PSAgraphics: An R Package to Support Propensity Score Analysis
Main Article Content
Abstract
Propensity score analysis is a technique for adjusting for selection bias in observational data. Estimated propensity scores (probability of treatment given observed covariates) are used for stratification of observations. Within strata covariates should be more balanced between the two treatments than without the stratification. PSAgraphics is an R package that provides flexible graphical tools to assess within strata balance between treatment groups, as well as how covariate distributions differ across strata. Additional graphical tools facilitate estimation of treatment effects having adjusted for covariate differences. Several new and conventional numerical measures of balance are also provided.