Linking SPA & DHS data [message #8855] |
Mon, 04 January 2016 01:04 |
Mercysh
Messages: 35 Registered: April 2014
|
Member |
|
|
Compliments of the season.
I am using sample domain linkage for linking Malawi 2010 DHS and 2013/4 SPA data. My analysis is at district level. I want to drop Likoma district, an island, and my problem is that in the DHS data is combined with Nkhata Bay, another district.
Please advise how I can go around the problem.
Thank you
Mercy
|
|
|
|
Re: Linking SPA & DHS data [message #9080 is a reply to message #8893] |
Tue, 09 February 2016 15:08 |
Matt-DHS
Messages: 18 Registered: July 2014
|
Member |
|
|
Unfortunately the sample was designed to only provide representative indicators for the combination of those two districts, so there is no way to split Likoma from Nkhata Bay using the representative indicators. If it is feasible to drop Nkhata Bay as well, you could drop the entire combination, but not just Likoma itself. GPS data are available for the Malawi 2010 DHS and the 2013-14 SPA surveys, which may allow you to do analysis on a scale different than the representative regions used for the survey. Guidance on how to use the GPS data within your analysis is available in our SAR8 report ( http://dhsprogram.com/publications/publication-SAR8-Spatial- Analysis-Reports.cfm).
|
|
|