Elementary school pupils can learn better when they help teach one another than they do in completely teacher-directed classrooms, according to researchers who analyzed more than 80 studies on peer-assisted learning.
An abstract of the report,“Peer-Assisted Learning Interventions With Elementary School Students: A Meta-Analytic Review,” is available from the Journal of Educational Psychology.
“This approach does seem to be effective,” said Cynthia A. Rohrbeck, one of the lead authors of a report published last month in the Journal of Educational Psychology. “And it particularly seems to be effective with younger kids, urban kids, low-income kids, and minority kids.”
The authors were quick to add, however, that their findings do not suggest that teachers should haphazardly assign students to work together and then simply sit back to watch what develops. Their study focused only on structured peer-tutoring programs with accumulated records of research.
The researchers used updated meta-analytic techniques to measure the average size of the test-score gains students made, giving more weight, for instance, to those that came from studies involving large groups of students.
Certain aspects of the peer-assisted learning programs seemed to be more effective than others, the review also found. Students in grades K- 6 made higher test-score gains when they took part in learning groups in which their grades depended to some extent upon how the entire group performed.
“There was usually some elaborate system where kids were given some points individually and some points for their group work,” said Ms. Rohrbeck, an associate professor of psychology at George Washington University in Washington. Her co-authors were Marika D. Ginsburg-Block of the University of Delaware, in Newark; John W. Fantuzzo of the University of Pennsylvania, in Philadelphia; and Traci R. Miller, also of George Washington University.
Stickers and Goals
Strategies that provided students with some autonomy—letting them choose their own reward stickers or set their own goals—also produced bigger learning gains.
On the other hand, the results showed it mattered less whether the student groups were working with curricula tailored to their individual learning needs.
While peer-assisted learning strategies have been described in education for 30 or more years, relatively few teachers use the kind of research-backed strategies the researchers studied. According to the report, surveys show that only about one- fifth of teachers use such strategies, even though two-thirds say they give students opportunities to interact on a weekly basis.
“They may be putting students together, but they’re probably using their own idiosyncratic ways of doing it,” Ms. Rohrbeck said.
The What Works Clearinghouse, the U.S. Department of Education’s center for vetting research on educational programs and practices, is planning to produce a similar review of studies on peer- assisted learning programs next year, according to Ms. Rohrbeck, who is consulting on that project.