The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Domestic Violence » MERGING DV CHARACTERISITCS TO CHILD OUTCOME
MERGING DV CHARACTERISITCS TO CHILD OUTCOME [message #28902] Tue, 26 March 2024 12:33 Go to next message
Noyel Sebastian is currently offline  Noyel Sebastian
Messages: 9
Registered: March 2024
Member
I am trying to understand the effects of domestic violence on certain health indicators in India. Health indicators I try to analyse include child mortality, child health, and maternal health. I am using NFHS 2-5 for this purpose. I tried to merge BR and IR files to have the women's DV characteristics with child data, but the exact merges turned to 0. I am confused here.

I learned that the BR file has five births, but the IR has 20. Is this creating the matching problem?

How could I navigate this? Can you help me sort this out? I want the DV characteristics of the mother to understand the health indicators. If someone could hint me with the STATA code, it would be a great help.

Can I create a child mortality dummy using this code from DHS GitHub (link)? Also, can this code generate a child mortality dummy from the KR or IR file, or can it only use the BR file?


Thank You
Re: MERGING DV CHARACTERISITCS TO CHILD OUTCOME [message #28924 is a reply to message #28902] Thu, 28 March 2024 15:20 Go to previous messageGo to next message
Janet-DHS is currently offline  Janet-DHS
Messages: 720
Registered: April 2022
Senior Member
Following is a response from DHS staff member, Tom Pullum:

What do you mean by "a child mortality dummy"? There is a variable b5 in the KR and BR files that is 1 if the child survived to the date of interview and 0 if died.

The IR and BR files go out to a maximum of 20 children ever born. The KR file goes out to a maximum of 5 births because it only includes children born in the past 5 years.

The following Stata lines will add the DV variables for the mother to the children's data.

* Specify a workspace
cd e:\DHS\DHS_data\scratch

* Save a version of the IR file restricted to the DV variables (or other variables
*   you want to add to the KR or BR file), plus the ID variables
 
use "...IAIR7EFL.DTA" , clear
keep v001 v002 v003 v024 d*
save dtemp.dta, replace

* Open the KR or BR file
use "...IAKR7EFL.DTA" , clear
gen in_file=1
tab in_file
merge m:1 v001 v002 v003 v024 using dtemp.dta

* Restrict to cases in the child file
tab _merge in_file,m
keep if in_file==1

* Save this file with a new name
Re: MERGING DV CHARACTERISITCS TO CHILD OUTCOME [message #28966 is a reply to message #28924] Wed, 03 April 2024 12:31 Go to previous message
Noyel Sebastian is currently offline  Noyel Sebastian
Messages: 9
Registered: March 2024
Member
Thank you for the response.

Thank You
Previous Topic: Trouble sorting DV questions
Next Topic: Intimate Partner Violence
Goto Forum:
  


Current Time: Sun May 26 12:09:38 Coordinated Universal Time 2024