This article was downloaded by: [Computing & Library Services, University of Huddersfield]

On: 27 December 2014, At: 22:04

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41
Mortimer Street, London W1T 3JH, UK

The Journal of Educational Research

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/vjer20

Cooperative Learning in the Secondary Mathematics

Classroom
a a b
Kristina M. Whicker , Linda Bol & John A. Nunnery
a
The University of Memphis
b
The Johns Hopkins University
Published online: 02 Apr 2010.

To cite this article: Kristina M. Whicker , Linda Bol & John A. Nunnery (1997) Cooperative Learning in the Secondary Mathematics
Classroom, The Journal of Educational Research, 91:1, 42-48, DOI: 10.1080/00220679709597519

To link to this article: http://dx.doi.org/10.1080/00220679709597519

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the
publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or
warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed
by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings,
demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly
in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Cooperative Learning in the
Secondary Mathematics Classroom
KRISTINA M. WHICKER JOHN A. NUNNERY
LINDA BOL The Johns Hopkins University
The University of Memphis
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

the results of seven experiments with third- through sixth-

ABSTRACT The effects of cooperative learning on student grade students, Slavin ( 1985) reported significantly positive
achievement and attitudes in a secondary mathematics class- results on mathematics achievement scores for the coopera-
room were investigated. In this quasi-experimental design, 2 tive learning conditions in six of the seven studies. In one
precalculus courses were compared. Students in 1 class stud-
ied the material in cooperative learning groups; students in the experiment, Madden and Slavin (1983) studied third and
2nd class studied the material independently. Three chapter fourth graders in a mathematics course in which coopera-
tests were used to measure student achievement, and a ques- tive strategies were used and found significant increases in
tionnaire was administered to the treatment group members achievement for all students in the experimental group.
after the study was completed to assess their attitudes toward Other researchers have reported similar findings that point
the cooperative learning procedure. The results obtained from
a repeated-measures multivariate analysis of variance (with to the achievement benefits associated with studying math-
pretest scores as the covariate) showed a significant Group x ematics in cooperative learning groups at lower grade levels
Time interaction. Students in the cooperative learning group (Edwards, Devries, & Snyder, 1972; Hamblin, Hathaway, &
had increasingly higher test scores than students in the com- Wodarski, 197 I ; Nattiv, 1994).
parison group and significantly outscored the comparison In the few studies conducted with students enrolled in
group on the 3rd chapter test. Survey results revealed primar-
ily favorable responses toward the cooperative learning proce- mathematics courses at higher grade levels, we found addi-
dure. Most students indicated that they liked working in tional evidence suggesting that cooperative learning pro-
groups and appreciated getting help from other students, espe- motes achievement. A recent study indicated that 7th- and
cially for learning difficult concepts. Some students disliked 8th-grade prealgebra students who used cooperative tech-
having groups preassigned and permanent, and they suggest- niques not only scored higher than the control group but
ed alternating group membership.
retained the information for a longer period of time (Duren
& Cherrington, 1992). These researchers also examined the
effects of cooperative learning in 9th-grade general mathe-
matics classes in 16 schools during the school year. Results
A lthough considerable evidence exists to support the
claim that cooperative learning promotes achieve-
ment as well as other positive affective outcomes at the ele-
supported the conclusion that cooperative learning posi-
tively affects student achievement; those students who
worked in cooperative teams scored significantly higher
mentary and middle-grade levels (see Johnson, Maruyuma,
than the control groups on the same test. Slavin and Kar-
Johnson, Nelson, & Skon, 1981; Slavin, 1985, 1991, for
weit (1984) also found that 9th-grade students who studied
reviews), few studies have examined the effects of cooper-
general mathematics in cooperative learning groups signif-
ative learning in secondary grade levels (Grades 10-12). In
icantly outperformed students who studied the material
his more recent reviews of this literature, Slavin ( 1 989-90,
individually. Similarly, Sherman and Thomas ( 1986)
1991) called for more studies on the effects of cooperative
reported significantly higher achievement gains among 9th
learning in high school. Among the studies conducted with
and 10th graders who studied general mathematics cooper-
students at these grades, even fewer have focused on sec-
atively rather than individually. These findings emphasize
ondary mathematics classrooms. Our purpose in this study
the academic benefits of using cooperative learning strate-
was to help address this gap in the research literature by
gies in secondary mathematics classrooms; however, the
investigating the effects of cooperative learning on
number of studies focusing on mathematics achievement at
achievement and students’ attitudes in the secondary math-
ematics classroom.
An examination of the research findings on the effects of Address correspondence to Linda Bol, Department of Counsel-
cooperative learning in mathematics classes in lower grade ing, Educational Psvchology and Research, The College of Edu-
levels suggests that it promotes achievement. Summarizing cation. The University of Memphis, Memphis TN 38/52.
42
 SeptembedOctober 1997 [Vol. 91(No. l)] 43

these grade levels is sparse relative to the number of stud- achievement benefits (Burns, 198 I ) . Similarly, King
ies at elementary grade levels. (1993) observed third graders who were learning mathe-
Research on the effects of cooperative learning has also matics in cooperative groups and found that high achiev-
focused on variables other than achievement. Cooperative ers dominated the work and decision making, while low
learning has been linked to other positive social or affective achievers remained generally passive. The latter type of
outcomes. One benefit is the increase in social skills of stu- cooperative structure employed neither group goals nor
dents who participate in group work (Slavin, 1091). By individual accountability. Group rewards with individual
working together, students learn to be tactful, to manage accountability means that each individual is rewarded only
conflicts effectively, and to respect the opinions of others when others in his or her group also succeed, which pre-
(Augustine, Gruber, & Hanson, 1989-90). Learning social vents high achievers from dominating the work (Johnson
skills may be particularly important in adolescence, a peri- & Johnson, 1987; Slavin, 1988). Thus, cooperative group
od when the need to belong conflicts with the need to be work that incorporates both group rewards and individual
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

recognized as an individual (Wood, 1987). Group work accountability not only forces students to take responsibil-
addresses this conflict provided the groups are small ity for their own mastery of the material, but also makes
enough for individual recognition. Cooperative leaming has the students responsible for their classmates’ mastery
also been linked to increases in self-esteem, attendance, (Slavin, 1990).
time on task, enjoyment of school and classes, and motiva- One method of cooperative group learning that involves
tion to learn, as well as a decrease in dependence on the both group reward and individual accountability is known
teacher (Augustine et al., 1989-90; Good, Reys, Grouws, & as Student Teams-Achievement Divisions (STAD). This
Mulryan, 1989--90; Slavin, 1990; Wood). Perhaps one of method of student team learning employs competition
the most important benefits of cooperative learning has between groups while encouraging cooperation within
been more positive intergroup relations. Improved race rela- groups (Slavin, 1990). All members of the groups are
tions, as well as increased acceptance of mainstreamed chil- required to learn the material because the accumulation of
dren, have frequently been reported (Augustine et al., points depends on the performance of the individual group
1989-90; Madden & Slavin, 1983, Slavin, 1989-90; members. The team score is based on individual improve-
Stevens & Slavin, 1989-90). ment on quiz scores, and the winning team is recognized
These positive social, affective, and behavioral benefits of and possibly rewarded. Specifically, individual points are
cooperative learning have also been specifically linked to awarded based on the percentage increase of a student’s
cooperative learning in mathematics classrooms (Davidson, grade from one test to the next. Group points are calculated
1985; Mulryan, 1994, 1995; Slavin, 1985). For example, by adding all of the individual points together and dividing
Mulryan (1994, 1995) found that fifth- and sixth-grade stu- by the number in each group. In this type of reward struc-
dents who studied mathematics in cooperative learning ture, even the low achievers have the opportunity to con-
groups spent more time on task, especially quality time, and tribute the maximum amount of points to the group. This
more time actively participating than they did in whole-group type of cooperative learning structure has been well
instruction. These students and their teachers stated that the researched and consistently linked to significant gains in
benefits of learning mathematics in cooperative groups were achievement and other favorable social or affective out-
collaboration, student-to-student help and support, social comes (Johnson & Johnson, 1981; Slavin, 1990).
interaction, and active participation (Mulryan, 1994). In this study, we used a modified version of STAD to
In contrast to some of the benefits associated with group assess the effectiveness of cooperative learning on achieve-
learning in mathematics classrooms, other researchers have ment in a secondary precalculus course. In this quasi-exper-
cited disadvantages of using group strategies in the class- imental design, the achievement of students in one section
room. Passivity among some students working in groups, of the course, who studied the material in cooperative
especially low-achieving students, has been noted (Good, groups, was compared with the achievement of students
Reys, Grouws, & Mulryan, 1989-90; King, 1993; Mulryan, enrolled in a second section, who studied the same material
1992, 1995). Other drawbacks identified were the amount individually. In addition, we administered a brief question-
of time required for some groups to form a cohesive bond, naire to students in the experimental group to assess their
which initially impeded progress, and the difficulty of attitudes toward studying in cooperative groups.
adapting the curriculum and materials for small-group Although there have been few studies that have focused
instruction (Good et al., 1989-90). on the effectiveness of cooperative learning in secondary
All forms of cooperative learning, however, are not nec- mathematics courses, the bulk of the evidence in other
essarily associated with achievement and other affective or grade levels led us to predict that students using the coop-
social benefits. According to Slavin (1988), both group erative learning strategy would score significantly higher on
rewards and individual accountability are necessary for the achievement measures than would students who studied
cooperative methods to be most effective. A method the material individually. The positive influence of cooper-
known as Groups of Four, which employs neither group ative learning strategies on affective and social variables
rewards nor individual accountability, has had few reported in the literature also led us to predict that students
 44 The Journal of Educational Research

would express positive attitudes about the cooperative Procedure

learning techniques.
Before beginning the treatment phase, students in the
experimental group were instructed in the rules for small-
Method group work. They were told to explain their answers to each
Participants and Design other and to direct questions to their teammates rather than
to the teacher. The students were told that they were not fin-
The participants were 31 students in a rural high school ished with their task until they were certain that all the
in a lower-middle-classarea in the mid-South. The students members of their group would score 100% on the test.
were 1 Ith and 12th graders enrolled in two sections of a The sequence of events in the treatment phase was teach-
precalculus course. One class was randomly designated as ing the lessons, group study, testing, and team recognition.
the treatment group and was composed of 15 students, 33% The teacher presented the material in the chapter in 5 to 8
girls and 67% boys. The other class served as a comparison
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

days, depending on the length of the chapter. The students

group and contained 16 students, 50% girls and 50% boys. spent two class periods studying the teacher-developed
The study lasted 6 weeks. During this time, the experi- review sheets. Correct answers were provided to the groups
mental group used a variation of STAD to study for three after they had reached a consensus on their answers.
chapter tests, whereas the comparison group studied indi- Students took individual tests on the chapter material the
vidually for the same tests. day after the two group study sessions. All the tests were
Within the treatment group, previous semester grades graded by the teacher. The students earned points for their
served as the basis for cooperative group membership. groups based on how much their test scores exceeded their
Cooperative groups consisted of 5 members and included 1 base scores. Initial base scores were computed as an aver-
student from the top fourth of the class, 1 student from the age of all tests taken the previous semester. Base scores
bottom fourth of the class, and 3 students from the middle were recalculated after each test. A student could earn 4
half of the class. extra points if his or her test score was 1-10 points above
the base score. Six extra points could be earned with a test
Measures score that was more than 10 points above the base score or
if the student achieved a perfect paper. Team scores were
Achievement computed by adding all the points earned by group mem-
bers and dividing the sum by the number of members in the
During the experimental phase, we assessed student
group that took the test.
achievement on three teacher-developed tests covering the
The comparison group was treated identically to the
content contained in three chapters from the textbook
experimental group except that group members were
Advanced Mathematics: A Precalculus Approach (Ryan,
instructed to complete the same review sheets individually
Doubet, Fabricant, & Rockhill, 1993). Topics covered
during the study phase; they received no team recognition.
included polynomial functions, inequalities and linear pro-
The same material was covered at the same time as in the
gramming, and exponential and logarithmic functions. The
experimental group. The comparison group also had the
tests consisted of 10 open-ended items that required com-
opportunity to receive the same additional points as
putation, graphing, and explanation. There were 50 points
described for the experimental group, but points were
possible for each test, and partial credit was awarded for
awarded based on individual improvement.
answering portions of the problems correctly.
At the conclusion of the study, students in the cooperative
Validity of these tests was strengthened by asking anoth-
learning group were asked to complete the questionnaire.
er mathematics teacher to review them and give suggestions
The teacher assured the students that their responses would
for improvement. Reliability of scoring was assessed by
be kept confidential and told them not to write their names
providing the scoring criteria to another mathematics
on their questionnaires or to otherwise identify themselves.
teacher and asking her to score a sample of completed tests.
The extent of agreement between scorers was 100%.
Data Analysis
Questionnaire Analysis of test scores
We designed a researcher-developed questionnaire to The first phase of the analysis consisted of using an inde-
assess students’ attitudes about the cooperative learning pendent t test to compare the pretest achievement scores for
procedure. The questionnaire consisted of five open-ended each class. This analysis was conducted to ensure that the
items asking students to identify what they liked and did not achievement levels were not statistically different at the
like about working in cooperative groups, what concepts beginning of the study.
were easier to learn in groups, what concepts were easier to Student achievement test scores from the three chapter
learn on their own, and what they would change about the tests were analyzed via a repeated-measures multivariate
cooperative learning groups. analysis of variance (MANOVA), with student pretest score
 SeptembedOctober 1997 [Vol. 91(No. l)] 45

as a covariate and cooperative learning versus individual over time (see Figure 1). Follow-up univariate tests also
study as independent variables. Univariate (Cochran’s C) indicated a growing difference between the groups; they
and multivariate (Box’s M) homogeneity of variance tests had similar means on the first test, somewhat different
were conducted to assess the homogeneity-of-dispersion- means (albeit not significantly different) on the second test,
matrices assumption underlying the MANOVA procedure. and significantly different means on the third test, F( I , 28)=
Because of its sensitivity to violations of the multivariate 4.57,p = .04.The descriptive statistics for each achieve-
normality assumption, Box’s M is also a good test for mul- ment test are reported in Table I .
tivariate normality. Of primary interest was the result of the The explanatory power of the model increased between
test of the Group x Time interaction effect; under the the first and third testing occasions, from a multiple R2 of
assumption of an ordinal interaction, a significant interac- .09 for the first test to a multiple RZof .IS on the third test;
tion effect would indicate a growing difference in the the explanatory value of the pretest decreased from
achievement levels of students based upon group member- accounting for almost 9% of the variance in test scores on
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

ship. Two procedures were used to aid in the interpretation the first test to accounting for only 1% of the variance in
of the MANOVA results: (a) Test scores for each group at scores on the third test. Thus, on the first test the pretest
each time were used to conduct univariate follow-up tests, and scores accounted for 96% of the explained variance, where-
(b) covariate-adjusted means were computed and graphed. as on the third test group membership accounted for 93% of
the explained variance.
Analysis of Questionnaire Data
Because the items called for open-ended responses, the Questionnaire Responses
analysis was qualitative in nature. A content analysis was The most frequently cited responses by category for each of
used to identify themes in students’ responses, to develop the five questionnaire items are reported in Table 2.Twelve
categories of responses based on those themes, and to tabu- students completed the questionnaire. Because some indi-
late the number and percentage of responses for each cate- vidual responses contained more than one theme or idea,
gory. Reliability in coding responses into categories was they were coded and counted in more than one category.
assessed by having another researcher independently code Therefore, the percentages shown in the table were calculated
the responses. There was a 96% agreement in categorization based on the number of different themes or ideas and not on
between the two coders across responses to the five questions. the number of students who completed the questionnaire.
When students were asked what they liked about the
Results cooperative learning groups, the most frequently occurring
response was that they liked receiving help from others
Achievement
(38%). For example, 1 student wrote that he liked working
An independent t test was used to compare the previous in groups because “If I did not understand something, I
semester achievement scores for each class, and the results could always ask someone in the group for help.” The sec-
revealed no significant differences in mean achievement ond most frequently occurring response was the more gen-
scores. The mean achievement score for the treatment group eral category of working with other students (1 9%). This
was 86.9;the mean achievement score for the comparison category encompassed the social benefits of group learning.
group was 88.0. Representative responses included “the time spent socializ-
Tests of MANCOVA assumptions. The results of the uni-
variate homogeneity of variance tests for the pretest and
achievement test scores indicated significantly heteroge- Figure 1. Adjusted Mean Achievement Scores for the
neous variance between treatment groups only on the first Cooperative Learning and Comparison Groups
achievement test (Cochran’s CI5, = 0 . 7 6 , =~ .03).Box’s M, Across the Three Prealgebra Chapter Tests
computed to assess multivariate normality and homogene- I
ity, indicated no significant differences between groups.
Given that the p values for the other univariate tests were all
high (above .80), and that the sensitive Box’s M test yield-
ed no significant difference between groups, the distribu-
tional assumptions underlying application of the repeated
measures MANCOVA did not appear to have been violated.
Time x Group interaction effect. The averaged Group x ”t
Time interaction effect was significant, F(3,86) = 29.41,
Ted 1 Ten 2 Ten 3
p < .001. A plot of the pretest-adjusted means suggested an
ordinal interaction favoring the cooperative learning group;
the difference in achievement test scores between the coop-
erative learning and independent study groups increased
 46 The Journal of Educational Research

Table 1.-Descriptive Statistics on the Pretest and Three Table 2.Students’ Most Frequent Responses to Open-Ended
Chapter Tests for the Cooperative Learning and Questionnaire Items
Comparison Groups

Itedcategory N %
Test M Adj. M SD
I . What did you like about the cooperative learning
~

Cooperative ( n = IS) groups?

Pretest 86.93 6.48 Getting help from others 8 38
Chapter test Getting to work with other students 4 19
I 42.53 42.7 1 7.53 Discussing, sharing ideas 3 14
2 41.53 4 I .54 5.78 Learning more 2 10
3 40.00 40.08 5.93 2. What did you not like about the cooperative
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

learning groups?
Compurison ( n = 16) Nothinglliked it all 5 29
Assigned, permanent groups 3 18
Pretest 88.00 4.4 I Uncooperative group members 2 12
Chapter test Feeling left behind 2 12
I 42.19 42.01 4.25
2 36.69 39.68 5.41 3. What concepts do you feel were easier to
3 35.94 35.86 5.27 understandkarn in groups?
Complex, difficult problems (hard to solve on own) 8 73
4. What concepts do you feel were easier to
ing about mathematics” and “getting to socialize and learn understandllearn on your own‘?
at the same time.” The opportunity to discuss and share Those requiring memorization 3 33
None 2 17
ideas was another frequently endorsed benefit (14%) of
group learning. Students responded that they liked being 5. What would you change about the cooperative
learning groups?
able to “share my ideas with others” and “to discuss prob-
Alternate group membership 5 36
lems with others.” Finally, 2 students (10%) said that they Nothing 4 29
learned more when working in cooperative groups. One
student commented that “I learned more from the group
than I do when you teach.”
When we asked the students what they disliked about the “Ones you just had to memorize,” and another said, “The
cooperative learning groups, they said that there was noth- easier ones like memorizing equations.” Two students
ing to dislike or that they liked it all (29%).One student ( 1 7%) responded that none of the material was easier to
wrote, “Everything about it was great. There was nothing to learn on their own. In elaborating on this response, 1 stu-
dislike.” Some students disliked having groups preassigned dent commented that “this subject is hard for me and I need
and permanent ( I 8%); others cited uncooperative group suggestions and advice on how to work out problems.” Two
members as a disadvantage of group learning (12%). In other related responses that did not address the question
addition, some students felt left behind when working in asked were that the students still had to study on their own
groups. For instance, 1 student complained of the times and that they could learn anything on their own if necessary.
“when people knew how to do a problem that I couldn’t and The final question posed to students was what they would
they left me behind.” change about the cooperative learning groups. The most fre-
Only one theme or category emerged in response to the quently occurring suggestion was to alternate group mem-
third item about what concepts were easier to understand bership (36%).An advantage of rotating group membership
and learn in groups. Students identified complex concepts cited by I student was “to help get fresh points of views.”
and difficult problems that were hard to solve on their own Some students responded to this question with “nothing” or
as easier to learn and understand in groups (73%). Repre- “I like them the way they are” (29%). In support of no
sentative comments in this category included “the concepts change, 1 student wrote, “Nothing because other people can
that were harder to grasp like proving identities”; “the con- show you how to do something and you also show others. It
cepts that were difficult to learn on your own”; and “the is an easier way to learn.” Individual suggestions included
things I couldn’t do.” One student wrote that it was “all” making the point system easier to understand, administering
easier; another said that detecting simple mistakes was eas- group tests, and allowing students to select their own groups.
ier in a group setting.
Whereas many students said that the more difficult con- Discussion
cepts were easier to learn in groups, 4 students said that the
content or material requiring memorization was easier to do The significant group differences on the achievement
on their own (33%).When asked what concepts were easi- measure support the first hypothesis that cooperative learn-
er to understand and learn individually, 1 student wrote, ing promotes mathematics achievement in the secondary
 September/October 1997 [Vol. 91(No. 111 47

grades. These results also replicate the achievement benefits among students, and the opportunity for social interaction.
in mathematics associated with cooperative groups that Nattiv (1994) reported that students liked to receive imme-
have been observed in the elementary grades (e.g., Slavin, diate help from their teammates rather than raising their
1985) as well as in the secondary grades (Duren & Cher- hands and waiting for the teacher to respond.
rington, 1992; Sherman & Thomas, 1986; Slavin & Karweit, Some students said that complex or difficult problems
1984). Moreover, these results extend previous research were easier to learn in groups, whereas material that simply
findings by illustrating that cooperative groups can also be required memorization was easier to learn on their own.
an effective strategy for learning advanced mathematical This pattern of results makes intuitive sense in that students
topics (precalculus) in the 1 1th and 12th grades. should particularly benefit from a classmate’s help on more
The significant Group x Time interaction suggests that it difficult or complex concepts compared with simpler concepts.
may take some time before the benefits of cooperative The only change consistently recommended by students
learning become apparent. Although there were increasing was to rotate group membership. Some students did not like
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

group differences on the second chapter test administered the fact that the groups were preassigned and permanent.
after 4 weeks, the groups did not significantly differ until Mulryan (1994) also reported that “most teachers and stu-
the scores on the third chapter test were compared, 6 weeks dents were in favor of relatively frequent changing of coop-
after the cooperative learning technique was initiated. In a erative small group membership” (p. 289). However, if the
survey of teachers who used small groups in mathematics assertion that the gradual development of group cohesion
instruction, they also observed that some groups required increases group productivity, then frequent changing of
more time to form a cohesive bond. In their view, this early group membership may impede this group dynamic, and ulti-
lack of cohesion initially deterred progress, but ultimately mately achievement.
the group dynamic increased productivity and achievement Although the present findings show that cooperative
(Good et al., 1989-90). The implication of this finding for learning promoted achievement and that students favorably
mathematics instruction is that teachers may have to be evaluated the group learning strategy, the study has some
patient in order to observe the advantages of group learning limitations. The first limitation is the potential teacher con-
relative to individual study. found. The same teacher taught both sections of the class;
We observed decreases in test scores for both groups therefore, her particular teaching style may have interacted
across time, but the decline in scores in the comparison with the treatment. A related limitation is the potential bias
group was greater than the decline in scores for the experi- on the part of the teacher that may have influenced both her
mental group. I n contrast, Sherman and Thomas (1986) teaching style across the two classes (e.g., enthusiasm) and
reported significant gains for both groups, with significantly the way that students responded to the questionnaire. Final-
greater gains for the cooperative learning group. A plausible ly, selection bias cannot be ruled out because the study was
explanation for the discrepancy may be the varying levels of a quasi-experimental design. Even though the groups did
subject difficulty and the samples of students included in the not differ significantly on the pretest achievement measure,
two studies. The instruction in the present study covered top- the students may have differed on other, unmeasured vari-
ics in precalculus taught to relatively advanced students, ables across the two classes.
whereas the students in the Sherman and Thomas study As Bassarear and Davidson ( 1992) emphasized, there
( 1986) were “primarily low academic achievers attempting have been recent calls for a fundamental reform in mathe-
to meet the state’s high school mathematics credit require- matics education in the United States. Small-group learn-
ment during their freshman and sophomore years” (p. 170). ing, in which students discuss ideas and solve problems, is
In this study, the decline in mathematics scores was likely a consistently described as one strategy to meet the challenge
function of the increasing difficulty of the course material. In of reform in mathematics education. This study provides
terms of the effectiveness of the cooperative learning proce- empirical evidence that cooperative learning can be an
dure, the important finding in each of the studies was the effective instructional strategy for learning advanced math-
increasing difference between groups, favoring the coopera- ematical topics at the secondary level.
tive learning condition over time.
The second prediction that students would favorably REFERENCES
evaluate the cooperative learning procedure was also con-
firmed. Most students reported that they liked receiving Augustine, D. K., Gruber, K. D., & Hanson, L. R. (1989-90). Cooperation
works! Educarionnl Leadership, 47, 4-7.
help from others or liked working with other students. Bassarear, & Davidson. N. (1992). The use of small group learming situa-
When asked what they did not like about the cooperative tions in mathematics instruction as a tool to develop thinking. In N.
learning group, the largest percentage said, “Nothing” or Davidson & T. Worsham (Eds.), Enhancing thinking thmugh cooperu-
five learning (pp. 235-250). New York: Teachers College Press.
that they “liked it all.” These survey results are consistent Burns, M. (1981). Groups of four: Solving the management problem.
with the student reactions to cooperative groups identified Learning, 12. 46-5 I .
by other researchers (Mulryan, 1994; Nattiv, 1994). In Mul- Davidson, N. (1985). Small-group learning and teaching in mathematics:
A selective review of the research. In R.E. Slavin, S. Sharon, S. Kagan,
ryan’s (1994) study, the predominate themes in student R. Hertz-Lazarowitz, C. Webb, & R. Schmuck (Fils.), Learning ro coop-
responses were collaboration and sharing, help and support erate, cooperaring to learn (pp. 21 1-230). New York: Plenum.
 48 The Journal of Educational Research

Duren. P. E., & Cherrington, A. (1992). The effects of cooperative group Mulryan, C. M. ( 1994). Perceptions of intermediate students’ cooperative
work versus independent practice on the learning of some problem-solv- small-group work in mathematics. The Journal of Educational
ing strategies. School Science and Mathematics, 92, 80-83. Research, 87, 28G29 I .
Edwards, K. J., Devries, D. L., & Snyder, J. P. (1972). Games and teams: Mulryan, C. M. (1995). Fifth and sixth graders’ involvement and partici-
A winning combination. Simulation and Games, 3, 247-269. pation in cooperative small groups in mathematics. Elementary School
Good, T. L., Reys, B. J., Grouws, D. A., & Mulryan, C. M. (1989-90). Journal. 95, 297-3 1 I .
Using work groups in mathematics instruction. Educational Leadership, Nattiv, A. (1994). Helping behaviors and math achievement gain of stu-
47, 56-60. dents using cooperative learning. The Elementary School Journal, 94,
Hamblin, R. L., Hathaway. C., & Wodarski, J. S. (1971). Group contin- 285-297.
gencies, peer tutoring, and accelerating academic achievement. In E. Sherman. L. W., &Thomas, M. (1986). Mathematics achievement in coop-
Ramp & W. Hopkins (Eds.). A new direction for education: Behavior erative versus goal structured high school classrooms. The Journal of
analvsis. Lawrence, Kansas. Educational Research, 79, 169-1 72.
Johnson, D. W., & Johnson, R. T. (1981). The internal dynamics of coop- Slavin, R. E. (1985). Learning to cooperate, cooperating to learn. New
erative learning groups. In R. E. Slavin, s. Sharan, s. Kagan, R. Hertz- York: Plenum.
Lazarowitz, C. Webb, & R. Schmuck (Eds.), Learning to cooperate, Slavin, R. E. (1988). Cooperative learning and student achievement. Edu-
cooperating to learn. New York: Plenum. cational Leadership, 46, 3 1-33.
Downloaded by [Computing & Library Services, University of Huddersfield] at 22:04 27 December 2014

Johnson, Maruyuma, Johnson, Nelson, & Skon. (1981). Effects of cooper- Slavin, R. E. (1989-90). Research on cooperative learning: Consensus &
ative, competitive and individualistic goal structures on achievement: A controversy. Educational Leadership, 47, 52-54.
meta-analysis. Psychological Bulletin, 89, 4 7 4 2 . Slavin, R. E. (1990). Cooperative Learning. New Jersey: Prentice-Hall.
Johnson, R. T., & Johnson, D. W. (1987). How can we put cooperative Slavin, R. E. (1991). Synthesis of research on cooperative learning. Edu-
learning into practice? The Science Teachec 54, 4 6 4 8 , 50. cational Leadership, 48, 5, 71-82.
King, L. H. (1993). High and low achievers’ perceptions and cooperative Slavin, R. E., & Karweit, N. L. (1984). Mastery learning and student
learning in two small groups. The Elementary School Journal, 93, teams: A factorial experiment in urban general mathematics classes.
399-4 16. American Educational Research Journal, 21. 725-736.
Madden, N. A., & Slavin, R. E. (1983). Effects of cooperative learning on Stevens, R. J., & Slavin, R. E. (1989-90). When cooperative learning
the social acceptance of mainstreamed academically handicapped stu- improves the achievement of students with mild disabilities: A response
dents. Journal of Special Education, 17, 171-182. to Tateyama-Sniezek. Exceptional Children. 57, 276-278.
Mulryan, C. M. (1992). Student passivity during cooperative small groups Wood, K. D. (1987). Fostering cooperative learning in the middle and sec-
in mathematics. The Journal of Educational Research, 85, 261-273. ondary level classrooms. Journal of Reading, 31. 10-1 8.