Journal of Behavioral Education, Vol. 1, No. 1, 1991, pp.

105-115

Cooperative Learning and Group Contingencies

Robert E. Slavin, Ph.D. 1,2

Accepted." August 13, 1990. Action Editor: Thomas G. Haring

This paper discusses the similarities and differences between cooperative learning and group contingencies. Cooperative learning refers to any methods" in
which students work together to help one another learn, while group contingencies refer to rewarding students based on the performance of a group. Research on the achievement effects of cooperative learning finds that these
methods are effective primarily when they incorporate group contingencies,
when groups are rewarded based on the average of their members' individual
learning performances. The use of group contingencies within cooperative
learning is hypothesized to motivate students to do a good job of explaining
concepts and skills to their groupmates, and elaborated explanation is the principal behavior found to account for achievement gains in cooperative learning.
KEY WORDS: cooperative learning; group contingencies; achievement gains; motivation.

INTRODUCTION
Cooperative learning methods are instructional techniques in which students work in small groups to help one another learn academic material. The
use of these methods has been increasing rapidly in use at all instructional
levels, from elementary school to college, and in every school subject.
There are several different forms of cooperative learning that are widely
used and/or extensively researched. Some cooperative methods only involve
IPrincipal Research Scientist and Director, Elementary School Program, Center for Research
on Effective Schooling for Disadvantaged Students, The Johns Hopkins University, Baltimore,
Maryland.
2Correspondence should be directed to Robert Slavin, Director, Elementary School Program,
Center for Research on Effective Schooling for Disadvantaged Students, The Johns Hopkins
University, 3505 N. Charles Street, Baltimore, Maryland 21218.
105
1053-0819/91/0300-0105506.50/09 1991 HumanSciencesPress, Inc.

106

Slavin

changes in the task structure of the classroom, allowing or encouraging students to work together on academic tasks but leaving intact the traditional
system of formal and informal incentives (e.g., grades and praise, respectively). Other cooperative methods also change the classroom reward structure.
This can be done in many ways. In our own Student Team Learning methods
(Slavin, 1986), students work in four-member, heterogeneous learning teams.
The teams can earn certificates or other forms of recognition based on the
learning of all team members, as demonstrated on individual quizzes, essays,
compositions, or other assessments taken without teammate help. In methods
developed by David and Roger Johnson (1987), students may be graded
based on the quality of one group product, the sum or average of individual
quizzes, or randomly selected individual work. Other cooperative methods
affect the classroom reward structure more indirectly. For example, such
methods as Jigsaw Teaching (Aronson, Blaney, Stephan, Sikes, & Snapp,
1978) and Group Investigation (Sharan & Shachar, 1988) have students
divide an overall group topic or task into individual subtasks. Even if the
reward structure is not directly changed in such programs, the fact that students must depend on their groupmates for critical information or a key element of a group project means that the classroom reward structure has
become more interdependent (and less competitive or independent).

GROUP CONTINGENCIES
When cooperative learning involves changes in the classroom reward
structure, it is using strategies familiar to behavior analysts: Group contingencies. Definitions of group contingencies vary (see Litow & Pumroy,
1975), but in the context of cooperative learning the term means incentive
systems in which reinforcement is provided to members of a group based
on the performance of the group as a whole. Group contingencies have
been studied extensively in the applied behavior analysis tradition. For example, one of the earliest classroom group contingencies to be studied was
the Good Behavior Game (Barrish, Saunders, & Wolf, 1969). In this program, students were randomly assigned to two large teams. When the
teacher saw any student disobeying class rules, the whole team received a
check mark on the chalkboard. If a team had fewer than five check marks
in a period, all team members could take part in a free-time activity at the
end of the day. If both teams got more than five check marks, the team
that got fewer marks would receive the free time. This program had an
immediate and dramatic effect on student behavior. Similar findings have
been obtained in dozens of classroom studies (Cavanagh, 1984; Hayes,
1976; Litow & Pumroy, 1975).

Cooperative Learning

107

The theory behind group contingencies hypothesizes a two-step

process (see Slavin, 1977). First, the group is rewarded if it collectively
meets some standard. Second, the members of the group apply social sanctions to one another to encourage group members to do what is necessary
to ensure that the group will be successful. It is actually these interpersonal
reinforcers and punishers that are hypothesized to have the most important
effects on student behavior.
Although group contingencies have most often been applied to increasing student compliance with class rules, a few studies have established the
effects of these methods on student achievement. Several (e.g., Axelrod &
Paluska, 1975; Lovitt, Guppy, & Blattner, 1969) used immediate recall or
accuracy in classroom tasks as dependent measures, but others (e.g.,
Cavanagh, 1984; Hamblin, Hathaway, & Wodarsld, 1971; Jacobs, 1970; VanHouten, 1980) found that group contingencies were more effective than individual contingencies or untreated control conditions for increasing student
performance on achievement tests. For example, Jacobs (1970) randomly
assigned fourth graders to five groups: no rewards, random rewards, individual rewards, group rewards (based on the behavior of the entire class),
and combined individual plus group rewards. All students used programmed
reading materials. After 11 weeks, students were assessed on the Stanford
Achievement Test. All of the reinforcement conditions resulted in greater
achievement than the control group, but the group rewards were considerably more effective than the individual rewards. Similarly, Cavanagh (1984)
compared Team Assisted Individualization (Slavin, Madden, & Leavey,
1984), an individualized mathematics program that uses cooperative teams
and group rewards, to a form of the program that was identical except that
it used individual rewards. Students who received group rewards finished
substantially more units and achieved more on a standardized mathematics
test than did the individually rewarded students.
Research in the applied behavior analysis tradition has also investigated peer tutoring (e.g., Greenwood et al., 1987). However, peer tutoring
is different from cooperative learning in one fundamental respect. In peer
tutoring, the tutor is assumed to know the material being taught, while in
cooperative learning, all students are equally assumed to be learners, and
therefore group activities usually follow a lesson from the teacher.

COOPERATIVE LEARNING AND GROUP CONTINGENCIES

As noted earlier, the degree of overlap between cooperative learning
and group contingencies depends on the extent to which cooperative
methods explicitly alter the classroom reward structure. Our own Student

108

Slavin

Team Learning methods most clearly employ group contingencies (see

Slavin, 1987a), and other cooperative methods do so as well. However, even
when the degree of correspondence is highest, there are significant differences in emphasis between group contingencies and cooperative learning. The most obvious is the concern of cooperative learning for task
structure; all cooperative learning methods place considerable emphasis on
the types of task-related interactions engaged in by students in their small
groups. Theories underlying group contingencies emphasize motivational issues, while those underlying cooperative learning emphasize cognitive as
well as motivational issues (see Slavin, 1987b, 1989). For example, cognitive
theories underlying the effectiveness of cooperative learning for increasing
student achievement include learning by teaching or cognitive elaboration
(Webb, 1985; Dansereau, 1988), Vygotskian or Piagetian theories emphasizing learning through cognitive conflict and students operating in each
others' proximal zones of development (Murray, 1982; Mugny & Doise,
1978), and opportunities for students to provide individualized assessment,
immediate feedback, and personally tailored assistance to groupmates
(Slavin, 1989, 1990). Cooperative learning theorists would hold that within
any particular reward structure, variations in task structure would have
major consequences for student achievement, and would emphasize interactions among students as an essential causal link in any theory relating
cooperative learning to achievement. In contrast, research in the applied
behavior analysis tradition has often studied group contingencies in which
students do not actually interact at all.

DO GROUP CONTINGENCIES CONTRIBUTE

TO EFFECTS OF COOPERATIVE LEARNING?
While cooperative learning and group contingencies are conceptually
distinct, they do usually overlap in practice when learning of school subjects
is involved. Much research and debate in cooperative learning focuses on
a set of questions which, in the language of applied behavior analysis, would
be phrased as "what effects do group contingencies have on the instructional effectiveness of cooperative learning, and which particular forms of
group contingencies are maximally effective?" Shifting back to the language
of cooperative learning research, this question focuses on the necessity of
group goals and individual accountability for achievement effects. Cooperative methods which use group goals are ones in which students are
rewarded based on the learning or performance of the group as a whole.
In other words, group goals are synonymous with group contingencies
(Slavin, 1987a). Individual accountability refers to the degree to which

Cooperative Learning

109

Table 1. Effect Sizes from Elementary and Secondary

Studies of Achievement Effects of Cooperative Learninga

Group goals and individual

accountability
Group goals only
Individual acct. only (task
specialization)
No group goals or individual acct.

Median effect
size

Numberof
studies

+.32

35

+.04
+.12

8
9

+.05

aAdapted from Slavin (1990). Effect sizes are the difference

between cooperative learning and control classes on achievement measures divided by the posttest standard deviation. Includes only methodologicallyadequate studies of at least four
weeks' duration.

group success depends on the individual learning of all group members.

For example, in our Student Team Learning methods, individual accountability is accomplished by having group success depend on the sum or
average of performances on individually administered quizzes or other assessments. The idea is that because the only way for teams to succeed is
for all team members to individually know the material, team members are
motivated to direct their efforts toward teaching each other, not simply
giving answers. Providing and receiving elaborated explanations are the behaviors within cooperative groups most positively associated with learning
(Webb, 1985).
The importance of group goals and individual accountability is clearly
shown in Table 1, which presents updated data from a recent review of the
cooperative learning literature (Slavin, 1990). The studies from which the
table was derived compared cooperative learning to randomly selected or
matched control groups on measures of the objectives pursued equally by
both groups. Study durations were at least four weeks (median = 10 weeks).
Table 1 shows that the success of cooperative learning in terms of
increasing student achievement depends substantially on the provision of
group goals and individual accountability. Methods which incorporate
group goals and individual accountability include Student Teams-Achievement Divisions (Slavin, 1986), Teams-Games-Tournament (De Vries &
Slavin, 1978), Cooperative Integrated Reading and Composition (Stevens,
Madden, Slavin, & F a r n i s h , 1987), and Team Assisted IndividualizationMathematics (Slavin, Madden, & Leavey, 1984).
In contrast to the relatively positive effects of methods which use
group goals and individual accountability, those which use group goals but

110

Slavin

not individual accountability have generally been ineffective in terms of

increasing student achievement. For example, in the Johnsons' (1987)
methods, students work together to complete a single worksheet and are
praised, rewarded, and/or graded on the basis of this common worksheet.
On measures of achievement, these methods produced no better achievement than individualistic or traditional methods (e.g., Johnson, Johnson,
& Scott, 1978). Two studies did find positive achievement effects of a form
of this approach in which students were graded not on the basis of one
worksheet, but on the 'average of individual quiz scores, which ensures individual accountability (Humphreys, Johnson, & Johnson, 1982; Yager,
Johnson, Johnson, & Snider, 1986). However, it is important to note that
these were highly artificial experiments. In them, teachers did not present
lessons to students, but only helped individuals with worksheets, so that in
the "individualistic" control groups students had no resources other than
the worksheets and occasional teacher assistance to help them understand
the material.
Another major category of cooperative learning methods uses task
specialization, which means that each student has a unique task within an
overall group objective. For example, Jigsaw Teaching (Aronson, Blaney,
Stephan, Sikes, & Snapp, 1978) assigns each student a topic on which they
are to become an "expert." This method has not generally been instructionally effective. However, a form of Jigsaw which does use group goals
and individual accountability called Jigsaw II (Slavin, 1986) has been instructionally effective (Mattingly & VanSickle, 1990). Another effective
form of cooperative learning which uses task specialization is Group Investigation (Sharan & Shachar, 1988), in which students take on subtasks
within an overall group task. In contrast to Jigsaw, Group Investigation
bases individuals' evaluations on the group's product or report, so this may
in fact be an instance of group goals and individual accountability.
Finally, studies of methods which provide neither group goals nor individual accountability find few achievement benefits for this approach.
One example is the Groups of Four mathematics program in which students
work together to solve complex math problems (Burns, 1981; Johnson &
Waxman, 1985).
Several component analyses have specifically examined the achievement effects of group rewards and individual accountability. Two studies
(Hulten & De Vries, 1976; Slavin, 1980b) found that providing recognition
to student teams based on the sum of their individual learning increased
student achievement even if students were not permitted to interact in class.
A German study (Huber, Bogatski, & Winter, 1982) found that providing
students an opportunity to study together did not increase their achievement, but adding group rewards based on individual learning did lead to

Cooperative Learning

111

enhanced achievement. Finally, Cavanagh (1984) found that students using

an individualized instruction method in which they were assigned to work
in small teams accurately completed more units and achieved more if they
received group rewards based on unit completion than if they received individual rewards.

WHY ARE GROUP GOALS AND I N D M D U A L

ACCOUNTABILITY ESSENTIAL?
Why are group goals and individual accountability essential to the
achievement effects of cooperative learning? There are several reasons.
Group goals are necessary to motivate students to help each other learn,
giving them a stake in one another's success. Without group goals, it is unlikely that students would engage for long in the elaborated explanations
that have been found to be essential to the achievement effects of cooperative learning (Webb, 1985). Further, group goals may help overcome
students' reluctance to ask for help or provide help to one another; without
an overriding group goal, it may be embarrassing to ask for or offer serious
help. Without individual accountability, it is likely that one or two group
members may do all the work, and that if group members perceived to be
low achievers contribute ideas or ask for help, they may be ignored. When
the group completes a single group worksheet or product, there is a danger
that some group members' efforts will not be needed or may even interfere
with the groups' success. For example, in a heterogeneous four-member
group, the two most able students could probably complete a group
worksheet by themselves as well as or better than if they actively involved
the two less able group members. In contrast, if the group's success depends
on the individual learning of each group member, then group members are
motivated to attempt to ensure that all group members master the material
being studied.

DOES PEER INTERACTION CONTRIBUTE TO EFFECTS

OF GROUP CONTINGENCIES?
The research on practical cooperative learning methods clearly supports the position that cooperative reward structure, or group contingencies, based on the individual learning of group members are necessary for
the success of these methods in improving student achievement. However,
does the peer interaction central to cooperative learning add to the effectiveness of group contingencies? Here the evidence is more indirect, but

112

Slavin
Table 2. Critical Features of Group Contingencies and Cooperative Learninga

Cooperative
Group contingencies
learning
Cooperative interaction
Group rewards
Individual accountability

Optional
Essential
Essential

Essential
Optional
Optional

Effective cooperative learning

methods
Essential
Essential
Essential

aFrom Slavin (1987a).

there are indications that peer interaction is important to the success of

cooperative strategies. As noted earlier, Webb (1985) and Peterson &
Janicki (1979) have found that the students who learn best from cooperative interaction are those who give and receive elaborated explanations (i.e.,
are not simply given answers or ignored by their groupmates). This finding
mirrors that of the Piagetian tradition, which finds that if two nonconserving children actively work together both can become conservers (Ames &
Murray, 1982), but if they simply accept a higher-quality answer to a conservation task they will not make cognitive progress (e.g., Mugny & Doise,
1978). However, what is critical in cooperative learning is the combination
of group contingencies and high-quality peer interactions. Students are
motivated to engage in elaborated, cognitively involving explanations and
discussions if the learning of their groupmates is made important by the
provision of group rewards based on individual learning performances
(Slavin, 1983b). For example, several studies (e.g., Hamblin, Hathaway, &
Wodarski, 1971; Slavin, 1980b) have established that active peer discussion
and peer explanation within cooperative groups are much more frequent
under conditions in which group rewards are based on individual learning
than under conditions in which collaborative work is encouraged but there
are no consequences based on group members' learning.
The relationship between group contingencies and cooperative learning is summarized in Table 2 (from Slavin, 1987a). The table notes that
group rewards and individual accountability are essential in group contingencies; group members must be aware of the individual contributions
made by each groupmate if they are to be able to apply the interpersonal
sanctions held to be central to the effectiveness of the group contingency.
In contrast, cooperative learning emphasizes cooperative interaction but
may or may not use group rewards or individual accountability. When all
three elements are present, the distinction between group contingencies
and cooperative learning is more or less semantic. Researchers from the
applied behavior analysis tradition would probably insist on the term
"group contingencies," while those from a social psychological or humanis-

Cooperative Learning

113

tic background would use "cooperative learning." Yet it is just this form
of cooperative learning/group contingencies, emphasizing interactions,
group rewards, and individual accountability, that has the greatest research
support in terms of student achievement.
The cooperative learning m o v e m e n t has created an interesting
phenomenon, in which humanistic educators and psychologists are championing classroom methods that could be described completely in behavioral language. For example, a book by the humanistic psychiatrist William Glasser
(1988) attacks behavioral learning theory but proposes widespread use of
cooperative learning teams. However, the attraction of cooperative learning
for many humanistic educators probably lies not so much in accelerating student achievement as in the consistently found positive effects of cooperative
learning on such variables as race relations, attitudes toward mainstreamed
classmates, self-esteem, and other nonacademic outcomes (see Slavin, 1990).
In contrast to achievement effects, these important outcomes do not appear
to depend on the use of group rewards for individual learning.
Despite the consistent evidence supporting the use of group rewards
based on group members' learning in cooperative learning, there are many
important questions yet to be resolved. Conclusions about the centrality of
these components are based on comparisons of achievement effects of alternative models, not on direct observation of changes in student behavior. To
confirm the arguments made in this article it would be important to contrast
groups working under group contingencies to those simply asked to work
together, to see if the quantity and quality of peer interactions are affected
by the reward structures under which they take place. Also, it is possible that
aspects of cooperative learning other than cooperation per se (e.g., clear objectives, frequent assessment) account for part of the achievement effect.
Cooperative learning provides a unique opportunity for collaborative
efforts between researchers in the applied behavior analysis tradition and
those in the social psychological or mainstream educational psychology
traditions. While differences in theoretical backgrounds, research methods,
and terminology are important between these different traditions, cooperative learning provides an interesting point of intersection. Much work still
lies ahead to understand fully how and why cooperative learning works and
to develop ever more effective strategies: collaboration among researchers
from different traditions is sure to enrich this work.

ACKNOWLEDGMENTS
This paper was written under funding from the Office of Educational
Research and Improvement, U.S. Department of Education (Grant No.

114

Slavin

OERI-R-117-R90002). However, any opinions expressed are those of the

author, and do not represent OERI positions or policies.

REFERENCES
Ames, G. J., & Murray, F. B. (1982). When two wrongs make a right: Promoting cognitive
change by social conflict. Developmental Psychology, 18, 894-897.
Aronson, E., Blaney, N., Stephan, C., Sikes, J., & Snapp, M. (1978). The jigsaw classroom.
Beverly Hills, CA: Sage.
Axelrod, S., & Paluska, J. (1975). A component analysis of the effects of a classroom game
on spelling performance. In E. Ramp & G. Semb (Eds.), Behavior analysis: Areas of research and application. Englewood Cliffs, NJ: Prentice-Hall.
Barrish, H. H., Saunders, M., & Wolf, M. M. (1969). Good behavior game: Effects of individual contingencies for group consequences on disruptive behavior in a classroom.
Journal of Applied Behavior Analysis, 2, 119-124.
Burns, M. (1981, September). Groups of four: Solving the management problem. Learning,
46-51.
Cavanagh, B. R. (1984). Effects of interdependent group contingencies on the achievement of
elementary school children. (Doctoral dissertation, University of Maryland, 1984). Dissertation Abstracts International, 46, 1558.
Dansereau, D. F. (1988). Cooperative learning strategies. In C. E. Weinstein, E. T. Goetz,
& P. A. Alexander (Eds.), Learnblg and study strategies: lssues in assessment, instruction,
and evaluation (pp. 103-120). New York: Academic Press.
DeVries, D. L., & Slavin, R. E. (1978). Teams-Games-Tournament (TGT): Review of ten
classroom experiments. Journal of Research and Development in Education, 12, 28-38.
Glasser, W. (1988). Control theoly in the classroom. New York: Harper & Row.
Greenwood, C. R., Dinwiddie, G., Terry, B., Wade, L., Stanley, S., Thibadeau, S., & Delaquadri, J. (1984). Teacher versus peer-mediated instruction: An ecobehavioral analysis.
Journal of Applied Behavior Analysis, 17, 521-538.
Hamblin, R. L., Hathaway, C., & Wodarski, J. S. (1971). Group contingencies, peer tutoring,
and accelerating academic achievement. In E. Ramp & W. Hopkins (Eds.), A new direction for education: Behavior analysis (pp. 41-53). Lawrence: University of Kansas.
Hayes, L. (1976). The use of group contingencies for behavioral control: A review. Psychological Bulletin, 83, 628-648.
Huber, G. L., Bogatzki, W., & Winter, M. (1982). Kooperation als Ziel schulischen Lehrens
und Lehrens. Tubingen, West Germany: Arbeitsbereich Padagogische Psychologie der
Universitat Tubingen.
Hulten, B. H., & DeVries, D. L. (1976). Team competition and group practice: Effects on student
achievement and attitudes (Report No. 212). Baltimore, MD: Johns Hopkins University,
Center for Social Organization of Schools.
Humphreys, B., Johnson, R., & Johnson, D. W. (1982). Effects of cooperative, competitive,
and individualistic learning on students' achievement in science class. Journal of Research
in Science Teaching~ 19, 351-356.
Jacobs, J. F. (1970). A comparison of group and bldividual rewards in teaching reading to slow
learners. Unpublished paper, University of Florida (ERIC Document Reproduction Service No. ED 044 265).
Johnson, D. W., & Johnson, R. T. (1987). Learnhlg together and alone (2nd Ed.). New Jersey:
Prentice-Hall.
Johnson, D. W., Johnson, R. T., & Scott, L. (1978). The effects of cooperative and individualized instruction on student attitudes and achievement. Journal of Social Psychology, 104,
207-216.

Cooperative Learning

115

Johnson, L. C., & Waxman, H. C. (1985, March). Evaluating the effects of the "groups of four"
program. Paper presented at the annual meeting of the American Educational Research
Association, Chicago.
Litow, L., & Pumroy, D. K. (1975). A brief review of classroom group-oriented contingencies.
Journal of Applied Behavior Analysis, 8, 341-347.
Lovitt, T. C., Guppy, T. E., & Blattner, J. E. (1969). The use of a free-time contingency to
increase spelling accuracy. Behavior Research and Therapy, 7, 151-156.
Mattingly, R. M., & VanSickle, R. L. (1990). Jigsaw H in secondary social studies: An experiment. Athens, GA: University of Georgia.
Mugny, G., & Doise, W. (1978). Socio-cognitive conflict and structurization of individual and
collective performances. European Journal of Social Psychology, 8, 181-192.
Murray, F. B. (1982). Teaching through social conflict. Contemporary Educational Psychology,
7, 257-271.
Peterson, P. L., & Janicki, T. C. (1979). Individual characteristics and children's learning in
large-group and small-group approaches. Journal of Educational Psychology, 71, 677-687.
Sharan, S., & Shachar, H. (1988). Language and learning in the cooperative classroom. New
York: Springer-Verlag.
Slavin, R. E. (1977). Classroom reward structure: An analytic and practical review. Review of
Educational Research, 47, 633-650.
Slavin, R. E. (1980b). Effects of student teams and peer tutoring on academic achievement
and time-on-task. Journal of Experimental Education, 48, 252-257.
Slavin, R. E. (1983b). When does cooperative learning increase student achievement?
Psychological Bulletin, 94, 429-445.
Slavin, R. E. (1986). Educational psychology: Theory into practice. Englewood Cliffs, N J: Prentice-Hall.
Slavin, R. E. (1987a). Cooperative learning: Where behavioral and humanistic approaches to
classroom motivation meet. Elementary School Journal, 88, 29-37.
Slavin, R. E. (1987b). Developmental and motivational perspectives on cooperative learning:
A reconciliation. ChiM Development, 58, 1161-1167.
Slavin, R. E. (1989). Cooperative learning and student achievement: Six theoretical perspectives. In M. L. Maehr & C. Ames (Eds.), Advances in motivation and achievement (Vol.
6) (pp. 161-178). Greenwich, CT:JAI.
Slavin, R. E. (1990). Cooperative learning: Theol3; research, and practice. Englewood Cliffs, NJ:
Prentice-Hall.
Slavin, R. E., Madden, N. A., & Leavey, M. (1984). Effects of Team Assisted Individualization
on the mathematics achievement of academically handicapped students and nonhandicapped students. Journal of Educational Psychology, 76, 813-819.
Stevens, R. J., Madden, N. A., Slavin, R. E., & Famish, A. M. (1987). Cooperative Integrated
Reading and Composition: Two field experiments. Reading Research Quarterly, 22, 433454.
Van Houten, R. (1980). Learning through feedback: A systematic approach for improving
academic performance. New York: Human Sciences Press.
Webb, N. (1985). Student interaction and learning in small groups: A research summary. In
R. E. Slavin, S. Sharan, S. Kagan, R. Hertz-Lazarowitz, C. Webb, & R. Schmuck (Eds.),
Learning to cooperate, cooperating to learn (pp. 147-172). New York: Plenum.
Yager, S., Johnson, R. T., Johnson, D. W., & Snider, B. (1986). The impact of group processing on achievement in cooperative learning. Journal of Social P~ychology, 126, 389-397.