Virtual globe programs such as Google Earth replicate real-world experiential learning of spatial and geographic concepts by allowing students to navigate across our planet without ever leaving campus. However, empirical evidence for the learning value of these technological tools and the experience students gain by exploration assignments framed within them remains to be quantified and compared by student demographics. This study examines the impact of a Google Earth-based exploration assignment on conceptual understanding in introductory geoscience courses at a research university in the US Midwest using predominantly traditional college-age students from a range of majors. Using repeated-measures ANOVA and paired-samples t tests, we test the significance of the activity using pretest and posttest scores on a subset of items from the Geoscience Concept Inventory, and the interactive effects of student gender and ethnicity on student score improvement. Analyses show that learning from the Google Earth exploration activity is highly significant overall and for all but one of the concept inventory items. Furthermore, we find no significant interactive effects of class format, student gender, or student ethnicity on the magnitude of the score increases. These results provide strong support for the use of experiential learning in virtual globe environments for students in introductory geoscience and perhaps other disciplines for which direct observation of our planet's surface is conceptually relevant.