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 #22173 is a reply to message #22167] Tue, 09 February 2021 08:49 Go to previous messageGo to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2630
Registered: February 2013
Senior Member
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

The clustering is by PSUs, not districts, so svyset does not need to include any reference to clusters. You can refer to a posted file (survey_strata.do) that describes the strata in all DHS surveys. In each survey, name the identifier "stratum_ID". Then construct the combined ID as, for example, stratum_ID_all, within "egen stratum_ID_all=group(survey stratum_ID)". Also "egen cluster_ID_all=group(survey v001)". That should work.

I'm not sure that your regression needs to include "i.district". That term could overlap too much with "i.post##i.Treatment". I would try regressions with and without that term. Good luck.

[Updated on: Tue, 09 February 2021 08:51]

Report message to a moderator

 
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 04:13:16 Coordinated Universal Time 2022