Weighting and state benefits
I am currently undertaking cross-sectional and longitudinal analyses using both BHPS and US data looking at life satisfaction/Mental Health Component Scores/GHQ as dependent variable (with independent variables including income, employment status and age).
For BHPS cross-sectional using indresp file, I am using w_xewght weights. And for US using indresp file I am using w_indscus/ub/ui weights.
For longitudinal using US only, I am using h_indinui_lw (as I am using indresp data from Waves 2-8). I wanted to check a) that these are correct weights to use and b) that if I'm doing longitudinal using BHPS and US, I am ok to still use h_indinui_lw?
Also, I am looking to split the group into those in receipt of state benefits, and those who are not. Is there a unified variable that give 'in receipt of benefits/not in receipt of benefits', or will I have to create that from all the various variables covering specific benefits (housing, child etc.)?
Updated by Alita Nandi over 1 year ago
- Status changed from New to In Progress
- Assignee set to Olena Kaminska
- % Done changed from 0 to 40
- Private changed from Yes to No
Many thanks for your enquiry.
We have assigned your query regarding weights to our Survey Statistician who will get back to you.
About your benefit indicator question: No there is no such variable. You will need to look at the “w_income.dta” files and see which benefits you want to use in the definition of benefit recipient and create this indicator variable based on that.
Understanding Society User Support Team