Weights in Covid-19 Study
Dear Support Team,
I tsset (pidp wave) the April and May Study. Can I only use the weight (betaindin_xw) cross-sectionally for April and May separately or can betaindin_xw be used for the collapsed (April and May) sample to derive representative descriptions?
Many thanks in advance.
#1 Updated by Alita Nandi 3 months ago
- Private changed from Yes to No
- % Done changed from 0 to 60
- Assignee set to Darja Reuschke
- Status changed from New to Feedback
ca_betaindin_xw = i_indinui_xw * non-response probability based on a model of non-response between main W9 &W1. This weight is non-zero for W9 respondents who also responded in W1
cb_betaindin_xw = i_indinui_xw * non-response probability based on a model of non-response between main W9 &W2. This weight is non-zero for W9 respondents who also responded in W2, but may not have responded to W1.
You will need to include in the sample all those with a wave one weight and use the wave one weight; or include all those with a wave two weight and use the wave two weight.
If you want to focus on respondents who did an interview at both waves one and two then you will need a longitudinal weight. The covid19 team is currently working to produce this weight and the plan is to include it in a future data release.
Information on weights for this survey is also available in the user guide: https://www.understandingsociety.ac.uk/documentation/covid-19/user-guide
Does this answer your question? If not, please let us know.
On behalf of Understanding Society User Support Team