Providing scientific visualization for spatial data analysis: Criteria and an assessment of SAGE

A consistent theme in recent work on developing exploratory spatial data analysis (ESDA) has been the importance attached to visualization techniques, particularly following the pioneering development of packages such as SPIDER and REGARD (Haslett et al 1990). The focus on visual techniques is often justified in two ways: (a) the power of modern graphical interfaces means that graphics is no longer a way of simply presenting results in the form of maps or graphs, but a tool for the extraction of information from data; (b) graphical, exploratory methods are felt to be more intuitive for non-specialists to use than methods of numerical spatial statistics enabling wider participation in the process of getting data insights. This paper briefly reviews a theoretical framework that has been suggested for developing visualization tools for ESDA that comprises two elements: (a) a data model, based on the distinction between rough and smooth properties of spatial data, that defines what an analyst is looking for in data (Haining et al 1998) and (b) a theoretical model for assessing the quality of visualisation tools (Cleveland 1994). The emphasis of this paper is the use of the theoretical framework to structure an assessment of SAGE, a software system that has been written for the spatial statistical analysis (including both exploratory and confirmatory data analysis) of area based data linked to a GIS. The aim of the assessment is to identify and illustrate what appear to be desirable features of the system (that might be employed in other systems) whilst also highlighting what the additional requirements are thereby contributing to the development of systems for ESDA that contain good quality scientific visualization tools for exploratory spatial data analysis.