A grid-based sampling design for household surveys in the absence of actionable sampling frames

Darin, Edith, Boo, Gianluca, Tatem, Andrew and Thomson, Dana (2019) A grid-based sampling design for household surveys in the absence of actionable sampling frames. GEOMED 2019: XI international, interdisciplinary conference on spatial statistics, geographical epidemiology and geographical aspects of public health, , Glasgow, United Kingdom. 27 - 29 Aug 2020. 1 pp . (doi:10.13140/RG.2.2.31686.83524).

Abstract

Household surveys are a cost-effective data source for estimating health and demographic characteristics in low- and middle-income countries. These surveys involve sampling from a frame listing all the members of the population of interest. However, the listing needs to be complete, accurate, and up-to-date to draw samples that can be generalized to the entire population. A sampling frame generally consists of the enumeration areas established during the last national census, together with the associated population counts. However, these counts are often inaccurate because the national census is carried out on a decade basis. Outdated sampling frames are particularly critical in Africa, where five countries had their last census more than fifteen years ago. To tackle the absence of actionable sampling frames, we propose an original grid-based sampling design, embedding spatial sampling concepts into household surveys. We demonstrate our framework with a case study developed in the western part of the Democratic Republic of Congo (DRC). This country had its last national census in 1984, and the resulting sampling frame is still currently used in household surveys.

This record has no associated files available for download.

Contributors

Author: Edith Darin

Author: Gianluca Boo

Author: Andrew Tatem

Author: Dana Thomson

Download statistics