The DHS Program User Forum: India » District-Level Analysis

Home » Countries » India » District-Level Analysis

Show: Today's Messages :: Show Polls :: Message Navigator

District-Level Analysis [message #14907]

Mon, 14 May 2018 11:21

slh277
Messages: 4
Registered: May 2018

Member

Hi,

I am trying to analyze the prevalence of anemia, stunting, wasting, and underweight in children in a specific age range (10-18 months) and in Mumbai from NFHS-4 data (on SAS).

I'm having trouble determining the variable for district (to specify Mumbai) as well as the codes for the different states.

Is it SDISTRI (as i saw on another forum posting here)? If so, which data set is it in? And is there a list of the codes and which district/state they represent? This variable doesnt seem to exist in the Recode Manual.

Thanks,
Samantha

Report message to a moderator

Re: District-Level Analysis [message #14917 is a reply to message #14907]

Tue, 15 May 2018 10:46

Bridgette-DHS
Messages: 3199
Registered: February 2013

Senior Member

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

Have you obtained access to the NFHS-4 data files? There are SAS versions on the website. The anthropometry variables are in both the PR and KR files. The PR file includes all children in the household. The KR file has slightly fewer children, those whose mother is living and in the same household as the child. We recommend using the PR file unless you want to relate the anthropometry variables to characteristics of the mother or other characteristics of the child, in which case you have to use the KR file. In the PR file, state is given by hv024 and district is given by shdistri. The labels for these variables give the codes and names of the states and districts, respectively.

Greater Mumbai includes two separate districts (Mumbai District and Mumbai Suburban District), so be sure to select both of those districts if you are interested in the results for Greater Mumbai.

Report message to a moderator

Re: District-Level Analysis [message #14925 is a reply to message #14917]

Wed, 16 May 2018 11:53

slh277
Messages: 4
Registered: May 2018

Member

Thank you so much for your kind response. I've downloaded the PR data and will take into account the two areas of Greater Mumbai during analysis.

However, i'm still a bit stuck.

The variable labels in SAS only say (for HV024): "State" and (for SHDISTRI) "District", there's no way I can know, from what I have downloaded and from perusing the recode manual, which state or district that each number in those two variables represents.

I appreciate your help on this.
Thanks,
Samantha

Report message to a moderator