the topic in groups, 75 percent of students answered the question correctly on a retest. When instructors used rudimentary models rather than standard lecture to introduce plate tectonics, 56 percent of students answered the question correctly the first time, and 84 percent answered correctly after discussion in groups.
McConnell also shared results from a study he and others conducted in his classes about the use of models to explain the seasons (McConnell et al., 2005). Students in two control classes learned about the seasons through standard lecture with some demonstration. Students in six experimental classes received rudimentary models—a foam ball on a skewer with a small flashlight—and instructions about how to model different scenarios related to the seasons. Students in the experimental classes had favorable views about using the models and showed greater gains in their conceptual understanding of the seasons than students in the control classes. In addition, students in the experimental classes made greater gains in their logical thinking skills as measured by the GALT ( McConnell et al., 2005).
Another way to address the challenges that large introductory classes can pose to academic success is to engage students in research. Research experiences allow students to work directly with, and learn from, individual science faculty. Noting that the best way to learn science is by doing science, committee member David Mogk introduced speakers to discuss two programs that provide research experiences for undergraduate students.
Sandra Gregerman (University of Michigan) discussed the Undergraduate Research Opportunity Program (UROP), which was launched in 1988 to increase the retention and academic success of underrepresented minority students at the University of Michigan. In this year-long program, first- and second-year students spend 6-12 hours per week conducting research on ongoing faculty projects in the sciences and other disciplines. The program contains academic and social support components, including peer advising, skill-building workshops, and research peer groups in which students discuss a variety of research-related issues. Each year, the program culminates in a symposium; in 2008, 750 students presented their research in poster form and 20 students delivered oral presentations on their research (Gregerman, 2008).
Gregerman and her colleagues have conducted many studies of the program over the years. Results of one longitudinal study with an experi-