
Supervised machine learning and predictive models have achieved an impressive standard today, enabling us to answer questions that were inconceivable a few years ago. Besides these successes, it becomes clear, that beyond pure prediction, which is the primary strength of most supervised machine learning algorithms, the quantification of uncertainty is relevant and necessary as well. However, before quantification is possible, types and sources of uncertainty need to be defined precisely. While first concepts and ideas in this direction have emerged in recent years, this paper adopts a conceptual, basic science perspective and examines possible sources of uncertainty. By adopting the viewpoint of a statistician, we discuss the concepts of aleatoric and epistemic uncertainty, which are more commonly associated with machine learning. The paper aims to formalize the two types of uncertainty and demonstrates that sources of uncertainty are miscellaneous and can not always be decomposed into aleatoric and epistemic. Drawing parallels between statistical concepts and uncertainty in machine learning, we emphasise the role of data and their influence on uncertainty.
In low- and middle-income countries (LMICs), accurate estimates of subnational health and demographic indicators are critical for guiding policy and identifying disparities. Many indicators of interest are proportions of binary outcomes and the task of estimating these fractions is often called prevalence mapping. In LMICs, health and vital records data are limited, so prevalence mapping relies on data from household surveys with complex sampling designs. However, estimates are often desired at spatial resolutions at which data are insufficient for reliable weighted estimation. We review two families of approaches to prevalence mapping: small area estimation (SAE) methods (from the survey statistics literature) and model-based geostatistics (MBG) methods (from the spatial statistics literature). SAE models can be “area-level” or “unit-level” and commonly use area-specific random effects and rely upon high-quality covariate data, often obtained from administrative sources. Unit-level models for binary responses are relatively underdeveloped. MBG approaches explicitly specify binary response models, incorporate continuous spatial random effects, and leverage alternative sources of data such as those arising from satellite imagery. These models are usually studied within a Bayesian framework. SAE methods often address the design by incorporating sampling weights or modeling the sampling mechanism. Two delicate issues arise when using MBG methods for prevalence mapping. First, aggregating unit-level predictions to create area-level summaries requires population-level information that is rarely directly available. Second, MBG approaches typically assume the sampling design is ignorable. We review both SAE and MBG approaches to prevalence mapping, and argue that binary response models can be improved using insights from both the survey sampling and the spatial statistics literature. We highlight these issues using household survey data from the Zambia 2018 Demographic Health Survey to estimate subnational HIV prevalence for women aged 15–49.
In low- and middle-income countries (LMICs), accurate estimates of subnational health and demographic indicators are critical for guiding policy and identifying disparities. Many indicators of interest are proportions of binary outcomes and the task of estimating these fractions is often called prevalence mapping. In LMICs, health and vital records data are limited, so prevalence mapping relies on data from household surveys with complex sampling designs. However, estimates are often desired at spatial resolutions at which data are insufficient for reliable weighted estimation. We review two families of approaches to prevalence mapping: small area estimation (SAE) methods (from the survey statistics literature) and model-based geostatistics (MBG) methods (from the spatial statistics literature). SAE models can be “area-level” or “unit-level” and commonly use area-specific random effects and rely upon high-quality covariate data, often obtained from administrative sources. Unit-level models for binary responses are relatively underdeveloped. MBG approaches explicitly specify binary response models, incorporate continuous spatial random effects, and leverage alternative sources of data such as those arising from satellite imagery. These models are usually studied within a Bayesian framework. SAE methods often address the design by incorporating sampling weights or modeling the sampling mechanism. Two delicate issues arise when using MBG methods for prevalence mapping. First, aggregating unit-level predictions to create area-level summaries requires population-level information that is rarely directly available. Second, MBG approaches typically assume the sampling design is ignorable. We review both SAE and MBG approaches to prevalence mapping, and argue that binary response models can be improved using insights from both the survey sampling and the spatial statistics literature. We highlight these issues using household survey data from the Zambia 2018 Demographic Health Survey to estimate subnational HIV prevalence for women aged 15–49.
In this discussion of Wakefield et al. (2026), I provide some intuition on why model-based geostatistical methods, based on a super-population (random field) assumption, can be robust even in finite-population settings, common in prevalence mapping. I provide a theoretical result showing that when there is true spatial structure in the prevalences across areas, these models enforce smoothing and can sometimes do better than purely design-based estimators.