Support #2271
open
difference between sample size and weighted population appears incorrect
50%
Description
Hi,
I am new to using UKHLS. I have used HILDA in the past so am aware of how to use weights. My question is this: I have used the following command in Stata:
svyset m_psu [pweight=m_indscui_xw], strata(m_strata) singleunit(scaled)
However, when I estimate totals, I see populations that are almost the same as the sample size. e.g.
svy: total pov2w
yields a sample size of 27,204 but population size of 26,942. Am I missing something?
Grateful for any help you can provide.
Updated by Understanding Society User Support Team 2 days ago
- Status changed from New to Feedback
- % Done changed from 0 to 50
- Private changed from Yes to No
Hello Jane,
Here is the feedback from our weighting team:
Understanding Society weights are scaled to a mean of 1 within each wave. This is the reason why the total you estimated is close to the sample size. To estimate a population total, you would need to scale up the weight (grossing up the weight) by multiplying the weight by the population size and dividing it by the sum of the weights.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team