Cooperative learning has demonstrated that children who help classmates understand problems learn more than children who simply tell classmates answers. To be remembered, new information must be related in meaningful ways to previous knowledge. One strategy for doing this is to have students explain material to others. The recipients of explanations learn more effectively if they have an immediate opportunity to apply this information to new material. All children learn more when they discuss differing viewpoints and seek to explain and justify their positions and to question others’ beliefs.
Research shows that students do not develop these effective interaction styles naturally, simply by participating in group work. Within a correctly structured group, however, children can learn these skills. In the current study, researchers Lynn S. Fuchs, Douglas Fuchs, Sarah Kazdan, Kathy Karns, Mary Beth Calhoon, Carol L. Hamlett, and Sally Hewlett, Peabody College of Vanderbilt University, examined the differing effects of group size and structure on student learning of complex mathematical tasks. Previous research has revealed inequalities in participation and learning among members of cooperative groups: Low-ability students tend to interact less frequently and are often relegated to non-thinking tasks. Since the amount and type of interaction predicts the amount of learning, it is important to equalize interaction in groups. This can be accomplished by providing a structured work environment for cooperative groups.
Fuchs et al. randomly assigned 36 third- and fourth-grade classrooms to three different study environments — individual work, unstructured group work, and structured group work where students were given goals, a role to perform, resources, and rewards for working together to solve complex tasks. All grouped classes were randomly assigned to either small groups of four students or pairs. These heterogeneous groups were based on teachers’ ranking of their students’ math skills. High- and low-achieving students were put together in pairs and groups.
Teachers in all 36 classrooms used the district’s basal program, Mathematics Plus, which emphasized solving problems related to everyday life as the primary focus of math instruction; reasoning about mathematics rather than memorizing rules and procedures to help children make sense of mathematics; mathematics as a way of thinking; manipulatives as a powerful tool to increase understanding; and computational proficiency as a necessary tool for successful problem solving. Regardless of their assigned structure, all classrooms received training in the performance assessments which were used in the study. In addition, unstructured groups received a 20-minute training session on cooperation, while structured groups received an hour’s training on the rules of participation, which required students to assume four roles:reader, monitor, checker, and writer.
Results and Conclusions
After four weekly performance assessments of complex math tasks, students in these different conditions were videotaped. No constraints were placed on groups for this session. While students were not prompted to use any particular style or method of interaction, they tended to follow the structure they had been practicing. Surprisingly, researchers found that the quality of interactions of the structured groups did not exceed those of students in unstructured groups. In fact, there was a modest positive effect for the unstructured groups. (Editor’s note: The relatively small difference in training given structured and unstructured groups may have contributed to the lack of difference in the outcomes.)
However, students with collaborative experience did better on the performance assessments than those who had worked individually. Contrary to previous research on routine tasks, explicit structured interaction may be unnecessary on more challenging, controversial tasks like the math problems in this study. Such complex tasks may provide a natural source of cognitive conflict that elicits discussion and motivates students to participate and cooperate. These findings are preliminary and additional research is warranted.
Pairs versus small groups
In contrast to this lack of difference for group structure, there was a more convincing pattern of findings for group size. Results favored pairs over small groups in several ways. There was significantly more procedural and conceptual talk in pairs, as well as more helpfulness and cooperation. The quality of the performance assessment was also higher in pairs than in groups. Low-achieving students benefited more from working in pairs, and their level of participation was higher. However, small groups generated more cognitive conflict, and this was beneficial for middle- and high-achieving students.
There are some limitations to this study. Results might be different if pairs were more homogeneous, consisting of either two high-achieving or a high- and middle-achieving student. The findings of this study apply only to heterogeneous pairs. In addition, students were videotaped outside the regular classroom environment, which may have influenced the results. More importantly, these researchers did not investigate the effects of group composition over the long term. They consider these findings important and practical, but tentative.
Overall, these results indicate that when tasks are challenging and complex, cooperative work may not have to be highly structured to be beneficial. Students in this study naturally engaged in productive interactions on cognitively demanding tasks. Findings also suggest that teachers should be prepared to vary groupings to benefit different kinds of learners. High achievers, when working on
complex material, should have opportunities to work with fellow high achievers. Low-achievers, however, learn better when paired with higher-achieving students.
“Effects of Structure and Size on Student Productivity during Collaborative Work on Complex Tasks” The Elementary School Journal Volume 100, Number 3, January 2000 pp. 183-212.
Published in ERN March 2000 Volume 13 Number 3