The study investigated the behaviors and interactions of children in structured and unstructured groups as they worked together on a 6-week social studies activity each term for 3 school terms. Two hundred and twelve children in Grade 1 and 184 children in Grade 3 participated in the study. Stratified random assignment occurred so that each gender-balanced group consisted of 1 high-, 2 medium-, and 1 low-ability student. The results show that the children in the structured groups were consistently more cooperative and they provided more elaborated and nonelaborated help than did their peers in the unstructured groups. The children in the structured groups in Grade 3 obtained higher reading and learning outcome scores than their peers in the unstructured groups.Research in recent years has shown that it is no longer necessary to defend cooperative learning as an instructional strategy that promotes learning (Cohen, 1994). Cooperative learning has been used successfully to promote learning achievements across a range of curriculum areas, from narrative writing in small groups (Zammuner, 1995), to problem solving in mathematical tasks (Webb & Farivar, 1994), to conceptual understanding in physics (Howe, Tolmie, Greer, & McKenzie, 1995). In the affective area, it influences the development of positive student attitudes toward other group members (Bennett, 1991;Shachar & Sharan, 1994) and motivation to learn (Sharan & Shaulov, 1990). In effect, cooperative learning has helped to enhance the achievement of a range of cognitive and affective goals (Bossert, 1988;Ross & Raphael, 1990). However, although support for cooperative learning is unequivocal, only a few studies have attempted to identify the variables that mediate the relationship between cooperative group experiences and learning outcomes (Knight & Bohlmeyer, 1990). Identifying these variables is crucial to understanding not only which ones influence learning in cooperative groups but also how they influence it. Whereas numerous studies have suggested potential mediating variables (see Cohen, 1994;Johnson & Johnson, 1989, this study focuses specifically on student interactions (Webb, 1985(Webb, , 1992 in groups in which the small-group procedures and the type of task have been structured to promote cooperation.
Student InteractionsIn a series of studies that examined children's verbal interactions as they worked together in groups, Webb (1982,