Choosing a Survey Sample when Data on the Population are Limited

Author(s)
Shannon, H. S., Hutson, R., Kolbe, A., Stringer, B., Haines, T., Nelson, N., Yang, L., Reilly, A., Hardin, J., & Hartley, D.
Publication language
English
Date published
01 Jan 2012
Publisher
Emerging Themes in Epidemiology
Type
Articles
Keywords
Conflict, violence & peace, Working in conflict setting, Evaluation-related

 

Methods proposed for sampling when data on the population are limited are often biased. This article proposes a new method for drawing a population sample using Global Positioning Systems and aerial or satellite photographs. Global Positioning System locations were randomly sampled in designated areas. A circle was drawn around each location with a radius representing 20 m. Buildings in the circle were identified from satellite photographs; one was randomly chosen. Interviewers selected one household from the building, and interviews were conducted with eligible household members. Participants had known selection probabilities, allowing proper estimation of parameters of interest and their variances.