[Unknown type: UNSPECIFIED]
Abstract
Statistical modelling approaches which produce fine spatial resolution population estimates have been developed to fill data gaps in resource-poor countries where census data are either outdated or incomplete. These population modelling methods often draw upon recent georeferenced sample population enumeration datasets to predict population density and distribution at both sampled and non-sampled locations, based on their correlation with a set of carefully selected geospatial covariates. These modelled population estimates are increasingly used to support governance, health surveillance, equitable resource allocation, and humanitarian response. However, methodological challenges remain. For example, the georeferenced sample enumeration data are usually disparate and patchy in their distributions, with a high proportion of non-sampled locations that result in highly uncertain estimates. Here, we present a model-based Bayesian geostatistical small area population estimation approach which simultaneously · Combines multiple sample population enumeration datasets and· Explicitly integrates spatial autocorrelation within a single modelling framework. Findings from a simulation study show varying levels of accuracy in the posterior parameter estimates over different levels of spatial variance and data missingness. The methodology, which was further validated using five nationally representative household listing datasets in Cameroon, provides a valuable methodological development in small area population estimation modelling from sparsely distributed sample enumeration data.
More information
Identifiers
Catalogue record
Export record
Altmetrics
Contributors
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.