How is sampling done for child development and child diet? [message #23689] |
Sat, 06 November 2021 04:07 |
jejimel
Messages: 2 Registered: November 2021
|
Member |
|
|
Hello DHS forum,
I'm working with several DHS datasets to examine the associations between child development and diet in children 36-59 months of age. Both child development and diet are not collected for all children in this age range. I'm trying to find information on how children are sampled for these two assessments. Since both are optional, there seems to be virtually no information on the DHS website or in final reports.
Thank you!
Lily
|
|
|
|
|
Re: How is sampling done for child development and child diet? [message #23754 is a reply to message #23707] |
Wed, 24 November 2021 13:54 |
Bridgette-DHS
Messages: 3167 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Specialist, Trevor Croft:
The variables v409-v414w in the women's data (the IR file) refer to the youngest child under 2 living with the mother. When the KR file is constructed, with one record for each child born in the last 5 years, those variables (along with other v variables in the IR file) are copied onto the records for all children born in the last 5 years, even when the variables do not apply to that child. Thus you must either use the IR file, or use the KR file after selecting just the youngest child under 2 living with the mother.
To correctly analyze the variables with the KR file, follow the selection documented in the Code Share Library:
Go to https://github.com/DHSProgram/DHS-Indicators-Stata; Select Chapt11_NT
*Selection of the youngest child under 2 years living with the mother is done with lines 96-100 in INTmain.do:
* Selecting for youngest child under 24 months and living with mother
keep if age < 24 & b9 == 0
* if caseid is the same as the prior case, then not the last born
keep if _n == 1 | caseid != caseid[_n-1]
For the Uganda 2016 survey, age can be replaced with b19.
|
|
|