Dynamic graphics in a GIS: Exploring and analyzing multivariate spatial data using linked software

Interactive and dynamic graphical methods provide powerful tools to interface a data analyst with data. For multivariate data, interactive techniques such as linked brushing and identification allow an analyst to query the data, isolate or mask subsets, and examine dependencies between many variables. Dynamic techniques such as the grand tour give the analyst a sense of the overall shape (clusters or nonlinearities) of the data. This paper reports on research into interactive and dynamic graphical methods applied to spatial data made available through a link between two software packages: ArcView 2.1(TM)(a Geographic Information System), and XGobi (a dynamic graphics program for exploratory multivariate data analysis).