Calculating standardized differences with complex survey data

[Catherine Bulka]

Hello all,

I am working with complex survey data similar to NHANES and have conducted a propensity score analysis. I then used my propensity score model to calculate inverse probability of treatment weights, and multiplied those by the survey weights. In propensity score analyses, I have commonly seen standardized differences used to show balance. Normally I would use pbalchk . My question is how can I calculate these in Stata, accounting for the primary sampling unit and strata clustering in addition to the multiplied IPTW/survey weights? I recognize I may need to manually code this using svy: tabulate or svy: mean commands.

Thanks in advance.
Catherine

[Neel Patel]

Hello Ms. Bulka,
I have the same issue, currently working on NIS database which contains weights. Please let me know if you have found a solution for the above question. Thank you.

[Adam Alhassan]

Hi Catherine Bulka Neel Patel ,
My understanding was to use pbalchk with the final weights (calculated as the product between ipw and survey weight). Curious to know if anyone found a solution.