The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Sample Weight for Merged Dataset (Which sample weight, strata, and cluster variables should be used to create an SPSS Complex Samples Plan File)
Re: Sample Weight for Merged Dataset [message #23747 is a reply to message #23736] Tue, 23 November 2021 10:09 Go to previous messageGo to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

We do not understand what you mean by "I am worried that there is no difference between the population estimate and unweighted counts from Complex samples frequencies." By "population estimate" do you mean the "weighted counts"? Your SPSS output does show a difference EXCEPT that the totals match. The total weighted number and the total unweighted number match--as they should, because DHS weights include a multiplier to force the mean weight to be 1 (with a multiplier of 1000000). The weighted counts are not population estimates. The weighted means, etc., are population estimates, but not the counts.

You have appended the files from the different surveys. There is a difference between merging and appending. What you have done is correct but should be described as appending, rather than merging.

The "v" variables for women almost always correspond exactly with "mv" variables for men. If you just drop the "m" for men, you can keep the "v" variables and not have to invent new names.

The cluster and stratum variables (v021 and v023) will repeat some values in the different surveys but have to be revised so the identifiers are distinct. This is done with "egen group" in Stata. In SPSS you have to do something else to renumber those variables. Let us know if you are not sure how to do that.

There are options for what to do with the weights when surveys are appended. Have you considered those options? Let us know if this is not clear.

 
Read Message
Read Message
Read Message
Read Message
Previous Topic: Using Normalized Weight in Analysis
Next Topic: Use of weights on subsetted data
Goto Forum:
  


Current Time: Fri Mar 29 09:40:58 Coordinated Universal Time 2024