Modeling Urban/Rural Fractions in Low- and Middle-Income Countries

TL;DR

This paper develops a model-based approach for estimating urban and rural population fractions in low- and middle-income countries, improving subnational health indicator estimates by accounting for stratification and using advanced classification algorithms.

Contribution

It introduces a novel stratification-aware aggregation method incorporating classification algorithms and pixel-level covariates for better subnational estimates.

Findings

Improved HIV prevalence estimates in Malawi's women aged 15-49.

Effective classification of urban/rural areas using multiple algorithms.

Enhanced predictive accuracy through pixel-level covariates.

Abstract

In low- and middle-income countries, household surveys are the most reliable data source to examine health and demographic indicators at the subnational level, an exercise in small area estimation. Model-based unit-level models are favored in producing the subnational estimates at fine scale, such as the admin-2 level. Typically, the surveys employ stratified two-stage cluster sampling with strata consisting of an urban/rural designation crossed with administrative regions. To avoid bias and increase predictive precision, the stratification should be acknowledged in the analysis. To move from the cluster to the area requires an aggregation step in which the prevalence surface is averaged with respect to population density. This requires estimating a partition of the study area into its urban and rural components, and to do this we experiment with a variety of classification algorithms, including logistic regression, Bayesian additive regression trees and gradient boosted trees. Pixel-level covariate surfaces are used to improve prediction. We estimate spatial HIV prevalence in women of age 15-49 in Malawi using the stratification/aggregation method we propose.