Peer and Cross-age Tutoring: An Overview

Peter L. Newcomb

City College of New York

Definition and Description

Peer tutoring and cross-age tutoring are teaching strategies that utilize one-on-one instruction. The usage of the terms peer tutoring and cross-age tutoring are sometimes confusing. Peer tutoring generally involves a dyad in which the tutor and tutee are of approximately the same age. Cross-age tutoring involves a tutor who is older than the tutee. However, the term peer tutoring is sometimes used to describe both types of tutoring (Gaustad 1993, http://eric.uoregon.edu/publications/digests/digest079.html). In this paper, "peer tutoring" will be used in the narrow sense of same-age tutoring only.

Educational researchers have developed various models for effective peer tutoring. In most, the tutor is more advanced in the subject area to be tutored than the tutee. However, this is not always the case. For example, the ASK your Partner to THink© tutoring model (King 1998) was developed for students of like ability and competence. This model trains students to use a sequenced series of questions and hints to scaffold their learning to higher levels. One student assumes the role of tutor, and the other that of tutee. The tutor asks questions, but does not provide explanation. The tutee answers the questions, but does not pose questions. The tutor begins by asking general questions to activate prior knowledge and assess the tutee’s understanding of the topic; she then gradually moves on to questions that require a higher degree of thinking. Researchers found the ASK your Partner to THink© tutoring model to be an "effective means of structuring tutorial interaction so that tutoring partners mutually scaffold each others' thinking and learning to progressively higher levels" (King 1998).

Peer tutoring is an adaptable strategy that can be used inside or outside the traditional classroom setting. One model that was designed for whole-class use is classwide peer tutoring, or CWPT. This system was developed by researchers at the Juniper Gardens Children's Project in collaboration with regular classroom teachers in an effort to improve instruction for children with learning disabilities or children from minority or disadvantaged backgrounds (Arreaga 1998). In CWPT, peer tutoring is used on a regular basis to reinforce concepts already presented by the teacher at the beginning of class, or in a previous class. CWPT involves using a game format in which the class is divided into two teams. Students are paired either randomly, or by similar skill levels, and each tutor-tutee pair is assigned to a team. The tutor asks a series of pre-determined questions, and awards points for the tutee’s correct answers. Each pair’s points are totaled to determine the team score. This method allows for competition between teams, but not directly between pairs. Tutoring pairs and teams are changed on a weekly basis to ensure all students have the chance to be on a winning team. While CWPT was developed for use with students with mild disabilities, it is appropriate for inclusive classes of mixed ability, as well as mainstream classes.

While present day school systems are marked by age stratification, this has not always been the case (Allen 1976). In the late eighteenth and early nineteenth centuries, educators Andrew Bell and Joseph Lancaster developed systems of school administration that used children as teachers. Bell first devised his system during his time as a superintendent of a school for orphans in India, while Lancaster drew on his ideas and actively promoted the system in Wales and England. The Bell-Lancaster system, as it came to be called, was used as an inexpensive method for schooling poor children, as it involved few paid teachers. The teacher drilled older students, who in turn taught groups of younger children, who would teach still younger children. Bell and Lancaster claimed that one advantage of the system was improved student behavior; younger students emulated the positive role models of older children placed in a position of trust and responsibility. The influence of the system waned, however, as the state began to provide money for public education in the nineteenth century.

In the American context, cross-age tutoring was used extensively in rural one-room schoolhouses, which were quite common well into the twentieth century (Allen, 1976). In these schools, older children were often assigned to help younger students with their lessons. As the sole teacher for several levels of students, the teacher had to consider the students themselves as valuable learning resources. In a 1974 study, researchers sent questionnaires to 110 one-room schoolhouses still operating in Nebraska (cited in Allen, 1976). Of these, 31% of schools used some form of tutoring on a regular basis, while an additional 25% had informal tutoring arrangements. Of students in these schools, 88% reported working with other students. Teachers reported positively on their school environments, in which children learned to work and play with children of different ages, with older children often taking responsibility for younger children. Some teachers reported a family-like closeness in their schools.

In the large, age-segregated schools prevalent today, opportunities for cross-age tutoring must be carefully designed and implemented. Cross-age tutoring not only benefits the younger learners, but can also be used to improve the skills of the older tutors. For example, cross-age tutoring has been used to improve the reading ability of struggling adolescent readers. A recent study involved using seventh grade remedial reading students to read to first and second graders (Fisher 2001). In this study, teachers developed lesson plans based on children’s books such as Hop on Pop and modeled the lessons for the seventh-graders. The seventh graders read to small groups of the first and second graders twice a week, while asking questions and performing activities from the lesson plan. The tutors kept a journal in which they recorded their tutoring experiences. They also wrote their own words for wordless books, and then read them to the younger students.

At the end of the study, "The literacy-related outcomes for the tutors, all of whom were struggling readers and/or identified with reading disabilities, all demonstrated significant increases" (Fisher 2001). They also outperformed a comparison group of students who did not engage in cross-age tutoring. Fisher identifies three key reasons for the success of the program: an authentic reason for literacy learning, extensive feedback, and the integration of writing into the curriculum. The seventh-graders wanted to improve their reading fluency in order to read to the younger children in a clear and entertaining manner. They also wanted to create interesting stories for the wordless books. Feedback came primarily from the younger students during the reading sessions; the readers could tell if they were boring their students, or if the students could not follow their reading. Finally, the program integrated the skills of reading, writing, listening and speaking. Tutors had to reflect on their reading by writing in their journals. In addition, they wrote their own stories, which they then read to the younger students. The seventh grade teacher also modeled the reading of adolescent literature, such as Harry Potter, for her class. By integrating the key language arts skills, students were able to make broad academic progress.

Cross-age tutoring can benefit the tutees as well as the tutors. One method designed to train tutors to help younger tutees to read is the "Pause, Prompt, and Praise" method (Wheldall 1990). The assumption behind this method is that students best learn to read through reading material at an appropriate level, rather than through exercises focusing on sub-skills. Briefly, tutors are trained to listen and respond to younger readers in a prescribed way. When the reader makes an error or hesitates, the tutor must wait at least five seconds or until the reader reaches the end of the sentence. If the reader does not correct himself, then the tutor prompts the reader with a question that may help him self-correct. If the student successfully self-corrects, the tutor reinforces him with verbal praise.

One example of the "Pause, Prompt, and Praise" method of tutoring involved 16-year-old tutors who were regarded as poor readers. They were trained to use the method to teach 12-year-old remedial readers (Wheldall 1990). Tutees taught by the trained tutors self-corrected around 20% of the time, while tutees in a control group self-corrected less than 10% of the time. Moreover, the experimental group of tutees gained a mean of six months in reading accuracy, while those in the control group averaged a 2.4-month gain. Wheldall went on to successfully replicate this study several times. He found that the "Pause, Prompt, and Praise" method works well for low-performing readers, but is not more beneficial than simply reading aloud to others for good and average readers.

Theory, Theorists, and Research

Theoretical frameworks that contribute to an understanding of the success of peer and cross-age tutoring include those of Piaget, Vygotsky, and Sternberg.

Piaget’s concepts of assimilation, accommodation, and equilibration are useful in explaining the peer learning process (Lisi 1999). Assimilation is a cognitive process in which an individual "files" new information under already existing schema. In contrast, accommodation requires cognitive schema to be changed in order to account for new information. However, "accommodation does not imply a permanent change or modification in a cognitive system or any of its components" (Lisi 1999, p.8). Rather, accommodation may take place in a situation such as accepting a teacher’s answers, without real understanding.

Equilibration is a process that can occur when an individual attempts to reconcile her current cognitive schema with an observation that does not fit. In some cases, a perturbation such as a failure to understand something leads the individual to restructure her cognitive schema in order to account for new information.

Researchers have studied the effect of contradiction and its resolution in pairs of children from a Piagetian perspective. Ames and Murray (1982, cited in Lisi, 1999) conducted an experiment to test the effect of social interaction of nonconserving children on their performance on conservation tasks. In one group, each child was paired with a nonconserving partner who had offered a conflicting, yet still incorrect, response to a conservation problem. A second group of children, in a model condition, listened to a peer respond to a conservation question. The peer’s response was wrong, yet different from the target child’s response. A third group of children were asked to pretend the opposite of their answer in the pretest. A fourth group of children were presented with two conservation items from the pretest which they had failed, with modifications to the items that suggested a contrasting nonconserving response. Finally, there was a no-treatment control group. Results showed that the social interaction group outperformed the other four groups on both an immediate and a delayed posttest. Apparently, the conflicting ideas of the two children resulted in a cognitive perturbation, which was resolved by restructuring of cognitive schema regarding conservation.

Piaget depicted children’s relationships with peers as different from adult-child relationships. Relationships with peers are often based on cooperation, while adults may dominate adult-child pairs. Children may accept the views of the adult authority figure without any real understanding. Researchers have investigated the potential for peer-peer interaction to promote cognitive change. Drawing from Piaget’s work on socialization, Damon and Killen studied peer interaction and young children’s reasoning about justice and fairness (1982, cited in Lisi, 1999). Following a pretest, children in peer triads worked together to reach consensus on new problem on fairness. On a posttest, they were shown to have improved their reasoning skills, suggesting that the peer interaction lead to cognitive growth.

Hogan and Tudge (1999) point out a key difference between Piaget’s and Vygotsky’s theories regarding cognitive development. Piaget stressed the value of social interaction with peers in a "socially symmetrical" relationship, while Vygotsky stressed the importance of collaborating with a more competent partner. For Vygotsky, age was not as important as the relative skills of the individuals. In a tutoring relationship, a more competent tutee can offer instruction at a level just beyond the tutee’s individual capability, that is, within the tutee’s zone of proximal development.

Some research has called into question the ability of more advanced peers to teach within the zone of proximal development of less advanced peers (Hogan et al., 1999). Ellis and Rogoff (1986) found that child teachers were not as good as adults in adjusting to learner’s levels of understanding. Hogan and Tudge (1999) caution that not all social interaction necessarily leads to cognitive growth. More competent peers, in the role of tutor, must be able to tailor their guidance to a level appropriate for less advanced peers in order to optimize the opportunity for cognitive growth.

Training tutors effectively is crucial to the success of peer and cross-age tutoring. Skilled tutors are capable of assessing tutees’ knowledge and structuring their tutoring accordingly (Person and Graesser 1999). Moreover, skilled tutors engage in dialogue with tutees that is qualitatively different from traditional classroom dialogue. Much classroom dialogue takes the form of a three-step dialogue frame, in which the teacher asks a question, a student answers, and the teacher gives feedback on the student’s answer. Person and Graesser (1999) characterize skilled tutoring discourse as occurring within a five-step dialogue frame. Particular importance is attached to step four, in which "tutor and student collaboratively improve the quality of the student’s answer" (p. 72). Techniques used by the tutor in step four include hinting, prompting, splicing (interrupting students to insert correct information), prompting, and summarizing. Person and Graesser conclude: "We believe that the learning advantages that one-to-one tutoring has over other learning environments is largely due to the conversational moves or strategies that tutors use in Step 4 of the dialogue frame" (p.86).

Advocates of peer and cross-age tutoring point out the positive effects of the experience on the tutors, not just the tutees. Sternberg’s triarchic theory of intelligence offers a framework for understanding the skills that tutors may develop from peer tutoring (Hartman 1990, cited in Topping 1998). These include the meta-cognitive skills of planning, monitoring, and evaluating peer tutoring sessions. Effective peer tutoring involves assessing and activating the learner’s prior knowledge in order to tutor at an appropriate level. Moreover, the effective tutor attends to feedback from the learner to monitor their level of understanding as the tutoring session progresses. Finally, the tutor should reflect on and evaluate the level of success of the tutoring session, and plan adjustments to future sessions accordingly.

Peer and cross-age tutoring are not only means of learning academic material. Briggs (1998) asserts that the primary advantage of peer and cross-age tutoring is the development of personal intelligence. In terms of Gardner’s theory of multiple intelligences, traditional classroom teaching methods do little to develop personal intelligence. Briggs argues that the neglect of personal intelligence in education is partly to blame for misbehavior and disobedience. Conversely, Briggs (1998) claims that there are great social benefits from peer teaching:

Mastery of teaching and success in the relationship brings spontaneous growth in both peer teacher and student. It strengthens and makes smoother transitions from childhood to adolescence to early adulthood. It transforms troublesome behavior into constructive behavior. (p. 63)

Indeed, research has shown that peer and cross-age tutoring can have a positive effect on students’ self-concept and their likelihood of dropping out of school (Roswal et al, 1995). In a large study involving 7^th grade students, students who participated in a sixteen-week collaborative peer tutoring program "demonstrated a significant improvement over both traditional classroom groups in both self-concept and attitudes toward school as demonstrated by the dropout scale" (Roswal et al, p. 279). Roswal and his fellow researchers used the Piers-Harris Self-Concept Scale to measure self-concept and the Demos D (Dropout) Scale to measure student tendency to drop out of school. This collaborative peer tutoring program was similar to the class-wide peer tutoring discussed above in that the group was divided into teams of 25 students. Subteams from these groups competed daily in a scholar-bowl format, using questions formulated by other subteams. Each team contained low, middle, and high-achieving students, and a mix of cultural and socioeconomic backgrounds.

Peer tutoring can have a significant positive effect on future school performance and the prevention of school failure (Greenwood & Delquadri 1995). A longitudinal study assessed the benefits of a classwide peer tutoring (CWPT) program conducted in the first through fourth grades. By the time they were in middle school, students who participated in the CWPT program were less likely than those in a control group to be receiving special education services. CWPT students also exceeded the achievement test levels of the control group on the California Test of Basic Skills (reading, language, and math). By 11^th grade, the students who participated in the CWPT program dropped out of school at a lower rate than students from a control group.

The researchers concluded that "Peer-mediated methods such as CWPT have relatively greater potential for increasing the active engagement of more diverse learners than do many conventional teacher-mediated methods at comparable levels of effort and cost" (Greenwood & Delquadri, p.25).

In fact, peer tutoring has been found to be a relatively cost-effective measure compared with common approaches to improving elementary school students’ reading and mathematics skills (Levine et al, cited in Topping, 1998). Peer tutoring was found to be a more cost-effective option than computer-assisted learning, reducing class size, or lengthening the school day.

Conclusion

Peer tutoring and cross-age tutoring offer many advantages over the traditional teacher-centered model of education. Students can spend much more time actively engaged with the subject matter by working in pairs. They may find motivation to learn in helping others, and gain a sense of responsibility for their learning. Various peer tutoring and cross-age tutoring schemes have been shown to boost academic performance, improve self-concept, and decrease the likelihood of dropping out. These methods have proven successful with students with learning disabilities and low academic achievement.

Why, then, aren’t these methods more widely used? It may be that evidence of their effectiveness was considered lacking until the late 1980s (Kalkowski 1995, http://www.nwrel.org/scpd/sirs/9/c018.html ). Teachers may feel their students aren’t capable of effective tutoring, or fear losing control of their classrooms. Another factor may be lack of training for prospective teachers and professional development for current teachers. I believe the most important underlying factor is the entrenchment of the idea of the teacher as the sole disseminator of knowledge in the classroom. In my fieldwork observations in a Bronx high school, every class I have seen has been teacher-centered, with no opportunity for students to communicate meaningfully with each other on the topic at hand.

It is clear from my review of the research that designing and implementing successful peer and cross-age tutoring programs takes time and careful planning. Tutors should receive training and feedback, and learn to evaluate and adjust their own tutoring to best serve the needs of the tutee. Students must clearly understand what they are expected to do in tutoring sessions. For a novice teacher, I believe it would be best to start by following an already established model of tutoring, such as the ASK your Partner to THink© tutoring model, or the Classwide Peer Tutoring model.

As a teacher of English as a Second Language in Japanese middle schools, I often grouped students in pairs for various games and activities. My main reason for doing so was to create a situation in which students could speak a maximum amount of English without the stress of speaking in front of the entire class. Students who understood the target English often helped partners who did not. It could be considered a simple form of peer tutoring. Researching this paper has opened my eyes to many more sophisticated methods of peer tutoring, in addition to cross-age tutoring, which I hope to use in my future teaching experience.

References

Allen, V. (1976). The helping relationship and socialization of children: Some perspectives on

tutoring. In Allen, V. (Ed.), Children as Teachers. (pp. 9-26). New York: Academic Press.

Arreaga-Mayer, C. (1998). Increasing active student responding and improving academic

performance through classwide peer tutoring. Intervention in School & Clinic, 34, 89-96.

Briggs, D. (1998). A class of their own: When children teach children. London: Bergin and

Garvey.

Fisher, D. (2001). Cross age tutoring: Alternatives to the reading resources room. Journal of

Instructional Psychology, 28, 234-241.

Gaustad, J. (1993). Peer and cross-age tutoring. ERIC Digest 79. University of Oregon.

Available: http://eric.uoregon.edu/publications/digests/digest079.html.

Greenwood, C. R., & Delquadri, J. (1995). Classwide peer tutoring and the prevention of school

failure. Preventing School Failure, 39(4), 21-25.

Hogan, D., & Tudge, J. (1999). Implications of Vygotsky’s theory for peer learning. In

O’Donnell, A., & King, A. (Eds.), Cognitive Perspectives on Peer Learning. (pp. 39-65).

Mahwah, NJ: Lawrence Erlbaum Associates.

Kalkowski, P. (1995). Peer and cross-age tutoring. Northwest Educational Research

Laboratory School Improvement Research Series, #18.

Available: http://www.nwrel.org/scpd/sirs/9/c018.html.

King, A., Staffieri, A., & Adelgais, A. (1998). Mutual peer tutoring: Effects of structuring

tutorial interaction to scaffold peer learning. Journal of Educational Psychology, 90, 134-

153.

Lisi, R., & Golbeck, S. (1999). Implications of Piagetian theory for peer learning. In

O’Donnell, A., & King, A. (Eds.), Cognitive Perspectives on Peer Learning. (pp. 3-37).

Mahwah, NJ: Lawrence Erlbaum Associates.

Person, N., & Graesser, A. (1999). Evolution of discourse during cross-age tutoring. In

O’Donnell, A., & King, A. (Eds.), Cognitive Perspectives on Peer Learning. (pp. 69-86).

Mahwah, NJ: Lawrence Erlbaum Associates.

Roswal, G., Mims, A., Evans, M., Smith, B., Young, M., Burch, M., Croce, R., Horvat, M.,

Block, M. (1995). Effects of collaborative peer tutoring on urban seventh graders. Journal of

Educational Research, 88 (5), 275-280.

Topping, K. (1998). The effectiveness of peer tutoring in further and higher education: A

typology and review of the literature. In Goodlad, S. (Ed.), Mentoring and Tutoring by

Students. (pp. 49-69). London: Kogan Page.

Wheldall, K., & Colmar, C. (1990). Peer tutoring for low-progress readers using ‘Pause, Prompt

and Praise’. In Foot, H., Morgan, M., & Shute, R. (Eds.), Children Helping Children. (pp.

117-134). Chichester, England: John Wiley & Sons.