treatment assignment, the variables associated with treatment assignment and the outcome should be included in the analysis (Rubin and Thomas 1996).”
Matching With Multiple Control Groups With Adjustment for Group Differences (2008, background): “Settings with maximum possible bias reduction greater than 100% indicate settings where 100% bias reduction is theoretically possible, and thus those are settings that are particularly conducive to matching (Rubin & Thomas, 1996).”