Proposal Description

Emerging well-being challenges in the Global South require statistical tools that go beyond traditional measures of income and wealth. Issues such as cooling poverty, food insecurity, and multidimensional well-being are strongly spatial: they vary across territories, interact with local infrastructure and environmental conditions, and are often insufficiently captured by conventional household surveys alone. At the same time, the growing availability of satellite imagery, remote sensing products, climate data, gridded population layers, geospatial infrastructure indicators, and administrative sources creates new opportunities for producing policy-relevant indicators at fine geographic scales.

This session discusses statistical modelling strategies for integrating traditional and non-traditional data sources to estimate emerging well-being indicators. Particular attention will be devoted to small area estimation, Bayesian and hierarchical models, and uncertainty quantification. The session addresses both methodological developments and applied examples, with a focus on low- and middle-income countries, where survey data are often sparse, spatial inequalities are substantial, and policy needs are urgent.

The contributions approach this agenda from complementary perspectives. The first contribution provides a methodological comparison of small area estimation models for integrating survey and remote sensing data. Using a synthetic population based on Maharashtra, India, it evaluates different Small Area Estimation strategies for estimating a wealth index under controlled scenarios. The simulation design combines individual characteristics, district effects, night-time lights, rural/urban classification, and a two-stage survey design, comparing area-level, unit-context, hierarchical, and fully Bayesian spatial models under different assumptions on signal strength, intra-cluster correlation, location jittering, and measurement uncertainty.

The second contribution focuses on food insecurity and food demand in Nigeria. Using recent household survey data and a combined Working–Leser model-SAE approach, it examines how agricultural seasonality and access to credit shape food demand elasticities. The results highlight the role of liquidity constraints in consumption smoothing and point to pronounced regional heterogeneity. This contribution shows why welfare-relevant indicators should account for both spatial and temporal variation, especially in contexts exposed to seasonal shocks and uneven market access.

The third contribution frames systemic cooling poverty as a small area estimation problem. Extreme heat is an increasing threat to health and welfare, but heat-related deprivation is not determined by temperature alone. It also depends on housing quality, access to water and sanitation, healthcare accessibility, labour conditions, education, information, infrastructure, and social protection. By combining household surveys, climate records, geospatial infrastructure indicators, and policy information, small area methods can help map cooling-related deprivation below the national level, identifying both where vulnerability is concentrated and which dimensions drive it.

Together, the papers show how modern statistical modelling can improve the precision, spatial resolution, and policy relevance of well-being indicators in data-limited settings, while explicitly accounting for uncertainty.