Re: Existence of Domestic Violence and impact on Children's Education Outcome [message #29612 is a reply to message #29591] |
Wed, 10 July 2024 09:49 |
Janet-DHS
Messages: 880 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
I was able to open the screenshots you sent. (For some reason I was unable to open the screenshots you sent earlier.) I see that you are using the HR and KR files, and I think that's your main problem. The cases in the KR file are children under 5, and the v variables refer to the mother. The cases in the HR file are households, and the hv variables (with subscripts such as _01 and _02) refer to all people in the household. That's why you are getting what appear to be impossible combinations of age and completed schooling.
You will get a file of children age 6-14 if you open the PR file (which is a reshaped version of the HR file and has individual household members as cases) and then "keep if hv105>=6 & hv105<=14". You will then want to merge that file with the IR file, which has the DV variables, matching the children with the mothers. This means matching hv001 hv002 hv112 in the reduced PR file (the children age 6-14) with v001 v002 v003 in the IR file (the file of mothers). There can be more than one child per mother, so this is an "m:1" merge. There will be quite a few cases in both files that do not merge, for children who are not living with their mother, and for women who do not have any children age 6-14, at least not in the same household. Let us know if you still have questions.
|
|
|