Sull'impiego di un indice di dissimilarità nello studio della disposizione di popolazioni straniere su un territorio urbano

Duncan and Duncan dissimilarity index is one of the most commonly used measures of segregation. It is well known that this index reflects random allocation as well as systematic group segregation; this leads to potentially large values of the segregation index when unit sizes or minority proportions are small, even if there is little or no systematic segregation. To overcome this bias issue a few methods have been proposed. In this paper it is presented a new bootstrap correction and Monte Carlo simulations prove it to perfonn better than the existing ones. This new correction is illustrated by comparing dissimilarity levels across the settlement patterns offoreign immigrants in the city of Catania.