The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Correct weight for a sub-sample
Re: Correct weight for a sub-sample [message #494 is a reply to message #491] Wed, 29 May 2013 22:19 Go to previous messageGo to previous message
adi.greif@yale.edu is currently offline  adi.greif@yale.edu
Messages: 5
Registered: May 2013
Member
Thank you so much for the fast response, it is tremendously helpful. I had considered the procedure that you wrote down but I wasn't sure it work--- I don't quite understand how STATA is estimating the standard errors using svyset, and was afraid this would introduce some sort of problem given all the warnings about using subpop rather than dropping observations.

I'll follow up on the cross-country question here -- again, thank you so much for offering to help. I am interested in estimating differences in variables (such as the average proportion of women working in agriculture) across former British and former French colonies. I'm taking one survey from each country to make weighing easier for me to understand. I want each survey to be considered as an observation (even though different countries have different population sizes). So I think in this case you are right that I need to renormalize the DHS weights so they all add up to 1, rather than using the total population size to renormalize. You wrote in another post that it depends on "What do you want the thing you are estimating to represent? An average of all the people that live in those five countries? An average that thinks of a country as an individual? An average that represents the people sampled in the DHS?"

I think I am going after the second estimation you mentioned "An average that thinks of a country as an individual" and that's why I should normalize to 1 rather than using the total population. Do you agree? I am not sure what you meant by it depending on the assumptions I am willing to make?

Many thanks,
Adi
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Weighting the KR for multiple countries
Next Topic: Comparing coefficients across years in the same country
Goto Forum:
  


Current Time: Thu May 23 10:53:53 Coordinated Universal Time 2024