Merging mother's data with children's data in SAS [message #11589] |
Thu, 12 January 2017 15:20 |
Benita
Messages: 4 Registered: January 2017 Location: Ann Arbor
|
Member |
|
|
Hello DHS team,
I am looking at the Rwanda's DHS 2014-2015 (My analysis is based on the association of socioeconomic differentials and child malnutrition) and I am having trouble merging the mother's ID with the child's ID (0-5 years), so that I can choose which mothers are eligible. Respondent's line numbers are confusing. How do I approach this?
**I am using SAS.
Thanks
Benita
[Updated on: Thu, 12 January 2017 15:34] Report message to a moderator
|
|
|
|
|
Re: Merging mother's data with children's data in SAS [message #11620 is a reply to message #11596] |
Tue, 17 January 2017 16:08 |
Bridgette-DHS
Messages: 3214 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Yes, caseid is always the id for the woman. It is a character string that includes v001 v002 v003. It is the same in the IR, KR, and BR files.
The child's record in the KR file (or the BR file) includes almost all the information about the mother that is in the IR file. There is no need to merge the KR and IR files.
A child in the KR file is uniquely identified by v001 v002 v003 bidx, where v001 v002 v003 identify the woman in the IR file and bidx identifies the child's order in the birth history (bidx=1 for the most recent birth).
Please ask again if this does not answer your question.
|
|
|