The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Merging data files » Merging Hiv Women Men and Couples
Merging Hiv Women Men and Couples [message #4165] Sat, 11 April 2015 07:01 Go to next message
Micaela_A is currently offline  Micaela_A
Messages: 1
Registered: April 2015
Location: Italia
Hello, I'm writing a paper on the impact of gender roles on the diffusion of HIV in Lesotho. I merged the hiv dataset with the IR and MR separately and then wanted to merge these two datasets to only have the couples. I know a Couples' Recode is already available but I was wondering which keys you use to know that person 1 is the partner of person 2. Because I merged the datasets and while most of the information that I got in the process was plausible, I also got that 150 couples or so had a big difference of 'age of respondent' and 'age of partner' answered by the other member of the couple. Is it believable that a couple has a 61 years of difference in actual age and age known by the partner? I checked this information on the CR and I got the same information regarding this and other individuals with less(though still big) difference...

[Updated on: Wed, 15 April 2015 17:26]

Report message to a moderator

Re: Merging Hiv Women Men and Couples [message #4282 is a reply to message #4165] Thu, 30 April 2015 13:02 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 697
Registered: January 2013
Senior Member
The key variables to use are:
Women: V001 (cluster number) V002 (household number) V034 (line number of husband)
Men: MV001 (cluster number) MV002 (household number) MV003 (line number of man)

See for more info.
Previous Topic: Survey identifiers
Next Topic: How to merge women and child data set?
Goto Forum:

Current Time: Tue Jun 2 10:04:20 Eastern Daylight Time 2020