The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » HIV » Reweighting after merging IR and AR files
Reweighting after merging IR and AR files [message #24318] Wed, 20 April 2022 20:12 Go to next message
dasanchezp is currently offline  dasanchezp
Messages: 11
Registered: February 2017
Location: San Sebastian
Member
Hello,

I want to compute ASFR and TFR by HIV status but I have problems with the weighting. For instance, in ZW TFR is 4.02 in 2015 (using IR file and v005). However, if I merge the IR and AR files, and then compute TFR (using hv005), TFR is 3.97, which is logical since the sample is smaller. How can I reweight v005 or hv005 so I can get TFR=4.02 from the merged file?

Thanks!
Re: Reweighting after merging IR and AR files [message #24324 is a reply to message #24318] Fri, 22 April 2022 08:49 Go to previous messageGo to next message
Janet-DHS is currently offline  Janet-DHS
Messages: 901
Registered: April 2022
Senior Member
Following is response from DHS Research & Data Analysis Director, Tom Pullum:

I don't understand your reference to hv005. This must be a typo and you are referring to hiv05, the weight in the AR file. I assume you are using either the fertility rates program on GitHub or Bruno Schoumaker's "tfr2" program. If you want to get the rates for a subpopulation, such as women who are HIV positive, the easiest way is to construct a data file for just that subpopulation, and run the program.

To get the TFR for all women, you would not use the merged IR/AR file. Just use the IR file with v005 as the weight. To get the TFR for women who are HIV positive, you do the merge, restrict to women with hiv03=1, set hiv05 as the weight variable, and run the program. If you want the TFR for women who are in the AR file, regardless of their test result (I'm not sure why you would want this), you do the merge, restrict to women with _merge==3, set hiv05 as the weight variable, and run the program. You can also (if needed) get the TFR for women in the AR file with hiv03=0 using hiv05 or women who are in the IR file but not the AR file with v005. What's crucial is that the weight is hiv05 for respondents in the AR file. Please let us know if this does not answer your question.
Re: Reweighting after merging IR and AR files [message #24355 is a reply to message #24324] Wed, 27 April 2022 17:37 Go to previous message
dasanchezp is currently offline  dasanchezp
Messages: 11
Registered: February 2017
Location: San Sebastian
Member
Hello.

Yes, it was a typo. I meant hiv05. Thank you for your answer.

Have a nice day.
Previous Topic: HIV Testing
Next Topic: How to treat missing data in analysis
Goto Forum:
  


Current Time: Wed Dec 11 18:40:47 Coordinated Universal Time 2024