Support #2285: Cross-sectional weights for having a household member interviewed in COVID-19 survey - Understanding Society User Support

Actions

Copy link

Support #2285

open

Cross-sectional weights for having a household member interviewed in COVID-19 survey

Added by Laura L 13 days ago. Updated 1 day ago.

Status:

Feedback

Priority:

Normal

Assignee:

Olena Kaminska

Category:

Weights

Start date:

10/15/2025

% Done:

80%

Description

Dear Olena,

I need to include in my analysis only individuals from wave 6 of the COVID-19 survey who have at least one household member interviewed. Because the COVID-19 study does not collect household enumeration grids, I am using mainstage wave 10 identifiers to reconstruct households. Can I still use the cross-sectional weights from wave 6 in my analysis, or should I compute specific weights for this subgroup?

Thank you in advance.

Actions

Copy link

Updated by Olena Kaminska 13 days ago

Laura,

Thank you for your question. You could use Covid-19 wave 6 weight as a suboptimal weight (with smaller sample size of course), or you could construct you own tailored weight using w10 enumeration base weight if you prefer. For this follow our course here: https://www.understandingsociety.ac.uk/help/training/creating-tailored-weights/

Hope this helps,
Olena

Actions

Copy link

Updated by Laura L 13 days ago

Dear Olena,

Thank you very much for your answer.

What do you mean exactly by "with smaller sample size"? Do you mean that the sum of weights is not going to be the total number of respondents in the wave, right?

Thanks again,
Laura

Actions

Copy link

Updated by Olena Kaminska 12 days ago

Laura,

No, I meant the number of respondents in your model. There will be more respondents that can be in your model and provided necessary information for you, but with Covid-19 weight you will limit only to those who provided full response in wave 6 of Covid study.

Hope this helps,
Olena

Actions

Copy link