Hey all,
I'm working on a Propensity Score Matching and I use the psmatch2 package for that (Stata 12). The problem is about balancing after, using the provided pstest command. I tried a lot of different assignment models and finally I managed to get a quite low Rubins B around 30. I know it should not be higher than 25. In my final model, %concern and %bad are zero, still the B is 30. I'm not really sure what to do here now, just accept it?
The second question is related. In the helpfile the authors write:
I checked the source (Rubin, D.B. (2001), "Using Propensity Scores to Help Design Observational Studies: Application to the Tobacco Litigation", Health Services & Outcomes Research Methodology 2, 169-188) but I could not find the original recommendation. Why is 25 such an important threshold and how bad are values that are larger? Maybe you guys have ideas. Where can I find more information about Rubins B? Thank you!
I'm working on a Propensity Score Matching and I use the psmatch2 package for that (Stata 12). The problem is about balancing after, using the provided pstest command. I tried a lot of different assignment models and finally I managed to get a quite low Rubins B around 30. I know it should not be higher than 25. In my final model, %concern and %bad are zero, still the B is 30. I'm not really sure what to do here now, just accept it?
The second question is related. In the helpfile the authors write:
Rubins' B (the absolute standardized difference of the means of the linear index of the propensity
score in the treated and (matched) non-treated group) and Rubin's R (the ratio of treated to (matched)
non-treated variances of the propensity score index). Rubin (2001) recommends that B be less than 25 and
that R be between 0.5 and 2 for the samples to be considered sufficiently balanced. An asterisk is
displayed next to B and R values that fall outside those limits.
score in the treated and (matched) non-treated group) and Rubin's R (the ratio of treated to (matched)
non-treated variances of the propensity score index). Rubin (2001) recommends that B be less than 25 and
that R be between 0.5 and 2 for the samples to be considered sufficiently balanced. An asterisk is
displayed next to B and R values that fall outside those limits.
