A new approach for measuring and analysing residential segregation

This work proposes a new approach for residential segregation analysis contributing to the methodological debate relating to the measurement of the phenomenon and its comparability between different urban contexts. The strategy of analysis involves the use of areal interpolation methods to create high-resolution population grids, a compositional data approach, and the implementation of factorial analysis to define a socio-economic class composition index based on categorical data, which is a common data type in social research. The latter, in combination with spatial autocorrelation tools and the adoption of a criterion based on temporal distances to define spatial relations between grid cells, enables the identification and mapping of segregated areas. To test our method, we rely on the latest UK census data (2021) for the metropolitan areas of Liverpool, Manchester, and Newcastle upon Tyne, employing social groups defined according to the National Statistics Socio-economic Classification provided by the Office for National Statistics as population data. Finally, the validity of the proposed methodology is demonstrated through case studies, and the results are interpreted within the broader theoretical framework on the topic.