Overweight to under 5 children [message #28904] |
Wed, 27 March 2024 04:54 |
Suleiman Chombo
Messages: 1 Registered: March 2024
|
Member |
|
|
Hello everyone
I am dealing with overweight to under5 in Sub Saharan African countries. So I am using the merged dataset with PR and KR dataset. My outcome variable is "hw72" from KR dataset, which is approximately the same as "hc72" from PR dataset. In most computations for most countries, I am having a difference of like 0.1%, say reported prevalence is 3.5%, but I am getting 3.6%.
What could be the source of the discrepancy that I get.
..AND the special case for Mali, prevalence of overweight to under 5 that I get is quite big, as compared to reported. Any inputs to resolve the issues.
I have defined overweight=1 if hw72>2SD and <=5SD
Overweight=0 if hw72<=2SD, and -5<=SD<=5, was the condition set for valid data.
[Updated on: Wed, 27 March 2024 05:12] Report message to a moderator
|
|
|
Re: Overweight to under 5 children [message #28923 is a reply to message #28904] |
Thu, 28 March 2024 15:17 |
Janet-DHS
Messages: 901 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
The measurements of height and weight are done in the household survey. All of the measured children appear in the PR file, and there the variables have the prefix hc. Later, women are interviewed and they list their children in a birth history. All of their children born in the past 5 years are included in the KR file. For children who were measured, their va;ies are copied into the KR file during data processing, with the prefix hw.
Most children in the PR file are also in the KR file and most children in the KR file are also in the PR file. For individual children who are in both files, there is no difference at all. However, some children in the PR file are not in the KR file. These are children whose mothers have died, or are not in the same household as the child, or were not interviewed. (Also some children in the KR file are not in the PR file--children who have died or are not living in the same household as the mother--but they have not been measured.)
Some other variables are in more than one file but have different names. For example, a woman's anemia status is ha57 in the PR file but v457 in the IR file, but for an individual woman who is in both files, her values are exactly the same.
|
|
|