The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » weighting
weighting [message #5531] Wed, 03 June 2015 11:13 Go to next message
asifalokozai is currently offline  asifalokozai
Messages: 1
Registered: June 2015
Location: Afghanistan
Member
Dear Experts

In surveys when the subject or sample has been selected with replacement, so are the best method to weight such data. e.g EPI coverage survey.

Thanks
Asif


M.Asif Alokozai
Re: weighting [message #5583 is a reply to message #5531] Wed, 10 June 2015 12:19 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 802
Registered: January 2013
Senior Member
We can't speak specifically for other surveys, but in general with most surveys you will need to use weights, even if the sample has been selected with replacement. Most surveys like DHS use two stage samples - selecting clusters at the first stage and selecting households within clusters at the second stage. The weights are used to adjust for the probability of selection of the clusters. The second stage of selecting households is where surveys other than DHS use replacement. You thus should use weights (in general - specific surveys may differ)
Previous Topic: How to set up sample weight for analysis
Next Topic: local government area variable
Goto Forum:
  


Current Time: Thu Nov 7 11:26:44 Coordinated Universal Time 2024