Equity issues in collaborative group assessment: Group composition and performance

This study investigated the effects of group ability composition on group p rocesses and outcomes in science performance assessments. Students in 21 ei ghth-grade science classes worked on science assessments first individually , then in groups, and finally individually again. Group composition had a m ajor impact on group discussion quality and on student achievement. Groups with above-average students produced more accurate and high-quality answers and explanations about how to solve the test problems than groups without above-average students. As a result, below-average students who worked with above-average students showed higher achievement than did below-average st udents who worked without above-average students. High-ability students gen erally performed better when they worked in homogeneous groups than when th ey worked in heterogeneous groups. The fact that heterogeneous groups provi de a greater benefit for below-average students than they impose a detrimen t on high-ability students is discussed.