Research on learning through collaboration has shown that students who teach their peers by providing explanations benefit more than the students who hear the explanations. Tutors benefit from cognitively reorganizing the material in order to explain to their peers. Researchers are trying to find ways to improve the learning of the receivers of information in collaborative settings such as cooperative learning, reciprocal teaching or peer tutoring.
Researchers at Peabody College of Vanderbilt University have been studying the quality and effectiveness of students’ mathematical explanations as a function of their achievement level. These researchers were looking for ways to optimize the benefits of collaborative learning arrangements in heterogeneous classrooms, specifically how to group students to improve outcomes for the greatest number of students. Lynn S. Fuchs, Douglas Fuchs, Kathy Karns, Carol L. Hamlett, Suzanne Dutka, and Michelle Katzaroff carried out a study in 20 elementary classrooms where students had been trained in constructive peer tutoring.
In each classroom, two pairs of students were compared. Each pair included the same learning-disabled math student. One pair had a high-achieving math tutor paired with the learning-disabled student, while the other had an average-achieving tutor paired with this same student.
In all, the study involved 60 students in twenty second-through-fourth-grade classrooms from four elementary schools in a Southeastern urban school district.
Students were identified as average or high achievers on the basis of their teachers’ nominations and their performance on a standardized math test. Each learning-disabled student was judged by the teacher to be chronically low-achieving in mathematics and had been classified as L.D. according to state regulations. The three students in each classroom were directed in peer-assisted learning methods by their teacher.
The researchers analyzed videotapes of all sessions to compare the quality and effectiveness of the explanations provided by the two children from each classroom who were serving as tutors. They were not aware that one tutor had been identified as high-achieving and the other as average-achieving.
The researchers identified high-quality explanations as including more verbal interaction between tutor and tutee, making more and longer task statements, explaining more often, providing more demonstrations, asking more questions and focusing on strategies for solving problems rather than on giving answers.
Both tutors in each classroom had been trained to interact effectively and productively in collaborative activities, and had practiced for almost six months.
This careful training and sustained practice created opportunities for all tutors to develop effective instructional patterns. Moreover, each tutee was required to be involved actively by applying explanations to novel problems.
Analysis and Results
High-ability tutors were rated significantly higher on quality-of-explanation items; they relied on more strategies and brought a more conceptual focus to their explanations. The learning-disabled students completed more problems correctly when working with the high-achieving tutors — 91 percent correct compared to 75 percent correct when working with average-achieving tutors. However, there was no difference in the ratings on social interactions and attention to task between average- and high-achieving tutors.
Average-achieving students less effective tutors
Fuchs et al. conclude that even with extensive training and practice, the average-achieving students were less effective tutors than high-achievers. The conceptual focus of their explanations and the greater variety of explanation strategies are believed to have produced better performances in low-achieving peers.
Fuchs et al. believe this has important implications for classroom practice. In formulating student groupings, teachers should vary the mathematical ability level of students so that the greatest number of children have opportunities for constructing explanations.
For example, groups should include high-achieving students to ensure the quality of explanations in that group. If a class is divided into quartiles of math achievement, then the highest-achieving students can be grouped with the lowest, and the second quartile can be grouped with the third quartile, so that the higher-achieving of the average students can help the low-average performers.
Despite the fact that high achievers will be explaining concepts well below their own mastery levels, impressive benefits have been documented for high achieving tutors in earlier studies.
Limitations of Study
These results should be viewed with the knowledge that these peer tutoring sessions were videotaped and took place outside the regular classroom. This study only used pairs of students.
Small groups used frequently in collaborative activities were not studied. Most important, this study provides no information about the long-term achievement of low-performing students when tutored by higher-achieving students.
“The Relation Between Student Ability and the Quality and Effectiveness of Explanations”, American Educational Research Journal, Volume 33, Number 3, Fall 1996, pp. 631-664.
Published in ERN January/February 1997 Volume 10 Number 1