Merging NFHS 3 and NFHS 4 (India) [message #28709] |
Wed, 28 February 2024 12:21 ![Go to next message Go to next message](/theme/default/images/down.png) |
Soumya Agrawal
Messages: 2 Registered: February 2024
|
Member |
|
|
Greetings.
I am analyzing the impact of a policy related to child anthropometry. However, since NFHS 4 contains information about children who are born after 2010, I wanted to get information about children born even before that. It would be great is you could help me in merging the two datasets. I know that the caseids are different in the two districts. I thought of merging at the district level but NFHS -3 does not contain a district-level dataset.
|
|
|
Re: Merging NFHS 3 and NFHS 4 (India) [message #28725 is a reply to message #28709] |
Thu, 29 February 2024 07:31 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Bridgette-DHS
Messages: 3151 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
You can put the two surveys into the same file by appending them. In Stata, the command is "append" and does not involve matching the ID code, region, district, or anything else. I think what you want to do is append the two files and construct a variable "NFHS". You could, for example, assign NFHS=3 for the NFHS-3 and NFHS=4 for the NFHS-4. You would then treat "NFHS" as a covariate, exactly the same way you would treat urban/rural residence (hv025 or v025) in the analysis of a single survey. Or you could think of "year of birth" as "birth cohort" and that would be your covariate. With this approach you could look at change within sub-populations. As you say, district is not included in the NFHS-3 or earlier rounds, so that limits what you can do with geography, but many other covariates are included in the NFHSs.
|
|
|