Survey Design and Results from UNICEF Project in Sierra Leone

• Activity Number: 250
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Theresa Diaz*+ and Sowmya R. Rao and Adam Wolkon and Gary Shapiro and Peter S. Bangura and John Baimba
• Companies: UNICEF and University of Massachusetts Medical School and Centers for Disease Control and Prevention and Statistics Without Borders and Statistics Sierra Leone and Statistics Sierra Leone
• Keywords: childhood mortality ; surveys ; cluster sample

Abstract

In Sierra Leone, UNICEF is implementing and evaluating Community Health Volunteers strategy to treat Malaria, Diarrhea, and Pneumonia in children< 5 years of age. As part of this evaluation, a baseline survey was conducted in 2010. A follow-up using the same sampling design and methodology is planned for 2012. This paper describes the sample design and preliminary analysis results. The design is a household cluster survey in two intervention and two control districts that were chosen based on pre-determined criteria. A cluster sample of 3000 households was then selected from each district using a probability proportional to size sampling scheme. Population-based interviews were conducted in person. Both direct and indirect mortality rates were computed. The results indicate that the intervention and control districts did not differ significantly in socio-economic factors, disease prevalence or health seeking behaviors. Mortality rates were lower than in previous surveys indicating declining rates of childhood mortality in Sierra Leone.