by Alfie Kohn
Goodlad ticks off what students are asked to do: passively listen to teachers’ ceaseless talking (in his survey, the average teacher “outtalked the entire class of students by a ratio of about three to one”), submit to close and constant monitoring, work separately and silently on textbooks and worksheets, and so on.
How would I react as an adult to these ways of the classroom? I would become restless. I would groan audibly over still another seatwork assignment. My mind would wander off soon after the beginning of a lecture. It would be necessary for me to put my mind in some kind of “hold” position. This is what students do.27
More to the point, students come to find boring28 and detestable not only the place called school but the subjects taught there and activities such as reading and solving problems. This is why CL, by virtue of making learning (and what is being learned) more enjoyable, can also make learning more successful.*
8. INTELLECTUAL INTERACTION: Most important, CL succeeds because “none of us is as smart as all of us.”29 A well-functioning group—and, of course, not all groups function well—can be more successful, particularly on open-ended, challenging tasks, than any member of the group could be on his own. Ten-year-old Jason, quoted at the beginning of this chapter, captured the phenomenon as succinctly as any researcher has. The whole of the group truly becomes more productive than the sum of its parts.
Cognitive and social psychologists have tried to figure out just what it is about groups that accounts for the heightened quality of performance.30 At the most basic level, everyone benefits from the sharing of talents and skills and resources, a process that is either discouraged or prohibited outright in noncooperative classrooms. When the different members of a group work separately to collect information and think through a problem, then reconvene to exchange the results, each individual has access to everyone else’s labors.
When one student catches on to something quickly and helps her teammate to understand, both tutor and tutee benefit. This, as Noreen Webb discovered in a series of studies, happens consistently, provided that the first student does not just announce the solution but explains how she got it and justifies her conviction that it is correct.31 As a group of British researchers put it, “Tutors may gain a deeper understanding of the material learned by virtue of having to teach it, and ‘learning how to learn’ strategies may spill over into learning contexts other than the immediate learning task.”32 Researchers also have found that simply telling a student that he will be asked to teach someone else what he is about to read (as opposed to reading it in order to perform well on a test) leads to higher interest in the material and greater conceptual understanding of the subject matter.33
Beyond explicit tutoring, when people work in groups there is a tendency for one person’s idea to evoke another idea from someone else. The second idea emerges as a reaction to the first and might not have occurred at all if its creator had been working on her own. In one study, fifth- and sixth-graders were given batteries, bulbs, and wire and asked to make a variety of different circuits so that the bulbs would light up. The students working in groups found the task more engaging, seemed to get less frustrated, and, by building on one another’s ideas, created both more circuits and more unusual kinds of circuits than their classmates who worked individually.34
This phenomenon applies not only to ideas about the topic being studied but also to ideas about how to examine the topic (meta-level or second-order ideas). Group work leads to “more frequent discovery and development of . . . higher quality cognitive reasoning strategies.” First-graders, for example, were more likely to figure out the abstract categories to which words belonged (in order to memorize these words more effectively) when they worked together.35
Finally, CL not only permits conflict, as I mentioned above, but actually relies on it to some extent for its success. The process of finding that someone else thought a story’s character had a motivation very different from the one you had inferred, or that another person assumed that dinosaurs became extinct for a reason that never dawned on you, nudges you to think through the problem in a new way, to take account of this brand-new perspective and try to reconcile it with your own.36
Notice that none of this analysis supports the idea that children should—or, with CL, do—become interchangeable members of a collective, relinquishing their selves to some amorphous blob of a group. It is competition that creates conformity ([>]); cooperation thrives on the diversity of its participants and the distinct contributions made by each. I do think teachers ought to make time for students to pursue independent work, too, but the more compelling point is that cooperative interaction may simply be the most powerful way to help each child find his own voice, make his own discoveries, devise his own connections to ideas and texts. “Talking is not merely a way of conveying existing ideas to others; it is also a way by which we explore ideas, clarify them and make them our own.”37
All of this impressive empirical evidence supporting cooperative learning—and, for that matter, the reliance on empirical evidence in general—needs to be placed firmly in perspective. Some researchers construct an imaginary brick wall, with those who respect the findings of science on one side and those who are in thrall to ideology on the other. But this dichotomy, which has its roots in a philosophy called positivism, confuses science with truth itself. It errs by overlooking the fact that all research, like all educational practice, is saturated in values, like it or not. But more than this, it suggests that there is something disgraceful about having and defending values; this is why many scientists label them “biases” or “ideologies” and treat them like last week’s seafood dinner.
Having just finished reviewing the data that bear on the use of cooperative learning, I am obviously not arguing that it doesn’t matter what the research turns up. It does matter, and it is with considerable relief that I can report that CL makes good sense when judged according to pragmatic and widely shared criteria. But the fact that I am relieved suggests that I am already drawn to CL because of other, deeper values that I hold. Frankly, it took me a few years to acknowledge this, to confess that I am not indifferent to how the evidence turns out (even though I hope I am able to assess it fairly).
Cooperative learning is worth defending even apart from its quantifiable consequences. Ideally, as Canadian educator Judy Clarke reminds us, the idea of interdependence is grounded in
the belief in every person’s worth. This belief frames people in relationships of positive interconnectedness and therefore reflects an ethical orientation to life. . . . Without this moral foundation, co-operative learning may simply appear to teachers as a set of techniques . . . to “master” and file in one’s repertoire.38
Without this moral foundation, CL likewise may appear to researchers as simply one arrangement among many to be evaluated dispassionately, much as one might test different thicknesses of glass for the classroom windows.
I believe there is something intrinsically preferable about having children work with each other as opposed to against or apart from each other. I am glad that I can reassure parents and teachers that it is a sensible arrangement from any number of perspectives, but I stand with Robert Bellah and his colleagues when they remark that
learning is never the result of the efforts of isolated, competitive individuals alone. . . . The evident weakness in American schools has much to do with the weakening of their community context. . . . Education can never merely be for the sake of individual self-enhancement. It pulls us into the common world or it fails altogether.39
Arguably, it has failed altogether. But while there is no shortage of critics willing to charge American education with failure, they often miss the point about what has gone wrong and why. The problem is not so much that students cannot find Turkey on a map but that they do not find themselves part of a community of learners; they do not find themselves, moreover, in a place where each person’s worth is affirmed. The fact that cooperative learning can alter these sad reali
ties is worth at least as much as anything the studies might tell us about achievement gains and the like.
THE PRACTICE OF COOPERATION
Cooperative learning takes many forms. It is present in a classroom where students edit each other’s essays or help each other to develop and practice vocabulary or multiplication skills. It may mean finding an ad hoc partner with whom to have a quick exchange of ideas about evolution before resuming a whole-class discussion. Or it may mean that each student belongs to a permanent team of four that meets daily to make sense of current events.
A cooperative lesson plan may come from the teacher’s notebook or it may be devised by the class. Perhaps the students find that their parents’ talk about politics has made them curious to learn more about political systems in other countries, which they proceed to do with their teammates. Or maybe a student’s announcement that her garage is now home to a litter of tiny, mewling kittens leads to an impromptu lesson on the subject, with children meeting in groups to read stories about cats, describe how a house looks from a cat’s point of view, research specific scientific questions that they want answered (How are cats both similar to and different from lions? When do cats fight with each other? Do cats dream?) and then present their findings to the rest of the class.
All these examples share the premise that learning is an active and an interactive process. To that extent, it is decidedly not a silent process. This is one very practical reason that it is difficult for individual teachers to decide on their own to switch to CL; if the whole school has not made the change, the teacher in the next room may not understand the noise level. Furthermore, the Johnsons have proposed, only half in jest, that principals wander through the halls of their schools, listening at each classroom door. Whenever they hear nothing, they ought to make a point of asking the teacher, “Why isn’t any learning going on in here?” As delicious as this reversal of customary practice may be, some trainers go even further, urging teachers to make sure their voices do not rise above and overpower the voices of students.
The latter idea, which suggests a pedagogical approach that would startle even some theorists and practitioners of CL, offers a useful point of departure for discussing different ways children might work together. Some of these issues are quite specific while others get right to the heart of a philosophy of learning; some are widely accepted among those who promote or use CL while others are controversial. The point here is to stimulate thought about the dimensions of cooperative learning, not to offer a comprehensive guide for implementing it.
ASSIGNING STUDENTS TO GROUPS: The size of CL groups will vary depending on the age of the students, their skill and experience working in teams, the type of assignment, and the time available. As a general rule, it is more challenging to integrate more people. As many as six students may be able to exchange information, but fewer—certainly no more than four—will be better able to produce a common product. Many tasks, particularly when attempted by younger children, can best be done in pairs.40
More complicated is the question of how to assign students to groups. Each of the most common methods has benefits and problems.
1. Letting students sort themselves offers a strong vote of confidence in their capacity to make decisions, but the temptation will be powerful for them to work with people they already know and like rather than learning to cooperate with those who are new to them. Moreover, some students will feel left out by a self-selection process, and the last thing teachers want is to introduce a competition for the most popular students, reminiscent of what happens when children choose up sides for team sports. One way around this, especially appropriate for open-ended cooperative assignments, is to let children form groups on the basis of the questions they are interested in exploring.
2. Teachers who construct groups may do so with an eye to making sure that each group includes students of different ethnic backgrounds and both genders. The advantages of CL that depend on working with those who are different are maximized when the groups are structured to be heterogeneous in this way.
3. Students with different levels of skill can be placed together deliberately, at least for some tasks, so that effective helping takes place. (One researcher finds that mixing students from two distinct ability levels works better than combining those of high, medium, and low ability in a single group.)41
4. Finally, students can be placed randomly in groups42 to avoid the contrived quality of assignments for heterogeneity. To make sure that each student has the chance to work with a variety of others, groups should probably be shuffled more frequently if assignments are random. Since students can belong to more than one group at a time, some assignments can last longer than others: they may belong to one team only for the duration of a short assignment, for example, and to another, more permanent, multipurpose “base” team for as long as a year.43 In any event, it is probably not a good idea for students to be able to switch groups just because they are having trouble resolving problems or getting along with their teammates. Such difficulties should be seen as learning opportunities—a chance to figure out (with the teacher’s guidance) how to untangle conflicts; students should not be encouraged to assume that it is possible to bail out at the first sign of trouble.
TEACHING SOCIAL SKILLS: When students do not seem to be working together effectively, some educators are inclined to blame them for not being cooperative—or to give up on CL as unworkable. Others, though, reason that children may have been presented with a task that requires more subtle interpersonal skills than are currently in their repertoires. Many champions of CL emphasize that teachers must pay explicit attention to the phenomenon of working together—what it means and how it can be improved. The Johnsons, for example, emphasize that “collaborative skills are directly taught in classrooms where teachers are serious about using cooperative learning”—not only because these skills are a prerequisite for realizing academic gains but also because they are valuable in their own right.44
One teacher who recalled her frustration when she first tried to introduce CL to a class of sixth-graders concluded:
We must not . . . make the mistake of expecting that cooperative group behavior and thinking for oneself will occur in the absence of classroom instruction and practice aimed at these specific goals. . . . We can’t order it to work; we have to make it work. We have to teach children the skills of working thoughtfully and responsibly together. . . . I have heard teachers give it up after a single attempt. . . . But these very same teachers would never say, “These children cannot read by themselves,” and thereafter remove any opportunity for them to learn to read.45
Social skills such as learning to listen carefully, to make eye contact, and to criticize someone’s ideas without being insulting can be taught just as reading can be taught. Each lesson, in fact, can be introduced by laying out an academic goal (mastering the future tense in Spanish) and a social goal (helping everyone in the group to feel included). But I would offer two caveats to this approach. First, students of every age should have a role in deciding which social skills they could benefit from Working on. It is important to let the specific needs of a given class determine how these issues are addressed rather than running through a packaged set of topics (If it’s Tuesday, it must be time to work on listening skills). Second, as I will argue below, in the long run we need to move beyond a focus on teaching discrete skills to individuals in favor of creating a caring classroom community as the context for all lessons.
PROCESSING: A concern with how we learn, not only what we learn, requires attention after each lesson as well as before. A certain block of time—which, again, will vary with the age of the students and their experience with CL—should be set aside following the completion of a unit for each group to talk about how successful they have been at working together. They can be asked to consider whether everyone contributed to the final project or one person did most of the work, whether someone dominated the discussion, whether everyone felt free to present ideas,
whether tasks were divided effectively. They can evaluate the process and, just as important, reflect on how it might be improved the next time they work together.46 If they decide they have done well, the team members should have the opportunity to celebrate their success, taking pleasure in what they have done and how they have done it.
INDIVIDUAL ACCOUNTABILITY: Many adults react skeptically to the idea of CL because their memory of doing group projects in school is that they ended up being stuck with most of the work while their teammates loafed. (Two social scientists have observed with evident amusement that everyone seems to remember having been the one who had to carry the load; it is always “the others” who did nothing.)47 To be sure, if one person in the group writes the whole report or works most of the problems, the benefits of CL will not be realized. The question, then, is how to ensure individual accountability.
Let us set aside the larger issue of how teachers can best find out, under any learning structure, what children have understood and what they need help with. Instead, let us consider how teachers can make sure that every group member is participating in the CL process. The simplest method is to give everyone a test on all the material or to choose someone from each team randomly to explain what his group came up with and how they came up with it.