Refining our use of cooperative learning

girl browse internetIn cooperative learning, students must work together through discussion of task and content to solve problems. Advocates of cooperative learning are generally agreed that encouraging this kind of cooperation provides group members with knowledge and skills they might not acquire otherwise. However, Michael S. Meloth and Paul D. Deering, University of Colorado-Boulder, contend that too few studies have compared the ways in which different cooperative learning conditions affect the type of discussions students engage in and the kind of learning they acquire.

Previous research on cooperative learning indicates that although group discussion is important for learning, the amount of discussion, in and of itself, does not guarantee increased achievement. However, certain kinds of group discussion do appear to increase learning.

For example, students who explain to other students how to solve problems perform better than students who simply tell each other the correct answers. In addition, Meloth and Ddering report growing evidence which suggests that rewards and direct social-skills training may not be necessary if students are taught and encouraged to use specific learning strategies in their groups.

Learning strategies in groups

To determine how different learning strategies affect achievement in cooperative groups, Meloth and Deering studied task-related talk, reading comprehension, and metacognition (student awareness of evaluation, planning, regulation and conditional knowledge) within two different cooperative learning contexts.

Eight 3rd grade classes, consisting of 219 predominantly white, middle class children participated in the study. At the outset, all students were tested with a reading comprehension test which included five brief passages followed by eight questions covering four comprehension strategies: prediction, inference, main idea, and summarization. In addition, students’ conscious knowledge of comprehension strategies was measured by the Index of Reading Awareness which consists of 20 questions in four subtests: evaluation, planning, regulation and conditional knowledge.

Rewarding achievement

Teachers reported that the reading programs in all eight classrooms were whole-language or literature-based and that these students had considerable experience with cooperative learning in which they receive rewards for effort and appropriate social behavior, but rarely for academic achievement.

Students in all eight classes were divided into mixed-ability groups of four or five and randomly assigned to either the “reward” or “strategy” condition. There were no significant pre-test differences on the comprehension or metacognition tests between the groups.

Students in the reward groups were encouraged to cooperate with each other through the use of social rewards, such as team recognition. Strategy group materials, on the other hand, were designed to encourage direct discussion about substantive task content without rewards or special recognition.

All teachers were given six hours of training in designing cooperative reading activities that emphasized the four target comprehension strategies. They were also given training in constructing “think sheets”; worksheets intended to direct group work toward important lesson content. In addition, teachers received another three hours of training in either the reward or strategy condition.

Teachers assigned to the reward condition were trained to provide detailed introductions to the information to be learned during group activities. Teachers were also trained to model the use of comprehension strategies and to lead students in guided practice in order to help them engage in academically-oriented talk.

In addition, they learned how to construct short quizzes which were used for group accountability. Team recognition was given for improvement in individual member’s scores. Team scores were recorded on a poster and teams were designated as “super”, “great”, or “good” on the basis of their improvement on weekly quizzes.

Teachers assigned to the strategy condition were trained to provide students with detailed information about the types of metacognitive strategies important in reading comprehension. They were taught to plan lessons that emphasized what students would learn during an activity, why such information would help their comprehension, and when and how to use this information in a strategic manner.

Teacher training stressed the importance of modeling comprehension strategies. Think sheets designed by teachers required students to explain their thinking to teammates. Students might be asked, for example, to explain why they had organized a character map in a particular way. Formal testing was eliminated and groups in the strategy condition received only verbal praise and feedback for their efforts.

All cooperative groups met for 35-50 minutes at least four time a week for a month. Randomly selected groups in each condition were tape recorded each week to allow researchers to analyze group discussion.

Strategy vs. reward conditions

Class observations and teacher and student interviews confirmed that students in both conditions enjoyed their group lessons and exhibited a very high degree of on-task behavior and discussion. Teacher intervention was rare. Although more than 80% of the talk in all groups was task-related, students in the strategy condition were on-task significantly more than students in the reward condition. Students in the reward condition appeared to enjoy the game-like team format and to value being designated a “super” team.

There were significant differences in the content of discussion between the two groups. In the reward groups, students offered assistance and gave directions frequently to their teammates. Student responses were focused mainly on task organization, resources, and correct answers. By contrast, students in the strategy groups made more requests for information from their teammates. Also, a higher percentage of their responses focused on concepts, strategies, and facts.

Post-test scores demonstrated that both groups of students significantly improved their comprehension. However, students in the strategy condition scored significantly higher on the comprehension subtests relating to prediction, inference and main ideas. No significant differences were found for summarization. Students in the strategy groups also scored higher than reward students on the metacognition test.

Encouraging metacognition

The purpose of this study was to examine the effects of two different cooperative conditions on group discussion and learning. The lessons and think sheets of these two cooperative models were specifically designed to create divergent learning contexts. Considering their age, it was remarkable that with little teacher intervention, groups in both conditions remained on task. Their familiarity with cooperative learning is assumed to have contributed to the high level of task-related talk.

Nevertheless, these researchers conclude that structured group activities do not guarantee that important content material will be discussed. Increases in talk devoted to facts, concepts, and strategies occurred only in the strategy groups when students were specifically trained and given worksheets which required them to talk about such things. These researchers concluded that this requirement results in improved reading comprehension scores.

It should be noted that this study is limited by its homogeneous sample of students and by its brief duration. Also, studying student verbalizations can provide only an indication of cognitive activity.

Nevertheless, these test results provide some evidence that focusing task talk on metacognitive strategies can have an important effect on the ability of students to use target strategies to improve reading comprehension. Providing elaborations and explanations of their work are thought to be beneficial because, in doing so, students are provided an opportunity to rehearse, organize, and clarify their thinking, receive feedback about the adequacy of their information, and become aware of gaps in their knowledge.

Both group conditions required students to discuss content, but students in the strategy condition were given directives and were asked questions about strategic knowledge – the why, what and how of what they were learning. This study suggests that metacognition is not an automatic by-product of collaboration and that more direct methods of encouraging specific metacognitive learning need to be designed into cooperative activities.


“Effects of Two Cooperative Conditions on Peer-Group Discussions, Reading Comprehension, and Metacognition” Contemporary Educational Psychology April 1992 Volume 17, Number 2, pp. 175-193.

Published in ERN September/October 1992 Volume 5 Number 4

Leave a Reply

  • (will not be published)