New Evidence on Cooperative Learning

The body of evidence documenting the effectiveness of cooperative learning is already impressive. The large and regularly cited meta-analysis of Johnson and Johnson published in 1987 reviews 378 studies that explore the use of cooperative learning groups in a wide range of settings. More than half of the studies reviewed favored cooperation in groups compared with only 10 percent favoring individual effort.

The new research highlighted here looked at a small but important variable not controlled in much of the previous work: time on task. Cooperative learning occurs in groups that typically meet in class or in controlled settings outside of class. But what if the members of those learning groups decide to study on their own without the aid of their group members? If they and others in their cooperative learning group do better than those exclusively studying alone, then it’s possible that individual study plus the group work is what’s making the difference. As the researcher describes it, “the present study argues that the learning method [in this case cooperative learning] and the time on task should be carefully monitored both in the classroom and during the students’ out-of-class studies to identify the true academic benefits of cooperative learning. Accordingly, the students participating in this study were randomly assigned to a cooperative learning condition or an individualistic learning condition and were compelled to remain in their assigned conditions for both the regular classroom sessions and a series of homework sessions scheduled on a weekly basis.” (pp. 100-101)

The procedures for ensuring that students in the cooperative learning groups did not study alone and studied about the same amount of time as those studying individually were elaborate and are described in detail in the article. The three-person cooperative learning groups were assembled using average test scores from the previous semester, and each group contained students with a range of test scores. Their learning of the content in this engineering course was assessed via seven homework tests and four unit exams.

The results “showed that given a sufficient period of time for the cooperative learning teams to mature, the students in the cooperative learning condition performed substantially better in both the homework and unit tests than those in the individualistic learning condition.” (p. 119) In other words, on the first homework and unit exams there were no significant differences between students studying individually and those studying in the cooperative learning groups. But gradually, across the course, the performance of students studying in groups improved to the point that at the end of the course they were outperforming those studying individually.

These findings not only verify the already established effectiveness of cooperative learning groups, they offer some interesting implications that are related to the characteristics of cooperative learning. This form of group work is based on several principles, including positive interdependence—students work together, therefore helping each other—and individual accountability. There are no group exams or group grades. Students do their own work and they perform individually, not as a group. But the group is committed to the success of each individual. They work together as a team.

The interesting question is why the students in the cooperative learning groups improved slowly and didn’t outperform those studying individually from the outset. The researcher hypothesizes that the students in the groups simply needed to learn how to work together and see the benefits of their collaboration. The implication: The benefits of group collaboration on problem-solving (in the case of this study) may not be realized if groups meet only once or twice or if membership in groups regularly changes. Groups need time to gel. Perhaps the process of getting groups to work together well can be expedited with certain activities. The researcher identifies this as an area for future research.

This is a solid piece of research with any number of difficult-to-control variables far more controlled than they are in most studies. It’s another convincing piece of evidence supportive of the power of students learning from and with each other.

Reference: Hsiung, C. (2012). The effectiveness of cooperative learning. Journal of Engineering Education, 101 (1), 119-137.

Reprinted from The Teaching Professor, 26.5 (2012): 5.