Re: create a weighted population variable [message #24526 is a reply to message #24504] |
Thu, 26 May 2022 16:39  |
Janet-DHS
Messages: 938 Registered: April 2022
|
Senior Member |
|
|
Following is response from DHS Research & Data Analysis Director, Tom Pullum:
I'm not clear what you mean by "a weighted variable "population" which calculates all the individuals interviewed".... The number of people interviewed is the unweighted number in the sample. Are you looking for an estimate of the population? For that, I would recommend that you get the estimated national population from the UN Population Division website (https://population.un.org/wpp/), and multiply that by the weighted proportion in the DHS sample in the province. The UN estimates are for July 1 of a specified year. The DHS estimates can be assigned to the median date of interview. You may want to interpolate.
|
|
|