The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » How would I use weights in this scenario?
Re: How would I use weights in this scenario? [message #11342 is a reply to message #11332] Tue, 06 December 2016 10:11 Go to previous messageGo to previous message
RASimmons is currently offline  RASimmons
Messages: 5
Registered: October 2016
Member
Thanks for the reply!

Do I need to follow the same sort of process for rescaling the weights as I do for the DHS weights? I would assume I do.

As for actual implementation of the weights, to my knowledge standard statistical software won't let me use two different sets of weights in one model (usually you just specify the weight variable in the PWEIGHT command, for example), so how would I actually perform this type of weighting? Can software apply different weights to different sets of variables (e.g. the DHS weights to the variables from the DHS, the SPA weights to the variables from the SPA)? If not, then how would I actually do this? Would I have to apply the weights manually to the data (so multiply each observation by the reciprocal of the weight value), and then use manually adjust the standard errors (or use robust standard errors, or bootstrap, etc.)?

[Updated on: Tue, 06 December 2016 10:17]

Report message to a moderator

 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Rural-Urban sample weights
Next Topic: Post-stratification for DHS data
Goto Forum:
  


Current Time: Thu Mar 28 16:18:27 Coordinated Universal Time 2024