The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Nutrition and Anthropometry » Children age Namibia DHS VI
Children age Namibia DHS VI [message #19682] Sun, 02 August 2020 00:54 Go to next message
relindwto6 is currently offline  relindwto6
Messages: 1
Registered: August 2020

I am using the DHS VI for the first time, my interest is Association between anemia and possible determinants among children aged 6-59 months in Namibia. I am using dataset KR. I am confused about following details:

1. Is hw1 the correct children age variable to use for my analysis?

2. The report indicates 2303 children where tested for anemia, but when i tab hw57 i get only about 1500 children. While hw1 give about 1900 and b4 give about 5000.

3. I would like to use malaria information for my analysis too, i find many questions regarding the use of bed nets and ant-malaria medicine, how can i use the malaria information in my analysis? do i need to merge all the questions regarding malaria?
VidMate Mobdro word counter

Thank you very much for your help.

[Updated on: Sun, 02 August 2020 23:26]

Report message to a moderator

Re: Children age Namibia DHS VI [message #19699 is a reply to message #19682] Tue, 04 August 2020 13:11 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3035
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

Please check the description of this survey in the main report. When any survey has large numbers of Not Applicable (NA) cases, indicated in Stata with a period, it is usually because the survey involved subsampling. It appears that the children in only about a third of the households were selected for hemoglobin testing. Those are the only children who have a valid (not NA) value of hw1.

Note that the KR file always includes all children born in the past five years. Some of them had died (b5=0) and were not tested even though their household may have been selected for the subsample.

The malaria variables are in the KR file. No merging is necessary to get, for example, ml101. But you need to distinguish between malaria variables that refer to the child and others that refer to the mother, whose data are on the child's record. I recommend that you look at the questionnaire, which is in an appendix to the main report.
Previous Topic: Stunting and wasting across countries and years
Next Topic: Website search and dataset mismatch
Goto Forum:

Current Time: Mon Apr 22 03:49:42 Coordinated Universal Time 2024