What is it about?
The choice of covariates used in statistical models should be based on influence on outcome not on association with allocation.
Featured Image
Why is it important?
Shows not just via algebraic development but also with the help of simple examples that there is no gain in efficiency in using the propensity score compared to using the covariates that are used to construct it. Furthemore, gains in validity compared to analysis of covariance are illusory and there may be some loss in efficiency.
Read the Original
This page is a summary of: Stratification for the propensity score compared with linear regression techniques to assess the effect of treatment or exposure, Statistics in Medicine, December 2007, Wiley,
DOI: 10.1002/sim.3133.
You can read the full text:
Contributors
The following have contributed to this page