Support #225
closed
Merging BHPS and USOC weighting.
Added by Nathan Hudson-Sharp about 11 years ago.
Updated about 9 years ago.
Description
Hi,
I have merged BHPS waves 1 - 18 and Understanding Society Waves 2 and 3 in order to construct consistent work-life histories for the the BHPS panel survey sample. My research is aiming to identify the differential impacts of work transitions on GHQ wellbeing. However, at this early stage in my analyses I am having some problems in deciding what weights to use. Since BHPS wave 18 has post-stratified design weights adjusted for successive waves of non-response, would it be appropriate to use this to form the base for weighting from USOC wave 2 onwards?
Nathan,
If you are conducting longitudinal analysis (BHPS sample) then there are weights for this: in wave 3 they are c_xxxxx91_lw and c_xxxxx01_lw where xxxxx depend on the data you are using (household vs. individual etc.). The c_xxxxx91_lw weight is for 1991+ analysis and c_xxxxx01_lw is for analysis since 1999 or 2001.
If you are pooling data the answer is more complex but can be found here:
https://www.understandingsociety.ac.uk/support/issues/228
Hope this answers your question,
Olena
- % Done changed from 0 to 50
- Status changed from New to Closed
- % Done changed from 50 to 100
- Category set to Weights
- Assignee set to Olena Kaminska
- Target version set to X M
Also available in: Atom
PDF