The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Data » Merging data files » Issues with Merging Congo 2009 HIV Data
Issues with Merging Congo 2009 HIV Data [message #11712] Thu, 02 February 2017 14:11 Go to next message
Bouboulina
Messages: 6
Registered: February 2017
Member
Hello,

I am trying to create a file that merges the women's and men's surveys with the HIV test results from the Congo 2009 survey.

I have done this for Burundi, the DRC, and Cameroon by stacking the women's and men's data files and then merging them with the HIV file.

The landing page for the 2009 Congo AIS survey indicates that the men's survey exists (see "Survey Characteristics"). However, there is no such file available for download.

I thought I might be able to merge the household member file with the HIV file and get the same result. But there are inconsistencies between the household member dataset and the women's dataset. I am using the cluster, household, and line variables to identify each person across datasets (hv001, hv002, and hvidx for the household member data and v001, v002, and v003 for the women's data). However, the women's dataset includes some persons who are listed as male in the household member dataset, based on the variable hv104.

Could you please take a look and let me know how to proceed?

Many thanks.
Re: Issues with Merging Congo 2009 HIV Data [message #11714 is a reply to message #11712] Thu, 02 February 2017 15:28 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1437
Registered: February 2013
Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:

In AIS surveys (most, if not all of them) there is no MR file. In such surveys the IR file includes both men and women, who are distinguished by a variable called "aidsex" that is 1 for men and 2 for women (same coding as hv104 in the PR file). If you want to handle the merges in the same way as for the other surveys, the easiest approach may be to construct a new IR file that contains only the cases with aidsex=2, and a pseudo MR file that consists only of cases with aidsex=1, and has the v variables renamed as mv variables. But you can definitely do your merge with the IR file as it is, so long as you are careful.

Re: Issues with Merging Congo 2009 HIV Data [message #11737 is a reply to message #11714] Sun, 05 February 2017 14:10 Go to previous message
Bouboulina
Messages: 6
Registered: February 2017
Member
Thank you, this is very helpful!
Previous Topic: Merging/Appending files into one combined dataset
Next Topic: Merging PR, KR and IR
Goto Forum:
  


Current Time: Sun Aug 19 07:39:24 Eastern Daylight Time 2018