In Astronomy 101 at the University of New Mexico, we carried out a repeated—items experiment on quizzes and tests to investigate the impact of cooperative testing. This trial was the only change in a reformed course format that had been refined over previous semesters. Our research questions were:
• Did cooperative quizzes result in gains for the class overall?
• Did these gains “stick” within the semester?
In the spring and fall semesters of 2000, students took quizzes individually and in cooperative learning teams, and tests individually. Normalized gain, 〈g〉, on the quizzes averaged about 0.4, and effect size about 0.8 (approximately a 10% increase in class mean score). Repeating selected quiz items on a subsequent test demonstrated that the gain was sustained over a month in both semesters. In addition, we compared demographics of UNM students with those of the National Astronomy Diagnostic Test project. We found that UNM students are similar to the national sample, except in ethnicity (more Hispanic American, fewer White). Based on these results, we judge that our cooperative quiz strategy will likely succeed in other “Astro 101” classes.