USOC cross-section household weight for BHPS sample: wave 3
Hello, this is a follow-up query related to a previous query (#291).
I need to estimate weighted medians for the BHPS sub-sample in UKHLS wave 3. What are the most appropriate weights to use for this?
Is it valid to use c_hhdenub_xw and c_indpxub_xw, even if I am restricting my analysis to just the BHPS sub-sample?
If not, then what would you advise I should do?
Updated by Redmine Admin about 7 years ago
- Category set to Weights
- Status changed from New to In Progress
- Assignee set to Alex Hurrell
- Target version set to M3
- % Done changed from 0 to 40
You should not use these wave 3 cross-section weights for the purpose of W3 cross-sectional analysis using only the BHPS sample. These weights cannot be used just for BHPS sample, all samples (GPS, EMB, BHPS) need to be used together. But why do you want to use the BHPS sample if you are doing W3 cross-section analysis. Why not use all samples? The questions that have been asked of the BHPS sample members have also been asked of the GPS and EMB samples.
On behalf of the team,