ABSTRACT
The introduction situates spatial data science relative to generic data science by stressing the unique spatial perspective. The focus is on exploring spatial dependence and spatial heterogeneity, two characteristics that distinguish spatial analysis from the standard statistical paradigm. Specifically, the methods presented provide insight into the location of interesting observations and 2structural breaks in the spatial distribution. This is achieved by means of the GeoDa software.
An overview is presented of the organization of the volume into five main parts and an epilogue. It offers a progression from basic data manipulation through description and exploration, to the identification of clusters and outliers by means of local spatial autocorrelation analysis.
A brief overview of the GeoDa software is presented, organized around subsets of the main toolbar menu: data entry, data manipulation, GIS operations, spatial weights, mapping, exploratory data analysis, space-time analysis, spatial autocorrelation analysis, and cluster analysis.
The chapter closes with a brief discussion of sample data sets used to illustrate the various methods. These data deal with carjackings and socio-economic determinants of health (Chicago) disease and socio-economic profiles for municipalities in Brazil, poverty and food insecurity indicators for municipalities in Mexico, and bank performance measures for community banks in Italy.
