Support #2080
open
Repeated cross-sectional data weighting
90%
Description
Hi!
I am a third year student writing my dissertation and I was hoping for some guidance on weighting. I am using waves 6-11 of the Understanding Society dataset as cross-sectional such that I am able to look at quarterly mental health data. If it is possible I would like some guidance on how to weight the data in the scenario.
To my knowledge the weights to use are f_indscui_xw, g_indscui_xw, h_indscui_xw, i_indscui_xw, j_indscui_xw and k_indscui_xw.
How can I weight these appropriately in stata? For the svyset command, what would be the psu value and strata value in this case?
Any guidance would be really appreciated!
Thank you,
Fran Reah.
Updated by Understanding Society User Support Team 9 months ago
- Status changed from New to Feedback
- Assignee deleted (
Olena Kaminska) - % Done changed from 0 to 90
- Private changed from Yes to No
Hello,
If you are using at least one self-completion variable in your analysis model then the correct weights are the self-completion weights, so w_indscui_xw you mentioned. For practical advice on weighting in Stata I would recommend to check example 6 in our online Introduction to Understanding Society course available here: https://open.essex.ac.uk/course/view.php?id=76 (you need to register first), check topic 9 "Working with data collected from surveys with complex survey designs": worksheet and do file for example 6.
Best wishes,
Piotr Marzec,
UKHLS User Support
Updated by Olena Kaminska 8 months ago
Fran Reah,
Thank you for your question. Question 13 here: https://www.understandingsociety.ac.uk/wp-content/uploads/working-papers/2024-01.pdf , may help you to analyse our data on quarterly basis. Psu and strata variables don't change over time, so take them from any wave. They have to be complete though (no missingness).
Hope this helps,
Olena