Keywords: household survey, gridded population data, sample frame

This study addresses the challenges faced by practitioners in generating gridded population sampling frames following the retirement of Flowminder’s GridSample tool in 2023. Sample frames based on gridded population data are often used in the absence of a recent, high-quality census as the basis for a sample frame, or when such a frame does not meet the needs of implementers. For several years the GridSample tool allowed users to specify parameters for a sample frame and then download selected PSUs from a frame based on gridded population data. The discontinuation of GridSample—which was accessible for users without programming or GIS skills—increased the barrier to entry for conducting a gridded population survey, but PC-based workflows for creating such frames have been developed. This presentation focuses on the gridEZ algorithm, a freely available R function for generating gridded population sample frames. The session provides an overview of the gridEZ workflow, including data requirements, common challenges with memory and runtime limitations, and post-processing steps to create shapefiles, maps, and other useful files. Survey practitioners with intermediate R and GIS skills will come away with practical insights on adapting the gridEZ workflow to generate sampling frames customized to their needs.

This presentation complements another paper on this panel that details how the gridEZ algorithm functions, algorithm options for customizing output frames, and new tools for refining gridEZ frames.