Dear Statalist users,

I have been digging previous posts regarding my problem and do not seem to find the answer. Currently I'm doing analysis using psmatch2 for a data that can be grouped into strata (province, in my case). I have 33 provinces in my dataset. I would like to perform matching within province, and as the helpfile suggested, I did this:

g att = .
egen g = group(groupvars)
levels g, local(gr)
qui foreach j of local gr {
psmatch2 treatvar varlist if g==`j', out(outvar)
replace att = r(att) if g==`j'
}
sum att

To assess the quality of the matching, one of the steps I did is I follow up the above command with pstest $controlvar, both hist.
The pstest command will only test for covariate balance in my latest strata, that is the 33rd province.
My question is, does it mean that I will have to check the quality of matching for 33 estimations separately? The reason I'm asking is that because the paper I refer to, who also does matching within province, seems to only provide single assessment on the overall quality of psmatch2.

I would love to learn more from other users who has done the same experience in the past. Thank you!