Dear SAS community,
I am trying to perform propensity score matching on a dataset, where treatment is not randomly assigned.
I have used the PSMATCH procedure to compute a dataset of matched observations including their propensity scores, and all the guides I've read now state that: "If you assume that no other confounding variables are associated with both the response variable and the treatment group indicator, then after the response variable is observed and added to the matched data set, you can use the same outcome analysis on this matched data set as you would have used on the original data set".
Can someone tell me how exactly such an outcome analysis would look in practice? The way I understand it, I can now simply regress the effect of X on Y using OLS and controlling only for the observations' propensity score - thus not controlling for the covariates I used to estimate the propensity score. Is this accurate? Or have I missed something?
I hope someone can help me.
Kind regards
