Journal of Social Issues, Winter 1999 v55 i4 p647

Improving Intergroup Relations: Lessons Learned
From Cooperative Learning Programs.

Robert E.Slavin;

Robert Cooper.

Full Text: COPYRIGHT 1999 Plenum Publishing Corporation

Robert Cooper [*]

This article discusses the need for cooperative learning groups in
integrated schools in order to promote more cross-race relationships
than might otherwise be the case. We review research on 8 cooperative
learning procedures. Evidence for the effectiveness of these programs in
facilitating cross-race peer interaction is presented.

Improving intergroup relations among diverse groups of students is
becoming a high priority among educators. As schools become more
diverse and destructive conflict and violence become more common in
schools, there is an increasing concern that schools not become the
battlegrounds for the next wave of racial unrest in this country.
Undeniably, the schooling context for America's youth is increasingly
multicultural (Heath, 1995), and conflicts will be defined along racial
and ethnic lines (R. Cooper, 1996). Unfortunately, the vast majority of
research about intergroup relations in schools is now 15 to 20 years old
and focuses mostly on improving relations between Whites and Blacks
(Schofield, 1995). Little empirical evidence exists about intergroup
conflict in settings in which many different racial and ethnic groups
coexist and in which the boundaries between groups are blurred by
overlapping categories (McLoyd, 1990; Quintanilla, 1995).

Given the enormous diversity found today in many public schools, racial
and ethnic relations are much more complicated than they were just a
decade ago. Public education is one of the few social institutions
through which the entire texture of the American diversity can be
experienced. Consequently, intergroup relations are no longer affected
just by the competition for resources and attention but must now
consider the relative power and status of the racial and ethnic groups
involved. If diversity is to be viewed as an asset to be built upon in
schools, rather than a problem to be solved, we must learn more about
how schools can foster positive social relationships among students of
different racial and ethnic backgrounds.

Schools play a vital role in helping children and adolescents understand,
through various representations and practices, the ways in which
difference is constructed (Giroux, 1992). The challenge for educators is
to create the conditions under which students are likely to cross the
borders that delimit their narrow personal and social worlds and
provide opportunities to experience the worlds of those different from
them. Positive cross-ethnic interactions help students expand their own
self-identify and build an appreciation of difference at the same time.

One of the most innovative widely prescribed strategies to manage and
build upon the strength of the increasing diversity found in classrooms is
the use of cooperative learning techniques (Slavin, 1995a). Cooperative
learning involves small teams of students of varying academic
achievement levels employing a variety of learning activities that
promote academic success for each team member. Research on the
effects of cooperative learning has consistently found that the use of
such methods improves academic achievement as well as intergroup
relations (Lopez-Reyna, 1997; Slavin, 1991b, 1992, 1995b). In many
cases, cooperative learning provides students an opportunity to be
grouped not only heterogeneously by academic performance, but also
by race, gender, and language proficiency. When using cooperative
learning methods, students are asked to work in heterogeneous groups
to solve problems and complete tasks. The intent of cooperative work
groups is to enhance the academic achievement of students by
providing th em with increased opportunity for discussion, for learning
from each other, and for encouraging each other to excel.

Because cooperative learning groups encourage positive social
interaction among students of diverse racial and ethnic backgrounds,
they have great potential to facilitate the building of cross-ethnic
friendships and to reduce racial stereotyping, discrimination, and
prejudice. When students work cooperatively, they have the
opportunity to judge each other on merits rather than stereotypes
(McLemore & Romo, 1998). This article, expanded from an earlier
review (Slavin, 1997), presents a summary of the research that suggests
that intergroup relations in desegregated schools can be improved by
applying a variety of cooperative learning classroom interventions that
are based on Allport's (1954) contact theory.

Intergroup Relations in the School Context

Social science inquiry on race and intergroup relations has been
dominated by tenets of Gordon Allport's research. Allport's The Nature
of Prejudice (1954) has served as the basis for the study of intergroup
relations since the mid-1950s. Allport cited evidence that asserts that
when students of diverse backgrounds have the opportunity to work
and get to know one another on equal footing, they become friends and
find it more difficult to hold prejudices against one another (Slavin,
1991a, 1995b).

Although Allport' s contact theory has been updated and expanded
over the years (Cook, 1978; Hewstone & Brown, 1986; Pettigrew,
1986), positive cross-ethnic relationships among students are an
anomaly rather than the norm on many desegregated school campuses
above the elementary school level. It was assumed after the Brown
decision that desegregation would improve relations between students
of different ethnic backgrounds. Despite efforts by educators, policy
makers, and researchers, however, youth from different backgrounds
still have limited interactions in school settings (Khmelkov & Hallinan,
this issue; Romo & Falbo, 1996; Schofield, 1995; Slavin, 1995b).

In many schools, cross-ethnic interaction between students is superficial
and competitive (Slavin, 1995b). Outside the classroom, students
compete for limited positions on athletic teams, newspaper staffs, and
student governments --organizations that are oftentimes racially
identifiable and fail to provide opportunities for positive cross-ethnic
interactions. The limited contact between students of diverse
backgrounds fosters harsh stereotypes, and racial tensions persist
(Crain, Mahard, & Narot, 1982; Qakes & Wells, 1995). Negative
stereotyping is often used to justify maintaining hostility, contempt, and
resentment toward others (Lilli & Rehm, 1990). Unfortunately,
research shows that children, rather than being taught how to value and
celebrate diversity, are more apt to be taught that intolerance is an
acceptable reaction to diversity (Schwartz, 1996), which can lay a
foundation for racism in adulthood.

Cooperative Learning

Cooperative learning is a well-documented and frequently
recommended strategy for enhancing academic (Cohen & Lotan, 1997;
Slavin, 1995a, 1995b; Sharan, 1994), cognitive (Lotan & Whitcomb,
1998), and social (Slavin, 1995b; Stevens & Slavin, 1995) outcomes
for students. The term applies to a set of instructional strategies that
involve students working collaboratively in groups with little teacher
supervision (Deering, 1989). Cooperative learning methods attempt to
reduce competition or individualism in classrooms by rewarding
students based on the performance of all individuals in their group
(Aronson, Blaney, Stephan, Sikes, & Snapp, 1978; Johnson &
Johnson, 1981; Slavin, 1983). In some cooperative learning methods,
the group is awarded points or recognition based on the average
academic performance of each member of the group. Teachers often
delegate authority and responsibility for group management and learning
to the students (Cohen, 1994). The instructional methods used are
structured to give each stude nt a chance to make substantial
contributions to the team, so that the teammates will be equal, at least in
the sense of role equity specified by Allport. It is important to note that
group work does not in itself constitute cooperative learning (Johnson,
Johnson, & Holubec, 1993), but that cooperative learning groups place
emphasis on the academic learning success of each individual member
of the group (Slavin, 1995a).

One review of research on cooperative learning (Slavin, 1995a)
identified 52 studies conducted over periods of at least 4 weeks in
regular secondary schools (grades 6-12) that have measured effects of
student achievement. These studies all compared the effects of
cooperative learning with effects of traditionally taught control groups
on measures of the same objectives pursued in all classes. Teachers
and classes were either randomly assigned to cooperative or control
conditions, or they were matched on pretest achievement level and
other factors. Of these studies, 33 (63%) found significantly greater
achievement in cooperative than in control classes. Sixteen (31%)
found no differences, and in only three studies did a control group
significantly outperform the experimental group.

Cooperative learning methods explicitly use the strength of the
desegregated school--the presence of students of different races or
ethnicities--to enhance intergroup relations (Slavin, 1995b). When
teachers assign students of different races or ethnicities to work
together, students are sent a strong positive message regarding
cross-group interaction. Although increasing positive intergroup
relations may not be explicitly stated by teachers as a goal of
cooperative learning, it would be difficult for students to believe that the
teacher supports racial separation when the teacher has assigned the
class to multiethnic teams. Slavin (1995a) suggests that, at least in
theory, cooperative learning methods satisfy the conditions outlined by
Allport for positive effects of desegregation on race relations:
cooperation across racial lines, equal-status roles for students of
different races, contact across racial lines that permits students to learn
about one another as individuals, and the communication of unequivoc
al teacher support for interracial contact.

There are eight principal well-researched cooperative learning methods
that embody the principles of contact theory. These methods are
relatively easy to implement, widely applicable in terms of subject
matter and grade level, and easily integrated into an existing school
without additional resources. In most cases, these methods have been
shown to improve achievement as well as intergroup relations (see
Slavin, 1995a). Four of the methods were developed and evaluated at
the Center for Social Organization of Schools at Johns Hopkins
University. These are Student Teams--Achievement Divisions (STAD),
Teams-Games-Tournament (TGT; Slavin, 1986), Team-Assisted
Individualization (TAI; Slavin, Leavy, & Madden, 1984), and
Cooperative Integrated Reading and Composition (CIRC; Stevens,
Madden, Slavin, & Famish, 1987). A fifth technique, Jigsaw Teaching
(Aronson, Blaney, Stephan, Sikes, & Snapp, 1978), has been
evaluated in several desegregated schools and is widely used both in its
original form and as modified by Sla vin (1986) and by Kagan (1995).
Methods developed and assessed at the University of Minnesota
(Johnson & Johnson, 1981, 1994) have been studied in desegregated
schools, and Group Investigation (Sharan & Sharan, 1992) has been
studied in Israeli schools that include European and Middle Eastern
Jews. Detailed descriptions of a few methods follow.

The Cooperative Learning Group Method

STAD is one commonly used form of cooperative learning. For
example, seventh- and eighth-grade students in one study (Slavin,
1979) were assigned to groups of four or five members who varied in
gender and ethnicity. The groups also had a mix of high-, average-, and
low-performing students, reflecting the composition of the larger class.
Teammates met for two 40-mm periods each week for 10 weeks to
receive instruction and then to discuss and learn English language arts
material. To prepare for a weekly quiz, teammates were encouraged to
help each other learn the material. Students then answered test items
individually.

Reward structures can be of three types: individual rewards for
individual achievement, group rewards for group achievement, or group
rewards for individual achievement. In the traditional classroom,
individuals are rewarded for their individual achievement. When groups
complete assignments, the group product receives an evaluation;
however, not all members may contribute equally, and in fact there is a
tendency to allow higher performing students to complete the
assignment. Consequently, the reward structure used by STAD (Slavin,
1979) is a group reward to which each student contributes as a function
of the weekly test score pro-rated in comparison with his or her
achievement division (i.e., in comparison with other students who have
in the past performed at the high, average or low level). Another
strategy for assigning group rewards for individual achievement is TGT
(DeVries, Edwards, & Slavin, 1978). Students take part in a
tournament of skill-testing games at the end of each week. Each student
competes, as a representative of his or her group, against similar-level
students from other teams. Thus, students win points for their team if
their performance compares favorably with others at their level.
Although some educators are concerned about the outgroup
competition aroused during tournaments, outcomes are as positive as
when the comparison group is a more abstract division of similarly
achieving students (DeVries et al., 1978), at least when implemented
with adolescent students.

Cooperative learning procedures have also been implemented and
evaluated by participants in elementary schools (e.g., Lampe, Rooze, &
Tallent-Runnels, 1996). For example, third-grade classes in a
heterogeneous school district were the focus of one study, in which
teachers and students completed questionnaires concerning their
experiences with the procedure, and classroom dynamics were
observed (McManus & Gettinger, 1996). Twenty-six teachers (half of
the district's third-grade teachers) completed the questionnaire, saying
that they used cooperative learning groups daily in their classrooms.
Most had received specialized training in the technique. They used it for
a variety of subjects, reading being the most common (92%) and
spelling the least common (46%) activity. Although the most effective
reward structure identified by researchers is group rewards for
individual achievement, it was used by teachers in this study less
frequently than group rewards for group products and individual
rewards for individual achievement. Most but not all teachers assigned
students to heterogeneous groups; some used homogeneous groupings.
Regardless of the reward structure or the heterogeneity of groups,
teachers generally gave high ratings to academic, social, and
motivational outcomes of cooperative group learning in their classes. In
other words, teachers who used the procedures frequently felt that
cooperative group learning was an effective academic and motivational
tool and that it increased peer instruction and cooperation. Students
also evaluated the academic outcomes (e.g., "easier for me to learn")
and the motivational outcomes positively (e.g., "it's fun to work with
other kids"), but some were bothered by the social conflicts that arose
during group activities (e.g., "kids don't always listen to each other and
get along"). Not surprisingly, the cooperative behaviors of helping and
listening to others may take some time to develop; it was for this reason
that the programs were developed in the first place.

Even though students were not particularly comfortable with conflict
and self-centered behavior, most of the behaviors exhibited by third
graders in their groups were productive. McManus and Gettinger
(1996) observed the groups in two classrooms over a 6-week period.
Behaviors of individual children were coded every 1-min interval.
Negative social behaviors such as conflicts and acting out occurred in
only 14% of the intervals. Listening and watching occurred in 68% of
the intervals; teaching behaviors, such as showing examples from a
book, demonstrating how to do something, and providing answers or
ideas, occurred 45% of the time. Positive, task-related social
interactions were also common. In short, students at this grade level
were communicating and cooperating with their peers.

Despite high levels of verbal interaction among group members,
concerns are sometimes raised about equal opportunities for all children
to make contributions. Cohen and Lotan (1997), for example, point out
that children often re-create the status differences of the larger society
in their groups. Students possessing high-status characteristics tend to
command more attention and participate more actively than those
possessing lower status characteristics. Individual differences in
participation, when they coincide with social status differences,
undercut the goal of creating equal status in cooperative groups. To
remedy this situation, Cohen and her colleagues train teachers on how
to raise the status of a child by making a pointed and public comment
on the child's skill. This has been effective in raising the levels of
participation in lower status children. Another concern has been the
training of teachers to use cooperative learning techniques in their
classroom. Teachers need to receive instruction and pr actice in the
technique in order to give them a sense of efficacy in being able to
promote the learning of students with different needs (e.g., Shachar &
Shmuelevitz, 1997; Sharan & Sharan, 1992).

Research on Cooperative Learning and Intergroup Relations

Many field experiments have evaluated the effects of cooperative
learning methods on intergroup relations. The current review
emphasizes studies in which the methods were compared to control
groups in elementary or secondary schools for at least 4 weeks (median
duration = 10 weeks) and in which appropriate research methods and
analyses were used to rule out obvious bias. Study ns ranged from 51
to 424 (median = 164), grade levels from 4 to 12, and percentage
minority students from 10% to 61%. Most of the studies used
sociometric indices (e.g., "Who are your friends in this class?"), peer
ratings, or behavioral observation to measure intergroup relations as
pairwise positive relations between individuals of different ethnic
backgrounds. Some studies measured intergroup relations in terms of
attitudes toward various ethnic groups. Several other studies used
self-reports for questions such as "Who have you helped in this class?"
Because only students in the cooperative learning classes were
instructed to help t heir classmates, such measures are biased toward
the cooperative learning treatments; thus, the results of these measures
are not discussed here. Also, observations of cross-racial interaction
during the treatment classes, another measure of implementation rather
than outcome, are not considered here.

The experimental evidence on cooperative learning has generally
supported the main tenets of contact theory (Allport, 1954). With only
a few exceptions, this research has demonstrated that, when the
conditions outlined by Allport are met in the classroom, students are
more likely to have friends outside their own racial groups than they
would in traditional classrooms, as measured by responses to such
sociometric items as "Who are your best friends in this class?"

Student Teams--Achievement Divisions. In STAD (Slavin, 1995a), the
teacher presents a lesson, and students then study worksheets in
four-member teams. Following this, students take individual quizzes,
and team scores are computed based on the degree to which each
student has improved over his or her own past record. The team scores
are recognized in newsletters.

The evidence linking STAD to gains in cross-racial friendships is strong.
In two studies, Slavin (1977, 1979) found that students who had
experienced STAD over periods of 10--12 weeks gained more in
number of cross-racial friendships than did control students. Slavin and
Oickle (1981) found significant gains in White friendships with African
Americans as a consequence of STAD but found no difference in
African American friendships with Whites. Kagan and colleagues
(Kagan, Zahn, Widman, Schwarzwald, & Tyrell, 1985) found that
STAD (and TGT) reversed a trend toward ethnic polarization of
friendship choices among Anglo, Latino, and African American
students. Sharan and colleagues (1984) found positive effects of STAD
on ethnic attitudes of both Middle Eastern and European Jews in Israeli
schools.

Slavin's (1979) study included a follow-up into the next academic year,
in which students who had been in the experimental and control classes
were asked to list their friends. Students in the control group listed an
average of less than 1 friend of another race, 9.8% of all of their
friendship choices; those in the experimental group named an average
of 2.4 friends outside their own race, 37.9% of their friendship choices.
The STAD research covered grades 2-8 and took place in schools
ranging from 13% to 61% minority.

Teams-Games-Tournaments. TGT is essentially the same as STAD in
rationale and method. However, it replaces the quizzes and
improvement score system used in STAD with a system of academic
game tournaments in which students from each team compete with
students form other teams of the same level of past performance to try
to contribute to their team scores (see Slavin, 1986).

DeVries et al. (1978) summarized data analyses from four studies of
TGT in desegregated schools. In three of these, students in classes that
used TGT gained significantly more friends outside their own racial
groups than did control students. In one, no differences were found.
The samples involved in these studies varied in grade level from 7 to 12
and in percentage of minority students from 10% to 51%. In addition,
Kagan et al. (1985) found positive effects of TGT on friendship choices
among African American, Mexican American, and Anglo students.

Team-Assisted Individualization. TAI combines the use of cooperative
teams (like those in STAD and TGT) with individualized instruction in
elementary mathematics (Slavin et al., 1984). Students work in four- to
five-member teams on self-instructional materials at their own levels and
rates. Students themselves take responsibility for all checking,
management, and routing and help one another with problems, freeing
the teacher to spend more time instructing small groups of students
working on similar concepts. Teams are rewarded with certificates if
they attain preset standards in terms of the number of units mastered by
all team members each week.

Two studies have assessed the effect of TAI on intergroup relations.
Oishi, Slavin, and Madden (1983) found positive effects of TAL on
cross-racial nominations on two sociometric scales, "Who are your
friends in this class?" and "Who would you rather not sit at a table
with?" No effects were found on cross-racial ratings of classmates as
"nice" or "smart," but TAI students made significantly fewer cross-racial
ratings as "not nice" and "not smart" than did control students. In a
similar study, Oishi (1983) found significantly positive effects of TAI on
cross-racial ratings as "smart" and on reductions in ratings as "not nice."
The effect on "smart" ratings was due primarily to increases in White
students' ratings of African American classmates.

Jigsaw. The original Jigsaw method (Aronson et al., 1978) assigned
students to heterogeneous six-member teams, and each member was
given a unique set of information to be discussed in "expert groups"
made up of students from different teams who were given the same
information. The "experts" returned to their teams to teach the
information to their teammates. Finally all students were quizzed and
received individual grades.

Jigsaw II modifies Jigsaw to correspond more closely to the Student
Team Learning format (Slavin, 1995a). Students work in four- to
five-member teams. All students read a chapter or story, but each team
member is given an individual topic on which to become an expert.
Students discuss their topics in expert groups and then teach them to
their teammates, as in original Jigsaw. However, quiz scores in Jigsaw
II are summed to form team scores, and teams are recognized in a class
newsletter as in STAD.

The effects of the original Jigsaw method on intergroup relations are
less clear than those for STAD, TGT, or TAI. Blaney and colleagues
(Blaney, Stephan, Rosenfield, Aronson, & Sikes, 1977) did find that
students in desegregated classes using Jigsaw preferred their Jigsaw
groupmates to their classmates in general. However, since students'
groupmates and their other classmates were similar in ethnic
composition, this cannot be seen as a measure of intergroup relations.
No differences between the experimental and control groups in
interethnic friendship choices were reported.

A. Gonzales (1979), using a method similar to Jigsaw, found that Anglo
and Asian American students had better attitudes toward Mexican
American classmates in the Jigsaw groups than those in control groups,
but he found no differences in attitudes toward Anglo or Asian
American students. In a subsequent study, A. Gonzales (1981) found
no differences in attitudes toward Mexican American, African
American, or Anglo students in Jigsaw and control bilingual classes.

The most positive effects of a Jigsaw-related intervention were found in
a study of Jigsaw II by Ziegler (1981) in classes composed of recent
European and West Indian immigrants and Anglo-Canadians in
Toronto. She found substantially more cross-ethnic friendships in the
Jigsaw II classes than in control classes, both "casual friendships"
("Who in this class have you called on the telephone in the last two
weeks?") and "close friendships" ("Who in this class have you spent
time with after school in the last two weeks?").

Johnson methods. In cooperative learning methods developed by David
Johnson and Roger Johnson (1994), students work in small
heterogeneous groups to complete a common worksheet and are
praised and rewarded as a group. Of all the cooperative learning
methods, these are the closest to a pure "cooperative" model. Two of
the Johnsons' studies have examined intergroup relation outcomes.
Cooper, Johnson, Johnson, and Wilderson (1980) found greater
friendship across race lines in a cooperative treatment than in an
individualized method in which students were not permitted to interact.
However, there were no differences in cross-racial friendships between
the cooperative condition and the competitive condition in which
students competed with equals (similar to the TGT tournaments).
Johnson and Johnson (1981) found more cross-racial interaction in
cooperative than in individualized classes during free time.

Group Investigation. Group Investigation (Sharan & Sharan, 1992),
developed by Shlomo and Yael Sharan and their colleagues in Israel, is
a general classroom organization plan in which students work in small
groups, using cooperative inquiry, group discussion, and cooperative
planning and projects. In this method, students form their own two- to
six-member groups. The groups choose subtopics from a unit being
studied by the entire class further break their subtopic into individual
tasks, and carry out the activities necessary to prepare a group report.
The group then makes a representation or display to communicate its
findings to the entire class and is evaluated based on the quality of this
report.

In a study in Israeli junior high schools, Sharan et al. (1984) compared
Group Investigation, STAD, and traditional instruction in terms of their
effect on relationships between Jews of Middle Eastern and European
backgrounds. They found that students who experienced Group
Investigation and STAD had much more positive ethnic attitudes than
students in traditional classes. There were no differences between
Group Investigation and STAD on this variable.

Weigel et al. method. Weigel, Wiser, and Cook (1975), working in
tri-ethnic (Mexican American, Anglo, and African American)
classrooms, conducted one of the largest and longest studies of
cooperative learning. They evaluated a method in which students in
multiethnic teams engaged in a variety of cooperative activities in
several subjects, winning prizes based on team performance. They
reported that their cooperative methods had positive effects on White
students' attitudes toward Mexican Americans, but not on
White-Black, Black-White, Black-Hispanic, Hispanic-Black, or
Hispanic-White attitudes. They also found that cooperative learning
reduced teachers' reports of interethnic conflict.

The effects of cooperative learning methods are not entirely consistent,
but 16 of the 19 studies reviewed here demonstrate that, when the
conditions of contact theory are fulfilled, some aspect of friendship
between students of different ethnicities improves. In a few studies
(e.g., A. Gonzales, 1979; Slavin & Oickle, 1981; Weigel et al., 1975),
improvements in attitudes were primarily improvements in the attitudes
of Whites toward minority classmates, but in most studies, attitudes
toward White and minority students were improved to the same
degree.

It is important to note that in addition to positive effects on intergroup
relations, cooperative learning methods have positive effects on student
achievement. This is particularly true of STAD, TGT, and TAI,
structured methods that combine cooperative goals and tasks with a
high degree of individual accountability (see Slavin, 1995a), as well as
Group Investigation (Sharan & Sharan, 1992). Thus it is apparent that
cooperative learning methods have positive effects on relationships
among students of different races or ethnicities while also increasing
their achievement.

Focus on Cross-Racial Friendships

Forces that promote the formation of homogeneous peer groups in
schools include geography, socioeconomic factors, and a preference
for particular activities (Lott & Lott, 1965). These factors can be
accentuated and lead to overt prejudice and interracial hostility when
race is considered. Given the many forces operating against the
formation of cross-racial friendships, it would seem that if cooperative
learning influences these friendships, it would create relatively weak
relationships rather than strong ones (see, for example, Schofield,
1991). It would seem unlikely that a few weeks of cooperative learning
would establish the trust and respect needed to build strong interracial
friendships.

A secondary analysis of the Slavin (1979) STAD study by Hansell and
Slavin (1981) investigated this hypothesis. The sample included 424
seventh and eighth-grade students in 12 inner-city language arts
classrooms. Classes were randomly assigned to cooperative learning
(STAD) or control treatment for a 10-week program. Students were
asked on both pre- and posttests, "Who are your best friends in this
class? Name as many as you wish," in a free-choice format. Choices
were defined as "close" if they were among the first six made by
students and "distant" if they ranked seventh or later. The reciprocity
and order of choices made and received were analyzed by multiple
regression.

The results showed that the positive effects of STAD on cross-racial
choices were primarily due to increases in strong friendship choices.
Reciprocated and close choices, both made and received, increased
more in STAD than in control group classes. Specifically, students
made on average slightly over 10 friend nominations; on the pretest,
STAD students averaged 3.54 cross-race friend choices compared to
3.87 on the posttest, whereas control students averaged 2.93 on the
pretest and 2.37 on the posttest. Thus, STAD students nominated
cross-race friends in proportions close to the class distribution. A
follow-up conducted a year later showed that students who
participated in the STAD classes continued to have more cross-race
friends than students from the control classes, controlling for total
number of friends.

A second way of assessing strong and weak friendship ties was used
by Hansell (1984) to assess peer relations among fifth and sixth graders
who had participated in cooperative learning groups. Students rated
each classmate on a 3-point scale in terms of liking; liking a lot
indicated a strong friendship tie, liking some indicated a weak friendship
tie. For these students, the effect of cooperative learning groups was to
increase their weak cross-race friendship ties without affecting strong
ties.

Similar conclusions were found when cooperative learning group
experiences were compared with other race relations programs in high
schools (Slavin & Madden, 1979). The dependent variables used in
this analysis included the race of students' top three friends at school
and peers they spoke with on the phone. Students who had
experienced mixed-race learning groups were more likely to have
cross-race friends and have spoken on the phone to a cross-race peer
than students who had experienced other programs such as a
multicultural curriculum.

Thus, cooperative learning appears to facilitate the development of
positive cross-race peer relations, whether at the level of weak
friendship ties or close, reciprocated friendship choices. The
implications of such research results are encouraging. These findings
suggest that positive social relations among students of differing racial
and ethnic backgrounds help students to transcend and transform
shared cultural norms and attitudes that can prohibit meaningful
cross-cultural interactions. Such transformation does not require
students to ignore or eliminate the differences that exist among their
classmates, in their histories, communities, and families, but rather to
understand them using a different cultural paradigm. The positive social
relations that are built between students of different racial and ethnic
backgrounds as they work collaboratively to solve complex problems
or to complete meaningful tasks are not simply a matter of students'
liking each other or having positive thoughts about each other . These
cross-cultural interactions are about broadening the cultural frames of
reference that define the social worlds and dictate social network
patterns for these students.

Conclusion

Cooperative learning is an instructional approach that has been shown
to promote a variety of positive cognitive, affective, and social
outcomes. The intent of cooperative learning is to enhance academic
achievement by providing students with increased opportunities for
discussion, learning from each other, and by allowing students to divide
up tasks in ways that tap into their academic strengths. Cooperative
learning promotes some of the most important goals in American
education: increasing the academic achievement of all students while
simultaneously improving intergroup relations among students of
different racial and ethnic backgrounds (Deering, 1989). With the
increasing racial diversity found in America's classrooms, instructional
strategies that can achieve these goals must be refined and widely
disseminated. It is undeniable that race and ethnicity are important ways
that students define themselves in schools, and racial intolerance and
hostility between students of different racial and ethnic backg rounds
still persists. Research shows that many youths still carry the legacy of
ethnic and racial hatred engendered by their parents, grandparents, and
community. Although acts of intolerance and racism, in most cases, are
more subtle today than they were 20 years ago (Vernay, 1996), we are
seeing a resurgence of overt racist and violent manifestations of
discrimination and prejudice on school campuses. If schools are to
serve as a safe haven from violence and a place for students to learn
how to be good citizens, the use of instructional strategies such as
cooperative learning will need to be more widespread.

The research presented in this article suggests that as students talk
andwork with each other in cooperative learning groups, they are not
only acquiring academic knowledge and skill but are also constructing a
shared cultural paradigm for defining the group, its work, and the social
identities of the participants. They are establishing a group culture--a
culture that sets the social context in which social relationships among
students are defined, established, and given value and meaning. The
hope is that students will carry this cultural paradigm into adulthood.
Although the research relating cooperative learning to intergroup
relations clearly indicates that cross-cultural friendships are developed
when students work in cooperative work groups, additional research is
needed to better understand intergroup behavior, particularly outside of
the schooling context. A few studies (Oishi, 1983; Ziegler, 1981) have
found positive effects of cooperative learning on self-reported
cross-racial friendships outside o f class, but behavioral observations in
nonclassroom settings are still needed. Such research will illuminate the
important role schools can play in reducing racism, prejudice, and
discrimination in the larger society.

ROBERT SLAVIN is currently Co-Director of the Center for
Research on the Education of Students Placed at Risk at Johns
Hopkins University, and Chairman of the Success for All Foundation.
Dr. Slavin has authored or co-authored more than 180 articles and 15
books, including Educational Psychology: Theory Into Practice (1994),
School and Classroom Organization (1989), Effective Programs for
Students at Risk (1989), Cooperative Learning: Theory, Research, and
Practice (1995), Preventing Early School Failure (1994), Every Child,
Every School: Success for All (Corwin, 1996), and Show Me the
Evidence: Proven and Promising Programs for America's Schools
(Corwin, 1998). He received the American Educational Research
Association's Raymond B. Cattell Early Career Award for
Programmatic Research in 1986, the Palmer O. Johnson award for the
best article in an AERA journal in 1988, and the Charles A. Dana
award in 1994.

ROBERT COOPER is an Associate Research Scientist for the Center
for Social Organization of Schools at Johns Hopkins University and
Assistant Professor of Education. He conducts research on the
implementation and scale up of school reform models. His research
focuses on the politics and policies of school reform, particularly as they
relate to issues of race and equity for at-risk students. Specializing in
the use of a mixed methods approach, he has published and presented
numerous papers on the varying aspects of school reform and school
change, including recent articles in Urban Education, Journal of Negro
Education, Education and Urban Society, and Journal of Education for
Students Placed at Risk.

(*.) This article is adapted from Slavin (1997). We wish to
acknowledge the contribution of Shelley C. Dennis.

Correspondence concerning this article should be addressed to Robert
E. Slavin, Co-Director, Center for Research on the Education of
Students Placed at Risk, Johns Hopkins University, Baltimore, MD
21204-5200 [e-mail: rslavin@successforall.net].

References

Allport, G. W. (1954). The nature of prejudice. Cambridge, MA:
Addison-Wesley.

Aronson, E., Blaney, N., Stephan, C., Sikes, J., & Snapp, M. (1978).
The Jigsaw classroom. Beverly Hills, CA: Sage Publications.

Bag]ey, C., Verma, G., Mallick, K., & Young, L. (1979). Personality,
self-esteem, and prejudice. London: Saxon House.

Blaney, N. T., Stephan, C., Rosenfield, D., Aronson, E., & Sikes, J.
(1977). Interdependence in the classroom: A field study. Journal of
Educational Psychology, 69, 121-128.

Braddock, J., Dawkins, A., & Wilson, G. (1995). Intellectual contact
and race relations among American youth. In W. D. Hawley & A.
Jackson (Eds.), Toward a common destiny: Improving race and ethnic
relations in America (pp. 237-256). San Francisco: Jossey-Bass.

Cohen, E. G. (1994). Designing groupwork: Strategies for
heterogeneous classrooms (2nd ed.). New York: Teachers College
Press.

Cohen, E. G., & Lotan, R. A. (1997). Working for equality in
heterogeneous classrooms: Sociological theory in practice. New York:
Teachers College Press.

Cook, S. W. (1978). Interpersonal and attitudinal outcomes of
cooperative interracial groups. Journal of Research and Development in
Education, 12, 7-113.

Cooper, L., Johnson, D. W., Johnson., R., & Wilderson, F. (1980).
Effects of cooperative, competitive, and individualistic experiences on
interpersonal attraction among heterogeneous peers. Journal of Social
Psychology, 111, 243-252.

Cooper, R., & William, E. (1999). The salience of race in America's
schools. Manuscript submitted for publication.

Cooper, R. (1996). Detracking in an urban California high school:
Improving the schooling experiences of African-American students.
Journal of Negro Education, 65, 190-208.

Cooper, R. (1999). A promising example of urban school reform:
Detracking in a racially mixed high school. Journal of Education for
Students Placed at Risk, 4, 259-275.

Cooper, R. (in press). Urban school reform from a student of color
perspective. Urban Education. Crain, R., Mahard, R., & Narot, R.
(1982). Making desegregation work. Cambridge, MA: Ballinger.
Datnow, A., & Cooper, R. (1996). Tracking. In D. Levinson, A.
Sadovnik, & P. Cookson (Eds.), Education and sociology: An
encyclopedia. New York: Garland Publishing.

Deering, P. D. (1989, October). An ethnographic approach for
examining participants' construction of a cooperative learning class
culture. Paper presented at the annual meeting of the American
Anthropological Association, Washington, DC. (ERIC Document
Reproduction Service No. ED 319 083)

DeVries, D. L., Edwards, K. J., & Slavin, R. E. (1978). Biracial
learning teams and race relations in the classroom: Four field
experiments on Teams-Games-Tournament. Journal of Educational
Psychology, 70, 356-362.

Finkelstein, N. W., & Haskins, R. (1983). Kindergarten children prefer
same-color peers. Child Development, 54, 502-508.

Giroux, H. (1992). Border crossing: Cultural workers and the politics
of education. New York: Routledge.

Gonzales, A. (1979, August). Classroom cooperation and ethnic
balance. Paper presented at the annual convention of the American
Psychological Association, New York.

Gonzales, A. (1981). An approach to independent-cooperative
bilingual education and measures related to social motives. Unpublished
manuscript, California State University at Fresno.

Gonzales, N., & Cauce, A. (1995). Ethnic identity and multicultural
competence: Dilemma and challenges for minority youth. In W. D.
Hawley & A. Jackson (Eds.), Toward a common destiny: Improving
race and ethnic relations in America (pp. 131-162). San Francisco:
Jossey-Bass.

Grattan, L., & Eslinger, P. (1989). Higher cognition and social
behavior: Changes in cognitive flexibility and empathy after cerebral
lesions. Neuropsychalogy, 3, 175-185.

Green, J. A., & Gerard, H. (1974). School desegregation and ethnic
attitudes. In F. Franklin & J. Sherwood (Eds.), Integrating the
organization. New York: Free Press.

Hallinan, M., & Williams, R. (1989). Interracial friendship choices in
secondary schools. American Sociological Review, 54, 67-78.

Hansell, S. (1984). Cooperative groups, weak ties, and the integration
of peer friendships. Social Psychology Quarterly, 47, 316-328.

Hansell, S., & Slavin, R. E. (1981). Cooperative learning and the
structure of interracial friendships. Sociology of Education, 54, 98-106.

Heath, S. B. (1995). Race, ethnicity and the defiance of categories. In
W. Hawley & A. Jackson (Eds.), Toward a common destiny:
Improving race and ethnic relations in America (pp. 39-70). San
Francisco: Jossey-Bass.

Hewstone, M., & Brown, R. (1986). Contact and conflict in intergroup
encounters. Oxford: Basil Blackwell.

Johnson, D. W., & Johnson, R. T. (1981). Effects of cooperative and
individualistic learning experience on interethnic interaction. Journal of
Educational Psychology, 73, 444-449.

Johnson, D. W., & Johnson, R. T. (1994). Learning together and
alone: Cooperative, competitive, and individualist learning (4th ed.).
Englewood Cliffs, NJ: Prentice Hall.

Johnson, D. W., Johnson, R. T., & Holubec, E. J. (1993). Circles of
learning: Cooperation in the classroom (4th ed.). Edina, MN:
Interaction Book.

Kagan, S. (1995). Cooperative learning. Boston: Charlesbridge.

Kagan, S., Zahn, G. L., Widman, K. F., Schwarzwald, J., & Tyrell, G.
(1985). Classroom structural bias: Impact of cooperative and
competitive classroom structures on cooperative and competitive
individuals and groups. In R. E. Slavin, S. Sharan, S. Kagan, R.
hertz-Lararowitz, C. Webb, & R. Schmuck (Eds.), Learning to
cooperate, cooperaring to learn (pp.277-312). New York: Plenum.

Lacy, W. B., Mason, E. J., & Middleton, E. (1983). Fostering
constructive intergroup contact in desegegrated schools: Suggestions
for future research. Journal of Negro Education, 52, 130-141.

Lampe, J. R., Rooze, G. E., & Tallent-Runnels, M. (1996). Effects of
cooperative learning among Hispanic students in elementary social
studies. Journal of Educational Research, 89, 187-191.

Lilli, W., & Rehm, J. (1990). Judgmental processes as a bases of
intergroup conflict. In W. S. Stroebe, A. W. Kruglanski, D. Bar-Tal, &
M. Hewstone (Eds.), The social psychology of intergroup conflict (pp.
29-46). New York: Springer Verlag.

Lopez-Reyna, N. A. (1997). The relation of interactions and story
quality among Mexican American and Anglo American students with
learning disabilities. Exceptionality, 7, 245-261.

Lotan, R. A., & Whitcomb, J. A. (1998). Groupwork in diverse
classrooms. New York: Teachers College Press.

Lott, A., & Lott, B. E. (1965). Group cohesiveness as interpersonal
attraction: A review of relationships with antecedent and consequent
variables. Psychological Bulletin, 64, 259-309.

Masson, C., & Verkuyten, M. (1993). Prejudice, ethnic identity,
contact, and ethnic group preferences among Dutch young adolescents.
Journal of Applied Social Psychology, 23, 156-168.

McDonald, A. (1970). Revised scale for ambiguity tolerance: Reliability
and validity. Psychological Reports, 26,791-898.

McLemore, S. D., & Romo, H. D. (1998). Racial and ethnic relations
in America (5th ed.). Boston: Allyn and Bacon.

McLoyd, V. (1990). Minority children: Introduction to the special
issue. Child Development, 61, 263-266.

McManus, S. M., & Gettinger, M. (1996). Teacher and student
evaluations of cooperative learning and observed interactive behaviors.
Journal of Educational Research, 90, 13-22.

Oakes, J. (1985). Keeping track: How schools structure inequality.
New Haven, CT: Yale University Press.

Oakes, J., & Guiton, G. (1995, March). Matchmaking: The dynamics
of high school tracking decisions. American Educational Research
Journal, 32, 3-33.

Oakes, J., & Wells, A. (1995). Beyond sorting and stratification:
Creating alternatives to tracking in racially mixed secondary schools.
Paper presented at the annual meeting of the American Educational
Research Association, San Francisco, CA.

Oishi, S. (1983). Effects of team assisted individualization in
mathematics on cross-race interactions of elementary school children.
Unpublished doctoral dissertation, University of Maryland, College
Park, MD.

Oishi, S., Slavin, R. E., & Madden, N. A. (1983, March). Effects of
student teams and individualized instruction on cross-race and
cross-sex friendships. Paper presented at the annual meeting of the
American Educational Research Association, Montreal, Canada.

Pettigrew, T. F. (1986). The intergroup contact hypothesis
reconsidered. In M. Hewstone and R. Brown (Eds.), Contact and
conflict in intergroup encounters (pp. 169-195). Oxford: Basil
Blackwell.

Phinney, J., & Cobb, N. (in press). Reasoning about intergroup
relations among Hispanic and Euro-American adolescents. Journal of
Adolescent Research.

Phinney. J., & Devich-Navarro. M. (1997). Variations in bicultural
identification among African American and Mexican American
adolescents. Journal of Research on Adolescence, 7, 3-32.

Phinney, J., Ferguson, D., & Tate, J. (1996). Intergroup attitudes and
interactions among adolescents from three ethnic groups. Manuscript
submitted for publication.

Quintanilla, M. (1995, November 17). The great divide. The Los
Angeles Times, pp. El, E7.

Romo, H. D. (1997). Improving ethnic and racial relations in the
schools. ERIC Clearinghouse on Rural Education and Small Schools.
(ERIC Document Reproduction Service No. ED 414 113)

Romo, H. D., & Falbo, T. (1996). Latino high school graduation:
Defying the odds. Austin, TX: University of Texas Press.

Schofield, J. S. (1991). School desegregation and intergroup relations:
A review of the literature. In G. Grant (Ed.), Review of Research in
Education (Vol. 17 pp. 335-409). Washington, DC: American
Educational Research Association.

Schofield, J. S. (1995). Improving intergroup relations among students.
In J. Banks & C. Banks (Eds.), Handbook of research on multicultural
education (pp. 635-645). New York: Macmillan.

Schwartz, W. (1996). Anti-bias conflict resolution curricula: Theory
and practice. ERIC Clearing-house on Urban Education. (ERIC
Document Reproduction Service No. ED 377 255)

Shachar, H., & Shmuelevitz, H. (1997). Implementing cooperative
learning, teacher collaboration and teachers' sense of efficacy in
heterogeneous junior high schools. Contemporary Educational
Psychology 22, 53-72.

Sharan, S. (1994). Handbook of cooperative learning methods.
Westport, CT: Greenwood Press.

Sharan, S., Kussell, P., Hertz-Lazarowitz, R., Bejarano, Y., Raviv, S.,
& Sharan, Y. (1984). Cooperative learning in the classroom: Research
in desegregated schools. Hillsdale, NJ: Erlbaum.

Sharan, Y., & Sharan, S. (1992). Expanding cooperative learning
through group investigation. New York: Teachers College Press.

Shrum, W., Cheek, N.H., & Hunter, S. M. (1988). Friendship in
school: Gender and racial homophily. Sociology of Education, 61,
227-239.

Slavin, R. E. (1977). How student learning teams can integrate the
desegregated classroom, integrated Education, 15, 5-58.

Slavin, R. E. (1979). Effects of biracial learning teams on cross-racial
friendships. Journal of Educational Psychology, 71, 381-387.

Slavin, R. E. (1983). Cooperative learning. New York: Longman.

Slavin, R. E. (1986). Using student team learning (3rd ed.). Baltimore:
Johns Hopkins University, Center for Research on Elementary and
Middle Schools.

Slavin, R. E. (1991a). Synthesis of research on cooperative learning.
Educational Leadership, 48, 71-82.

Slavin, R. E. (1991b). Are cooperative learning and untracking harmful
to the gifted? Educational Leadership, 48(6), 68-71.

Slavin, R. E. (1992). When and why does cooperative learning
increase achievement? Theoretical and empirical perspectives. In R.
Hertz-Lazarowitz & N. Miller (Eds.), Interaction in cooperative groups:
The theoretical anatomy of group learning (pp. 145-173). New York:
Cambridge University Press.

Slavin, R. E. (1995a). Cooperative learning: Theory, research, and
practice (2nd ed.). Boston: Allyn & Bacon.

Slavin, R. E. (1995b). Cooperative learning and intergroup relations. In
J. Banks and C. M. Banks (Eds.), Handbook of research on
multicultural education (pp. 628-634). New York: Macmillan.

Slavin, R. E. (1997). Cooperative learning and student diversity. In R.
Ben-Ari & Y. Rich (Eds.), Enhancing education in heterogeneous
schools (pp. 215-247). Ramat-Gat, Israel: Bar-Ilan University Press.

Slavin, R. E., Leavey, M., & Madden, N. A. (1984). Combining
cooperative learning and individualized instruction: Effects on students'
mathematics achievement, attitudes, and behaviors. Elementary School
Journal, 84, 409-422.

Slavin, R. E., & Madden, N. A. (1979). School practices that improve
race relations. American Educational Research Journal, 16, 169-180.

Slavin, R. E., & Oickle, E. (1981). Effects of cooperative learning
teams on student achievement and race relations: Treatment by race
interactions. Sociology of Education, 54, 174-180.

Sprinthall, N., & Collins, W. (1995). Adolescent psychology: A
developmental view (3rd ed.). New York: McGraw-Hill.

Stephan, W. (1978). School desegregation: An evaluation of the
predictions made in Brown v. Board of Education. Psychological
Bulletin, 85, 217-238.

Stevens, R. J., Madden, N. A., Slavin, R. E., & Farnish, A. M.
(1987). Cooperative integrated reading and composition: Two field
experiments. Reading Research Quarterly, 22, 433-454.

Stevens, R. J., & Slavin, R. E. (1995). The cooperative elementary
school: Effects on students' achievement, attitudes, and social relations.
American Educational Research Journal, 32, 321-351.

Vernay, M. (1996). Curriculum and instruction to reduce conflict.
ERIC Clearinghouse on Urban Education. (ERIC Document
Reproduction Service No. 64 ED 322 274)

Waters, M. C. (1990). Ethnic options: Choosing identities in America.
Berkeley and Los Angeles: University of California Press.

Weigel, R. H., Wiser, P. L., & Cook, S. W. (1975). Impact of
cooperative learning experiences on cross-ethnic relations and attitudes.
Journal of Social Issues, 31, 219-245.

Wheclock, A. (1992). Crossing the tracks: How "untracking" can save
America's schools. New York: New Press.

Ziegler, S. (1981). The effectiveness of cooperative learning teams for
increasing cross-ethnic friendship: Additional evidence. Human
Organization, 40, 264-268.