Mismatch between NFHS 4 Factsheets and Output of Stunting, Wasting [message #17987] |
Sat, 10 August 2019 17:15 |
marian
Messages: 13 Registered: August 2019
|
Member |
|
|
Hi,
I want to do cross tabulations of anthropometry indicators (stunting, wasting, underwt) for districts of UP according to caste, tribe, gender. The same problem persists for child anemia estimates. The SPSS output (after using state weights) is different from the prevalence reported in district factsheets of NFHS 4. Please note, I have used both KR and PR recode file (HW 70-72, HC70-72) to check the data. This is problematic since other variables are generating same estimates as given in factsheets. What should I do?
Please note, the PR recode is currently removed from DHS website. Is there a new file for this data? or do I have to merge certain recode files.
|
|
|
|
|
Re: Mismatch between NFHS 4 Factsheets and Output of Stunting, Wasting [message #26427 is a reply to message #26419] |
Mon, 20 March 2023 09:51 |
Bridgette-DHS
Messages: 3199 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
We just posted (in response to a question from Manasa) a program that calculates the rates for stunting, etc., by state. It can be adapted to other subpopulations.
If you are having difficulty matching with the spreadsheets, the most important thing to check is whether the factsheets used the children in the PR file or the KR file. Also, if they used the PR file, check the title for whether there was a restriction of any kind, perhaps related to de jure residence.
The files for the NFHS-5 went through several updates, and some changes affected the weights. There are some tables for which we cannot get an exact match with the reports and factsheets.
[Updated on: Mon, 20 March 2023 10:00] Report message to a moderator
|
|
|