New Research Topics

I am interested in exploring the effect of a variety of instructional interventions on learning in a range of class settings including large lectures and smaller labs. Instructional interventions account for a range of strategies that may be employed to improve learning. This may include the use of:

• Cooperative Exams and Multiple Testing Opportunities - students would complete and submit an individual exam before working with peers on a cooperative exam that would include some of the same and/or similar questions. These exams have been used sparingly in higher education but emerging research suggests that, when used appropriately, cooperative exams can improve student conceptual learning. Ongoing reserach in introductory classes is examining the impact of this exam format. One semester of data reveals that all students improve their scores as a consequence of these exams, not just the students with the lowest initial grades. Likewise, research suggests that more frequent testing promotes the long-term retention of information better than less frequent testing. Further, the sooner the testing opportunity occurs subsequent to the learning experience, the more information is retained. How can we create learning environments that use this strategy without turning the classroom into a constant high-stress testing facility. I am exploring the creation of online, self-paced assessments that allow students to take advantage of unlimited testing opporunities to promote mastery of basic concepts. These out-of-class learning modules will free more time for in-class discussion of more challenging concepts.

• Physical and Visual Models - We are interested in comparing how well students learn new concepts with or without the use of physical models. These models include simple paper models of faults, brick-and-pulley models of earthquake behavior, or dry-erase cubes. Is there a difference in learning that occurs when building simple models versus simply manipulating models? Do students learn more from the use of three-dimensional (e.g., models) vs. two-dimensional representations of features (e.g., topographic maps)? Do students learn more if we spend more time teaching concepts using simple, accessible bench-top models or is it worth investing that time in training students to use more sophisticated computer animations that they can maniplate at their own pace and time outside of class?

• Novice-Expert Continuum - For several years we have been administering the Geoscience Concept Inventory (GCI) as a pre/post test in introductory geology or earth science classes. Many classes that have employed this instrument have shown no learning gains over the course of a semester (see Libarkin and Anderson, 2005; see also Figure 1). We anticipate using this instrument more widely as an assessment of pre-existing knowledge at the beginning of the semester in a range of undergraduate and graduate classes. By asking freshmen, sophomores, juniors, and seniors, as well as graduate students, the same suite of questions, we will conduct a quasi-longitudinal study of student learning of key geoscience concepts. This will be matched with semi-guided interviews of selected students to investigate the evolution of understanding about specific geologic concepts such as landscape development, geologic time, or plate tectonics.

Figure 1. Left: Normalized gain scores (post-pre/1-pre) for 40 classes using the GCI. Data from unpublished Libarkin and Anderson data, published articles, and authors classes. Right: Normalized gain scores for classes with paired data (same students taking pre/post tests) and showing statistically significant improvement (p<0.01) in scores over the course of the semester. Note trendline shows decreasing gain with increasing class size. Nearly half of the original classes did not show a significant improvement in scores between the early and late administrations of the GCI.

Ongoing Research

If you are interested in working on any aspects of this research, contact me at david_mcconnell@ncsu.edu.