The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Couple weight
Couple weight [message #23820] Wed, 08 December 2021 02:06 Go to next message
anjalibansal6693 is currently offline  anjalibansal6693
Messages: 20
Registered: February 2018
Dear all

I was doing some analysis based on couples in India. Which weight should be used for the analysis as no couple weight is available for India DHS.

Also, if couple weight is to be used, how can we compute the same for India DHS, 2015-16.
Re: Couple weight [message #23821 is a reply to message #23820] Wed, 08 December 2021 08:51 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2932
Registered: February 2013
Senior Member

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

As we have said in earlier posts, we recommend using mv005, the man's weight, because it is corrected for nonresponse by men, which is usually more serious than nonresponse by women. Stan Becker proposed a couple weight in a Demography article, but it can only be constructed in some surveys. We are currently working on a possible couple weight that could be constructed in all surveys, including the NFHS's. Preliminary evidence is that it has a correlation of .99 (or better) with mv005. You are "safe" using mv005.
Next Topic: Tabulate Data at Lower Level
Goto Forum:

Current Time: Wed Nov 29 22:52:08 Coordinated Universal Time 2023