Support #2009
Covid survey - longitudinal weights with telephone responses
Added by Marina Kousta 5 months ago.
Updated 3 months ago.
Description
To whom it may concern,
I would appreciate some help/guidance with using longitudinal weights for the understanding society covid survey, especially as this is my first time conducting longitudinal analysis. I went through both main and covid survey's user guides but I am still unsure of how to handle the issue I'm facing. I have created a longitudinal dataset using covid waves 1,2,4,5,7. For covid wave 2 I am using both the telephone and web responses (as a single combined wave). I am using a fixed effects analysis to assess causal estimates. The only wave with a longitudinal weight which accounts for the telephone respondents, is wave 6 which is not part of may analysis, but even if it was, it would still exclude wave 7.
Could you please provide me with some guidance on how to best handle this?
Sincere thanks for your time and consideration.
Best wishes,
Marina
Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can. We aim to respond to simple queries within 48 hours and more complex issues within 7 working days.
Best wishes,
Understanding Society User Support Team
- Status changed from New to In Progress
- Status changed from In Progress to Feedback
- % Done changed from 0 to 90
- Private changed from Yes to No
Dear Marina,
Here is the answer from our Covid-19 data weighting expert:
"You aren’t going to be able to use the telephone respondents in your longitudinal analysis. The wave 6 web and telephone longitudinal weights map web and telephone respondents to wave 1 then wave 6 to the population, so don’t consider non-response to most of the waves you are interested in. The wave 6 web longitudinal weights map web to wave 1 then wave 2 then wave 3 then wave 4 then wave 5 then wave 6 then wave 7 to the population, so telephone respondents aren’t weighted. I’m not sure why you’d want to anyway: none of them have more than a wave 1 response. The correct approach would be to focus on the web respondents only, and use the wave 7 longitudinal weight. In theory, you may have a few respondents with zero weights: those that responded to wave 1,2,4, 5 and seven, but not 3, and those who responded to waves 1,2,4,5 and seven, but not 6. However, I can’t imagine these being many."
Best wishes,
UKHLS User Support Team
- Status changed from Feedback to Resolved
- % Done changed from 90 to 100
Also available in: Atom
PDF
Updated by