Are cluster values the same across different years (versions)? [message #12470] |
Tue, 23 May 2017 10:59 |
SantiagoCC
Messages: 4 Registered: May 2017 Location: Bogota
|
Member |
|
|
Hi, I wanted to ask if the cluster values relating specific regions in a country are the same in different versions of the DHS program.
I mean, if in version 2000 a cluster with value 1 made reference to region A, then if in version 2005 a value of 1 for the same variable also refers to region A or to another one.
Thanks!
|
|
|
|
|
Re: Are cluster values the same across different years (versions)? [message #12485 is a reply to message #12474] |
Fri, 26 May 2017 11:19 |
Bridgette-DHS
Messages: 3190 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Thank you for posting this question on the forum, where other users can see it. Unless a question is very specific, we prefer that it be put on the forum rather than sent to the archive email address. It also happens that we did not receive the email you sent to the archive. Perhaps there was a typo in the address....
In general, clusters are enumeration areas (EAs) from the most recent census. After the survey has been done, DHS does not have any way to identify which EAs appeared in the sample, let alone to link cluster id codes to specific EAs in the census sampling frame. The lists used during sampling and data collection are always destroyed. This is part of the agreement between DHS and the countries, in order to protect the privacy of the respondents. This is also the reason why the GPS coordinates of the clusters (in the more recent surveys that include them) are always displaced. DHS tries to make it impossible for anyone to identify the selected EAs.
|
|
|