Visualizing Multivariate Spatial Correlation with Dynamically Linked Windows1

Several recent efforts have focused on adding exploratory data analysis functionality to geographic information systems (GIS) by integrating established statistical software with a GIS. In this paper, we outline an alternative approach, where the functionality is built from scratch, using a combination of small libraries of dedicated functions, rather than relying on the full scope of existing software suites. The suggested approach is modular and freestanding. Within an overall framework of dynamically linked windows, it combines a cartographic representation of data on a map with traditional statistical graphics, such as histograms, box plots, and scatterplots. It extends earlier work on the visualization of spatial autocorrelation to a multivariate setting, introducing a Moran Scatterplot Matrix and Multivariate LISA Maps. The new program (DynESDA2) works on both point and polygon coverages, implements true brushing of maps, as well as the usual linking and brushing between maps and statistical graphs.