The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Sampling » Do clusters change over time?
Do clusters change over time? [message #26373] Tue, 14 March 2023 07:11 Go to next message
flore6614 is currently offline  flore6614
Messages: 1
Registered: March 2023
For my thesis, I am currently using the DHS from Kenya and Uganda.
I was wondering, are the same clusters revisited every 5 years for the survey? Or do clusters (and their geographical position) change depending on which households are sampled?
In other words, is cluster number 1 (variable v001) from Kenyan DHS 2014 on the same geographical location as cluster number 1 (variable v001) in Kenyan DHS 2008-2009?

If not, how can I find out which cluster from 2014 corresponds to a cluster from 2008-2009?

Many thanks in advance.

Kind regards,
Re: Do clusters change over time? [message #26375 is a reply to message #26373] Tue, 14 March 2023 08:19 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2742
Registered: February 2013
Senior Member

Following is a response from Senior DHS staff member, Tom Pullum:

The clusters are sampled from a very long list of census enumeration areas. The samples from different surveys are completely independent. The cluster ID numbers (1, 2, 3, etc.) have nothing to do with the ID numbers in the sampling frame or the numbers assigned in earlier surveys. Specific clusters are NOT revisited.

The clusters are selected within strata (v023, which is usually the same as v022). The strata are usually the same in successive surveys, although sometimes there are changes in administrative boundaries. Uganda has had major changes of this kind.

Previous Topic: SE calculation stratifications
Next Topic: Malawi Micronutrient Survey design issue
Goto Forum:

Current Time: Mon Mar 27 19:19:56 Coordinated Universal Time 2023