COLLABORATIVE GROUP ASSESSMENTS IN THE SCIENCE CLASSROOM USING THE 1/3 MODEL

This Study consisted of both individual and collaborative group testing in a secondary educational science setting. Data were gathered over three years from multiple grade levels and >500 students in order to test the hypothesis that group evaluations generally improve scores for students of all achievement levels (e.g. high-performing, average performing, and low-performing students). These classifications were based on criteria such as reading and math achievement levels relative to peers and overall performance within the science class setting, including students’ current grade. Students were grouped into three using the 1/3 model of combining a top, average, and low performing student within each group. Learning outcomes were measured for students after individual and group testing. Initially, students were tested individually, and then combined into groups to take the identical test with the ability to discuss questions and possible answers. Student test performance, for all achievement levels, was generally enhanced through collaborative testing (P > 0.05). However, lower performing students tended to benefit more significantly from collaboration.