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 #3196 is a reply to message #3188] Wed, 05 November 2014 08:20 Go to previous messageGo to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3214
Registered: February 2013
Senior Member
Following is a response from senior DHS Specialist, Tom Pullum.

I really don't think there is an easy answer to this question. If the software doesn't allow you to use weights--and there are several complex statistical procedures for which a weighted version has not yet been written--then your results for each country, let alone all countries combined, will be biased toward the over-sampled sub-populations.

I recommend that you include fixed effects for country or survey, and fixed or random effects for all the strata, in your models. Most of the variation in the weights (v005 and the survey-specific factors that you could calculate) will be explained by those effects. Perhaps other users have different solutions to this problem.
 
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: Fri Dec 27 07:03:17 Coordinated Universal Time 2024