Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Matching using estimated propensity scores: relating theory to practice

D B Rubin1, N Thomas

  • 1Department of Statistics, Harvard University, Cambridge, Massachusetts 02138, USA.

Biometrics
|March 1, 1996
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Geographic information in the French national health data system: Key challenges for territorial health studies.

Journal of epidemiology and population health·2026
Same author

Management of women with abnormal cervical cytology: Update of French guidelines after the implementation of HPV screening.

European journal of obstetrics, gynecology, and reproductive biology·2025
Same author

Counternull sets in randomized experiments.

The American statistician·2025
Same author

The Europa Imaging System (EIS) Investigation.

Space science reviews·2024
Same author

A Global Dataset of Potential Chloride Deposits on Mars as Identified by TGO CaSSIS.

Scientific data·2024
Same author

A resonant sextuplet of sub-Neptunes transiting the bright star HD 110067.

Nature·2023

This study refines analytic approximations for propensity score matching in observational studies. The findings offer practical guidance for designing matching studies and computing standard errors for estimators.

Area of Science:

  • Statistics
  • Observational Studies
  • Causal Inference

Background:

  • Matched sampling is crucial for evaluating treatments in observational studies.
  • Propensity score matching is vital when numerous matching variables are involved.
  • Theoretical work on matching methods provides a framework for evaluating operating characteristics.

Purpose of the Study:

  • To bridge theoretical approximations and practical applications of propensity score matching.
  • To refine analytic approximations for propensity score matching under various distributional assumptions.
  • To provide guidance for the design and analysis of matching studies.

Main Methods:

  • Completed and refined normal-based analytic approximations for propensity score matching.

Related Experiment Videos

  • Conducted Monte Carlo evaluations of analytic results under normal and nonnormal ellipsoidal distributions.
  • Applied analytic approximations to real-world data with nonellipsoidal distributions.

    • Main Results:

    • Confirmed the accuracy of analytic approximations under normal and nonnormal ellipsoidal distributions.
    • Demonstrated predictable deviations from simulation results when normal assumptions are violated.
    • Showed that theoretical expressions provide useful practical guidance even with nonellipsoidal data.

    Conclusions:

    • Matching on estimated linear propensity scores performs well across a wide range of settings.
    • Theoretical approximations serve as benchmarks for expected performance in matching studies.
    • The study provides variances for computing valid standard errors in matching analyses.