The DHS Program User Forum: Weighting data

Home » Data » Weighting data » Pooled Cross sections

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

Re: Pooled Cross sections [message #10658 is a reply to message #10089]

Fri, 26 August 2016 10:28

CKAllen
Messages: 7
Registered: April 2016
Location: London, UK

Member

Quote:

If I follow the suggestion below and eliminate NW then I'm not denormalizing and creating the country specific weight (CSW) that I thought was the recommended process for pooled surveys? Did I miss something obvious?

Thanks again.

I have the same question... I have created de-normalized weights after creating unique strata and clusters for 6 appended surveys (three countries, two surveys each). However when I run my logistic regression my 'population size' is gigantic (see below). I've read through most of the threads on this site and cannot seem to understand why my weighting is wrong. However the post above suggests that I do not need to de-normalize the weights? There are so many differing opinions and methods on this, but I think I'd like to de-normalize my weights in order to account for very different country sizes. This is my first analysis I've done with DHS, so I realize there may be something simple I am missing. Any help is appreciated! thank you!

number of obs = 36,349
population size = 11,882,122,587,765

Report message to a moderator

[Message index]

		Pooled Cross sections By: cbdolan on Tue, 14 June 2016 09:35
		Re: Pooled Cross sections By: Reduced-For(u)m on Wed, 15 June 2016 15:48
		Re: Pooled Cross sections By: cbdolan on Thu, 23 June 2016 10:22
		Re: Pooled Cross sections By: Bridgette-DHS on Thu, 23 June 2016 15:18
		Re: Pooled Cross sections By: cbdolan on Fri, 24 June 2016 09:33
		Re: Pooled Cross sections By: Bridgette-DHS on Thu, 30 June 2016 09:46
		Re: Pooled Cross sections By: CKAllen on Fri, 26 August 2016 10:28
		Re: Pooled Cross sections By: Bridgette-DHS on Mon, 29 August 2016 10:05
		Re: Pooled Cross sections By: id709nvz on Tue, 06 October 2020 21:45
		Re: Pooled Cross sections By: Bridgette-DHS on Fri, 09 October 2020 11:57
		Re: Pooled Cross sections By: id709nvz on Fri, 09 October 2020 12:04
		Re: Pooled Cross sections By: joemer on Mon, 22 August 2016 02:56

Previous Topic:	Antenatal visit with skilled providers (Ethiopian DHS 2000, 2005, 2011 AND 2016))
Next Topic:	Using R to weight data for domestic violence

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Dec 2 12:59:05 Coordinated Universal Time 2024