The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » When weights are not supported
Re: When weights are not supported [message #3199 is a reply to message #3196] Wed, 05 November 2014 18:54 Go to previous messageGo to previous message
Yohannes is currently offline  Yohannes
Messages: 6
Registered: July 2014
Member
Thank you for the suggestion. Just a quick follow up question. Actually, I was going to fit random effects at country and strata levels but for a different reason - i.e. to capture unobserved strata and country level differences in my response variable -, so looking form the comment does this now mean that the random effects will not measure these effects, but rather show only the 'variation in weights'? As part of my random effect model I was also going to include some explanatory variables on the variance function (such as GDP for the country level random effect that are not necessarily related to sample size), and given the comments on the links between random effects and weights, not sure how these things play out.
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Using weights for small geographical locations
Next Topic: Weight for studying specific states or regions?
Goto Forum:
  


Current Time: Tue Apr 16 17:55:58 Coordinated Universal Time 2024