The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Municipality sampling weight
Municipality sampling weight [message #19840] Wed, 19 August 2020 16:48 Go to next message
j_lewis98 is currently offline  j_lewis98
Messages: 7
Registered: July 2020

For the SADHS 2016 I have managed to identify municipality-level location of each cluster through the matching of GADM administrative boundary shapefiles to DHS cluster GPS data; in ArcGIS. I want to use municipalities as level 3 in a multilevel logistic model: with level 1 as individuals and level 2 as households. How can I calculate the separate finalised sampling weights for each level (in particular municipality) and how would I present this in STATA code? Additinoally, would I need to account for the actual PSU (clusters) in my analysis or not?


Re: Municipality sampling weight [message #19858 is a reply to message #19840] Fri, 21 August 2020 12:58 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2872
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

Yesterday we posted a reply to a similar question from you. Perhaps you did not have a chance to read it. We cannot offer any guidance with respect to the use of municipality in a multi-level model. Those units were not part of the sampling design. You can of course include covariates that are municipality-specific even if that is not a level in a multi-level model.
Previous Topic: Province-level summary stats for Kenya 2014
Next Topic: PAB EDHS-2000 BR file
Goto Forum:

Current Time: Tue Oct 3 16:53:15 Coordinated Universal Time 2023