The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Pakistan: Standard DHS, 1990 Dataset (Question about Stunting Estimates)
Pakistan: Standard DHS, 1990 Dataset [message #22284] Tue, 23 February 2021 17:13 Go to next message
yiningw is currently offline  yiningw
Messages: 4
Registered: February 2021
Member
Dear DHS Representative,

I am working with the Pakistan 1990 dataset, specifically PKHW21FL.DTA and had a question about the stunting indicators (hc70). Using the DHS definition to find the share of stunted, it shows a close to 0% (the value for later years are higher). The hunch is that the stunting variable wasn't recorded for all the individuals and is maybe missing for some sub-groups in 1990, but wanted to see if there is any documentation about how we should understand how the survey data was generated.

Thank you!
Re: Pakistan: Standard DHS, 1990 Dataset [message #22384 is a reply to message #22284] Thu, 04 March 2021 21:23 Go to previous messageGo to next message
yiningw is currently offline  yiningw
Messages: 4
Registered: February 2021
Member
Dear DHS Representative,

I wanted to follow up with a more specific question. I was able to look at the Merge.doc documentation under PKHW21FL.dta and realized that the household and individual recodes have to be merged with the PKHW21FL.dta separately before combining them, which I feel may be the source of my issue. However, when following the household steps from the Merge.doc:
• Use HWHHID from the Height and Weight data file with HHID from the Household Recode file to merge it with the household data where the child was measured.

• Use HWHHID and HWLINE from the Height and Weight file with HHID and HVIDX from the Members Recode file to merge it with the household member data.

HWHHID is not an existing variable in this dataset. Further, HWLEVEL observations are all individual. I am wondering if I may be missing something here or if there is a different dataset I should be taking from this.

Thank you in advance!
Re: Pakistan: Standard DHS, 1990 Dataset [message #22390 is a reply to message #22384] Fri, 05 March 2021 08:21 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2372
Registered: February 2013
Senior Member
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

This is an old survey, and dates to the time when height and weight were measured as part of the survey of women, rather than the household survey. Therefore only the children whose mother was alive and in the same household as the child were measured. In the HW file, I see that the id's are badly labeled. "hwcaseid" is actually the mother's "caseid" and "hwline" is bidx, NOT the household line number of the child. This survey predates the use of b16.

So for this survey the HW file is to be merged with the KR file, not the PR file. The lines to do this are below. There is a 1 to 1 match, that is, _merge=3 for all cases in the merged file. You only need to change the paths.


use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\PKHW21FL.DTA" 
rename hwcaseid caseid 
rename hwline bidx

sort caseid bidx
save e:\DHS\DHS_data\scratch\PKtemp.dta, replace

use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\PKKR21FL.DTA" 
sort caseid bidx

merge caseid bidx using e:\DHS\DHS_data\scratch\PKtemp.dta
tab _merge

[Updated on: Fri, 05 March 2021 08:22]

Report message to a moderator

Previous Topic: Indonesian Special DHS 2002-03
Next Topic: Philippine 2008-2017 data - strata ID
Goto Forum:
  


Current Time: Sat Sep 18 19:29:22 Coordinated Universal Time 2021