The DHS Program User Forum: General

Home » Topics » General » Couple weight

Show: Today's Messages :: Show Polls :: Message Navigator

Couple weight [message #23820]

Wed, 08 December 2021 02:06

anjalibansal6693
Messages: 24
Registered: February 2018

Member

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.

Report message to a moderator

Re: Couple weight [message #23821 is a reply to message #23820]

Wed, 08 December 2021 08:51

Bridgette-DHS
Messages: 3202
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.

Report message to a moderator

Previous Topic:	SCHOOL ATTENDANCE
Next Topic:	Tabulate Data at Lower Level

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Dec 2 17:46:25 Coordinated Universal Time 2024