ANALYZING STUDENT LEARNING USING THE GEOSCIENCE CONCEPT INVENTORY (GCI): MISCONCEPTIONS, DODGY QUESTIONS, AND THE POTENTIAL FOR CHANGE

A 15-question Geoscience Concept Inventory (GCI; Libarkin and Anderson, 2005) was assigned as pre- and post-tests for students in 62 Physical Geology lab sections taught during the Fall 2009 and Spring 2010. We used fifteen questions - 1, 2, 6, 7, 13, 20, 28, 37, 51, 61, 65, 71, 73 - arranged from difficult to easy; seven of the fifteen contained some type of diagram. A total of 781 students completed both pre- and post-GCIs and gave permission to for their data to be analyzed (74% of all students taking the pre-test). Lab instructors did not discuss the GCI questions with the students between the administration of the pre- and post-tests in an effort to ensure that the instrument represented an unadulterated gauge of student learning. Here we consider the aggregate responses to individual questions between the pre- and post-tests over both semesters.

Data analysis revealed that responses to two of the most challenging questions (2, 6) showed little change or a slight decrease in the number of correct responses. We have concerns about the reliability of these questions as a large proportion (>75%) of students who initially answered either question correctly, selected an incorrect response on the post-test. Several other questions showed significant improvement with normalized gains of more than 20% (28, 35, 51, 61, 65, 71). In these cases, a significant proportion (>75%) of students who identified the correct response on the pretest were also able to identify the correct response on the post-test.

The GCI helped us identify the labs that successfully addressed students’ misunderstandings about key geoscience concepts and the labs that may need to be modified to address more robust misconceptions that resisted change. Student perceptions about the character of the early Earth (7), the distribution of active volcanoes (13), Earth structure (20), and the rates of plate motions (37) showed only small positive changes (normalized gains of 2-9%). These topics were addressed, some more directly than others, as part of the lab experience but will need to be given greater attention in future labs. In contrast, there were considerable gains in correct responses to questions addressing change in life on Earth (28), the location of tectonic plates (35), plate tectonic processes (51, 65, 71, 73), and the definition of density (61).