Small Area Estimation Methods to Monitor the Progress towards the Sustainable Development Goals at the Subnational Level in the European Union (SAESDGs-EU)

forthcoming

The United Nations (UN) General Assembly highlights the need to disaggregate Sustainable Development Goals (SDGs) indicators by different socio-economic characteristics e.g., income, sex, age, race, ethnicity, migratory status, disability, and small geographic location, or other relevant characteristics (General Assembly resolution 68/261). However, large scale national sample surveys, i.e., the EU Statistics on Income and Leaving Conditions (EUSILC), which collect variables that are needed to produce such indicators, are not designed to be reliable at those disaggregation levels. Small area estimation (SAE) methods are thus needed to produce accurate and precise estimates at a subnational level. SAE methods combine different auxiliary data sources (e.g., the Census with survey data), thus ‘borrowing strength’ from related areas to make a better-quality small area estimates. The focus of SAESDGs-EU is on economic well-being, which is closely related to the SDGs, as improving quality of life is a core objective of sustainable development. Many of the 17 SDGs explicitly target well-being by addressing economic, social, and environmental factors. Economic well-being is multidimensional, hence a variety of indicators should be considered when measuring it. In this project, we will develop a single-step method to produce small area composite indicators of economic well-being based on the EUSILC. This is a ground-breaking innovation, since currently first composite indicators are produced and then SAE methods are applied (or vice-versa), in a two-step approach. The literature has highlighted problems with this approach which impact the data quality of the final estimates. Given that SAE requires auxiliary information, we will also explore pitfalls and potentials of using non-traditional data sources as auxiliary information (e.g., web-data, satellite data). These are well known data sources that provide granular spatial information.