The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Merging data files » Merging SPA and DHS data
Merging SPA and DHS data [message #2847] Tue, 02 September 2014 08:28 Go to next message
leo2015 is currently offline  leo2015
Messages: 4
Registered: June 2014
Member
I would like to merge the SPA and DHS (HR) data sets for one country in STATA. Can you please explain how DHS researchers have linked the household and facility level data for analysis?

Thank you for your help.

Dan
Re: Merging SPA and DHS data [message #2848 is a reply to message #2847] Tue, 02 September 2014 08:34 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1763
Registered: February 2013
Senior Member
It is not possible to merge the DHS and SPA surveys, since the analysis unit is not the same (households vs facilities).
Re: Merging SPA and DHS data [message #2849 is a reply to message #2848] Tue, 02 September 2014 08:38 Go to previous messageGo to next message
leo2015 is currently offline  leo2015
Messages: 4
Registered: June 2014
Member
Is there a way to link the data at least based on administrative level such as region?
Re: Merging SPA and DHS data [message #2918 is a reply to message #2849] Tue, 09 September 2014 15:58 Go to previous message
ClaraB-DHS is currently offline  ClaraB-DHS
Messages: 38
Registered: March 2013
Location: Maryland
Member
Yes there are ways to link the DHS and SPA data using geographic information system (GIS) software. For SPA surveys since 2009 the GPS location of the Facilities are available for download from the DHS data request system. The GPS location of the SPA facility and the DHS cluster can then be linked by sub-national administrative areas or by other geographic analysis methods.

This approach may not always be appropriate depending on your research question.
1) DHS cluster locations are displaced to protect confidentiality. See Spatial Analysis report 7 for more information on this: http://www.dhsprogram.com/publications/publication-SAR7-Spat ial-Analysis-Reports.cfm
2) Measure direct distances to facilities and other analysis methods might be impacted by the displacement, see Spatial Analysis Report 8 for more information on this: http://www.dhsprogram.com/publications/publication-SAR8-Spat ial-Analysis-Reports.cfm
3) Many SPA surveys are samples which makes the linkage at small geographic scales problematic. A recent per reviewed publication on this topic look at the misclassification bias due to the sample. You can find that article at: http://www.pophealthmetrics.com/content/11/1/14

A forth coming Spatial Analysis Report 10 (Release date ~September 28, 2014) will look at issues around the linkage more in depth.
A recent publication by MEASURE Evaluation also looks more generally at the linkage of facility and population based surveys, that report can be found at: http://www.cpc.unc.edu/measure/publications/ms-14-88

Previous Topic: variables hv001 hv002 hv003 do not uniquely identify observations in the using data
Next Topic: Merging child's malaria results from HH member recode dataset with other child data in Child recode
Goto Forum:
  


Current Time: Sat Mar 28 19:06:21 Eastern Daylight Time 2020