The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Merging data files » Merging IR and MR files India
Merging IR and MR files India [message #20127] Mon, 28 September 2020 09:46 Go to next message
DHSUser232
Messages: 4
Registered: September 2017
Member
Hello,
I am working with the DHS 2015-2016 datasets for India. I am trying to merge women with men to create a dataset with couples as the unit of measurement. I know a couple recode is available but I would like to do it manually to retain all of the variables I need. Please find my Stata code for this merge attached.

Is this the correct way to do this? If yes, why do I have 63,996 matched couples while the couple recode dataset provided by DHS has 63,696 observations?

Thank you in advance for your help.
Re: Merging IR and MR files India [message #20143 is a reply to message #20127] Mon, 28 September 2020 14:35 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

Constructing couples is not easy. I suggest that you DO use the CR file. If you want more variables for women, then merge the CR with the IR and keep cases with _merge=3. Then drop _merge and merge the CR with the MR and keep cases with _merge=3. This will be much easier than trying to reproduce the matching of partners.

Re: Merging IR and MR files India [message #20148 is a reply to message #20127] Wed, 30 September 2020 05:44 Go to previous messageGo to next message
DHSUser232
Messages: 4
Registered: September 2017
Member
Thank you for the quick response. Can you tell me which variables I should be using to do the merges for men and women to the couple dataset please? I am confused because the variables mv001 mv002 and mv003 together do not uniquely identify men in the couples recode. Also mcaseid does not exist in the couples recode. Here is my Stata command and output:

duplicates report mv001 mv002 mv003

Duplicates in terms of mv001 mv002 mv003

--------------------------------------
copies | observations surplus
----------+---------------------------
1 | 63512 0
2 | 184 92
--------------------------------------

[Updated on: Wed, 30 September 2020 05:47]

Report message to a moderator

Re: Merging IR and MR files India [message #20196 is a reply to message #20148] Wed, 07 October 2020 07:47 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

Following is another response from DHS Research & Data Analysis Director, Tom Pullum:

With the India files, you need to include region in the identifiers. For women, you merge with v001 v002 v003 v024 and for men you merge with mv001 mv002 mv003 mv024. Sorry for forgetting that. The India surveys are the only ones for which you need to include region.
Previous Topic: Merging Couple File with KR File NFHS-4 (India)
Next Topic: Merging Egypt 2000 KR and HW files
Goto Forum:
  


Current Time: Thu Mar 28 20:42:40 Coordinated Universal Time 2024