Software Integration for Multivariate Exploratory Spatial Data Analysis

This paper describes a decade’s worth of evolution of integrating software to support exploratory spatial data analysis (ESDA) where there are multiple measured attributes. The multivariate graphics tools we use are XGobi, and more recently, GGobi. The paper is divided into two parts. In the first part, we review early experiments in software linking for ESDA, which used XGobi, different Geographic Information Systems (GIS), and the statistical analysis packages S and XploRe. We discuss applications, capabilities and deficits of these existing links. In the second part we describe GGobi, the descendant of XGobi, which was designed with software integration as a fundamental goal. GGobi lends itself to broader possibilities for linking with other software for richer ESDA.