The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Do I really need to Weight the DHS data
Do I really need to Weight the DHS data [message #11455] Thu, 22 December 2016 11:27 Go to next message
r01efa16
Messages: 10
Registered: November 2016
Location: UK
Member

Hello everyone,

I know that here we discussed deeply and several times the issue of weighting the data from DHS before using them in the analysis.

My observation is I employ the data in a regression analysis, and I did run the regression once with weighted data set and another time without, using the wriiten lines 'below' from the website ( http://dhsprogram.com/data/Using-DataSets-for-Analysis.cfm#C P_JUMP_14042)

generate wgt = v005/1000000
tab var [iweight=wgt]


But I did not find a big difference either in the regression estimators or in the standard errors

So my question is Do I really need to Weight the DHS data before using in the analysis?

I really need your comments

Thanks
Re: Do I really need to Weight the DHS data [message #11463 is a reply to message #11455] Wed, 28 December 2016 14:06 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:

The position of DHS is that you should always make the survey design adjustment, even if they do not seem to make much difference.
Re: Do I really need to Weight the DHS data [message #11640 is a reply to message #11463] Fri, 20 January 2017 08:26 Go to previous message
r01efa16
Messages: 10
Registered: November 2016
Location: UK
Member

Well Thank you so much
Previous Topic: Disaggregating to lower Administrative Division
Next Topic: Interrupted time series analysis using DHS data and applying sample weights
Goto Forum:
  


Current Time: Thu Mar 28 08:59:06 Coordinated Universal Time 2024