The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Clustered Standard Errors
Re: Clustered Standard Errors [message #22292 is a reply to message #22281] Wed, 24 February 2021 08:09 Go to previous messageGo to previous message
analyst_till is currently offline  analyst_till
Messages: 10
Registered: January 2021
Member
Thanks a lot!

I have a last concluding question: Since we discussed the clustering of standard errors here, I was confused about how the svy-adjustments directly relate to the clustering of standard errors? In this thread above, you mentioned that one can make cluster adjustments either via svy or via "reg, cluster(v021)". Could you tell me which part of "svyset" defines that I will cluster SEs at a certain level, is it "svyset PSU...." Or the "singleunit(centered)" option at the end?

Here, since I am working with repeated cross-sectional data over several years, you recommended to include a grouped PSU variable, namely "egen cluster_ID = group(v001 survey)".
Does this mean that I cluster standard errors at PSU-level, or at the PSU-time level (since I interact PSU with survey) here? And if yes, does it always has to be the PSU-time level at which I cluster standard errors?
 
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
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Confirming correct survey design
Next Topic: What is the difference between v005 and sweight DHS India?
Goto Forum:
  


Current Time: Sat Nov 26 04:52:08 Coordinated Universal Time 2022