1 

Cooperative Learning and Achievement: 

Methods for Assessing Causal Mechanisms
GEORGE P. KNIGHT 
AND ELAINE MORTON BOHLMEYER

The research on cooperative learning environments has generally focused upon 

several well-developed classroom structures. These environments substantially 
modify the nature of the classroom in an attempt to foster cooperation. Further, 
this research has focused upon the effects of cooperative learning on academic 
achievement and interpersonal relationships (cf. Sharan et al. 180, 1984; Slavin 
et al. 1985). In the present chapter we will be concerned only with the influ- 
ence of cooperative learning environments upon academic achievement; how- 
ever, much of our analysis of the research and our recommended research di- 
rections may be applied to the effects on interpersonal relationships as well. 
We will briefly review, first, some of the most commonly used cooperative 
learning environments and second, the research on the hypothesized causal 
mechanisms through which cooperative learning environments may influence 
academic achievement. We will also discuss what we believe are critical limi- 
tations of this research which seriously limit our ability to make confident in- 
ferences regarding the causal mechanisms. Finally, we will describe an ex- 
ample research approach that we believe will dramatically improve our ability 
to infer the causal mechanisms through which cooperative learning affects ac- 
ademic achievement.

COOPERATIVE LEARNING METHODS

A number of cooperative learning methods have been developed and are 
being used. We will focus upon those cooperative learning methods that have

-1-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 1.

een most widely adopted by educators and which have stimulated considerable 
research. Thus, our review of cooperative learning methods will be by no means 
exhaustive and some well-developed methods will be omitted. However, we 
believe the set of issues described in subsequent sections of this chapter is 
applicable to all cooperative learning methods. 
Circles of Learning (Learning Together)
When Johnson and Johnson ( 1975) developed their method of cooperative 
learning, often called Learning Together, it was quite general in terms of im- 
plementation. A cooperative goal structure was described as one in which there 
is a group goal, sharing of ideas and materials, a division of labor when appro- 
priate, and group rewards. In the research reports of this method, the typical 
description was that students worked as a group to complete a single group 
product, shared ideas and helped each other with answers to questions, made 
sure all members were involved and understood group answers, and asked for 
help from each other before asking the teacher, and the teacher praised and 
rewarded the group on the basis of group performance ( Johnson and Johnson 
1979; Johnson, Johnson, and Skon 1979; Johnson et al. 1983).Recently, Johnson et al. ( 1984)
have called their method Circles of Learning 
and have delineated the following 18 specific steps for implementation (some 
of which are optional):
1. Clearly specify instructional objectives.
2. Limit group size to no more than six. (Students new to cooperative learning should 
be in smaller groups to help ensure that everyone will participate.)
3. Structure groups to achieve heterogeneity in terms of ability, sex, and ethnicity. 
(Occasionally, homogeneous groups may be used to master specific skills.)
4. Arrange groups in circles to facilitate communication.
5. Use instructional materials to promote interdependence among students. Several 
alternatives are suggested, such as giving only one copy of the materials to a group 
so that students will have to share, giving each student in the group access to only 
one part of the lesson, and structuring competition among groups so that students 
will have to depend upon each other for their group to win.
6. Assign roles to ensure interdependence. Suggested roles are summarizer-checker, 
to summarize the lesson and to quiz group members; encourager, to solicit and 
encourage contributions from each member; recorder, to write down group deci- 
sions or a group report; and observer, to check for collaboration among group 
members.
7. Explain the academic task.
8. Structure positive goal interdependence. This can be accomplished by having the 
group produce a single product or by providing group rewards based on the indi- 
vidual performance of each group member.
9. Structure individual accountability for learning so that all group members must
-2-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 2.

contribute. For example, the teacher may give individual tests, randomly
select 
members to explain a group project, have members edit each other's work, or 
randomly select one member's work on which to base a group grade.
11. Structure inter-
 group
cooperation.
12. 11. Explain
criteria for
success.
Grading must
be objective
rather than on a
curve. With 
heterogeneous
groups, criteria
for earning
points for one's
group may need
to be 
individually
determined.
13. 12. Specify
desired
behaviors.
Suggested
beginning
behaviors are to
stay with the
group, 
use each other's
names, and take
turns. More
advanced
behaviors
include making 
sure each group
member
participates in
discussions and
understands and
agrees with 
group answers.
14. 13. Monitor
students'
behavior
continually for
problems with
the task or with
collab- 
orative efforts.
15. 14. Provide task
assistance. The
teacher will
need to
intervene at
times to clarify
in- 
structions,
answer
questions,
encourage
discussions, and
to teach
academic skills.
16. 15. Intervene to
teach
collaborative
skills of
effective
communication,
building a
trust- 
ing
environment,
and
constructive
management of
controversy.
17. 16. Provide
closure to the
lesson, with
summaries by
students and
teacher.
18. 17. Evaluate the
students' work.
A variety of
methods of
evaluation are
permitted. 
There may be
only a
cooperative
incentive, with
each person in a
group
 receiving 
the same grade.
There may be
both an
individualistic
and a
cooperative
incentive, 
with individual
grades for each
student and a
group reward
based on the
combined 
grades of group
members, or
students may
receive
individual
grades with
bonus 
points based on
how many
members of
their group
reached a
criterion.
19. 18. Assess
group
functioning
through
ongoing
observation and
discussion of
group pro- 
cess.

Jigsaw Methods
In original Jigsaw, developed by Aronson and colleagues ( 1978), interdepen- 
dence among students is promoted by giving each student in a learning group 
access to information comprising only one part of a lesson. Students are then 
accountable to their Jigsaw group for teaching that part of the lesson to the rest 
of the Jigsaw group members. In addition, the students from the different groups, 
each having the same material to learn, meet in counterpart groups to discuss 
and learn their part of the lesson before attempting to teach the material to the 
students in their Jigsaw groups. In this way, cooperation among students occurs 
not only within each Jigsaw group but also within the counterpart group.

Cooperative skills are emphasized and taught directly when using the Jigsaw 
method. In addition, students are encouraged constantly to evaluate group pro- 
cesses; however, there is no specific group reward for achievement or for the 
use of cooperative skills. The incentive structure in Jigsaw is individualistic: 
students' grades are based on individual examination performance. Therefore,

-3-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 3.

students are individually accountable for learning the entire lesson, although 
there is no group incentive for doing so.

Jigsaw II ( Slavin 1980) is an adaptation of original Jigsaw in which there is 

competition among groups called learning teams. There are both cooperative 
and individualistic incentives in Jigsaw II. The cooperative incentive operates 
as teams compete for specific group rewards such as group recognition in a 
class newsletter and special privileges provided to all members of a winning 
team. These group rewards are based on individual performance. Points are 
earned for one's team by improving performance relative to individual perfor- 
mance on previous quizzes. Each student, therefore, has an equal chance to 
contribute to the team score, and each student is individually accountable for 
performance. The individualistic incentives are grades based on actual scores 
obtained on the quizzes.

Jigsaw III, described by Kagan ( 1986), is a recent adaptation of Jigsaw de- 

veloped by Gonzalez and Guerrero ( 1983). This adaptation is uniquely de- 
signed for bilingual classrooms to increase interaction among students of vary- 
ing degrees of English proficiency. Each Jigsaw III group contains a student 
who speaks English, a student who does not speak English, and a student who 
is bilingual. All materials are bilingual. Team building and evaluation of group 
processes are particularly emphasized.

Student Teams -- Achievement Divisions (STAD)

An essential component of STAD, developed by Slavin ( 1980), is competi- 
tion among groups. Students work in their groups to drill and tutor each other 
to prepare for the competition. The competition factor has been advocated as a 
component of STAD that facilitates peer support and group norms for achieve- 
ment ( Slavin 1980). More recently, however, Slavin (1983a) has suggested that 
a specific group reward given for individual learning is the crucial factor in 
facilitating peer norms and sanctions for achievement. Slavin's more current 
view is that group competition is no more effective in providing a group reward 
than are other ways of providing team recognition.

In STAD the typical group reward for winning the competition is recognition 
in a class newsletter, but more tangible rewards may be given. Because of the 
competition it is essential that teams be matched evenly according to ability of 
team members. Slavin ( 1980) has outlined an objective procedure for group 
assignment that ensures even matching. Individual accountability to the group 
is accomplished in STAD by having each member's score on a quiz contribute 
to the team score. In order for each student to have an equal opportunity to 
contribute to the team score, students' scores are adjusted so that points con- 
tributed to the team are based on improvement over previous performance. The 
unadjusted scores are the basis for individual grades.

-4-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 4.

Teams-Games-Tournaments
TGT was developed by DeVries and Slavin ( 1978) and is similar to STAD 
except in the method of competition among teams and the resulting accounta- 
bility to teams. Students of comparable ability from different teams compete 
face to face in tournaments. Comparable ability among competitors is main- 
tained by having tournament winners compete with students of higher ability 
in the next tournament, while tournament losers compete with students of lower 
ability in the next tournament. Quizzes and tests are given to obtain individual 
grades for students.

Team-Assisted Individualization or Team-Accelerated Instruction

TAI ( Slavin 1985) was designed to combine the motivational incentive of 
group rewards with an individualized instructional program appropriate for the 
level of skills possessed by each student. In this method, as in most other 
methods of cooperative learning, small learning groups are composed of stu- 
dents of varying ability. The unique aspect of this method is that each student 
works on an individualized unit of mathematics instruction. Team members use 
answer sheets to check each other's worksheets and practice tests, and are re- 
sponsible for making sure that their teammates are prepared to take the final 
test for each unit. Discussion and peer tutoring occur because students are re- 
quired to ask their team members for help before they ask for help from the 
teacher. The teacher, besides acting as a resource person for the cooperative 
learning groups, takes students out of their teams for five to six minutes daily 
to give instruction to groups of students who are at about the same level in the 
curriculum.

Both individual accountability and specific group rewards are built into TAI. 
Team scores are computed from the average number of units covered by team 
members in a four-week period and on their scores on their final unit tests. 
Teams whose scores meet an objective criterion receive a group reward in the 
form of a certificate.

Group Investigation
The Group Investigation method, developed by Sharan and colleagues ( Sharan 
et al. 1984; Sharan and Hertz-Lazarowitz 1980; Sharan and Sharan 1976), em- 
phasizes student self-regulation of learning activities. Group investigation in- 
cludes both a cooperative group process, or what Sharan and Hertz-Lazarowitz 
( 1980) call cooperation in means, and a cooperative goal structure. This em- 
phasis is consistent with Slavin's ( 1983a, 1983b) conclusions that both a co- 
operative task structure and a cooperative incentive structure are necessary for 
improving achievement in a cooperative learning situation. There are four ma-

-5-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 5.

or dimensions characteristic of this method. First, the class is divided into a 

number of groups, with each group studying a different aspect of a general 
topic. Second, the topics for study are multifaceted for a meaningful division 
of labor that promotes interdependence among group members. Third, there is 
multilateral communication among students who, in addition to simply gather- 
ing information, must plan, coordinate, evaluate, analyze, and integrate their 
work with other students. Fourth, the teacher must adopt an indirect style of 
classroom leadership, acting as a resource person, providing direction and clar- 
ification as needed, and creating a stimulating classroom environment.Group investigation is
a highly structured method with six specific stages of 
implementation. Student involvement occurs in every stage, from selection of 
the topics for study to evaluation of student learning. The stages are as follows:
1. The teacher delineates a general topic area, and subtopics are identified through class 
discussion. Students then form small groups of two to six students. Group formation 
is based upon student interest in a particular subtopic, but heterogeneity of gender, 
 ethnicity, and ability level is strongly encouraged. In this stage three types of goals 
need to be considered. First, instructional goals are pursued as students research the 
topic, delineate the most crucial subtopics for understanding the general topic area, 
and classify suggestions for group study. Second, organizational goals are pursued 
as students form groups and make decisions about how the various groups should 
maintain and exchange information. Third, social goals are pursued as students help 
each other to match subtopics to individualized interests and establish patterns of 
mutual assistance and support.
2. Students collaborate in planning how to carry out the investigation of their subtopic. 
Division of labor is encouraged to promote interdependence and individual account- 
ability to the group. Students are taught to focus on the purpose of the investigation. 
There is an emphasis on possibilities for application of the knowledge acquired, as 
well as on what and how to study. A premise of this method is that focusing on the 
what, the how, and the why of an investigation will stimulate personal involvement 
and help students to perceive the relevance of learning.
3. Students implement their plans. The teacher arranges a wide variety of informational 
sources, both within and outside of school. Important sources of information are 
learning centers that the teacher has developed and which contain resources relevant 
to the general topic area. Frequent group discussions are promoted for social facili- 
tation, cognitive stimulation, and clarification of information. This will probably be 
the longest stage of the investigation, and the allotted time may need to be adjusted 
so that students have sufficient opportunity to complete their work plans. In addition, 
the teacher may need to stop a group's work intermittently to teach necessary group 
process skills or study skills.
4. Students collaborate in analyzing and evaluating the information they have gathered. 
Groups must abstract the essential aspects of their investigation and plan a presen- 
tation that will be instructive as well as interesting to the rest of the class. To coor- 
dinate the presentations, a steering committee is formed that is composed of a rep- 
resentative from each group. The steering committee coordinates the time schedules
-6-
Questia, a part of Gale, Cengage Learning. www.questia.com 

Publication Information: Book Title: Cooperative Learning: Theory and Research. Contributors: Shlomo Sharan - editor.
Publisher: Praeger. Place of Publication: Westport, CT. Publication Year: 1990. Page Number: 6.