North-Central Section - 43rd Annual Meeting (2-3 April 2009)

Paper No. 7
Presentation Time: 10:00 AM

HOLLYWOOD VOLCANOES CAUSING CONCEPTUAL DAMAGE: THE InVEST VOLCANIC CONCEPT SURVEY AND STUDENT SOURCES OF KNOWLEDGE


PARHAM Jr, Thomas L.1, CERVATO, Cinzia2, GALLUS Jr, William A.1 and LARSEN, Michael D.3, (1)Dept. of Geological & Atmospheric Sciences, Iowa State University, 253 Science I, Ames, IA 50011, (2)Dept. of Geological and Atmospheric Sciences, Iowa State University, 253 Science I, Ames, IA 50011, (3)Dept. of Statistics, Iowa State University, Snedecor Hall, Ames, IA 50011, tparham@gmail.com

The InVEST (Interactive Virtual Earth Science Teaching) Volcanic Concept Survey was recently administered to over 400 students at six institutions with the goal of identifying areas of greatest misconception relating to volcanoes. Both mean and median scores on the instrument were exceptionally low, indicating that students generally possessed minimal understanding of volcanic systems. Students held many misconceptions about eruptive controls and hazards, did not perceive a global pattern of volcanic activity, and often claimed that volcanoes were “always near water” or “in hot places.”

Data collected via an accompanying demographic questionnaire suggests that students' source of knowledge (SoK) may be a key factor in the generation and proliferation of these misconceptions. Over 41% of students surveyed reported that they had acquired most of their knowledge about volcanoes from non-traditional and non-peer-reviewed sources, such as the popular media and Hollywood films. Recent research has found that even limited viewing of popular media and science-fiction films can be a significant impediment to the conceptualization of Earth science processes and the development of a scientifically literate citizenry. Statistical analysis of the InVEST survey data quantitatively supports this hypothesis.

Application of a multiple linear regression model found that SoK accounted for 7% of total score variability and that non-traditional sources were not significant predictors of high score at the p < .01 level. An expanded model controlling for key demographic factors accounted for 35% of the variability among all scores, but still found non-traditional sources to be insignificant predictors of high score even at the p < 0.05 level. In contrast, traditional SoK's (in-class learning, learning from textbooks) were highly significant predictors of high score in both models.