Cooperative Learning: Richard M. Felder, and Rebecca Brent
Cooperative Learning: Richard M. Felder, and Rebecca Brent
Cooperative Learning: Richard M. Felder, and Rebecca Brent
*
Richard M. Felder
1
, and Rebecca Brent
2
1
Department of Chemical Engineering, N.C. State University,
Raleigh, NC 27695-7905
2
Education Designs, Inc., Cary, NC 27518
Introduction
Many students who have worked in a team in a laboratory- or project-based course do not have fond
memories of the experience. Some recall one or two team members doing all the work and the others
simply going along for the ride but getting the same grade. Others remember dominant students, whose
intense desire for a good grade led them to stifle their teammates efforts to contribute. Still others recall
arrangements in which the work was divided up and the completed parts were stapled together and turned
in, with each team member knowing little or nothing about what any of the others did. Whatever else
these students learned from their team experiences, they learned to avoid team projects whenever
possible.
Cooperative learning is an approach to groupwork that minimizes the occurrence of those unpleasant
situations and maximizes the learning and satisfaction that result from working on a high-performance
team. A large and rapidly growing body of research confirms the effectiveness of cooperative learning in
higher education (1-4). Relative to students taught traditionallyi.e., with instructor-centered lectures,
individual assignments, and competitive gradingcooperatively taught students tend to exhibit higher
academic achievement, greater persistence through graduation, better high-level reasoning and critical
thinking skills, deeper understanding of learned material, greater time on task and less disruptive behavior
in class, lower levels of anxiety and stress, greater intrinsic motivation to learn and achieve, greater ability
to view situations from others perspectives, more positive and supportive relationships with peers, more
positive attitudes toward subject areas, and higher self-esteem. Another nontrivial benefit for instructors is
that when assignments are done cooperatively, the number of papers to grade decreases by a factor of
three or four.
There are several reasons why cooperative learning works as well as it does. The idea that students
learn more by doing something active than by simply watching and listening has long been known to both
cognitive psychologists and effective teachers (5, 6) and cooperative learning is by its nature an active
method. Beyond that, cooperation enhances learning in several ways. Weak students working individually
are likely to give up when they get stuck; working cooperatively, they keep going. Strong students faced
with the task of explaining and clarifying material to weaker students often find gaps in their own
understanding and fill them in. Students working alone may tend to delay completing assignments or skip
*
P.A. Mabrouk, ed., Active Learning: Models from the Analytical Sciences, ACS Symposium Series 970, Chapter 4,
pp. 3453. Washington, DC: American Chemical Society, 2007.
The term cooperative learning (CL) refers to students working in teams on an
assignment or project under conditions in which certain criteria are satisfied,
including that the team members be held individually accountable for the complete
content of the assignment or project. This chapter summarizes the defining criteria
of cooperative learning, surveys CL applications, summarizes the research base
that attests to the effectiveness of the method, and outlines proven methods for
implementing CL and overcoming common obstacles to its success.
2
them altogether, but when they know that others are counting on them, they are motivated to do the work
in a timely manner.
The proven benefits of cooperative learning notwithstanding, instructors who attempt it frequently
encounter resistance and sometimes open hostility from the students. Bright students complain about
begin held back by their slower teammates; weak or unassertive students complain about being
discounted or ignored in group sessions; and resentments build when some team members fail to pull
their weight. Knowledgeable and patient instructors find ways to deal with these problems, but others
become discouraged and revert to the traditional teacher-centered instructional paradigm, which is a loss
both for them and for their students.
In this chapter we describe cooperative learning methods that have been proven effective in a variety
of instructional settings. We then suggest ways to maximize the benefits of the approach and to deal with
the difficulties that may arise when cooperative learning is implemented.
What is Cooperative Learning?
Several definitions of cooperative learning have been formulated. The one most widely used in higher
education is probably that of David and Roger J ohnson of the University of Minnesota. According to the
J ohnson & J ohnson model, cooperative learning is instruction that involves students working in teams to
accomplish a common goal, under conditions that include the following elements (7):
1. Positive interdependence. Team members are obliged to rely on one another to achieve the goal.
If any team members fail to do their part, everyone suffers consequences.
2. Individual accountability. All students in a group are held accountable for doing their share of
the work and for mastery of all of the material to be learned.
3. Face-to-face promotive interaction. Although some of the group work may be parcelled out and
done individually, some must be done interactively, with group members providing one another
with feedback, challenging reasoning and conclusions, and perhaps most importantly, teaching
and encouraging one another.
4. Appropriate use of collaborative skills. Students are encouraged and helped to develop and
practice trust-building, leadership, decision-making, communication, and conflict management
skills.
5. Group processing. Team members set group goals, periodically assess what they are doing well
as a team, and identify changes they will make to function more effectively in the future.
Cooperative learning is not simply a synonym for students working in groups. A learning exercise only
qualifies as cooperative learning to the extent that the five listed elements are present.
Cooperative Learning Structures
Cooperative learning can be used in for any type of assignment that can be given to students in lecture
classes, laboratories, or project-based courses. Following are some of the structures that have been used,
with some recommendations for how they may be effectively implemented. (Additional suggestions are
given at the conclusion of the chapter.)
3
Problem Sets
Students complete some or most of their homework assignments in teams. The teams are encouraged to
include only the names of actual participants on the solution set that they hand in. The students are
initially disinclined to leave anyones name off, but eventually they get tired of letting nonparticipants
(hitchhikers, in cooperative learning parlance) get good grades for work they didnt do and begin to
omit names, at which point many hitchhikersunhappy about getting zeroes on assignmentsstart
cooperating.
The team gets a grade for the assignment, but eventually the performance of each team member
should be assessed and the results used to adjust the average team homework grade separately for each
team member. Adjusting team grades for individual performance is one of the principal ways of assuring
individual accountability in cooperative learning, second only in importance to giving individual exams.
Later in this chapter we will describe systems for performing the performance assessments and making
the adjustments.
We recommend using a mixture of individual and team assignments in a lecture course rather than
having all assignments completed by teams. One obvious reason is to provide another measure of
individual accountability. Another is that if there is a lot of dropping and adding in the first one or two
weeks of the course, it is better to wait until the class population stabilizes before forming teams.
We also suggest advising teams not to simply meet and complete each assignment together. One team
member is usually the fastest problem solver and begins almost every homework problem solution in the
group sessions, and the other members then have to figure out how to get the solutions started for the first
time on the individual tests, which is not a good time for them to have to do it. We recommend instead
that all team members outline solutions individually before meeting to work out the details. On the first
few assignments we require team members to sign and hand in their outlines to help them acquire the
habit.
Laboratories and Projects
Laboratories and projects may be carried out by teams (as they often are in traditional curricula), except
that again the team grades should be adjusted for individual performance.
The problem with team labs and projects as they are normally conducted is that there is no individual
accountability at all. The result is the familiar situation in which some team members do the bulk of the
work, others contribute little and understand little or nothing about the project, everyone gets the same
grade, and resentment abounds. Adjusting the team project grades for individual performance goes a long
way toward correcting these injustices. In addition, it is good practice to include some individual testing
on every aspect of the project and have the results count toward the final course grade. If this is done,
hitchhikers who understand either nothing or only the little they did personally will be penalized and
perhaps induced to play a more active role in subsequent work.
Jigsaw
Jigsaw is a cooperative learning structure applicable to team assignments that call for expertise in
several distinct areas. For example, in a laboratory exercise, areas of expertise might include experimental
design, equipment calibration and operation, data analysis (including statistical error analysis), and
interpretation of results in light of theory, and in a design project the areas might be conceptual design,
process instrumentation and control, safety and environmental impact evaluation, and cost and
profitability analysis.
4
Suppose four such areas are identified for a project. The students are formed into teams of four, and
either the instructor or the team members designate which member will be responsible for each area. Then
all the experts in each area are given specialized training, which may involve getting handouts or
presentations by the course instructor, a faculty colleague, or a graduate student knowledgeable in the
area in question. The students then return to their home teams and complete the assignment. The teams
count on each member to provide his or her expertise, and if an expert does a poor job, the quality of the
final project is compromised and everyones grade suffers. Moreover, if the students are tested on all of
the areas of expertise, the overall learning from the assignment improves dramatically. The tests require
all students to understand the entire project, and not just the part that they were the experts in (individual
accountability), and the experts have the responsibility of transmitting their expertise to their teammates
(positive interdependence).
Peer Editing
When teams turn in written lab reports and/or give oral presentations, the usual procedure is for the
instructor to do the critiquing and grading. A powerful alternative is peer editing, in which pairs of groups
do the critiquing for each others first drafts (written) or run-throughs (oral). The groups then revise their
reports and presentations taking into account the critiquing teams suggestions and then submit or present
to the instructor. This activity lightens the grading load for instructors, who end up with much better
products to grade than they would have without the first round of critiquing.
If a grading checklist or rubric is to be used for grading the team reports (which is always a good
idea), it should be shared with the students before the reports are written and used for the peer editing.
This practice helps the students understand what the instructor is looking for and invariably results in the
preparation of better reports, and it also helps assure that the peer critiques are as consistent and useful as
possible. If several rounds of peer editing are done and the instructor collects and grades the checklists or
rubrics for the first one or two rounds, the students will end up giving much the same rubric scores as the
instructor gives, and in good classes the instructor may only have to do spot checks of peer grades instead
of having to provide detailed feedback on every report.
Peer-Led Team Learning
In peer-led team learning (PLTL), lectures are supplemented by weekly 2-hour workshops in which
students work in six- to eight-person groups to solve structured problems under the guidance of trained
peer leaders. The problems must be challenging and directly related to the course tests and other
assessment measures. The course professor creates problems and instructional materials, assists with the
training and supervision of peer leaders, and reviews progress of the workshops. The materials prompt
students to consider ideas, confront misconceptions, and apply what they know to the solution process.
The peer leaders clarify goals, facilitate engagement of the students with the materials and one another,
and provide encouragement, but do not lecture or provide answers and solutions (8, 9).
PLTL was developed by chemistry educators in the 1990s and may be the most prominent group-
learning strategy in chemistry education. (We will later describe illustrative implementations of the
approach.) It is not a cooperative learning strategy by definition, but as Tien et al. (10) point out, it shares
a number of elements with CL. The students are confronted with difficult problems and must rely
primarily on one another to develop solutions, which promotes positive interdependence, and face-to-face
interaction is crucial to the workshop format. Students are tested individually on the knowledge required
to solve the problems, and a function of the peer leader is to get team members to explain their
understanding to their teammates, both of which provide individual accountability. There is no formal
instruction in teamwork skills in PLTL, but informal instruction invariably occurs as the peer leaders
facilitate the group interactions. The only CL criterion that does not appear to be addressed as part of the
5
PLTL model is regular self-assessment of team functioning, and it would be trivial to add that in PLTL
implementations.
Applications in Chemistry Education
The literature of applications of cooperative learning in science, technology, engineering, and
mathematics is quite large, and a comprehensive review of it is well beyond the scope of this chapter. We
will confine ourselves here to describing several examples of applications in chemistry courses.
A bibliography assembled by Nurrenbern and Robinson (11) cites references to roughly 30 studies of
team-based learning in chemistry lecture and laboratory courses, and a search for articles in recent issues
of the Journal of Chemical Education that included cooperative learning among the key words revealed
47 articles published in 2004, 2005, and the first half of 2006. In the remainder of this section we describe
several of these studies.
Hinde and Kovac (12) discuss two courses that introduced team-based learning in different ways. In
the second semester of a physical chemistry course for chemistry and chemical engineering majors,
biweekly computer-based group work sessions supplemented traditional lectures, and in the the second
semester of a biophysical chemistry course taken primarily by biochemistry majors, an approach based on
group work with occasional supplementary mini-lectures was used. The group sessions in both courses
were inquiry-based. The self-selected teams of three or four in the biophysical chemistry course were
given guidelines on effective teamwork, and both peer ratings and self-ratings of student performance on
teams contributed to the final course grades. In the physical chemistry course there was little difference in
performance between the class in question and previous classes that had been taught without group work,
but this result is not surprising in view of the fact that the group activities were infrequent and most of the
defining criteria for cooperative learning were not met. In the biophysical chemistry course the
instructors assessment was that the students gained considerable conceptual understanding and problem-
solving ability as well as critical thinking and teamwork sklls, but no comparison with a control group
was carried out that would elevate the assessment of the course beyond the anecdotal level. The author
concludes that the course would have been improved by providing more structure and feedback,
maintaining a better balance between individual and group work, and doing more to promote individual
accountability (e.g., give more individual tests) and positive interdependence (e.g., establish and rotate
assigned roles within teams).
A better example of cooperative learning implementation and assessment is provided by Tien et al.
(10), who conducted peer-led team learning in a first-semester organic chemistry course over a three-year
period and compared the performance of the students with the that of students who had taken a traditional
version of the course in the preceding three years. The course instructor, text, examination structure, and
grading system were the same for both the treatment and comparison groups. While instruction in
teamwork skills is not necessarily a component of PLTL, in this case the peer leaders were trained in
group dynamics and group skills and used their training to help the student teams learn to function
effectively. It is therefore fair to say that the PLTL implementation described in this study fully qualifies
as cooperative learning. On average, the workshop students significantly outscored their traditionally-
taught counterparts on individual course exams, final course grades, retention in the course, and
percentage earning the minimum acceptable grade of C for moving on to the second semester organic
chemistry course. Similar results were obtained specifically for female students and underrepresented
minority students. The treatment group found the workshops and workshop problems their most
important aids to learning in the course. Similar findings have been reported for PLTL programs in an
organic chemistry class at another institution (13) and in a biology course (14), as well as for a
cooperative learning implementation in organic chemistry (15).
6
A classical implementation of cooperative learning in chemistry is that of Hanson & Wolfskill (16),
who used a process workshop format in the general chemistry class at SUNY-Stony Brook. Students
worked in teams of three or four on activities that involved guided discovery, critical thinking questions
that help provide the guidance, solving context-rich and sometimes open-ended and incompletely defined
problems, and metacognitive reflecting. Most activities focused on a single concept or issue and could be
completed in a 55-minute session. Following each workshop, students completed an individual quiz on
the workshop content, thus promoting individual accountability. The use of this approach led to
substantially improved examination grades relative to the previous year, in which the course was
conventionally taught, as well as increased attendance at recitation and tutorial sessions and
improvements in student self confidence, interest in chemistry, and attitudes toward instruction. The same
authors report on an interactive computer-assisted learning model that supports and enhances the process
workshop format by providing immediate feedback on student efforts, networked reporting capabilities,
and software tools for both peer assessment and self-assessment (17).
Research Support for Cooperative Learning
Hundreds of research studies of team-based learning in higher education have been conducted, with most
of them yielding positive results for a variety of cognitive and affective outcomes. Analyses of the
research support the following conclusions:
Individual student performance was superior when cooperative methods were used as compared
with competitive or individualistic methods. The performance outcomes measured include
knowledge acquisition, retention, accuracy, creativity in problem solving, and higher-level
reasoning. Other studies show that cooperative learning is superior for promoting metacognitive
thought, persistence in working toward a goal, transfer of learning from one setting to another,
time on task, and intrinsic motivation For example, students who score in the 50
th
percentile when
learning competitively would score in the 69
th
percentile when taught cooperatively (1, 3).
Similar positive effects of group interactions have been found specifically for chemistry courses.
In a meta-analysis of research on cooperative learning in high school and college chemistry
courses, Bowen (18) found that students in the 50
th
percentile with traditional instruction would
be in the 64
th
percentile in a cooperative learning environment.
Several studies of active/collaborative instruction report positive effects on a variety of cognitive
and affective outcomes. In a compilation of pre-post test gains in force concept inventory scores
obtained by students in introductory physics courses, the use of instruction involving interactive
engagement led to an average gain two standard deviations greater than was observed for
traditionally-taught courses (19). Students in engineering capstone design courses taught with
active and collaborative approaches outperformed traditionally-taught students in acquisition of
design skills, communication skills, and teamwork skills (4). The use of collaborative methods
had significant positive effects on understanding science and technology, analytical skills, and
appreciation for diversity, among other outcomes (20).
Affective outcomes were also improved by the use of cooperative learning. Relative to students
involved in individual or competitive learning environments, cooperatively taught students
exhibited better social skills and higher self-esteem (3), as well as more positive attitudes about
their educational experience, the subject area, and the college (7). Towns et al. (21) used field
notes and survey data to analyze students attitudes toward group activities in a physical
chemistry class. The students viewed the group work as a positive force in their learning, and they
also valued the interactions for promoting a sense of community in the classroom.
7
Implementing Cooperative Learning
The benefits of using cooperative learning are well supported by theory and well established by classroom
research, but the method is not without its problems, most of which have to do with individual student
resistance and dysfunctional teams. Many techniques have been developed that minimize the problems,
most of which involve addressing one or more of the five criteria for cooperative learning. The
suggestions that follow are drawn primarily from Felder and Brent (22, 23), J ohnson et al. (7), Oakley et
al. (24) and Smith et al. (2).
Before we offer the suggestions, we should make clear that implementing cooperative learning
successfully does not require adopting every one of them. In fact, trying to do so all at once might be a
serious mistake: the instructor would have to juggle many unfamiliar techniques and end by doing none
of them well, and the students would be deluged by an array of unfamiliar demands and many might rise
up in rebellion. Rather, instructors new to cooperative learning should take a more gradual approach,
choosing mainly the methods with which they feel most comfortable and adopting additional methods
only when they have had time to get used to the current ones. If they do that, they will never stray too far
from their comfort zones and will become increasingly adept at defusing student resistance long enough
for the students to see the benefits of this new form of instruction for themselves.
Forming teams
Instructors should form teams rather than permitting students to choose their own teammates. When
students self-select into teams, the best students tend to cluster, leaving the weak ones to shift for
themselves, and friends cluster, leaving some students out of groups and excluding others from cliques
within groups. Moreover, when graduates go to work in industry or business, they will be required to
work in teams and will have no voice in the team formation, and their job performance evaluation will
depend as much on their ability to work with their teammates as on their technical skills. Since thats
what theyll be doing then, the job of their instructors is to prepare them for it now.
The following criteria are recommended for team formation:
1. Form teams of 34 students for most tasks. When students work in pairs, the diversity of ideas
and approaches that leads to many of the benefits of cooperative learning may be lacking. In
teams of five or more, some students are likely to be inactive unless the tasks have distinct and
well-defined roles for each team member.
2. Make the teams heterogeneous in ability level. The unfairness of forming a group with only weak
students is obvious, but groups with only strong students are equally undesirable. The members of
such teams are likely to divide up the homework and communicate only cursorily with one
another, avoiding the interactions that lead to most of the proven benefits of cooperative learning.
In heterogeneous groups, the weaker students gain from seeing how better students approach
problems, and the stronger students gain a deeper understanding of the subject by teaching it to
others.
3. If the assignments require work being done outside class, form teams whose members have
common blocks of time to meet during the week.
4. When students in a particular demographic category are historically at risk for dropping out,
dont isolate members of that category in a team. Students belonging to at-risk populations are
also at risk for being marginalized or adopting passive roles when they are isolated in teams (22,
25, 26). Once they reach the third year, however, they are very likely to graduate. The focus
should then shift to preparing them for the professional world where no one will be protecting
them, and so this criterion may be dropped.
8
There are three principal ways to get the information needed to form teams using those rules:
1. On the first day of class, have the students fill out a survey containing several questions and an
hour-by-hour matrix of the week, similar to the form at <http://www.ncsu.edu/felder-
public/CL_forms.doc>. On the form, the students write their grades in selected prerequisite
courses, times they are not available to meet outside class with their teams, andif the criterion
related to at-risk minorities is to be usedtheir gender and ethnicity. Use the surveys to form the
groups, following the guidelines given above and using grades in prerequisite courses as the
measure of ability.
2. Use Team Maker