What is Collaborative Learning?

Barbara Leigh Smith and Jean T. MacGregor

Collaboration. Collaborative learning. Community. Communities of learners.

Notions of collaboration and community have been informally linked to the learning
process for many years, but they have become catch phrases in education in the1980's
and the 1990's.Collaborative learning is now finding prominence in college view books,
at conferences, and in journals on higher education. Although its various approaches
are known by different names, collaborative learning is occurring in every discipline at
every level of education. While these strategies are often called "innovative" and "new,"
they have engaged students and teachers throughout much of this century. We are
simply developing new forms and adapting them to new contexts.

Collaborative learning is particularly timely now. In the1980's an avalanche of

reports underscored the problems of undergraduate education: the distance between
faculty and students, the fragmentation of the curriculum, a prevailing pedagogy of
lecture and routinized tests, an educational culture that reinforces student passivity,
high rates of student attrition, and a reward system that gives low priority to teaching. In
many ways, the academy mirrors larger social trends of fragmentation, lack of civic
involvement, and undercurrents of alienation. Collaborative learning, with its emphasis
on social and intellectual engagement and mutual responsibility, aims to counteract
many of these educational and societal trends.

Collaborative learning holds enormous promise for improving student learning

and revitalizing college teaching. It is a flexible and adaptable approach appropriate to
any discipline. Nonetheless, teachers who adopt collaborative learning approaches find
it challenging. They inevitably face fundamental questions about the purposes of their
classes, teacher and student roles and responsibilities, the relationship between
educational form and content, and the nature of knowledge itself. Collaborative learning
represents a radical departure from contemporary practices in postsecondary
education.

1
 In this article, we describe collaborative learning and identify some of its
underlying assumptions and goals. We describe some the collaborative learning
approaches most widely used in higher education, and we conclude with some
observations on the challenges and opportunities that teachers encounter as they work
to build collaboration and community into their classrooms.

Characterizing Collaborative Learning

"Collaborative learning" is an umbrella term for a variety of educational

approaches involving joint intellectual effort by students, or students and teachers
together. In most collaborative learning situations students are working in groups of two
or more, mutually searching for understanding, solutions, or meanings, or creating a
product. There is wide variability in collaborative learning activities, but most center on
the students' exploration or application of the course material, not simply the teacher's
presentation or explication of it. Everyone in the class is participating, working as
partners or in small groups. Questions, problems, or the challenge to create something
drive the group activity. Learning unfolds in the most public of ways.

However practiced, collaborative learning represents a significant shift away from

the typical teacher-centered or lecture-centered milieu in college. In collaborative
classrooms, the lecturing/listening/note-taking process may not disappear entirely, but it
lives alongside other processes that are based in students' discussion and active work
with the course material. Teachers who use collaborative learning approaches tend to
think of themselves less as expert transmitters of knowledge to students and more as
expert designers of intellectual experiences for students--as coaches or mid-wives of a
more emergent learning process (Belenky, Clinchy, Goldberger, & Tarule, 1985;
Schon,1983, 1987; Whipple, 1987).

2
Assumptions about Learning

Though collaborative learning takes on a variety of forms and is practiced by

teachers of different disciplinary backgrounds and teaching traditions, the field is tied
together by a number of important assumptions about learners and the learning
process.

Learning is an active, constructive process. To learn new information, ideas,

or skills, students have to work actively with them in purposeful ways. They need to
attach this new material to, or integrate it with, what they already know--or use it to
reorganize what they thought they knew. In collaborative learning situations, students
are not simply taking in new information or ideas. They are creating something new with
the information and ideas. These acts of intellectual processing—of constructing
meaning or creating something new--are crucial to learning.

Learning depends on rich contexts. Recent research suggests that learning is

fundamentally influenced by the context and activity in which it is embedded (Brown,
Collins, & Duguid, 1989).Collaborative learning activities immerse students in
challenging tasks or questions. Rather than beginning with facts and ideas and then
moving to an application, collaborative learning activities frequently begin with
problems, for which students must marshal pertinent facts and ideas. Instead of being
distant observers of questions and answers, or problems and solutions, students
become immediate practitioners. Rich contexts challenge students to practice and
develop higher order reasoning and problem-solving skills. They invite students to join
what Bruffee calls the conversation of the discipline with knowledgeable peers (Bruffee,
1984. See page XXX of this sourcebook).

Learners are diverse. Students bring multiple perspectives to the classroom--

diverse backgrounds, learning styles, experiences, and aspirations; teachers can no
longer assume a one-size-fits-all approach. When students work together on their
learning in class, teachers get a direct and immediate sense of how students are

3
learning, and what experiences and ideas they bring to their learning. The diverse
perspectives that emerge in collaborative activities are clarifying not just for teachers;
they are illuminating for students as well.

Learning is inherently social. As Jeff Golub points out, "Collaborative learning

has as its main feature a structure that allows for student talk: students are supposed to
talk with each other....and it is in this talking that much of the learning occurs." (Golub,
1988).

In collaborative learning, there is the intellectual synergy of many minds coming

to bear on a problem, and the social stimulation of mutual engagement in a common
endeavor. This mutual exploration, meaning-making, and feedback often leads to better
understanding on the part of students, and to the creation of new understandings as
well.

Learning has affective and subjective dimensions. Collaborative tasks build

connections between learners and ideas and between students and teachers. Listening
to and acknowledging diverse perspectives, working in a cooperative spirit, becoming a
peer teacher or a peer learner--all these activities are socially involving, as well as
emotionally demanding. Such intense social interaction stimulates learners and
learning. In collaborative learning situations, students generally experience a shift in
their intellectual development as they learn to articulate their own point of view and
listen to the views of others. They begin to see themselves not just as recipients of
truths from textbooks or faculty members, or procedural knowers (going through the
motions called for by the teacher), but as responsible creators of their own knowledge
and meanings--a change that is essential to life-long learning and true intellectual
development.

4
Goals for Education

While faculty members use collaborative learning because they believe it helps
students learn more effectively, many of them also place a high premium on teaching
strategies that go beyond mere mastery of content and ideas; they believe that
collaborative learning promotes a larger educational agenda. Still, there isn't just one
rationale for collaborative learning, but rather several intertwined rationales.

Involvement. Today's college students are increasingly diverse in terms of

background, prior experience, skills, and goals; they are commuter students with busy
lives, full of distractions and multiple responsibilities. It should not surprise us that many
of these students have little sense of connection to each other or the academic
community as a whole. Calls to involve students more actively in their learning are
coming from virtually every quarter of higher education (Astin,1985; Bonwell & Eison,
1991; Kuh et al., 1991; Study Group on the Conditions of Excellence in Higher
Education, 1984).These exhortations are repeatedly borne out by studies both of
students who leave college and those who stay, and by studies on what students find
most important and meaningful to their learning(Light, 1990, 1991; Tinto,
1987).Involvement in learning, involvement with other students, and involvement with
faculty are factors that make an overwhelming difference in student retention and
success in college. By its very nature, collaborative learning is socially and intellectually
involving. It invites students to build closer connections to other students, to their
faculty, to their courses, and to their learning.

Cooperation and team-work. In collaborative endeavors, students inevitably:

encounter difference and must grapple with recognizing and working with it. Building the
capacities for tolerating or resolving differences, for building agreement that honors all
the voices in a group, for caring how others are doing--these abilities are crucial aspects
of living in a community. Too often the development of these values and skills are
relegated to what is called the "Student Life" side of the campus. Cultivation of team-

5
work and leadership skills are legitimate and valuable classroom goals, not just extra-
curricular ones.

As Alexander Astin points out in "Competition or Cooperation: Teaching

Teamwork as a Basic Skill" (1987), there is both an implicit and an explicit curriculum
embedded in the content and pedagogy of any course. Often, the implicit values are
unexamined. Many educational reform efforts are unsuccessful because they fail to deal
with the implicit values in the educational environment. Astin believes there is an
underlying culture of individualism and competition that gets in the way of many current
reform efforts. Collaborative learning represents a new and different value system, one
that regards teamwork, cooperation, and community as just as important as academic
achievement.

Civic responsibility. These collaborative skills and values are essential

components in a larger civic landscape. If democracy is to sustain in any meaningful
way, our educational system must foster habits of participation and a sense of
responsibility to the larger community. Collaborative learning encourages students to
acquire an active voice in shaping their ideas and values and a sensitive ear in hearing
others. Dialogue, deliberation, and consensus-building out of differences are strong
threads in the fabric of collaborative learning, and in civic life as well.

Collaborative Learning Approaches

Collaborative learning covers a broad territory of approaches, and there is wide

variability in the amount of in-class or out-of-class time built around group work.
Collaborative activities can range from classroom discussions interspersed with short
lectures, through entire class periods, to study on research teams that last a whole term
or a year. There is also enormous variability in the goals and processes of collaborative
activities. Some faculty members design small group work around specific sequential
steps, or tightly structured tasks. Others are comfortable with a more spontaneous
agenda developing out of student interests or questions. In some collaborative learning

6
settings, the task for students is to create a clearly delineated product; in others, the
task is not to produce a product, but rather to participate in a process, an exercise of
responding to each other's work or engaging in analysis and meaning making.

In the next section, we describe a number of widely used collaborative learning

approaches. Some of these approaches, such as Guided Design and peer writing,
evolved in a particular discipline and then spread to others. Others, such as seminars,
peer teaching, and cooperative learning, have been used in many disciplines. Learning
communities are a structural approach to curriculum reform that embraces multiple
courses or disciplines.

While the approaches we describe are referred to by their distinctive names,

there are myriad other small group teaching approaches that also constitute
collaborative learning that we will not describe in detail. For example, many faculty
punctuate their lectures with questions to student pairs or threesomes. (Johnson,
Johnson, & Smith, 1991a).Others create "worksheet workshops" like those Finkel and
Monkdescribe in a later article in this sourcebook (pp. XXX).In numerous lab and field
courses, student pairs or student teams gather data together and produce reports. In
every discipline, teachers are inventing more extended collaborative projects through
presentations or debates, dramatizations and research papers. The possibilities are
endless.

Cooperative Learning

Cooperative learning represents the most carefully structured end of the

collaborative learning continuum. Defined as "the instructional use of small groups so
that students work together to maximize their own and each other's learning,"(Johnson,
Johnson, & Holubec, 1990) cooperative learning is based on the social interdependence
theories of Kurt Lewin and Morton Deutsch (Deutsch, 1949; Lewin,1935).These theories
and associated research explore how the structure of social interdependence influences
individual interaction within a given situation which, in turn, affects the outcomes of that

7
interaction (Johnson & Johnson, 1989).Pioneers in cooperative learning, David and
Roger Johnson at the University of Minnesota, Robert Slavin at Johns Hopkins
University, and Elizabeth Cohen at Stanford University, have devoted years of detailed
research and analysis to clarify the conditions under which cooperative, competitive, or
individualized goal structures affect or increase student achievement, psychological
adjustment, self-esteem, and social skills.

Cooperative learning structures small group learning around precisely defined

tasks or problems. Although numbers of cooperative learning strategies are workable in
any discipline, there are several essential elements. Positive interdependence of effort
is crucial. Cooperative learning activities are designed so that every learner contributes
to the collaborative task. There is "promotive interaction"; students work constructively,
talking face-to-face, helping each other complete the given task. At the same time,
however, careful attention is given to individual accountability and personal
responsibility to achieve the group's goals. Within the framework of group work, each
student's performance is still individually assessed and each student is held responsible
for contributing to the group's success.

In cooperative learning, the development of interpersonal skills is as important as

the learning itself. The development of social skills in group work--learning to cooperate-
-is key to high quality group work, and many cooperative learning tasks are put to
students with both academic objectives and social skills objectives. Many of the
strategies involve the assigning of roles within each small group (such as recorder,
participation encourager, summarizer) to ensure the positive interdependence of the
group participants and to enable students to practice different team-work skills. Built into
cooperative learning work is regular group processing, a "debriefing" time where
students reflect on how they are doing in order to learn how to become more effective in
group learning settings (Johnson, Johnson, & Holubec, 1990).

For years, researchers in the cooperative learning field have focused their work
on comparing cooperative learning contexts with competitive and individualized ones.

8
As the Johnsons' summary and analysis of hundreds of studies concludes, cooperative
learning situations foster more intrinsic motivation, more continuing interest and
commitment to achievement, greater persistence, and the incentive for everyone to
succeed together. On the other hand, the motivational environment associated with
competitive or individualized learning situations fosters more extrinsic motivation, less
continuing interest in achievement, and lower persistence on tasks. Moreover,
competition seems to motivate only "winners," students with high ability to achieve in
competitive situations (Johnson & Johnson, 1989).Current cooperative learning
research is now turning to the internal dynamics of cooperative learning groups, to
understand more about the qualities of an effective learning group. Research findings in
higher education, though less well explored, appear in more detail in Cooper and
Mueck's (pp. XXX) and Slavin's (pp. XXX) articles which follow in this sourcebook.

Under the leadership of the Johnsons at the Cooperative Learning Center at the
University of Minnesota, and David DeVries, Keith Edwards, and Robert Slavin at the
Study for Social Organization of Schools at Johns Hopkins, cooperative learning has
developed in the past 25 years into a forceful movement in K-12 education. Growing
numbers of practitioners in higher education are adopting cooperative learning methods.
The International Association for the Study of Cooperation in Education (IASCE)
publishes the magazine Cooperative Learning and holds triennial conferences. More
recently, with support from the Fund for the Improvement of Postsecondary Education,
Jim Cooper and his colleagues at California State University Dominguez Hills
established a Center for Cooperative Learning in Higher Education which disseminates
and researches cooperative learning at the college level and publishes Cooperative
Learning in College Teaching.

Problem-Centered Instruction

Problem-centered instruction, widely used in professional education, frequently is

built around collaborative learning strategies. Many of these spring from common roots,
especially the work of John Dewey in the early part of this century. Dewey endorsed

9
discussion-based teaching and believed strongly in the importance of giving students
direct experiential encounters with real-world problems. Guided Design, cases, and
simulations are all forms of problem-centered instruction which immerse students in
complex problems' that they must analyze and work through together. These
approaches develop problem solving abilities, understanding of complex relationships,
and decision-making in the face of uncertainty. While problem-solving has long been a
focus of professional education, it is increasingly regarded as an important aspect of the
liberal arts as well. Our focus here is on problem-centered instruction that involves
collaborative learning.

Guided Design. Guided Design is the most carefully structured approach to

problem-centered instruction. The approach asks students working in small groups to
practice decision making in sequenced tasks, with detailed feedback at every step.
Developed in the late 1960's in the engineering program at West Virginia University, the
Guided Design approach has since been adopted in many disciplines and professional
programs, most notably in engineering, nursing, and pharmacy, but in many liberal arts
and sciences courses as well (Borchardt,1984; Day, Macy, & Jackson, 1984;deTornyay
& Thompson, 1987;Miller, 1981; Roemer, 1981; Vogt, Cameron, & Dolan, in
press).Each Guided Design activity presents a large and open-ended problem to
students, but the problem is broken down into the following steps: (1) situation or
problem definition, (2)statement of the goal, (3) generation of possible solutions,
(4)evaluation of solutions, and (5) development of a plan of action. Each of these steps
themselves involve open-ended questions. To answer them, students must marshal
both information and the thinking skills of analysis, synthesis, and evaluation.

They also must build their social skills to work in a team, reconcile differences,
and reach a common decision. After developing their response to each step of the
"design," each student team receives written feedback from the faculty member, the
"professional," about the strengths, weaknesses and implications of their decision. At
each step of the process, there is an interplay between novice problem-solver and

10
expert problem-solver. (Wales, Nardi, & Stager, 1987)Some designs take about a week
to complete, while others run over several weeks of a course.

Careful guidance underpins this approach; it develops from the sequenced steps,
from related homework assignments, from the thinking of other students, and from
detailed feedback from the faculty member at each step in the process (Wales et al.,
1978).Charles Wales at West Virginia University, Director of the Center for Guided
Design, and Robert Stager at the University of Windsor co-developed this approach.
Guided Design practitioners share their work under the auspices of the International
Society for Exploring Teaching Alternatives (ISETA), an organization that promotes a
variety of alternative teaching approaches.

Cases. Case studies have long been a staple for teaching and learning in the
professions, particularly in the fields of business, law, and education, and they are now
being used in many other disciplines as well. A case is a story or a narrative of areal life
situation that sets up a problem or unresolved tension which the students analyze and
resolve. The use of cases does not necessarily imply collaborative learning or small
seminar discussion. However, case method teaching frequently asks small groups of
students to tackle cases in class or in study group sessions.

Harvard University's Business School pioneered the development of the case

method in the early part of this century. The dean of Harvard's first business program
saw the case method as especially 'appropriate to educating managers and decision-
makers. As one of Harvard's early professors put it, "[Businesspeople must be able] to
meet in action the problems arising out of new situations of an ever-changing
environment. Education, accordingly, would consist of acquiring facility to at in the
presence of new experience. It asks not how a man may be trained to know, but how a
man be trained to act" (Dewing, 1931, 23).

More recently, in The Reflective Practitioner: How Professionals Think in Action,

Donald Schon examines how professionals solve, problems and how they develop a

11
highly valuable type of knowledge through reflection-in-action. He believes that
education must be designed to promote this reflective practice, by immersing students
in the "complexity, uncertainty, instability, uniqueness, and value conflicts which are
increasingly perceived as central to the world of professional practice" (Schon, 1983,
14).Cases provide a kind of classroom apprenticeship for professional decision-making.

Cases can describe an actual event or composites of several events. They can
be developed from almost any materials- -letters, business reports, legal documents, or
descriptions of actual historical events. Effective cases are complex and realistic, with a
strong sense of plot and character. Case narratives compress time and space but
otherwise mirror real life in all its provocative complexity and ambiguity. Cases can be
very brief, as short as several paragraphs, or quite lengthy. As Boehrer and Linsky point
out, the definition of a case is quite elastic and the form of cases is changing: "today,
video and computer technology come into increasing use, separately and together, both
to present cases and to engage students in working through them" ( Boehrer & Linsky,
1990, 56).

Harvard's professional schools have spent many years refining the case method
and developing new ways of supporting it in the classroom through the development of
new cases and faculty training seminars. Harvard remains the richest source of
published cases on a wide variety of subjects in business, law, education, and public
policy (Christensen & Hansen, 1987; McNair& Hersum, 1954).There is now a new
renaissance of interest in teaching with cases, especially in schools of education and
many professional graduate schools.

Problem- centered instruction in medical education. Problem-centered

instruction has also emerged in recent decades in the field of medical education. This
work began in England, then spread to Canada and ultimately to the United States.
M.L.J. Abercrombie's research in England in the 1950's had a profound impact on
collaborative learning in medical education both in England and North America
(Abercrombie, 1970, 1961).She made a compelling case for discussion methods of

12
teaching, contending that when people work in teams, they make more valid
judgements than when working alone. McMaster University in Canada was one of the
early pioneers in problem-centered medical education(Barrows & Tamblyn, 1980),
followed by Western Reserve University, the University of New Mexico, and others.

In 1985, the Harvard Medical School adopted a problem-based curriculum

entitled "New Pathways" that has garnered national attention. It was so successful in its
pilot years that the program was quickly extended to all students. While several medical
schools have ambitiously mounted whole curricula and extensive teaching support
around problem-based instruction, many other campuses have embarked on more
modest efforts, building individual courses around these approaches.

Simulations. Simulations are complex, structured role-playing situations that

simulate real experiences. These complex scenarios provide one of the most open-
ended forms of collaborative learning and often the most exciting way to get students
involved. Most simulations ask students, working individually or in teams, to play the
roles of opposing stakeholders in a problematic situation or an unfolding drama. Taking
on the values and acting the part of a stakeholder usually gets students emotionally
invested in the situation. The key aspect of simulations, though, is that of perspective-
taking, both during the simulation exercise, and afterward. Following the simulation,
there is usually a lengthy debriefing process, a discussion period where students reflect
on the simulation and explore their own actions and those of others. This is where
important concepts and lessons emerge.

When used in a carefully planned syllabus with a clear purpose, simulations

enrich the learning process and provide a tangible underpinning to more theoretical
material. A colleague of ours in a social science class asked students to read
Machiavelli's The Prince, and write an essay about their ideal society. Then, the class
session played Starpower, a trading game about the distribution of power and authority
in society. After a debriefing of the simulation, the students tackled a second writing

13
assignment in which they were asked to juxtapose Machiavelli's analysis of power
against their personal vision and their experience in the simulation.

There are now a large number of simulations or educational games, as they are
sometimes called, relating to many disciplinary areas (Abt, 1987; Bratley, 1987). Some
are quite extensive, taking from four hours to an entire quarter to
complete. And a number of simulations utilize computers.
Simulations can also be easily developed from everyday events, and many teachers
find it useful to have groups of students develop their own simulations (Glazier,
1969).Some widely used commercially designed simulations are CLUG: the Community
Land Use Game (Feldt, 1978); games designed to model prejudice and inter-group
cultural communication and relationships such as Bafa and Barnga: A Simulation on
Cultural Clashes (Intercultural Press, 1989); and simulations designed to study power
and societal relationships such as Starpower, SimSoc:Simulated Society, and What's
News? A Game Simulation of TV News. (Gamson,1978, 1984).And for some time
business schools have used a variety of simulation games called "operational games."

Writing Groups

Both in theory and practice, the most concentrated effort in undergraduate

collaborative learning has focused on the teaching of writing. The writing group
approach (known variously as peer response groups, class criticism, or helping circles)
has transformed thousands of college writing classes. Through the spread of writing
across the curriculum initiatives, writing groups increasingly are appearing in other
courses as well. While many proponents of peer writing think of this approach as
innovative, writing groups are actually as old as the nation. Anne Ruggles Gere's (1987)
fascinating book on the subject describes how writing groups enjoy an extensive history
in this country, both within and beyond the academy. Literary societies and writing
clubs, developing in early American universities in the late 18th and early 19th
centuries, met regularly to debate ideas, and to hear and respond to members' work.
Benjamin Franklin and countless aspiring and established writers have met to share and

14
critique their work. By the early part of this century, many educators were leading writing
groups in high school and college classrooms and were convinced that these processes
improved critical thinking as well as writing skills(Gere, 1987).

Using writing groups as a vehicle for reforming the teaching of college English
visibly surfaced in the late 1960's, when American writing teachers learned about writing
group approaches in Great Britain. Indeed, three seminal books advocating writing as a
social process' appeared in 1968 (Macrorie, 1968; Moffett,1968; Murray, 1968).In the
decades since, a large body of literature about theory and practice has helped writing
teachers move to more active, student-centered, sharing classrooms.

Writing teachers at both the secondary and undergraduate level have embraced
peer writing because it helps students see writing as an emergent and social process.
As Peter Elbow puts it, "Meaning is not what you start out with but what you end up
with. Control, coherence, and knowing your mind are not what you start out with but
what you end up with. Think of writing then not as a way to transmit a message but as a
way to grow and cook a message" (Elbow, 1973,14-15).

Peer writing involves students working in small groups at every stage of the
writing process. Many writing groups begin as composing groups: they formulate ideas,
clarify their positions, test an argument or focus a thesis statement before committing it
to paper. This shared composing challenges students to think through their ideas out
loud, to hear what they "sound like," so they will know "what to say" in writing.

Writing groups also serve as peer response groups. Students exchange their
written drafts of papers and get feedback on them either orally or in writing. This is a
challenging process, one that requires students to read and listen to fellow students'
writing with insight, and to make useful suggestions for improvement. Word processors
have helped peer writing enormously: In many writing labs, students share their drafts
and revise them right on the screens.

15
 Getting and giving feedback helps students understand that writing is a social
process, not a solo performance. The mutual support of peer writing groups attempts to
make the processes of composing and drafting less lonely and alienating (Spear,
1988).Sharing their writing with peers not only gives student writers an audience, it
helps them understand the idea of audience(Maimon, 1979).John Bean puts it this way:
"Good writing grows out of good talking. "And, "Good talking means focused dialectical
conversation where students can practice creating and testing their own arguments on
an audience of peers" (Bean, 1991,1990).

Peer writing also makes better writers. A major research study from the
University of Chicago compared results of all the major approaches in teaching
composition. It concluded that "having students work independently in small groups on
purposefully designed and sequenced tasks produces significantly better results, as
measured by the quality of thinking revealed in the writing, than does the lecture
method, whole class discussion methods, or open-ended group work" (Hillocks, 1984,
as summarized in Bean, 1991, 90).

Peer Teaching

With its roots in our one-room schoolhouse tradition, the process of students
teaching their fellow students is probably the oldest form of collaborative learning in
American education. In recent decades, however, peer teaching approaches have
proliferated in higher education, under many names and structures. Many of these
approaches have drawn on the peer teaching methods and studies developed by the
Goldschmids at McGill University. Student pairs, called "learning cells, "practice
structured approaches for completing out-of-class assignments, as well as for teaching
and quizzing each other on new material. In studies comparing the learning cells
approach to seminars, discussion and independent study, the learning cell students at
McGill not only outperformed others, but they preferred learning cells to the other
approaches (Goldschmid & Goldschmid, 1976).

16
 In his recent book surveying the literature on peer teaching, Neal Whitman offers
a helpful typology of peer teaching approaches (Whitman, 1988)."Near-peers" are peer
teachers who are slightly more advanced than the learners. They may be
undergraduate teaching assistants who successfully complete a class and then return to
assist the instructor in teaching it by leading discussion groups or help sessions.
Another "near-peer" might be a tutor, also a previously successful student who works in
one-to-one situations with fellow students in need of help in a specific course.
Counselors is Whitman's term for near-peers who also work one-on-one with fellow
students, but instead of being attached to a specific course, they offer broad help,
perhaps on writing, study skills, or academic advising. A second type of peer tutor is the
"co-peer," a student at the same level who helps another. Students may work in two-
person partnerships or in larger work groups that share a common task.

Peer teaching designs and programs are prolific and naturally quite variable. The
following examples represent three of the most successful and widely adapted peer
teaching models.

Supplemental instruction. The supplemental instruction approach is an

undergraduate teaching assistant model developed by Deanna Martin at the University
of Missouri-Kansas City. It has been adopted at hundreds of colleges in the United
States and abroad. This urban campus recognized the need to offer tutoring help to
students, but budgetary constraints made one-to-one tutoring too expensive. Their
search for an alternative approach led to "Supplemental Instruction." This approach
focused not on at risk students," but rather on "at-risk classes," entry level in health
sciences, and later in general arts and sciences classes where more than 30% of the
students were either withdrawing or failing. The university invited advanced
undergraduates who had done well in those classes to become "SI leaders." These
students are paid to attend the class and to convene Supplemental Instruction sessions
at least three times a week at hours convenient to students in the class. All the students
in the class are welcome to attend the SI sessions.

17
 The course instructor works closely with the SI student leader to assess what
students need to master the content of the class and to help the SI leader develop
sessions to facilitate learning. Still, the SI leader is presented as a "student of the
subject," not an expert of the subject--an approach meant to close the perceived gaps
between teacher and student and student and subject matter. Evaluations of
Supplemental Instruction at the University of Missouri-Kansas City and elsewhere have
shown that if students attend the SI sessions consistently, their grades and their
persistence in college are significantly higher, regardless of whether they are strong or
weak academically(Blanc, DeBuhr, & Martin, 1983; "Supplemental Instruction,"1991).

Writing fellows. The writing fellows approach, pioneered by Tori Haring-Smith at

Brown University, is a peer teaching approach somewhat parallel to supplemental
instruction. The writing fellows are upper division students who are strong writers. After
extensive training, these students are each deployed to an undergraduate class
(generally in the discipline of their major) where they read and respond to the papers of
all the students. Haring-Smith calls this a "bottom-up approach" to sustaining writing
across the curriculum initiatives, particularly in large classes where many faculty flag at
assigning writing because there simply are too many papers to which they must
respond. Over 50 colleges and universities have created writing fellows programs.

Mathematics workshops. A third peer teaching approach that spread rapidly in

the late 1980's is the intensive mathematics workshops program developed by Uri
Treisman while he was at the University of California at Berkeley. Treisman wanted to
address the drawbacks of traditional tutoring models--particularly those geared to
minority students in academic difficulty. Finding that study groups made a difference in
student success, he created a co-peer teaching approach called the Professional
Development Program. The program assumes the culture of an honors program rather
than a remedial program. Graduate instructors (usually doctoral candidates) lead math
workshops built around small group problem-solving with an explicit emphasis on peer
teaching. These workshops supplement the regular lecture and discussion sections of
mathematics courses. This intensive small group workshop approach, which

18
emphasizes developing strength rather than remediating weakness, and peer
collaboration rather than solo competition, completely reversed the prevailing patterns
of failure in calculus classes by Hispanic and African American students at Berkeley
(Treisman, 1985).This intensive math workshop approach has since spread widely in
the mathematics community in high schools, and in both two- and four-year colleges.

These peer teaching approaches and many others like them depart from many
tutoring models that focus on the remediation or rescue of the drowning. Many of these
newer models require all students to participate as teachers and learners in turn, or they
invite all students to participate voluntarily. The tutors are available to all, and the
learning context is one of collaboration and success. These programs lead to better
learning and higher motivation both for the tutors and the learners. Also, peer teaching
introduces countless undergraduates to the stimulation, challenge, and satisfaction of
teaching--an important investment in developing the future professorate.

Discussion Groups and Seminars

The terms discussion group and seminar refer to a broad array of teaching
approaches. In college settings we usually think of discussions as processes, both
formal and informal, that encourage student dialogue with teachers and with each other.
These are spaces within classes, where "instructors and groups of students consider a
topic, issue, or problem and exchange information, experiences, ideas, opinions,
reactions, and conclusions with one another" (Ewens, 1989).Seminar has several
connotations; historically the seminar has been thought of as a course where advanced
students take turns presenting research for discussion and critical feedback from
student peers as well as the teacher. Seminar also refers to an extended discussion in
which students and teacher examine a specific text or common experience.

While the terms group discussion and seminar are often used interchangeably, it
is interesting to note that discussion derives from the Latin words meaning breaking
apart, while the word seminar comes from words having to do with nurseries and seed

19
plots. As the etymologies suggest, both these settings involve the interplay between the
dissection of ideas and the cultivation of new ones, analysis and synthesis, the
acknowledgment of diverse perspectives, and the creation of community. These are
powerful arenas for collaborative learning, spaces in the curriculum where the
conversation turns to mutual search for understanding.

All the approaches we have described above involve discussion. However most
have distinct protocols, goals, or structures framing the activity. What we are describing
here- -more open-ended discussion or seminars--puts the onus on the teacher or the
students to pose questions and build a conversation in the context of the topic at hand.
There is enormous variability, then, in terms of who sets the agenda, who organizes and
monitors the discussion, and who evaluates what. Some discussions or seminars may
be heavily teacher-directed, others much more student-centered. There are a myriad
possibilities for discussions, and many good resources on strategies (Christensen,
Garvin, & Sweet, 1991; Eble, 1976;McKeachie, 1986; Neff & Weimer, 1989).

Learning Communities

Collaborative learning practitioners would say that all collaborative learning is

about the building of learning communities. However, we are using the term learning
community here in a broader and more specific sense, in terms of an intentional
reconfiguration of the curriculum. In the past 15years, a number of colleges have
recognized that deep-seated structural factors weaken the quality of undergraduate
learning and inhibit the development of community. These schools have attacked this
problem directly by developing learning communities, a "purposeful restructuring of the
curriculum to link together courses so that students find greater coherence in what they
are learning and increased interaction with faculty and fellow students"(Gabelnick,
MacGregor, Matthews, & Smith,1990).As such, learning communities are a delivery
system and a facilitating structure for the practice of collaborative learning.

20
 Learning community curriculum structures vary from campus to campus, and can
serve many different purposes, but they have two common intentions. They attempt to
provide intellectual coherence for students by linking classes together and building
relationships between subject matter, or by teaching a skill (e.g., writing or speaking) in
the context of a discipline. Second, they aim to build both academic and social
community for students by enrolling them together in a large block of coursework. While
the learning community approach goes back 60 years or more (Meiklejohn, 1932), we
have seen a recent proliferation of learning community approaches on all sizes and
types of campuses. Learning communities directly confront multiple problems plaguing
undergraduate education: the fragmentation of general education classes, the isolation
of students (especially on large campuses or commuter schools), the lack of meaningful
connection-building between classes, the need for greater intellectual interaction
between students and faculty, and the lack of sustained opportunities for faculty
development.

Some learning community models are quite modest. In the Freshman Interest
Group (FIG) model used at several large universities, cohorts of 25-30 freshman
students enroll in three classes that are an appropriate introduction and platform for a
major. In addition, the FIG group meets in a discussion group once a week with a peer
advisor. The faculty of the three classes teach them in the usual way, but they rapidly
discover that the FIG students become the most active students in their class.

Other learning community models are more complex in terms of both pedagogy
and curriculum redesign. In many linked classes, or three-course clusters, the faculty
members co-plan their syllabi to address common themes or develop common
assignments. Still other learning community models are completely team taught and
involve a more ambitious reconfiguring of coursework around broad interdisciplinary
themes. Not only are these closely integrated models exciting for students, they are
revitalizing for faculty. Team teaching creates a unique opportunity for learning from
each other's disciplinary perspectives and for creating and sharing teaching
approaches.

21
 By altering the curricular structure to provide larger units of student learning
communities frequently provide more time and space for collaborative learning and
other more complicated educational approaches. Small group workshops and book
seminars are staples of most learning communities. Peer writing groups and team
projects associated with labs and field work are also fairly common. Study groups
emerge in learning communities, both intentionally and spontaneously. These programs
provide a unique social and intellectual glue for students that result in high rates of
student retention, increased student achievement, and more complex intellectual
development (MacGregor, 1991).

Collaborative Learning: Challenges and Opportunities

In the past decade, collaborative learning approaches quietly have begun to

proliferate. Specific strategies are spreading across campuses and through disciplinary
and professional networks. And as more faculty members use collaborative learning,
the design and analysis of these approaches are becoming more diverse and more
sophisticated. Research and evaluation on collaborative learning strategies are
sharpening our definitions of student outcomes and giving us a clearer understanding of
when collaborative activities do and do not work.

Creating a collaborative classroom is full of challenges and dilemmas. Few of us

experienced collaborative work in our own undergraduate settings, and much of our
graduate school training reinforced the teacher-centered, lecture-driven model of
college teaching. For the individual teacher, stepping "out of the center" and engaging
students in group activity is hard work, especially at first. For students and teachers
alike, every collaborative activity is new and unpredictable in the way it unfolds.
Everyone involved must take some risks.

And designing collaborative learning situations requires a demanding yet

important rethinking of one's syllabus, in terms of course content and time allocation. If

22
some (or a great deal) of the classroom time is considered an important social space for
developing understandings about course material, or if some of the out-of-class time is
devoted to study groups or group projects, how then should the rest of the class time
(lectures, assignments, examinations) be designed? How does the teacher ensure that
students are learning and mastering key skills and ideas in the course, while at the
same time addressing all the material of the course? Teaching in collaborative settings
puts the tension between the process of student learning and content coverage front
and center.

As teachers become more involved in using collaborative learning, they discover

what radical questions it raises. Collaborative learning goes to the roots of long-held
assumptions about teaching and learning. Classroom roles change: Both teachers and
students take on more complex roles and responsibilities. (Finkel & Monk, 1983,
available in this sourcebook pp. XXX; MacGregor, 1990).The classroom is no longer
solo teacher and independent students--it becomes more an interdependent
community. This degree of involvement often questions and reshapes assumed power
relationships between teachers and students, a process that at first can be confusing
and disorienting (Romer & Whipple, 1990).Not only is course content reshaped, so are
definitions of student competence. The public nature of group work makes the
demonstration of student learning continuous. Thus, for teachers and students,
collaborative learning both complicates and enriches the evaluation process.

Challenges to collaborative learning at the classroom level are compounded by

the traditional structures and culture of the academy, which continue to perpetuate the
teacher-centered, transmission-of-information model of teaching and learning. The
political economy of the academy is set up to front load the curriculum with large lower
division classes in rooms immutably arranged for lectures, usually in classes limited to
fifty-minute "hours." Student-student interaction; extended, careful examination of ideas;
the hearing-out of multiple perspectives; the development of an intellectual community--
all these are hard to accomplish under these physical and time constraints.

23
 The lecture-centered model is reinforced (both subtly and blatantly) by
institutional reward systems that favor limited engagement in teaching and give greater
recognition to research. Achievement for teachers and students alike is assumed to be
a scarce honor, which one works for alone, in competition with peers. This assumption
of scarcity is the platform for norm-referenced grading, or "grading on the curve, "a
procedure that enforces distance between students and corrodes the trust on which
collaborative learning is built. Moreover, our definitions of ourselves as teachers, as
keepers and dispensers of disciplinary expertise, are still very much bound up in the
lecture podium. As a young colleague of ours just beginning to use collaborative
learning in her class acidly observed, "I know this works, but my colleagues don't
respect it as real teaching. They associate group work with lazy, unprepared faculty
members."

And there are compelling reasons to believe our colleagues. Lectures, the
prevailing mode of classroom teaching in college, have only limited efficacy (Blackburn,
Pellino, Boberg, &O'Connell, 1980; Costin, 1972, 1980; McKeachie, 1986; Penner,1984;
Thielens, 1987; Verner & Dickinson, 1967).The myths about interpersonal competition--
that it is motivating, enjoyable, character-building, and necessary for success in a
competitive workplace and world--have been debunked increasingly in the past twenty
years, both in theoretical terms (Astin, 1987; Bricker,1989; Nichols, 1989; Palmer, 1983)
and through extensive research(Johnson & Johnson, 1989a; Kohn, 1986).Most
troubling of all, more than 50% of the students who begin college leave, often never to
return. Much of this student leaving has to do with feelings of isolation and a lack of
involvement with the college environment (Tinto, 1987).Whether we measure these
losses in wasted resources, in thwarted aspirations, or in workplace unpreparedness,
the costs of this kind of attrition are too high.

While these reasons may motivate some teachers, what really propels teachers
into collaborative classrooms is the desire to motivate students by getting them more
actively engaged. Nonetheless, wanting to be a facilitator of collaborative learning and
being good at it are very different things. As with all kinds of teaching, designing and

24
guiding group work takes time to learn and practice. Most teachers start with modest,
efforts while others may work with colleagues, designing, trying, and observing each
other's approaches.

Several years ago, two colleagues of ours embarked on collaborative learning

because they were dissatisfied with their introductory biology course. Because it
seemed students were having difficulty grasping the concepts in the textbook, these
teachers found themselves devoting too much class time to re-explaining the text
material. At the same time, they noticed how engaged students were with their
occasional problem-solving exercises and small group seminars on journal articles.

Over a period of a year, these two biologists began to shift their lectured-
centered course to one involving small group problem-solving workshops. They
developed these workshops as applications and extensions of the textbook reading and
required students to complete reading assignments in order to participate in class
workshops. At the same time, these faculty members built support for their new
approach with their biology department colleagues by asking for their help in defining
the knowledge and understandings essential to completion of Introductory Biology. The
rewards were immediate: The completion rate of the course soared, student
achievement rose significantly, and the course became much more exciting to teach.
These teachers have continued their collaboration, refining the workshops in the course
and developing new ones.

The story of our biology colleagues is not anomalous. Faculty collaboration

seems to be an important ingredient in the design of and experimentation with
collaborative learning approaches. Learning new moves in the classroom need not be a
lonely enterprise. Faculty development initiatives at the departmental or college-wide
level need to acknowledge this as they work to create a supportive climate for
dissemination of approaches as well as a forum for work on questions that arise. At
several universities, collaborative and cooperative learning "users groups" have sprung
up and become valuable structures for sharing approaches and problems. The team-

25
teaching that is embedded in many learning community programs is a powerful strategy
for enabling faculty to build their repertoires and confidence. Research and evaluation,
from modest faculty-designed "classroom research" (Cross & Angelo, 1985) to more
formal studies, can also help develop approaches and clarify their results. Sourcebooks
like this one and growing networks, such as AAHE's Action Community on Collaborative
Learning, will also continue to share resources and build momentum.

There is no getting around the challenging nature of collaborative learning. But

when collaborative work becomes a regular feature of their class, faculty members
usually find it enormously energizing and liberating. The specter of teaching becoming
repetitive or routinized simply isn't an issue for these teachers. Every course and every
class presents an intriguing opportunity. Teachers relish the intellectual challenges of
creating (and re-creating) activities or problems that really engage students. They enjoy
those moments when the class becomes a community. And they often speak of the new
lens they gain on their students, which comes from watching them struggle with ideas
and build meaningful connections to previous learning or personal experiences. They
also remark on the fresh perspectives they gain on their subject matter, as it is enriched
and challenged by continuous and diverse student examination and re-shaping of it.

Ideally, collaborative learning leads students to become much more directly

immersed in the ideas of the class. They will develop confidence and skills at
entertaining ideas on their own while learning to raise questions, to listen carefully, and
to respond to others' questions. They will develop the ability to stay focused, sustain an
idea, build rapport with fellow students and learn the art of disagreeing with others with
respect and courtesy. They may learn to recognize and acknowledge the limitations of
their own points of view. These intellectual and interpersonal skills don't come easily to
college students, not to mention college graduates! Their development requires
extended and focused practice. As Finkel and Monk point out in one of the following
articles, students' awkwardness and tentativeness can often discourage teachers, and
drive them back into the comparative ease of lecturing (Finkel & Monk,
1983).Developing successful collaborative learning activities challenges teachers to

26
become coaches and facilitators of complex social processes, but these are deeply
important ones for true learning.

Ultimately, collaborative classrooms stimulate both students and teachers. In the

most authentic of ways, the collaborative learning process models what it means to
question, learn and understand in concert with others. Learning collaboratively demands
responsibility, persistence, and sensitivity, but the result can be a community of learners
in which everyone is welcome to join, participate, and grow.

Authors' note: We Are grateful to the following people for their helpful comments on this
article: John Bean, Jim Cooper, Thad Curtz, Anne Goodsell, Deanna Martin, Sharon
McDade, Roberta Matthews, Kelly Parsley, David Paulsen, Karl Smith, Maryellen
Weimer, Vincent Tinto.

Ed.'s note: In a very thoughtful, reasoned way Bruffee makes the case for collaborative
learning. He traces its history in terms of our current interests in it, defines it, and
through that definition justifies it as an instructional method of merit. Although the paper
was originally written for an English faculty member audience, the rationale it
establishes for collaborative learning crosses disciplinary boundaries and is relevant in
most fields. To emphasize that interdisciplinary relevance (and to make the paper a
more manageable reading length), we have deleted some of the passages where the
further and more specific application to English is made.

This version is true to the author's intent: "This essay is frankly an attempt to encourage
other teachers to try collaborative learning and to help them use collaborative learning
appropriately and effectively. But it offers no recipes. It is written instead on the
assumption that understanding both the history and the complex ideas that underlie
collaborative learning can improve its practice and demonstrate its educational value."

Please visit http://files.eric.ed.gov/fulltext/ED357705.pdf for the original publication and

references.

27