Support #836
closed
Weighting for general population
Added by Ray Storry over 7 years ago.
Updated over 7 years ago.
Description
I have run a query on persons employed who expect to move in the next year Yes/No. I have weighted the data in Wave 6 with f_indpxui_xw. The total population I am getting along with don't knows etc. is 23557.862. But this looks to me like the total number of employed respondents in the sample. There are around 31.4 millions employed persons in the UK.
So how do get from the weighted figures shown above to come anywhere near the real world figures?
- Category changed from Data inconsistency to Weights
- Status changed from New to In Progress
- Assignee changed from Alita Nandi to Peter Lynn
- % Done changed from 0 to 10
- Private changed from Yes to No
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.
Best wishes,
Stephanie Auty - Understanding Society User Support Officer
- Assignee changed from Peter Lynn to Ray Storry
- Target version set to M6
- % Done changed from 10 to 50
Hello Ray,
Yes, all weights are designed for estimation of ratio estimates (proportions, means, regression coefficients, etc). They do not include any element of "grossing up" to the population, which would be needed if you want to estimate the number (rather than proportion) of people in a category (I infer this is what you want to do?)
But if you know the population total, it is a simple matter to gross up your weighted estimate. Just multiply the weighted sample number (base in table) by N/nw, where N is the population total (31.4m?) and nw is the weighted base (23,558?).
Peter
Ray,
Apologies, where I wrote "(base in table)", I meant "(cell count in table)".
Peter
- Status changed from In Progress to Closed
- Assignee deleted (
Ray Storry)
- % Done changed from 50 to 100
Also available in: Atom
PDF
Updated by Stephanie Auty 
Updated by