Tall Buddies: Peer-Assisted Learning for Elementary Schools

Introduction and Rationale

The ability to read well contributes to virtually all other academic pursuits, and it is little wonder that a great deal of attention has been paid to this aspect of education in recent years. In this regard, Otaiba, Fuchs, Fuchs et al. (2001) emphasize that "the problems associated with limited reading competence are serious. Students with poor reading skills have lower self-esteem, pose greater discipline problems, and are less likely to complete school than more skillful readers" (p. 15). One initiative that has been shown to be highly effective in helping young students with reading problems achieve improved ability across the board is peer-assisted learning (Otaiba et al., 2001). According to Dufrene, Duhon, Gilbertson, and Noell (2005), "Peer tutoring has been shown to increase significantly students' time spent in academic instruction and engagement and student outcomes. Additionally, peer tutoring has been used effectively to address individual needs in remedial programs for at-risk and mildly disabled students" (p. 74).

Peer-assisted learning initiatives are especially useful alternative instructional approaches because they take advantage of a relatively abundant resource — students — and place comparatively limited demands on scarcer resources such as educators. Moreover, such programs provide a framework in which students can learn cooperative work skills in addition to the academic skills being targeted (Dufrene et al., 2005). The beneficial outcomes of peer tutoring programs on students' reading performance have been previously documented (Dufrene et al., 2005). While a wide range of potential positive outcomes are possible for students engaging in peer tutoring programs, academic progress is typically the area that has received the most attention. However, besides gains in scholastic achievement, researchers have suggested that there are a number of other beneficial student outcomes that result from peer tutoring initiatives, including improved academic self-efficacy and persistence (Hartman, Mencke, Uribe, & Xu, 2001). These authors add that "tutoring also brings a social component to the learning process, given that educators now believe more effective learning takes place when there are two or more people involved" (Hartman et al., 2001, p. 22).

In reality, such learning approaches are not new. Some of the earliest research from which cooperative learning techniques were developed goes back to the early 1900s, and the foundation of peer tutoring can be traced to Andrew Bell and Joseph Lancaster in the late eighteenth century, or perhaps even earlier (Falchikov, 2001). This study provides a critical review of the relevant peer-reviewed and scholarly literature concerning peer-assisted learning in general and how these programs have been used to help young readers in particular, as well as a case study of the effectiveness of this initiative at an elementary school that provides services for at-risk sixth graders with emotional, social, or behavioral problems and first-grade struggling readers. The product of this study is a best practices guide to help educators at all levels identify which peer-assisted technique might be best suited for their individual needs, how to implement it, and how to monitor its effectiveness to ensure long-term success.

Importance of the Study

Today, many teachers are overwhelmed by the special challenges they face. Not only are there escalating numbers of students categorized as emotionally or behaviorally disordered, and escalating demands on teachers to fill non-teaching roles in children's lives, but teachers today also face an increasingly diverse set of learners in any given classroom. For example, Fuchs, Fuchs, and Saenz (2005) emphasize that "American schools of the 21st century face the challenge of educating the world's most diverse student body. This diversity is reflected in variations in achievement, socioeconomic status, cultural background, and linguistic background" (p. 231). Current estimates suggest that more than 2 million students come from non-English-speaking backgrounds, and that by the year 2020 this number would reach 6 million (Fuchs et al., 2005). There is also a pressing need to address individual differences in capacity among an increasingly diverse set of learners in the general population, with many educators turning to alternatives to whole-class instruction — alternatives that share the common feature of peer-assisted learning (Keogh & Speece, 1996).

It is this researcher's belief that helping troubled sixth-grade students feel capable, connected, and contributing members of a school will help decrease behavioral problems and increase students' self-esteem. This belief is congruent with Ma (2003), who reports that young students' sense of belonging to their school plays an important role in achieving successful academic outcomes. The principal investigator also believes that training sixth-grade students to tutor low-achieving first graders will give those first graders a chance to make accelerated growth. With the Tall Buddy program, three fundamental positive outcomes are possible for the sixth graders: (a) behavior can improve, (b) student self-esteem increases — which is necessary if students are to behave more responsibly — and (c) reading skills improve.

Beyond these contributions, another important outcome of Tall Buddy tutoring is to improve the reading skills of both Tall Buddies and Small Buddies to ensure that all students become proficient readers. First graders who have difficulty in reading quickly fall behind their classmates. First-grade teachers can predict with some confidence that students with considerable reading deficits by the end of the school year will likely have long-lasting challenges in reading throughout their school years. Indeed, as Paterson and Elliott (2006) emphasize, "As students progress from elementary school through middle school to high school, they confirm their attitude toward reading. Many high school readers who have struggled with reading along the way carry deeply entrenched negative beliefs about the reading process and, consequently, construct barriers to protect themselves against feelings of failure. These attitudes push reading achievement into a downward spiral" (p. 378).

All teachers involved should select academic measures to track students' reading levels both before Tall Buddies begins and during the tutoring program. It is also important to note that when a Tall Buddy is chosen to participate in the program, the privilege is non-revocable — their position is not held over their head or threatened to be ended due to poor behavior, incomplete homework, or unfinished assignments. It takes time to break habits that have been in place for years. This researcher believes in progress toward the desired behavior, not perfection.

Review of Literature

According to Ehly and Topping (1998), "Peer tutoring is where tutor and tutee are clear about their respective and separate roles and goals. However, peer-assisted learning (PAL) also encompasses peer education and counseling, and peer modeling, monitoring, and assessment" (p. xiii). These authors suggest that many educators may be aware of the general concepts of peer tutoring; however, reciprocal tutoring, class-wide tutoring, paired learning methods, and the deployment of students with special needs as tutors may be relatively unfamiliar (Ehly & Topping, 1998). This point is also made by Otaiba and colleagues (2001), who report that "one promising alternative to conventional instructional methods is a collaborative arrangement whereby children work together to support each other's learning. Research demonstrates that in the elementary grades, children's reading competence can improve when they work collaboratively on structured learning activities" (p. 15). Using such peer-assisted learning approaches, groups of young students are able to operate at different levels of curricula using alternative instructional procedures; moreover, teachers can create and simultaneously implement different lessons to address a greater range of individual learning needs (Otaiba et al., 2001).

Despite the increased interest in peer-assisted learning initiatives, there are still many things that remain better described than understood in the scholarly literature on tutoring and mentoring, and on which further research needs to be conducted (Falchikov, 2001). There is, however, a growing body of research concerning the effectiveness of peer-assisted learning, with studies conducted with students with mild disabilities enrolled in general education classes and from studies carried out in separate special education classes for students with severe disabilities (Fister, Mathot-Buckner, McDonnell, & Thorson, 2001).

In particular, peer tutoring has been extensively examined as an alternative learning strategy for providing students with severe disabilities with instruction tailored to their individual needs (Fister et al., 2001). Research studies have repeatedly demonstrated that students without disabilities can be effective in teaching a wide range of academic and developmental skills to this group of students; however, there remains a paucity of studies that have specifically examined the effectiveness of these strategies in meeting the educational needs of students with severe disabilities in general education classes (Fister et al., 2001). The review of the relevant peer-reviewed and scholarly literature for this study provides the background and an overview of peer tutoring in general and the potential benefits that accrue from such initiatives, followed by a discussion of implementation considerations. Methods for evaluating and monitoring the effectiveness of peer-assisted learning programs are discussed as well, followed by a summary of the literature review.

The growing body of scholarly evidence concerning peer tutoring has consistently emphasized the powerful effects that children can exert on the academic and interpersonal development of their classmates and other students (Ehly & Topping, 1998). For example, Bloom (1984) reported early on that one-on-one tutoring by a fully skilled peer was more effective than both conventional (i.e., teacher-led lecture) and mastery learning (i.e., student-regulated) methods of teaching. Across several replications involving different academic content and student age levels, Bloom (1984) reported that peer tutoring programs produced effect sizes on the order of 2 standard deviations above the mean of the control group — students receiving conventional lecture-based instruction — compared with 1.3 standard deviations for mastery learning. Effect sizes larger than .25 of one standard deviation were described as educationally significant (Ehly & Topping, 1998). Likewise, Slavin (1990) enthuses that "one-to-one tutoring is the most effective form of instruction known" (p. 44). Besides these benefits to the tutee, peer tutoring programs have been shown to provide improved academic outcomes when used on a same-age, cross-age, and class-wide basis; moreover, many empirical investigations have documented that mutual benefits accrue from the systematic use of peer tutoring programs, wherein children who serve as tutors frequently achieve academic gains comparable to those who receive tutorial assistance (Ehly & Topping, 1998).

Generally speaking, peer tutoring programs create alternative teaching arrangements in which students act as instructional agents for one another (Harper, Maheady, & Mallette, 1994). Peer tutoring also provides opportunities for students to discuss the content and processes of reading, and studies to date suggest that specific methods can enhance students' ability to monitor their own processes and that the social context plays a major role in supporting students' understanding of these strategies (Afflerbach, Baumann, Duffy-Hester, Hoffman, McCarthey, & Ro, 2000). The potential advantages of peer tutoring programs include providing a framework that allows the teacher to customize instruction to the needs of individual students and to provide a higher number of instructional trials in one-on-one or small-group teaching formats (Fister et al., 2001).

One theoretical model assigns the main sub-processes of peer tutoring into five categories. The first includes organizational or structural features of the learning interaction, such as the press toward increased time on task (t.o.t.) and time engaged with task (t.e.t.) inherent in peer-assisted learning, the need for both helper and helped to elaborate goals and plans, the individualization of learning and immediacy of feedback possible within the one-on-one situation, and the variety offered by a different kind of learning interaction.

From a cognitive perspective, peer-assisted learning alternatives involve both conflict and challenge; they also involve support and scaffolding from the more competent participant, within the so-called "Zone of Proximal Development" of both the tutor and the tutee (Harrison, 2001, p. 157). According to Harrison, "The cognitive demands upon the helper in terms of detecting, diagnosing, and correcting errors and misconceptions is substantial — and herein lies much of the cognitive exercise and benefit for the helper" (p. 157).

Peer-assisted learning also places heavy demands on the communication skills of both helper and helped, but in so doing these initiatives also serve to develop those skills. Participants might never have truly grasped a concept until they had to explain it to another, embodying and crystallizing thought into language. The affective component of peer-assisted learning can also prove very powerful. A trusting relationship with a peer who does not occupy a position of authority might well facilitate self-disclosure of ignorance and misconception, thereby enabling subsequent diagnosis and correction (Harrison, 2001). Likewise, modeling of enthusiasm and competence by the tutor — and the simple possibility of a successful outcome — can influence the self-confidence of the tutee, while a sense of loyalty and accountability to each other might provide the motivation needed to keep both participants focused and on-task (Harrison, 2001). Previous research seeking to identify why and how peer tutoring is effective has pointed to the role of specific interactions in promoting learning during tutoring and other peer-directed small-group learning (Adelgais, King, & Staffieri, 1998).

Some of the peer interactions identified as effective are those that support students' engagement in higher-order cognitive processes. Such interactions include:

Providing elaborated explanations; asking appropriate questions (question-asking during tutoring has also been found to facilitate learning when the questions require higher-order cognitive processes); providing sufficient time for the partner to think before being expected to respond; and using supportive communication skills such as listening attentively to a partner's response and giving feedback and encouragement (Adelgais et al., 1998).

The following rationales are identified by Falchikov (2001) for using same-level peer-tutoring techniques that require little preparation:

Cooperative note-taking pairs aim to help improve students' note-taking skills and to encourage students to engage with new material. Peer coaching seeks to strengthen collegial relationships, increase participant confidence, and foster individual growth. Peer monitoring aims to increase levels of performance in large classes and investigate the effects of a group contingency procedure on academic performance. Think-pair-share and Think-pair-square encourage students to discuss responses to questions and increase participation. Three-step interview aims to improve academic achievement, encourage modeling of effective practice, provide opportunities for cooperation, and improve social outcomes. Flashcard tutoring develops relatively low-level cognitive skills such as learning definitions, memorizing concepts, or vocabulary building (Falchikov, 2001, p. 15).

Falchikov (2001) also identifies rationales for cross-level peer-tutoring techniques within an institution. The Supplemental Instruction (SI) system, in which second-year students act as leaders to help first-year students in at-risk courses, aims to support students, reduce drop-out and failure, and encourage cooperative learning. Mentoring — the relationship between a less experienced person and a more experienced partner who guides and supports in a variety of contexts — seeks to provide guidance, advice, feedback, and support, improve overall academic performance, and encourage mentee personal growth. Proctoring or Keller's Personalized System of Instruction (PSI), in which an experienced undergraduate helps a beginner under the guidance of an academic, aims to help beginners achieve mastery and become part of the academic culture, and to help proctors develop leadership, team-building, and communication skills. Parrainage (students counseling students) aims to provide counseling support to new students, help them adapt to the new educational environment, and improve their practical problem-solving and study skills (Falchikov, 2001, p. 38).

Group peer-tutoring techniques identified by Falchikov (2001) include: Guided Reciprocal Peer Questioning (RPQ), which helps students comprehend and remember lecture content and facilitates encoding and retrieval of information; Structured Academic Controversy (SAC), which uses disagreement and conflict to aid learning and increase student motivation; The Jigsaw Classroom, which aids integration of children from different ethnic backgrounds, improves self-esteem and liking for school among minority-group children, and improves students' academic and social learning; the Syndicate method, which helps cope with differing needs and expectations of international students; and Team learning, which strengthens students' comfort level with and skills in teamwork (Falchikov, 2001, p. 61).

Traditional procedures used to improve the implementation of peer-tutoring reading program initiatives have been based on direct observation and feedback from consultants. Such consultants have improved program quality in a number of ways: recommendations are contextually specific and appropriate to the needs of the teacher, based on observational and quantitative data; consultants are capable of diagnosing and prescribing changes in the teacher's implementation using students' pre- and post-peer-tutoring progress information; and consultants can model and show the teacher how to use student progress data as a basis for making decisions about program changes and improvements (Arreaga-Mayer, Gavin, Greenwood et al., 2001). Based on such assessments, advice, and feedback, measurable improvements in program quality and student progress have frequently been achieved (Arreaga-Mayer et al., 2001).

Because resources are by definition scarce, many schools may not be able to afford such professional assistance, and obligatory consultation services may serve as a roadblock to gaining approval for such initiatives in the first place. Therefore, regardless of the specific type of peer-assisted reading program used, educators are advised to take advantage of insights from past initiatives:

Children need preparation and practice if group work is to pay dividends, so immediate success should not be expected — in the case of peer tutoring, tutors need some training. The structuring of work groups or tutoring pairs by ability and gender is not a crucial concern and may have only marginal impact. Successful learning outcomes are correlated with rich interaction; therefore, learners need to be encouraged to talk, deliver instructions, and exchange information and ideas. Interactions need to be managed to ensure that dialogue takes as many forms as possible: pooling of information, agreements, questioning, clarification and elaboration, and giving feedback — as well as conflict, argument, disagreement, challenging, and defending. A mix of these interaction types is probably desirable for maximum benefit, though conflict and argument should be focused on issues and problems rather than being personal and destructive. Finally, learners need time and encouragement to reflect on their interactive experience, as much of the cognitive work that stimulates learning comes after the interaction has concluded (Ehly & Topping, 1998).

A study by Wright (2003) concerning the affective influences on male student attitudes toward peer tutoring found that gender-specific sociolinguistic characteristics as well as cultural pressures impacted male students' willingness to seek academic assistance and may affect the quality of the peer relationship. In addition, as Dufrene and colleagues (2005) emphasize, implementation of peer-assisted learning initiatives is typically facilitated by naturally occurring routines and the consequences of participation, which may be sufficient for most students to maintain accurate implementation throughout the program.

The efficacy of any classroom innovation will clearly be mediated by the seriousness of students' learning problems; therefore, based on the increasing diversification of the student population within general education, it is important to examine effects separately by learner type (Keogh & Speece, 1996). Finally, and just as importantly, teachers themselves also need to become increasingly involved in designing and implementing peer-assisted methods: "Their enthusiasm for successfully deploying such techniques in the classroom crucially depends on our ability to produce and refine creative learning models" (Ehly & Topping, 1998, p. 41).

It is important that all teachers involved select academic measures to track students' reading levels both before the Tall Buddies initiative begins and during the tutoring program. In addition, careful planning, teacher modeling, and supervision are required to ensure that peer tutor and tutee relationships work (Heller, 1999). According to Heller, "Tutors should never feel that they are being denied the opportunity to read and write connected text at the expense of time spent schooling a peer in a particular area" (1999, p. 328). The evaluation of such programs requires feedback from participants and educators alike to ensure that tutors are receiving the support they require to succeed while simultaneously ensuring that tutees are receiving the quality of tutoring needed to improve their reading ability. In this regard, Heller advises that "peer-tutoring programs work best when the tutors are highly motivated to teach their friends (or younger students) and are rewarded intrinsically for their efforts. At the same time, the tutees should feel good about working with a peer" (1999, p. 328).

In the Fister et al. (2001) peer-tutoring initiative, general education teachers were assigned the following responsibilities for monitoring and oversight, with good results reported by the researchers: monitoring peer tutoring activities including students' performance of the tutoring roles; providing feedback to the teams about their performance; and providing assistance to teams when questions arose about assigned tasks.

Conclusion and Summary

Because good reading abilities contribute to gains in virtually all other academic pursuits, it just makes good sense to help students become as skilled at reading as possible during the elementary school years. The review of the relevant peer-reviewed and scholarly literature concerning peer-assisted learning alternatives showed that peer tutoring can be quite effective in improving children's reading ability as well as their academic performance in other areas, and that both the tutor and the tutee achieve academic and social gains as a result of the tutoring experience (Pearson, 2002). These mutual gains have been shown to take place in both reading and mathematics, and with children from different socioeconomic and racial groups; therefore, involving low-socioeconomic children in tutoring programs could increase the value they place on reading and other academic skills.

The research also showed that careful evaluation of learner needs is required when designing peer tutoring programs. In order to be successful, tutoring programs should use tutors who are several years older than tutees, tutors should be trained and supervised closely, and the intervention should be implemented for a relatively sustained period of time (Pearson, 2002). The teacher feedback gathered from the Tall Buddies initiative confirmed these findings: students demonstrated improved reading skills, increased self-esteem, stronger school connectedness, and reduced behavioral problems across both the tutor and tutee populations.

On a final note, the research showed that peer-assisted learning techniques are particularly worthwhile for heterogeneous groups of students because they encourage mutual respect and learning among students of various backgrounds and academic abilities (Keogh & Speece, 1996). The Tall Buddies program stands as a practical and replicable model for schools seeking a low-cost, high-impact intervention that benefits the whole school community.