ASSESSING GROWTH IN CONCEPTUAL UNDERSTANDING AND THINKING IN INTRODUCTORY COURSES

Development of the Geoscience Concept Inventory (GCI) with questions calibrated through item response theory (IRT) offers geoscience instructors an unprecedented opportunity to modify and assess introductory courses to more effectively promote reflective judgment and evaluative thinking. At Idaho State University (ISU), I utilized the GCI items as pre-post- measures on both knowledge surveys and tests. As shown in the "Private Universe" project in the late 1980s (http://www.learner.org/resources/series28.html), erroneous preconceptions about the physical world can resist even a Harvard education. Conventional lectures do not carry the power needed to replace such preconceptions. In my course, students required engagement through graded active learning projects to displace erroneous geoscience preconceptions. IRT shows that it is difficult to design questions (quizzes/projects) of high reliability when these address open-ended challenges. Quantifying their reliability requires methods too cumbersome for routine classroom use. However, calibrated GCI items offer a way to reliably track students' progress in self assessment of conceptual understanding along with performance on open-ended tests. Students in the ISU class engaged an understanding of "concept," reflected on their own preconceptions, analyzed selected CGI items through considering the larger global concept (change-through-time, water-cycle, rock-cycle, etc.) each represented and addressed how these could become candidates in constructing a "private universe." Resulting gain in conceptual understanding was profound from a pre-class test average of 45% to a post-class average of 97%. Distributions between post-course knowledge survey ratings and tests were nearly identical. High-level thinking is a desirable outcome of every college education, and introductory courses can promote this early through focusing on the nature of concepts and preconceptions.