Dr Garrido in her supplementary presents a table showing a summary following her matching (see screenshot from article below)
However I do not know how to produce the 'Summary of the distribution of the abs(bias)' table - it does not part of the output following pstest. Can you help? See my code below
Methods for constructing and assessing propensity scores - PubMed (nih.gov)
I have managed to
set seed 1234
gen sort_id = uniform()
sort sort_id
//Calculate psscore using psmatch2
psmatch2 $treatment $xlist, outcome($ylist) neighbor(3) bw (0.06) common logit
//Checking for balance using:
pstest $xlist, treated($treatment) both
However with the pstest I do not get her table which is the summary.
I also tried - again pstest does not show the summary.
//Checking for distribution of pscores between treatment and control and checking for balance
*If not balanced, it is split into blocks until the means between treatment and control and the same
pscore $treatment $xlist,pscore(myscore)blockid(myblock)comsup detail
//myscore - name of PS scores
//myblock - blocks of the propensity scores
*Which is shown by:
psgraph, treated($treatment) pscore(myscore)
//Check for balance between covariates in treatment and comparison groups within blocks of propensity scores. Ensures the propensity score distribution is similar within each quintile group.
*Bias (not bias reduction) between treated and control need to be <10%
*Cariance between 0.5 to 2 - variance of 1 indicates good balance
*SMD recommend is 0.1 and 0.5
pstest $xlist, treated($treatment) both
However I do not know how to produce the 'Summary of the distribution of the abs(bias)' table - it does not part of the output following pstest. Can you help? See my code below
Methods for constructing and assessing propensity scores - PubMed (nih.gov)
I have managed to
set seed 1234
gen sort_id = uniform()
sort sort_id
//Calculate psscore using psmatch2
psmatch2 $treatment $xlist, outcome($ylist) neighbor(3) bw (0.06) common logit
//Checking for balance using:
pstest $xlist, treated($treatment) both
However with the pstest I do not get her table which is the summary.
I also tried - again pstest does not show the summary.
//Checking for distribution of pscores between treatment and control and checking for balance
*If not balanced, it is split into blocks until the means between treatment and control and the same
pscore $treatment $xlist,pscore(myscore)blockid(myblock)comsup detail
//myscore - name of PS scores
//myblock - blocks of the propensity scores
*Which is shown by:
psgraph, treated($treatment) pscore(myscore)
//Check for balance between covariates in treatment and comparison groups within blocks of propensity scores. Ensures the propensity score distribution is similar within each quintile group.
*Bias (not bias reduction) between treated and control need to be <10%
*Cariance between 0.5 to 2 - variance of 1 indicates good balance
*SMD recommend is 0.1 and 0.5
pstest $xlist, treated($treatment) both
