The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » India » District level Individual panel data
District level Individual panel data [message #14967] Tue, 22 May 2018 12:23 Go to next message
anirban.chattopadhyaya@gm is currently offline  anirban.chattopadhyaya@gm
Messages: 4
Registered: May 2018
Member
Hi,

I am trying to create panel dataset at individual across all 4 NFHS rounds.
I have encountered certain issues as I am going over the datasets.
1. For the 1st round, birth records and children level data don't have district level identifiers.
2. Recoded Individual dataset does not have district level identifier either. But the raw individual dataset has. But on trying to merge using "caseid" to locate common units, the match result is 0.

Help will be appreciated.
Re: District level Individual panel data [message #14976 is a reply to message #14967] Wed, 23 May 2018 07:49 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

Are you trying to link individuals across surveys? That would be the normal implication of the term "panel". This cannot be done. Probably no one has appeared in more than one round of the NFHS, and even if someone has appeared more than once, it would be impossible to link the separate appearances. The individual id code (caseid) normally consists of the state code, cluster code, household code, and line number within the household. The only one of these that applies to successive surveys is the state code.
Re: District level Individual panel data [message #14979 is a reply to message #14976] Wed, 23 May 2018 08:02 Go to previous messageGo to next message
anirban.chattopadhyaya@gm is currently offline  anirban.chattopadhyaya@gm
Messages: 4
Registered: May 2018
Member
Thank you for the reply.

No, I am not trying to link the same individuals across different rounds. What I am trying to create pseudo-panel dataset at district levels(say by taking a mean, median, etc.).

To ask the question on caseid more specifically, how can one match observations amongst the 'individual raw' and 'individual recoded' from NFHS 1?

Re: District level Individual panel data [message #14980 is a reply to message #14979] Wed, 23 May 2018 09:13 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member

Following is another response from Senior DHS Stata Specialist, Tom Pullum:

For NFHS-1 I see files IAIR23FL.dta and IAPR23FL.dta (the Stata versions). These are the recode files for individual women and individuals in the households. I don't know of any "raw" data files. Perhaps you have access to something I don't know about. What files are you using?
Re: District level Individual panel data [message #14983 is a reply to message #14980] Wed, 23 May 2018 09:26 Go to previous messageGo to next message
anirban.chattopadhyaya@gm is currently offline  anirban.chattopadhyaya@gm
Messages: 4
Registered: May 2018
Member
I am referring to IAIR23DT(.dta) and IAIQ21DT(.dta) where in the directory they are mentioned as Individual Recode and Individual Raw respectively.

Re: District level Individual panel data [message #15007 is a reply to message #14983] Thu, 24 May 2018 17:49 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member

Following is another response from Senior DHS Stata Specialist, Tom Pullum:

You are right--our data processing archives for India include "raw" data files for NFHS-1. However, (1) they are not user-friendly (we in the analysis group never touch those files) and (2) they would not contain more information about geography than is in the recoded files. What you propose to do--tracking districts across the successive NFHS rounds--sounds interesting, but I don't see how we can help you with it.
Re: District level Individual panel data [message #15074 is a reply to message #14967] Wed, 30 May 2018 09:21 Go to previous message
boyle014 is currently offline  boyle014
Messages: 78
Registered: December 2015
Location: Minneapolis
Senior Member
IPUMS-DHS includes harmonized geography across all the India surveys.

If you use IPUMS-DHS, I'd recommend clicking on "Select Samples" to limit the interface to just the India surveys. The Indian integrated geography variable is GEO_IA1992_2015. You will find it under Topics -> Geography -> Integrated Geography.

Since you've started with the regular DHS files, you'll probably want to change the default to see the original DHS variable names in IPUMS-DHS. This option is in the upper right corner of the variable discovery screen:

index.php?t=getfile&id=888&private=0

Good luck with your research.


Professor Elizabeth Boyle
Sociology & Law, University of Minnesota, USA
Principal Investigator, IPUMS-DHS
Previous Topic: Occupation Codes (NFHS - 4 India)
Next Topic: Malaria
Goto Forum:
  


Current Time: Thu Mar 28 09:04:21 Coordinated Universal Time 2024