Identification of Hot Spots of Social and Housing Difficulty in Urban Areas: Scan Statistic for Housing Market and Urban Planning Policies

The objective of the present work is to use statistical data to identify territorial zones characterized by the presence of urban poverty related to property ownership and the availability of residential services. Poverty clusters have a high concentration of poor people, but that does not mean that everyone living in them is poor. While poverty is widely accepted to be an inherently multi-dimensional concept, it has proved very difficult to develop measures that both capture this multidimensionality and make comparisons over time and space easy. With this in mind, we attempt to apply a Total Fuzzy and Relative (TFR) approach, based on a fuzzy measure of the degree of association of an individual to the totality of the poor and an approach of semantic distance (Munda 1995), based on the definition of a “fuzzy distance” as a discriminating multidimensional reference to rank the availability to property in real estate market, as complement of urban poverty, in the specific case of the City of Bari. These approaches have been improved using the SaTScan methodology, a circle-based spatial-scan statistical method (Kulldorff 1997; Patil and Taille 2004; Aldstat and Getis 2006). It concerns geoinformatic surveillance for poverty hot-spot detection, used as a scientific base to lead urban regeneration policies.