ASSESSMENT OF THE USE OF CONCEPTESTS IN PROMOTING PEER INSTRUCTION DURING LECTURE

Conceptests were incorporated into two traditional introductory geology classes in an attempt to promote peer instruction during lecture. Conceptests are comprised of questions specifically designed to assess a student's understanding of the principal concepts underlying lecture material. They are multiple-choice questions that feature one correct answer and reasonable distractor answers that typically reflect student misconceptions. One or two conceptests were given during lecture and anonymous responses were collected from the students using the PRS system. Approximately 30 questions were used during the course of the semester. Most questions used during the semester generated between 40 and 70 percent correct responses the first time the class was polled. In nearly every case the number of students answering the question correctly increased with the second polling of the class. This suggests that peer instruction resulted in an increased understanding of geologic principles.

We chose the GALT (Group Assessment of Logical Thinking) and the Libarkin tests as formal assessment tools to determine the effectiveness of conceptests in promoting peer instruction. The GALT test measures the students' logical thinking skills and the Libarkin test is a geoscience specific test. The initial results of the GALT and Libarkin test indicated that incoming students had varying degrees of cognitive skills and basic knowledge of the geosciences. The GALT test classifies student as concrete thinkers, transitional thinkers and formal thinkers. The initial GALT distribution of the study group was 17 percent concrete, 26 percent transitional and 57 percent formal thinkers. At the end of the semester the concrete group dropped to 12 percent, the transitional group dropped to 24 percent and the formal thinker group grew to 64 percent. In addition approximately 50 percent of the study group increased their raw score on the GALT exam. The average GALT score improved from 7.4 (beginning of semester) to 8.0 (end of semester). The average Libarkin raw score improved from 9.5 (beginning) to 11.3 (end of semester). We believe that peer instruction during lecture contributed to the observed increase in the average scores.