Support #2262
open
Analysing debt: how to deduct mortgage debt and which weights to apply
Added by Gabriele Mari 2 days ago.
Updated about 6 hours ago.
Category:
Data documentation
Description
Hi,
I was wondering if you have suggestions on best practices for an analyses of debt using the "debty" variable asked in waves 4, 8, 13. Specifically,
1. Is there a way you'd suggest to calculate non-mortgage debt amounts? I see that mortgage information is recorded in a number of variables outside of the wealth modules. Would there be a (best) way to go about this?
2. I was also wondering which weights are best for a pooled cross-sectional analyses of waves 4, 8, and 13. Would it be via indinub (W4) and indinui (W8, W13) or should I perhaps rely on the weight for the "extra five minutes" questions (which seem to include "debty")?
Thanks so much for your help. I deeply appreciate it.
All the best,
Gabriele
- Category set to Data documentation
- Status changed from New to Feedback
- % Done changed from 0 to 50
- Private changed from Yes to No
Hello Gabriele
Regarding your first question, I’ve shared it with the Income Team to get their input. I’ll forward their response to you as soon as I hear back.
As for your second question: when conducting pooled cross-sectional analyses, you need to use the wave-specific weights. The best approach is to carefully consider whom you want your analysis to represent, as this determines the correct weighting strategy.
Issue #2167 (https://iserredex.essex.ac.uk/support/issues/2167) addresses a similar question, though in that case the pooled analysis involved both BHPS and UKHLS. Nonetheless, the underlying logic applies in your case as well.
I’d also recommend reviewing Working Paper 2024-01: Weighting FAQ: https://www.understandingsociety.ac.uk/wp-content/uploads/working-papers/2024-01.pdf In particular, take a look at Questions 14 to 16, which are especially relevant to your query.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team
Also available in: Atom
PDF
Updated by