The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Senegal regions
Senegal regions [message #2475] Wed, 25 June 2014 17:07 Go to next message
mledger is currently offline  mledger
Messages: 8
Registered: June 2014
Location: Southampton
Member
I am trying to recreate a graph which shows the breakdown of contraceptive use in Senegal from 1986 to 2005 by region.

However I seem unable to find the variable which breaks down the regions past west, central, south and north east and into the 10 named regions shown in this report: http://www.acquireproject.org/archive/files/3.0_program_effe ctively/3.2_resources/3.2.2_studies/senegal_case_study_final .pdf

I would normally have used the region variable v101 but for 1986 - 1997 this variable doesn't seem to be broken down, as it has been in 2005 and 2010.

Do you any suggestions as to what I could do?
Re: Senegal regions [message #2506 is a reply to message #2475] Tue, 01 July 2014 16:19 Go to previous message
Thea-DHS
Messages: 28
Registered: March 2013
Member
The 1986, 1993 and 1997 Senegal DHS were designed to give representative results at the level of four survey regions which consisted of groupings of the 10 (at that time) administrative regions. Since the 1999 DHS, the surveys have been representative at the level of the individual administrative regions. To view how the DHS regions have changed over time or to download GIS files of the regions for the Senegal surveys, please visit the country's page on our Spatial Data Repository site http://spatialdata.dhsprogram.com/boundaries.html?country=SN .

The main geographic variable in the DHS datasets is the survey region classification, which is v101 in the 1986 dataset and hv024 in the later surveys. The reason that it is not broken down to administrative region level in the first three surveys is that those surveys are not representative at that level. However, the 1993 and 1997 datasets do contain the variable shregion, which gives the name of the administrative region in which each cluster is located. You can use these to match up the clusters with the regions in the later surveys, however, we do advise caution when conducting your analysis at administrative levels below those at which the surveys were designed to be representative.

The 1986 dataset does not contain any geographic variables below the four survey regions, and GPS data was not collected for that survey. Therefore it is not possible to match the clusters to the 10 administrative regions for that survey. If you do need to include the data from the 1986 survey in your study, we suggest that you aggregate the data from the other surveys up to the 4 large regions and conduct the analysis at that level.

If this has not adequately answered your question, please reply.
Previous Topic: Peru 2010-2012
Next Topic: Ghana 2003 - Height and Weight Scores Data
Goto Forum:
  


Current Time: Thu Oct 31 19:54:29 Coordinated Universal Time 2024