Support #1646
open
Weight for analysis of household income restricted to complete households
100%
Description
I'm undertaking analysis to understand if pensioners who enter/exit poverty do so because their household receipt of particular social benefits changes, making use of the data in the w_income datafile. Hence the analysis should only use households where all adults in the household completed an interview.
https://www.understandingsociety.ac.uk/documentation/mainstage/user-guides/main-survey-user-guide/household-income-variables says that income components are imputed for all proxy and within household non-respondents. Hence in Understanding Society household income estimates are available for all households – including where some household members non-respond. Users may decide to drop cases based on such imputed data but they would then need to adjust their results to take into account the consequent sample selection i.e. their results would no longer be representative of the UK population.
It is recommended that if the analysis of household income is restricted to complete households then results should be reweighted. Is a specific weight for this purpose available? If not, would you be able to advise on how to construct one please?
Thank you.
Updated by Understanding Society User Support Team almost 3 years ago
- Status changed from New to In Progress
- Assignee set to Olena Kaminska
- % Done changed from 0 to 10
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. While we will aim to keep to this response times due to the current coronavirus (COVID-19) related situation it may take us longer to respond.
Best wishes,
Understanding Society User Support Team
Updated by Understanding Society User Support Team almost 3 years ago
- Private changed from Yes to No
Updated by Understanding Society User Support Team almost 3 years ago
- Status changed from In Progress to Feedback
- % Done changed from 10 to 50
Matt,
Thank you for your question. Yes, you are right - our weights are either for people who respondent to interviews or for responding households (where some people may be missing).
Can I just clarify: do you have imputed information for those who didn't respond within a household but where household information is available (from another person in the household)? If this is the case, you could analyse data with a household weight (it's only cross-sectional) and the imputed information.
If you want to restrict your analysis to only completed households you would need to create your own tailored weight. Email usersupport to request access to our course. If your analysis is cross-sectional I suggest you use either hhden??_xw weight for a household-level analysis (on hhresp.dta) or psnen??_xw weight for a person-level analysis (from an indall.dta file).
Hope this helps,
Olena
Updated by Matt Barnes almost 3 years ago
Understanding Society User Support Team wrote in #note-3:
Matt,
Thank you for your question. Yes, you are right - our weights are either for people who respondent to interviews or for responding households (where some people may be missing).
Can I just clarify: do you have imputed information for those who didn't respond within a household but where household information is available (from another person in the household)? If this is the case, you could analyse data with a household weight (it's only cross-sectional) and the imputed information.
If you want to restrict your analysis to only completed households you would need to create your own tailored weight. Email usersupport to request access to our course. If your analysis is cross-sectional I suggest you use either hhden??_xw weight for a household-level analysis (on hhresp.dta) or psnen??_xw weight for a person-level analysis (from an indall.dta file).
Hope this helps,
Olena
Thanks Olena. I want to restrict the analysis to complete households, and then compare changes in income sources (type of source and amount, from the w_income datafiles) across two waves (e.g. w10 - w11), to see what happens as older people move into (and out of) poverty. So your suggestion is to attend a USoc sampling course, which will advise on how to do this?
Best wishes
Matt
Updated by Olena Kaminska almost 3 years ago
Matt,
Thanks. For the situations likes yours we have prepared an online training course for creating your own tailored weight. Please request access to it via email to usersupport@understandingsociety.ac.uk .
Thanks,
Olena
Updated by Matt Barnes almost 3 years ago
Great - thanks very much for your help.
Updated by Understanding Society User Support Team over 2 years ago
- Status changed from Feedback to Resolved
- % Done changed from 50 to 100