How to deal with missing DHS Clusters? [message #1224] |
Fri, 24 January 2014 07:57 ![Go to next message Go to next message](/theme/default/images/down.png) |
baremma2002
Messages: 20 Registered: April 2013
|
Member |
|
|
Dear DHS users,
I am using Côte d'Ivoire DHS 2011-2012 data and there is no cluster 225. However, it is reported that there were 352 DHS clusters. Is there a mistake in the geographic dataset? Some DHS clusters are missing but are not empty. How to deal with those clusters? If I used the centroids for those last ones, can I displace them up to 1 km for example to avoid duplicate points?
Thank you.
Emmanuel Barankanira
|
|
|
Re: How to deal with missing DHS Clusters? [message #1242 is a reply to message #1224] |
Mon, 27 January 2014 10:52 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Josh-DHS
Messages: 19 Registered: November 2013
|
Member |
|
|
Emmanuel,
Cluster 225 was inaccessible, so no data was collected from that cluster, and it was dropped from the sample. Please refer to page 7 of the final report (http://www.measuredhs.com/pubs/pdf/FR272/FR272.pdf) for additional information.
In the geographic dataset, clusters that have 0,0 as their coordinates are those for which no usable GPS data was collected. They are designated as MIS for "missing" under the SOURCE column of the table. If you are doing a spatial analysis, you should exclude these clusters as their coordinates do not represent their true locations.
Hope this helps.
|
|
|