Propensity Scoring
Keywords
Covariate adjustment; Optimal matching; Propensity score adjustments; Propensity score matching; Propensity scores; Quasi-experiments; Selection bias; Standardized bias estimate; Stratifying; Weighting
Abstract
A propensity score is the conditional probability that a participant will be assigned to a treatment condition given a set of observed covariates. These scores are used as a single, aggregated covariate to make statistical adjustments through matching, weighting, stratifying or covariate adjustments to reduce selection bias in nonrandomized designs. Assuming that propensity scores are computed from all relevant covariates, propensity score adjustments should balance the treatment and control groups in an experiment and provide an unbiased estimate of the treatment effect.
Publication Date
3-26-2015
Publication Title
International Encyclopedia of the Social & Behavioral Sciences: Second Edition
Number of Pages
140-146
Document Type
Article; Book Chapter
Personal Identifier
scopus
DOI Link
https://doi.org/10.1016/B978-0-08-097086-8.10557-4
Copyright Status
Unknown
Socpus ID
85043435627 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85043435627
STARS Citation
Clark, M. H., "Propensity Scoring" (2015). Scopus Export 2015-2019. 1336.
https://stars.library.ucf.edu/scopus2015/1336