| How to account for dependency and weighing [message #11662] |
Sun, 22 January 2017 20:34  |
Benita
Messages: 4
Registered: January 2017
Location: Ann Arbor
|
Member |
|
|
I would like to know if you know of any way to account for dependency, if you have two people in a household? Do you use Proc Survey ? or?
And how do you approach the weighing of children? I am doing a secondary analysis, looking at the association that exists between socioeconomic differences and child malnutrition in Rwanda. I am using SAS and I am using the RDHS 2014-2015 (the recent one)
Also, if anyone knows of a way to calculate a cut-off (let's say for weight-for-age) using the same data, please let me know.
Thank you
[Updated on: Sun, 22 January 2017 20:35] Report message to a moderator |
|
|
|
| Re: How to account for dependency and weighing [message #11666 is a reply to message #11662] |
Mon, 23 January 2017 10:57  |
Bridgette-DHS
Messages: 3035
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
The weight variable in an IR, KR, or BR file is v005. We adjust for clustering (dependency) by the PSU or cluster, which is v001 or v021. (They are almost always the same.) We do not adjust for clustering at the household or maternal level. If you want to try, you can, of course, but in a multi-level model that includes clustering at the PSU level.
We (the DHS analysis staff) do not support SAS (we mostly use Stata). For specifically SAS questions you must hope for replies from other forum users.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
