which STATA weighting option should I use for IPUMS international datasets? Can I use iweight ie [iweight=perwt]?

INTERNATIONAL
1

Good morning,

I am analysis a total of 40 datasets downloaded from IPMUS’s website. I read that for some countries (e.g. Mexico 2010, South Africa 2011) weights (PERWT in my case) must be used to generate representative statistics. I am using STATA which has many weighting options (fweight, iweight…) and I was wondering which one is suitable for my analysis.

In order to estimate the prevalence of disability (proportion of disabled people in the population) in South Africa, is it correct if I use the following codes: sum DISABLED [iweight=perwt]

Thanks,

2

The various weighting options available in Stata are each “correct” in specific situations, based on details of the analysis. We generally do not provide technical guidance on these topics. Our focus is on providing access to harmonized data. I can point you toward a few existing and helpful resources: the first from Stata and the second from the Carolina Population Center. However, as long as I am understanding what you are trying to do correctly, it sounds like you might want to be using the fweight option.