The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Service Provision Assessment (SPA) » 2012-2013 Senegal SPA: Sampling Strategy
2012-2013 Senegal SPA: Sampling Strategy [message #6965] Fri, 07 August 2015 08:33 Go to next message
cjuan is currently offline  cjuan
Messages: 1
Registered: August 2015
Location: London, UK

For the family planning client observation and exit data in the 2012-2013 Senegal SPA, would you kindly provide details on the sampling strategy? For example, is there stratification to account for? In the Senegal SCSPA 2012-2013 report, it notes that "the sample encompassed 50 percent of hospitals, 50 percent of health centers including garrison medical centers (GMC), 20 percent of health posts including infirmaries..."

Also, does the client weight found in the FP observation/exit data account for health provider and health facility weights from the other SPA data files?

Thank you in advance,
Christina Juan
Re: 2012-2013 Senegal SPA: Sampling Strategy [message #7095 is a reply to message #6965] Fri, 21 August 2015 18:03 Go to previous messageGo to next message
Messages: 1393
Registered: February 2013
Senior Member
Dear User,
We have submitted your post to one of our sampling experts. As soon as we have an answer we will post. Thank you!
Re: 2012-2013 Senegal SPA: Sampling Strategy [message #8115 is a reply to message #7095] Tue, 25 August 2015 10:23 Go to previous message
Messages: 1393
Registered: February 2013
Senior Member
Here is a response from Senior Sampling expert, Dr. Ruilin Ren:
The SPA sample of Senegal 2012-2013 was a stratified sample. In each region, the Health Facilities (HF) were classified by crossing the management authority (Government/private) with the health facility type (Hospitals, Health Centers and Health Posts). The approximate sampling fraction at national level was given in the final report as you cited. But at sampling stratum level, the fraction varies a little depending on availability of HF.

As for the client weight (sick child or family planning) in the 2012-2013 survey, because it was not able to link the client with the provider, the provider weight is not taken into count. The client weight uses the HF weight as base weight.
The SPA survey still evolves, the sampling weight calculation has been changed since the 2012-2013 survey. A general provider weight was calculated (not taken the provider type into account) as base weight for clients which we believe improves a little the client weight. Since the field procedure will not guarantee that for every interviewed client, their provider is interviewed, so the client weight is not an "exact" weight but an approximate weight. Hope this will continue to improve if there is an exact "interviewed provider"-"interviewed client" relation for all interviewed clients.

Hope this is helpful.
Previous Topic: Linking DHS and SPA datasets
Next Topic: Malawi SPA xc112
Goto Forum:

Current Time: Fri May 24 19:55:28 Eastern Daylight Time 2019