It was for this dimension that a new index was developed.

From those available from the software, one, the Neighbourhood Sorting Index was only global. And without the local indexes it was useless.  

And the Spatial Dissimilarity Index was not linear, it was high both in the richest and the poorest areas, and could not be used in the regression analysis.

So, to measure the segregation in the evenness/clustering dimension we developed a new measure based on the Local Indices of Autocorrelation.