This paper proposes an original methodology for the analysis of socioeconomic residential segregation. The strategy involves employing areal interpolation methods to create population grids, applying a compositional data approach to quantify categorical distributions, and utilising principal component analysis to define an index of socio-economic class composition for each cell in the study area. By combining index values with spatial autocorrelation tools, it is possible to identify and map segregated areas. To test our method, we rely on the latest UK census data (2021) for the Liverpool metropolitan area, using social groups defined according to the National Statistics Socio-economic Classification.

KEYWORDS: residential segregation analysis, grid cells, compositional data analysis,PCA, spatial analysis