The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Topics » HIV » Computing Cluster Sample weights in stata
Computing Cluster Sample weights in stata [message #12809] Sun, 16 July 2017 19:40 Go to next message
lucion is currently offline  lucion
Messages: 2
Registered: May 2017
Member
Dear All,
Formula for obtaining sample weights are provided in "sampling and household listing manual, page 23" for cluster level (1st stage) and household level (2nd stage).
So far I did not see cluster sample weight, only household sample weights and idividual sample weights.
Can anyone explain me how to compute cluster sample weights (e.g. in Stata).

Thank you,

Lucioni
Re: Computing Cluster Sample weights in stata [message #12814 is a reply to message #12809] Mon, 17 July 2017 16:58 Go to previous messageGo to next message
Liz-DHS
Messages: 1046
Registered: February 2013
Senior Member
A response from senior sampling expert, Dr. Mahmoud Elkasabi:
Quote:

The cluster sample weight and the HH/Individual weight are not two separate weights. As illustrated in pages 23 and 24 in the sampling manual, the household weight and the individual weights are calculated taking into consideration the two stages of selection, the clusters selection and the households selection. In addition to the selection probabilities, adjustment for household and individual nonresponse are accommodated in the calculations.
If your question is about the selection probabilities for clusters and how to calculate them, you should be able to calculate these selection probabilities using the equation in page 23.


Re: Computing Cluster Sample weights in stata [message #12816 is a reply to message #12814] Tue, 18 July 2017 07:50 Go to previous message
lucion is currently offline  lucion
Messages: 2
Registered: May 2017
Member
Many thanks
Previous Topic: Missing test statistics
Next Topic: Accepting attitude towards people living with HIV/AIDS
Goto Forum:
  


Current Time: Thu Sep 21 06:37:30 Eastern Daylight Time 2017