| Merging Zambia LCMS and DHS [message #13110] |
Thu, 21 September 2017 15:01  |
sarnoff
Messages: 1
Registered: September 2017
|
Member |
|
|
Hello,
I am trying to link two sets of surveys together on standard enumeration area (SEA):
(1) 2006 Zambia LCMS to 2007 Zambia DHS
(2) 2015 Zambia LCMS to 2014 Zambia DHS.
In theory, these should be matchable, since the DHS uses the census supervisory areas (CSA) and SEAs that Zambia's Central Statistical Office draws for the LCMS.
However, SEA is numbered differently in DHS and LCMS. In DHS, the numbers (variable: hv001/hv004) are unique: in 2007, they run from 1 to 320, and in 2014 they run from 1 to 722. In LCMS, the numbers are not unique and repeat across districts. Does a concordance file exist to match the SEAs in LCMS to DHS for the aforementioned years?
Thank you,
Kim
|
|
|
|
| Re: Merging Zambia LCMS and DHS [message #13115 is a reply to message #13110] |
Fri, 22 September 2017 08:26  |
Bridgette-DHS
Messages: 3214
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
It is DHS policy never to identify the specific enumeration areas that appear in a survey. This is part of the promise of confidentiality to the respondents. In fact, we do not have this information. Any links between the data and the sampling frame are destroyed after the data files have been constructed.
We had no involvement in the LCMS's and do not know whether this policy applied to them as well.
It is conceivable that the Zambian CSO can help with the LCMS, but supposedly they, as well as DHS, cannot link the survey cluster codes to the enumeration areas.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
