Multivariate Data Visualization and Usability: Preliminary Notes

Abstract

Effective and understandable communication of geographic information is currently under discussion by the scientific community in general and cartographers in particular. Multivariate mapping techniques used for representation of three or more variables are even more challenging for individuals with a limited cartographic background. Several types of preferred visualization techniques exist for multivariate spatial patterns. However, only preliminary discussion with users has been held regarding their ability to use such representations. In this empirical study we compared the effectiveness and efficiency of two multivariate mapping methods—ring maps and small multiples—for both spatial and temporal data. An empirical test was conducted which included three basic groups of tasks. The first task focused on the user’s ability to read the particular representations and to compare the legend and map data. The second task dealt with the ability to interpret the spatial characteristics (attributes) under pre-defined criteria. The final task concentrated on temporal characteristics and the user’s ability to interpret spatio-temporal changes within the depicted information.

Preliminary results based on a limited group of users show that the ring map depiction achieved faster (more efficient) responses for both the map reading and spatial information inference. Overall accuracy of answers (effectiveness) was closely connected with particular tasks.