Stars, Polygons, and Clusters,

One technique for displaying a set of quantitative variables is to represent the values graphically in the form of a polygon or star, allowing the observer to visually handle complex information quickly, as a whole. Such displays have been used in various areas, for data exploration and presentation, and in status displays they are also suited to categorization and identification tasks. For polygon displays to be reliably used, they should be capable of being interpreted consistently. An experimental investigation was undertaken to ascertain the effect of variation in certain visual features of the display on the consistency with which untrained subjects categorize information presented as polygons. Variables included background information, shading of figure, and form. Subjects performed a categorization task on several sets of data the results were analyzed on the basis of consistency between individuals and consistency with certain standard clustering algorithms. Implications for display design include the effects of distinctive patterns on similarity judgements, of the nature of the data represented, and of the interaction of variables. Author