Exploring the spatial variation of regional social inequality allows the design of public policies; therefore, this study analyzes this phenomenon in the Metropolitan District of Quito, Ecuador (MDQ). To achieve this goal, an index was constructed to measure social inequality for each of the census sectors in the MDQ, using the Kernel nonlinear principal component analysis (KPCA) method. Two scenarios were considered while applying this method in order to select the one that best explained the social conditions of the census sectors according to the index. The first scenario used the method with no geographic weighting, and the second scenario involved the spatial component of the KPCA method. For the latter, appraisals of the properties in each census sector were considered. The results showed that by involving the spatial component in the method, the index reflects the MDQ reality better, and therefore, it could be seen that social inequality is highly spatially heterogeneous in the MDQ. © 2022, The Author(s), under exclusive licence to Springer Nature B.V.