This study represents a collaborative school university partnership. Using a mixed-method approach, the authors report on the motivational and psychological consequences of students choosing their groupmates in cooperative learning triads. 139 students in five science classes participated in this study. Classes were randomly assigned to condition: Teacher-selected or student-selected. In teacher-selected classes, the teacher chose the members of each group; in student-selected classes, the students chose their groupmates. Results revealed a decrease in willingness to choose one's groupmates. Focus group data indicated that students felt obligated to choose friends as groupmates, and low-achieving students questioned the value of working with similar others. Teachers should be aware that when permitting students to choose their groupmates that friendships and status hierarchies exert strong influences on choice of partner.
Grouping students has become a standard instructional approach for many classroom teachers at all levels (Cohen, Manion, & Morrison, 1996; Cooper, 1999; Orlich, Harder, Callahan, & Gibson, 1998). By encouraging children and adolescents to learn and work together, cooperative learning attempts to create a shift from the paradigm of knowledge transfer from an active teacher to passive pupils, to one of social constructivism, where knowledge is actively created by students through social interaction on academic tasks (Johnson, Johnson, & Holubec, 1992).
Most cooperative learning models call for teachers to create groups and select members according to criteria that maximize diversity of learning styles, gender, race, culture, achievement, and other relevant qualities (Cohen, 1984, 1994; Cohen & Lotan, 1995; Johnson at al., 1993; Kagan & Kagan, 1994; Slavin, 1995). Heterogeneous groupings are recommended since they encourage the acceptance of diverse styles and points of view, promote achievement in mixed ability classes, and produce benefits in socio-emotional domains (Abrami et al., 1995; Slavin, 1990, 1993, 1995).
Heterogeneous cooperative grouping is also seen as the antidote to the systematic alienation of women and students of color in math and science courses. Oakes, Ormseth, and Camp (1994) emphasized the importance of forming groups with mixed abilities that statistically reflect the overall gender and racial mix in a classroom, so that the majority may understand minority approaches. Slavin (1990) cautioned that too thin a distribution of minority students might actually be harmful, especially in nontraditional areas, since this may, in effect, be isolating for the minority students (i.e., the only one in the group). Other research highlights the effects of alienation on achievement scores some groupings may produce (Johnson, 1997; Rosser, 1997), as groups may not demonstrate the underlying assumption of cooperative learning, i.e. that students' voices are heard and valued simply because they are members of the group (Evans, 1996). Women, for example, are more likely to drop out of a group if they are the only female, especially in non-traditional settings (Light, 1990). Etzkowitz, Kemelgov, Newschatty, Uzzi, and Alonzo (1994) argued for groupings of several women or people of color within cooperative learning groups in order to decrease isolation and "spotlighting" of their differentness.
Selecting Cooperative Partners
Once a teacher has decided to employ group work, he or she is faced with a number of practical questions concerning cooperative groupings. These questions include the issue of group composition, appropriate tasks and roles, and the methods of formation. One especially problematic question many teachers face is "Should students be allowed to choose their own groupmates?" Students can place a great deal of pressure on teachers to form their own groups. This pressure stems from the notion common in childhood and adolescence that one works with friends, rather than the reality of adult life in which one is not necessarily friends with coworkers (Cohen, 1994). …