SVYSET with HIV data [message #18891] |
Thu, 05 March 2020 15:33 ![Go to next message Go to next message](/theme/default/images/down.png) |
behayes4
Messages: 24 Registered: February 2020
|
Member |
|
|
When using weights with svyset command in Stata with HIV data -- do you use HIVCLUST for the PSU? Would v023/v022 then be used for Strata? Are you not supposed to use the HIV weight information?
|
|
|
Re: SVYSET with HIV data [message #19107 is a reply to message #18891] |
Tue, 21 April 2020 17:41 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Liz-DHS
Messages: 1516 Registered: February 2013
|
Senior Member |
|
|
A response from Dr. Tom Pullum:
Quote:
Usually, when analyzing the HIV data, you merge that file with the PR or IR or MR file. You match on the cluster id, household id, and line number. Yes, hivclust is the PSU, but after the merge it will probably be named hv001 or v001 or mv001 (or hv021). The stratum id will be v023 or v022 (maybe with prefix h or m). If your analysis of the merged file includes hiv03, then you use hiv05 as the weight. For the cases that are in the AR file but not the PR (or IR or MR) file, hiv05 will be undefined, i.e. NA
|
|
|