Do Instrumental Variables Belong in Propensity Scores?
Open accessAbstract
Propensity score matching is a popular way to make causal inferences about a binary treatment in observational data.The validity of these methods depends on which variables are used to predict the propensity score.We ask: "Absent strong ignorability, what would be the effect of including an instrumental variable in the predictor set of a propensity score matching estimator?"In the case of linear adjustment, using an instrumental variable as a predictor variable for the propensity score yields greater inconsistency than the naive estimator.This additional inconsistency is increasing in the predictive power of the instrument.In the case of stratification, with a strong instrument, propensity score matching yields greater inconsistency than the naive estimator.Since the propensity score matching estimator with the instrument in the predictor set is both more biased and more variable than the naive estimator, it is conceivable that the confidence intervals for the matching estimator would have greater coverage rates.In a Monte Carlo simulation, we show that this need not be the case.Our results are further illustrated with two empirical examples: one, the Tennessee STAR experiment, with a strong instrument and the other, the Connors' (1996) Swan-Ganz catheterization dataset, with a weak instrument.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Adjusting for an instrumental variable Z amplifies the bias created by U.
“Recent work has shown, however, that such adjustment would amplify the bias created by U [25–27].”
From Linear Models: A Useful "Microscope" for Causal Analysis (2013) · cites this paper as [25]
Section 2.2 shows that regression-based propensity-score adjustment using an instrument multiplies the naive estimator’s bias by a factor greater than or equal to one. The claim omits this method-specific qualification: Section 2.3 says the differences in inconsistency cannot be signed in general for other propensity-score methods. (LLM, conf=0.86)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 17 · Cited by 81
Research works printed but not linked: 7
- [2]2. Brookhart, M. A., Schneeweiss, S., Rothman, K. J., Glynn, R. J., Avorn, J., and Sturmer, T. (2006). Variable selection for propensity score models. American journal of epidemiology, 163(12):1149-1156.
- [4]4. D'Agostino, Jr., R. B. and D'Agostino, Sr., R. B. (2007). Estimating treatment effects using observational data. Journal of the American Medical Association, 297(3):314-316.
- [8]8. Heckman, J. J. and Robb, R. (1985). Alternative methods for evaluating the impact of interventions: an overview. Journal of econometrics, 30(1-2):239-267.
- [9]9. Hirano, K. and Imbens, G. W. (2001). Estimation of causal effects using propensity score weighting: an application to data on right heart catheterization. Health services & out- comes research methodology, 2(3-4):259-278.
- [12]12. Murphy, D. and Cluff, L. (1990). Support: Study to understand prognoses and preferences for outcomes and risks of treatments-study design. Journal of Clinical Epidemiology, 43(suppl):1S-123S.
- [16]16. Rosenbaum, P. R. and Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal efffects. Biometrika, 70(1):41-55.
- [22]22. Weitzen, S., Lapane, K. L., Alicia Y. Toledano, A. L. H., and Mor, V. (2005). Weakness of goodness-of-fit tests for evaluating propensity score models: the case of the omitted confounder. Pharmacoepidemiology and drug safety, 14(4):227-238.
- Authors
- Bhattacharya, J., & Vogt, W.
- Affiliations
- National Bureau of Economic Research (US)
- RAND Corporation (US)
- ORCID
- Bhattacharya, Jay
- Venue
- National Bureau of Economic Research
- Publisher
- National Bureau of Economic Research
- Published
- 2007-09
- OpenAlex
- W1507480792
- Cited by
- 81
- References
- 17
- Citations by year
- 2026: 1 · 2025: 3 · 2024: 4 · 2023: 7 · 2022: 3 · 2021: 3 · 2020: 5 · 2019: 8 · 2018: 6 · 2017: 6 · 2016: 6 · 2015: 2 · 2014: 10 · 2013: 7 · 2012: 3
- Type
- report
- Language
- en
- Open access
- green
- doi.org
- Locations
- National Bureau of Economic Research (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- SSRN Electronic Journal (submittedVersion)
- Funders
- National Science Foundation
- National Institute on Aging
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Education › School Choice and Performance
- Keywords
- propensity score matching, instrumental variables, binary treatment, causal inference, ignorability, estimator inconsistency, stratification, instrument strength, confidence interval coverage, observational data
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
