Evaluating the Effectiveness of Thematic Mapping on Virtual Globes

Abstract

Virtual 3D global representations of the earth (such as Google Earth) have over the past seven years become nearly as ubiquitous as traditional 2D ("flat") maps. Because of their novelty and popularity, it is necessary for cartographers to evaluate the potential of virtual globes as a thematic mapping medium. Through a series of thematic map comparison tests and surveys, I evaluate participant performance using two map media (2D flat maps and 3D virtual globes) and two quantitative symbolization methods (choropleth and prisms), and participant preference for each variable combination. I test hypotheses regarding the effects dimensionality, interactivity, and dynamism have on participant map reading, value identification and estimation accuracy, task completion times, and preference. The results indicate that symbolization dimensionality, not map medium dimensionality, is responsible for significant differences among participant accuracy rates and completion times; in particular, symbol dimensionality increases result in significantly lower accuracy rates and significantly higher completion times. However, the results also show that participants perform best using the 2D flat map and 2D choropleth map combination and worst using the 3D virtual globes and 3D prism map combination.