Re: De-normalizing weights and svyset command in Stata [message #12620 is a reply to message #8573] |
Fri, 23 June 2017 14:25 |
DaniD
Messages: 13 Registered: November 2014
|
Member |
|
|
Thanks so much Bridgette and Tom for all your help!
Sorry, what I meant by "using all one gender's variables for the weighting process" was in reference to creating the svyset. I am using the men's hiv weight for my data (CR merged with AR) so I was wondering if I should be using the men's variables for the cluster and strata identifiers as well or if it's fine to use the women's variables for that part.
To clarify, my understanding is that it shouldn't matter whether I use the men's or women's cluster and strata variables in the couple's file because the unit of analysis is the couple, is this correct?
Thanks!
Dani
[Updated on: Fri, 23 June 2017 19:24] Report message to a moderator
|
|
|