| Webinar Question on Burundi [message #6892] |
Wed, 29 July 2015 09:29 |
Sarah-DHS
Messages: 54
Registered: February 2013
|
Senior Member |
|
|
We wanted to provide an answer to a question that was asked by July 28th webinar participant Emmanuel that we were not able to get to during the discussion.
Emmanuel: Where can we link SPA and DHS clusters in Burundi?
DHS: Currently there is no SPA in Burundi.
Emmanuel: DHS clusters are enumeration areas (EA) that are represented by their centroids. How can we say that an HIV service is for example close or not to a DHS cluster given that they have not the same form?
DHS: You are correct DHS cluster data is a centroid of an area of unknown size and additionally that point is displaced to protect confidentiality which further masks the true location of any given respondent. The SPA is a point that represents the facility' s exact location. This means in linking these two data points together we are making some assumptions, these include (1) That on average the area of the cluster is fairly small and uniform in size (not oblong), (2) That the displacement of the point remains more or less in the same area of the original EA, and (3) that access from the displaced cluster location is similiar to access in the original cluster location. This is why we think of the linkage more as service environment and less as linking a DHS cluster to its "closest" health facility.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
