The DHS Program User Forum: Weighting data » Weights for regression analysis

Home » Data » Weighting data » Weights for regression analysis

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

Weights for regression analysis [message #11947]

Sat, 11 March 2017 08:33

ale_sovera
Messages: 10
Registered: March 2017
Location: Milan, Italy

Member

I am working with female and male individual datasets and I have read that I have to use the corresponding weights to do the tabulations of the variables contained. What I am wondering is if I am supposed to use weights also when doing regressions and more generally, when trying to do estimations.

Thanks

Report message to a moderator

Re: Weights for regression analysis [message #11973 is a reply to message #11947]

Tue, 14 March 2017 10:29

Bridgette-DHS
Messages: 3230
Registered: February 2013

Senior Member

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

Yes, you should include weights in regression analysis. In Stata, these would be pweights and you would include something like "[pweight=v005]" before the comma in an estimation command. There have been many forum posts on weights.

Report message to a moderator

Previous Topic:	Weighing SPA data Tanzania
Next Topic:	Multiple Occurring Variable

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Tue Sep 16 05:08:06 Coordinated Universal Time 2025