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 message Go to previous message](/theme/default/images/up.png) ![Go to next message Go to previous message](/theme/default/images/down.png) |
Bridgette-DHS
Messages: 3151 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
|
|
|
Goto Forum:
Current Time: Sat Aug 31 07:14:20 Coordinated Universal Time 2024
|