The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Geographic Data » Pakistan GPS Dataset for DHS 2006-07
Pakistan GPS Dataset for DHS 2006-07 [message #2487] Sat, 28 June 2014 09:26 Go to next message
owraza is currently offline  owraza
Messages: 36
Registered: December 2013
Location: Karachi, Pakistan
Member
In DHS Spatial Analysis Report 7, it is mentioned that, after 2008 GPS data in all DHS surveys have used displacement procedures... So can you guide me what method was applied for confidentiality in GPS data during Pakistan DHS 2006-07 survey? and whether GPS data for this survey contain x,y coordinates of each cluster (as I have understood from 'GPS Cluster Position Form') or each household?

Regards
Re: Pakistan GPS Dataset for DHS 2006-07 [message #2507 is a reply to message #2487] Tue, 01 July 2014 16:26 Go to previous messageGo to next message
Thea-DHS
Messages: 28
Registered: March 2013
Member
Thanks for your post. All of the GPS data that The DHS Program releases from household surveys is displaced according to the procedure described in the Spatial Analysis Report 7 ( http://dhsprogram.com/publications/publication-SAR7-Spatial- Analysis-Reports.cfm). For most surveys conducted since 2008, the displacement is restricted to prevent cluster coordinates from being displaced across second level administrative unit boundaries. The Pakistan 2006-07 survey was conducted before then so the displacement of the GPS data was only restricted at the level of the DHS survey region. This means that it was possible for the coordinates to be displaced across a lower level administrative boundary (e.g. divisions or districts).

The coordinates in the GPS dataset are the displaced location of each cluster center, not the location of individual households. For more information, please see this other user forum thread on the topic of displacement http://userforum.dhsprogram.com/index.php?t=msg&th=19&am p;start=0&S=55ef2e974e7abbb8fbb42e52854af55e.

If this has not adequately answered your question, please reply.
Re: Pakistan GPS Dataset for DHS 2006-07 [message #2508 is a reply to message #2507] Tue, 01 July 2014 20:16 Go to previous messageGo to next message
owraza is currently offline  owraza
Messages: 36
Registered: December 2013
Location: Karachi, Pakistan
Member
Thanks for your reply. If I have understood your post correctly, displacement has been done in all DHS surveys, just the method (level) of displacement has been modified in post-2008 DHS surveys. So, is it possible that somehow I can get information about the amount of displacement? (Let's say in terms of percentage of displaced cluster across the administrative boundaries). Such information if made available, can be used in discussion of study's result.
Re: Pakistan GPS Dataset for DHS 2006-07 [message #2516 is a reply to message #2508] Thu, 03 July 2014 09:05 Go to previous message
Thea-DHS
Messages: 28
Registered: March 2013
Member
You are correct, all GPS data for DHS Program household surveys are displaced. For most surveys after 2008, the displacement has been further restricted to the administrative two level, to ensure that displaced GPS clusters remain in the same administrative two unit as the original GPS cluster.

The DHS Program is not able to release information that details how many clusters were displaced across lower level administrative boundaries. However, the Guidelines on the Use of DHS GPS Data publication will provide you with guidance on how to account for the displacement procedure in your analysis: http://dhsprogram.com/publications/publication-SAR8-Spatial- Analysis-Reports.cfm. Chapter 5, which discusses misclassification of GPS clusters as a result of the displacement, should be applicable to your study.

If this has not adequately answered your question, please reply.
Previous Topic: Import GPS Data - Manuals
Next Topic: Mapping Z-scores
Goto Forum:
  


Current Time: Thu Mar 28 15:24:07 Coordinated Universal Time 2024