The Study Of Influence Of Ability Group Composition On Cognitive Interaction And Cognitive Outcomes In Cooperative Learning

Cooperative learning, or collaborative learning(CL) as a widespread teaching strategy, has been studied abroad for several decades. It is a kind of situated learning, reflecting the demand on teaching and learning under constructive perspectives. Its advantages and features are to make full use of the resources of interaction among peers. In this report some theories about CL introduced by Johnson brothers and Slavin (who are leading authorities in the field of CL) are reviewed. CL's theory foundation can be classified to two sorts, non-cognitive and cognitive perspective. Johnson brothers and Slavin are representatives of non-cognitive, while Piaget and Vygotsky's theories are cognitive-oriented. Extending from the Piaget and Vygotsky's view, some contemporary psychologists intend to think CL's cognitive interaction composed of cognitive support and cognitive conflict. Some factors that affect interaction were identified, especially about ability group composition (i.e., ability homogeneous or heterogeneous group) on cognitive interactions and cognitive outcomes of CL. In addition, concept of interaction, method about measurement and relationship between interaction and cognitive effects of CL were introduced.The experiment was conducted in four-grade mathematical classroom in an elementary school. The aim of this experiment is to examine if the factor of ability group composition will affect the student's learning process and outcomes when they were asked to work with peers in a 4 or 5 member cooperative learning team. The assigned learning task is to carry out two complex and multiple-steps word problems. The purpose of the investigation is to find if we can see various interaction and effects between ability homogeneous and heterogeneous group, the students with various ability level benefit differently from CL group with different ability group composition by analyzing the videotape recorded and scores they got in the CL experiment.What can we get from this experiment? At first we know interactions and effects between homogeneous and heterogeneous group are different. From total score of view, there is no evident difference between homogeneous and heterogeneous group. The average score of homogeneous group is lower than homogeneous group but variance of homogeneous group is greater than homogeneous group. From strategy score of view, there is still no evident difference between homogeneous and heterogeneous group but the average score of homogeneous group is higher than homogeneous group. When high ability students work in homogeneous group, they exhibit excellent involvement, they discuss and inquire or ask for reason or refinements, they can actively check or monitor others progress, and they get highest strategy scores. The individual final performance of students with high ability in heterogeneous group is worse than their same performance in homogeneous group. Their total score (significance level=0.2) and strategy score (significance level=0.05) in heterogeneous group are lower than them in homogeneous group. When low ability students work in homogeneous group, due to their low ability level, they often are not able to express their ideas, even if they do, the questions posed are not dealtwith critical aspect of the problem, and their attentions are seldom paid to others, when they are in heterogeneous group they behave better than in homogeneous group. When medium ability students work in homogeneous or heterogeneous group, there are no clear interaction mode or pattern we can find. According to the degree of participate of the member, who exhibit more positive attitude toward CL, who will get higher individual transfer test grades. When medium ability students work in heterogeneous group, their strategy scores (significance level=0.2)are higher than their strategy scores in homogeneous group, but there is no evident difference between their total scores in homogeneous and heterogeneous group. There is evident relation between the common score of total group and the average score of group member individual...