The National Cancer Institute awarded a contract to Dr. Alan MacEachren and Robert Edsall of the GeoVISTA Center to develop and evaluate the dynamic parallel coordinate plot for spatiotemporal analysis of health statistics data. The dynamic parallel coordinate plot has been demonstrated to be a powerful tool for visual multivariate data analysis (Edsall 1999), but has not been applied to health statistics. The plots employ a novel methodology to visualize more than three dimensions by representing each observation not as a point in a scatter plot but as a series of unbroken line segments connecting parallel axes, each of which represents a different variable. By representing variables (X1, X2, etc. on the figure right) as parallel - as opposed to orthogonal - axes, the representation breaks the bonds of two or three-dimensional representations such as scatter plots (Inselberg 1985; Wegman 1990). This project is designed to explore the potential of dynamic parallel coordinate plots for depicting multivariate health data linked to maps depicting the geographic aspects of those data. Attention is also directed to the use of PCPs for exploring time series of these data.

The research plan specifies a review of relevant literature in visualizing multivariate geo-referenced health statistics, the development of an environment with a dynamic parallel coordinate plot linked to geographic and other statistical graphics (like scatter plots and histograms), and the execution of an evaluation of the usability of the environment for problem solving and data exploration.

We've made significant progress on this research: the dynamic PCP has been developed (figure 2) directly within a modified ArcView GIS environment (HealthVis), created in previous GeoVISTA research activities (funded by the National Center for Health Statistics) to facilitate interactive visual analysis of health statistics. The original and updated versions of HealthVis were created using ArcView's object-oriented scripting language, Avenue, with an eye toward possible future redevelopment in other object-oriented languages with more flexibility and extensibility (Java, Visual Basic). In June, a usability test was carried out with students and researchers at Penn State. Analysis of that data is ongoing. For more information and progress updates, contact Rob Edsall.

Download a copy of HealthVis PCP by clicking here (~9.22 MB). You must have ArcView GIS 3.x and an unzip utility to use this program. A user manual for installation and use of HealthVis PCP can be viewed here.

To read more about the use of this software, see the following publication:

Edsall, R.M., MacEachren, A.M. and Pickle, L.J. 2001. Case Study: Design and Assessment of an Enhanced Geographic Information System for Exploration of Multivariate Health Statistics. In: K. Andrews, S.Roth and P.C. Wong,Proceedings of the IEEE Symposium on Information Visualization 2001, San Diego, CA, October 22-25, 2001.

Related Information: