Weighting for longitudinal analysis of USOC waves 2 to 5
Hi, I just wanted to clarify which weight I should use for a longitudinal analysis of USOC waves 2 to 5 please, in order to specify the inclusion of the BHPS, EMBS and GPS samples in this cohort. Would the best weight be indinub_lw, which becomes available at wave 3 (presumably indscub_lw in analyses which use self-completion data) or should we be using indinus_lw and indscus_lw?
If our analysis using self-completion data in some parts but not others, is it best to always use the self-completion weight, for consistency or better to use the indinub/indinus weight when no self-completion variables are involved?
#1 Updated by Victoria Nolan over 3 years ago
- Status changed from New to In Progress
- Assignee changed from Olena Kaminska to Jane Lakey
- % Done changed from 0 to 10
- Private changed from Yes to No
Many thanks for your query, which has been passed on to the weighting team to look into.
Best wishes, Victoria.
#2 Updated by Peter Lynn over 3 years ago
Always use the "ub" weights (not "us") for longitudinal analysis starting at wave 2, otherwise the BHPS samples will get excluded.
Whether sticking with the "sc" weights for consistency is preferable really depends on the nature of your analyses, e.g. how closely related (substantively) are the analyses that use only interview data to the ones that use self-comp too. You should get more precise estimates of the former type if you use the "in" weights.