The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » stunting calculation
stunting calculation Sat, 18 March 2023 10:04
 Manasa Messages: 4Registered: March 2023 Member
sir i have combined pr and kr data set of NFHS 5. however im not able to calculate statewise stunting and underweight percentage of children. How to approach it sir.
Re: stunting calculation [message #26425 is a reply to message #26417] Mon, 20 March 2023 09:20
 Bridgette-DHS Messages: 3064Registered: February 2013 Senior Member

Following is a response from Senior DHS staff member, Tom Pullum:

You do not need to combine the PR and KR files for this purpose. You use one file or the other. In the PR file, the relevant variables are hc70-hc72. In the KR file they are hw70-hw72. If you use the PR file you get a few more children--those who are in the household but whose mother is NOT in the household. I will paste the Stata lines below, illustrating the "tabstat" command, which some users may not know about. You can run the same lines in the KR file by changing hc to hw.
```use "...IAPR7DFL.DTA", clear

keep if hc1<.
gen haz=hc70 if hc70>=-600 & hc70<=600
gen waz=hc71 if hc71>=-600 & hc71<=500
gen whz=hc72 if hc72>=-500 & hc72<=500

gen stunted=0
replace stunted=100 if haz<-200

gen underweight=0
replace underweight=100 if waz<-200

gen overweight=0
replace overweight=100 if whz>200

gen wasted=0
replace wasted=100 if whz<-200

tabstat stunted underweight overweight wasted [fweight=hv005], statistics(mean) by(hv024) format(%6.1f)

```
Re: stunting calculation [message #27241 is a reply to message #26425] Wed, 05 July 2023 13:20
 Hajra Adil Nawaz Messages: 2Registered: July 2023 Member
what syntax have to be added if we want to assess province wise stunting prevalence
Re: stunting calculation [message #27254 is a reply to message #27241] Fri, 07 July 2023 10:05
 Bridgette-DHS Messages: 3064Registered: February 2013 Senior Member
Following is a response from Senior DHS staff member, Tom Pullum:

In the Stata lines I gave, "by(hv024)" is the syntax for getting the prevalence of stunting, etc., within each category of hv024, which is "region", usually the first administrative level of a country. In India, hv024 is states. The original question was about India and the NFHS-5. In Pakistan hv024 is provinces. The syntax works for any other variable, such as wealth quintiles, education levels, etc.

[Updated on: Thu, 19 October 2023 08:44]

Report message to a moderator

Re: stunting calculation [message #27896 is a reply to message #26425] Wed, 18 October 2023 08:13
 eceaydin Messages: 3Registered: October 2023 Member
I would like to calculate the number of stunted children in the Uganda 2016 dataset. At the moment, because I needed to merge the IR and PR datasets, I will be using the PR to calculate this. I cannot use the hc variables because they are all missing values after the merging. So like this post mentions, I wanted to use the hw70 variable. However, in my dataset I have hw70_1, hw70_2 etc. for the different children. Do I need to create a new variable combining the children together? Or how else do I do this calculation? Any help is appreciated!

Re: stunting calculation [message #27899 is a reply to message #27896] Thu, 19 October 2023 08:44
 Bridgette-DHS Messages: 3064Registered: February 2013 Senior Member

Following is a response from Senior DHS staff member, Tom Pullum:

You should use either the hc variables in the PR file or the hw variables in the KR file. You do not need to use the IR file or to merge the IR and PR files. In the KR file, individual children are cases and the mother's information is on the child's record.
 Previous Topic: command for Cluster Average of Women's employment Next Topic: Birth order and Antenatal Care NFHS-5 India
Goto Forum:

Current Time: Tue May 28 01:14:29 Coordinated Universal Time 2024