Resumen

GARROCHO, Carlos  y  CAMPOS-ALANIS, Juan. Requiem for no spatial indicators of residential segregation. Pap. poblac [online]. 2013, vol.19, n.77, pp.269-300. ISSN 2448-7147.

This paper is based on deliberately simple numerical examples for: i. To explain the serious problems of using no spatial indicators of residential segregation (as the traditional indicators of Dissimilarity, Isolation and Interaction, intensively reported in the Latin American literature); and, ii. Conceptually and operationally demonstrate the superiority of the genuinely spatial segregation indicators, as Moran's Global and Local Spatial Autocorrelation indexes. No spatial indicators of segregation register, at least, four fundamental flaws: produce the same results for different spatial patterns of segregation (Flaw 1), are unable to disclose what happens with the segregation within the study area (Flaw 2), their results depend entirely on how the data are grouped and organized (Flaw 3), and do not provide information on the statistical confidence (reliability) of the results (Flaw 4). Genuinely spatial indicators (such as Moran's Global and Local Spatial Autocorrelation indexes) solve or minimize these problems. It is time to abandon the use of traditional no spatial indicators of residential segregation originated in sociology and advance in the utilization of genuinely spatial indicators developed in spatial analysis: Requiem for no spatial indicators of residential segregation.

Palabras llave : Residential segregation; spatial segregation indicators; key flaws of no spatial segregation indicators; global and local spatial autocorrelation.