The DHS Program User Forum: Weighting data » Weighting Couples HIV data

Home » Data » Weighting data » Weighting Couples HIV data

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

Re: Weighting Couples HIV data [message #8545 is a reply to message #8498]

Thu, 12 November 2015 09:44

Bridgette-DHS
Messages: 3114
Registered: February 2013

Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

If you are using couples with the HIV data attached, DHS would recommend that you use hiv05 from the men as the weight variable. We always have worse nonresponse for men than for women, especially for HIV tests. This is contrary to your suggestion. Sorry about that.

Just "off the top of my head" I would suggest that in the couples file you take the log of hiv05 for the women and hiv05 for the men. I would predict that in most surveys the standard deviation of log(hiv05) will be greater for men than for women. That's a way of quantifying the variability in nonresponse.

Stan Becker at Johns Hopkins University has proposed a way to synthesize the weights from men and women in a couples file. Here's a reference: http://uaps2015.princeton.edu/abstracts/151669.

Report message to a moderator

[Message index]

		Weighting Couples HIV data By: DaniD on Mon, 19 October 2015 12:10
		Re: Weighting Couples HIV data By: Bridgette-DHS on Wed, 21 October 2015 13:05
		Re: Weighting Couples HIV data By: DaniD on Mon, 02 November 2015 15:52
		Re: Weighting Couples HIV data By: DaniD on Wed, 04 November 2015 12:04
		Re: Weighting Couples HIV data By: Bridgette-DHS on Thu, 12 November 2015 09:44

Previous Topic:	Weighting sanitation data
Next Topic:	Pakistan 2014 - Using correct weights for domestic violence

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Jul 17 11:38:31 Coordinated Universal Time 2024