Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Making Retention Count: The Power of Becoming a Peer Tutor

by Leigh Mesler - 2009

Background/Context: A review of the literature demonstrates that grade retention often fails to improve the academic and socioemotional outcomes of retained students. Although little empirical work on peer tutoring has focused specifically on retained students, the literature suggests that those students who act as peer tutors often experience improved school performance and self-concepts.

Purpose of Study: This work developed out of a concern that elementary school students being held back to repeat a grade, or retained, were not benefiting academically from nonpromotion. The purpose of this action research study was to identify and implement an intervention that would improve the academic and socioemotional outcomes of a twice-retained third-grade student.

Setting: This study took place in a New York City public elementary school.

Intervention: The intervention involved implementing a 12-week peer tutoring program in which a retained third-grade student tutored a struggling classmate in mathematics.

Research Design: This is an action research study in which the author conducted research and implemented an intervention in her own classroom.

Results: After serving as a peer tutor, this student experienced increased math achievement, an improved self-concept, and better classroom behavior. These results suggest that having struggling students serve as peer tutors may be effective in improving both their academic achievement and socioemotional outcomes.

“Good morning, Ms. Mesler,” Christopher grumbled as he entered my classroom on the first day of school. It was clear that he was not looking forward to spending a second year in third grade. Nonetheless, it was territory with which he was all too familiar, because he had also spent 2 years in second grade. Christopher was a student whose reputation preceded him. His former teachers described him as a lazy student with a bad attitude who struggled to keep up in class. Beginning on that first day of school, I wondered if Christopher’s deflated attitude had contributed to his spending 4 years in two grades, or vice-versa.

During my 3 short years in the classroom, I asked many questions about how students benefited from being “left back” to repeat a grade, or retained. Based on my own observations, most teachers at my school did not go out of their way to ensure a student’s success during the repeated year. Like many of my colleagues, I, too, was guilty of such an offense. More often than not, retention seemed like a way of punishing students for not working hard enough the year before. Once their 1-year sentence was served, they were moved ahead to the next grade regardless of whether they had met grade-level standards. Few of them made significant academic improvements, and all the retained students with whom I was familiar continued to struggle in subsequent years. I often wondered how retention affected students’ academic and socioemotional outcomes. Even though I had not yet witnessed many benefits to retaining students, I was determined to find something that would work for Christopher and make his second year in third grade marked by progress and change.

This article outlines my experiences as a teacher working with Christopher in a New York City public school. It explores one intersection of policy and practice, as I sought to understand the socioemotional and academic outcomes of a student affected by New York City’s grade retention policy. In addition, it explores how the self-image and academic performance of this retained student were affected when he acted as a peer tutor while repeating third grade.

I begin with descriptions of the school in which I worked and the student who was the focus of my case study. These descriptions are followed by the rationale for pursuing this work and a review of the literature on grade retention and peer tutoring—the two lines of education research directly relevant to my action research. Next, I present the research design and the data collected during this study. Finally, I discuss the results, which support the notion that becoming a peer tutor may be a useful intervention for struggling students, and I conclude with recommendations for policy and practice.


It is important to learn about the environment in which my research took place—to see our “school portrait,” in a sense—and to understand Christopher’s previous school experiences in order to appreciate the specific context in which this case study took place. I taught third grade at a small public elementary school in New York City, which had approximately 400 students in Grades pre-K to 5. Eighty-five percent of the student population was Hispanic, and 90% of the students were eligible for free lunch.

During the 2005–2006 school year, 4 of the 22 students in my third-grade class had repeated a grade. Christopher was the one student who had been left back more than once. He was repeating third grade because he received 1s, the lowest possible grade, on the citywide English Language Arts and mathematics tests the previous year. Scoring a 1 on these tests indicates that a student has serious academic deficiencies and is unable to demonstrate an understanding of grade-level content (“Assessment and Accountability,” n.d.). Christopher was also retained in second grade, so he would spend a total of 4 years in only two grades. When a child failed a grade at my school, he or she did not automatically receive academic intervention services. Even those students designated to receive additional services were rarely seen by their intervention specialists because those staff members were repeatedly asked to tend to the administrative tasks of the school or cover classes when self-contained classroom teachers were absent. Further, there was an unspoken policy that a student would not be retained more than once in the same grade, so neither teacher nor student was compelled to ensure the student’s academic success during the repeated year.

When I decided to study retention, I chose to focus on Christopher because he had failed multiple times and had a history of both academic and behavioral problems in school. I had heard other teachers openly discuss his personal shortcomings with colleagues when he was within earshot, in addition to reporting that he was a lazy troublemaker who was incapable of completing his assignments. However, while teaching Christopher in an after-school program the previous year, I learned that he would apply himself with some encouragement and that he had many redeeming qualities, including his sense of humor and willingness to help others. I also had an existing relationship with Christopher’s family because I taught his brother—now also in third grade—in first grade 2 years earlier. Although they could be tough on him at him at times, I knew that Christopher had two supportive parents who would likely be willing to accommodate the extra time I wanted to spend working with him.

I thought it was critically important that Christopher’s second year in third grade be markedly different from his first. I started the year by letting him know that I had requested that he be in my class. I also went out of my way to create a positive relationship with him from the beginning of the year. This was something I tried to do with all my students, but especially those whose reputations preceded them because they had problems with other teachers. The previous year, it had been difficult to see Christopher in my after-school class struggling with an incredibly poor self-image that was constantly reinforced by negative experiences in school. I let Christopher know that I was confident he would be successful this year and that I recognized that he had the potential to be a good student. I tried to start the year off on the right foot with Christopher, but I knew that I needed to find out more about retention to get a better understanding of the challenge that he and I faced.


Before finding out how to best help Christopher become a successful third grader, I sought a better understanding of grade retention. Grade retention and related topics, including social promotion and high-stakes testing, have long been discussed in the world of education. In recent years, social promotion has been frequently discussed in the New York City media. In January 2004, New York City mayor Michael Bloomberg announced that the practice of social promotion would no longer exist for the city’s third graders: “This year,” he proclaimed, “for third graders, we're putting an end to the discredited practice of social promotion. We're not just saying it this time. This time, we're going to do it” (Herszenhorn, 2004a). Considering the popularity of Bloomberg’s policy and my own reservations about the impact of grade retention on student outcomes, I looked to the retention literature to find out what researchers had discovered.

Of the recent studies conducted, most showed that retention had little or no long-lasting effects on student academic achievement. In 19 different studies conducted during the 1990s, comparisons of academic achievement in math and reading between retained and matched comparison students found negative effects of grade retention across all areas of achievement (Jimerson, 2001a, 2001b). Another recent study on the Chicago Public Schools’ program that ended social promotion tracked the long-term effects that retention had on third- and sixth-grade students (Roderick & Nagaoka, 2005). The authors found that retaining third graders had no short-term effect on their academic achievement and that retained sixth-grade students performed almost 25% worse on standardized reading tests than low-achieving sixth graders who were promoted. Further, retained students faced significantly higher rates of special education placement than similar students who were promoted. In addition to having negative impacts on student academic performance, research suggests being retained in any grade makes students significantly more likely to drop out of school (Allensworth, 2004; Jimerson, 2001a; Jimerson, Anderson, & Whipple, 2002; Roderick, 1995) and results in lower rates of school attendance (Holmes & Matthews, 1984; Jimerson, 2001b).

Although most studies of grade retention suggest that it is associated with negative school outcomes, other studies have indicated that retained students may experience some benefits from being retained. One study showed that retained students made larger relative gains than students who could have been retained but received exemptions because they were English language learners or were promoted based on portfolio assessment (Greene & Winters, 2006). Although the research highlighting the positive effects of retention is heavily outweighed by studies reporting negative effects, a number of the studies that found retention to be effective suggested that remedial academic interventions or bolstered literacy support were responsible for retained students’ success (Alexander, Entwisle, & Dauber, 1994; Greene & Winters). This finding made me optimistic that pinpointing an appropriate intervention for Christopher might substantially improve his chances of benefiting from retention.

Despite this kernel of optimism, the predominantly negative relationship between retention and academic outcomes, coupled with research on the socioemotional effects of retention, made me feel an even greater pressure to make Christopher’s second year in third grade a positive one. Most studies addressing the effects of retention on nonacademic outcomes have indicated that retention has negative effects on students’ socioemotional outcomes. Retained students were found to have poorer self-concepts, poorer social and emotional adjustment, and more behavior problems in school than comparable students who were not retained (Holmes & Matthews, 1984; Jimerson, 2001a, 2001b). Research also found that students have a deep fear of being retained: A 1980 study found that being held back was the third most prevalent worry of children, behind only the death of a parent and becoming blind. When this study was replicated in 2001, sixth-grade students cited repeating a grade as the single most stressful life event, above both losing a parent and going blind (Anderson, Jimerson, & Whipple, 2005). These findings strongly support the notion that being retained may be damaging to the socioemotional well-being of young students.

As mentioned, one of the first things I noticed about Christopher before he ever entered my third-grade classroom was his low self-esteem. According to Maslow’s (1970) hierarchy of needs, if children cannot satisfy one of their needs, they will never reach their full potential. These needs—physical, social, and emotional—include the need for self-esteem. A child who does not realize self-esteem will not be successful in school. The need for approval and self-esteem involves both valuing oneself and being valued by others as a part of a community. Christopher did not appear to have this need satisfied. I spoke with many colleagues and a number of professors at graduate schools of education about possible interventions that might help raise a student’s self-esteem. When one veteran teacher suggested having him help another student in the class, I turned to research on peer tutoring to determine if it might be a worthwhile intervention for Christopher.


Research on peer tutoring has documented many positive effects for both tutors and tutees participating in peer tutoring. Given that I hoped such participation would help Christopher, I focused on understanding the benefits to the tutor as discussed in the literature. Bernard (1990) found that peer tutoring can be instrumental in improving the academic performance of students who act as tutors. Further, in meta-analyses conducted by Cohen, Kulik, and Kulik (1982) and Robinson, Schofield, and Steers-Wentzell (2005), the authors reported that the vast majority of studies (nearly 90%) found that students experienced academic gains after acting as peer tutors.

Research has also suggested that student tutors experience improved attitudinal and socioemotional outcomes. Hedin (1987) found that for the tutor, “the experience of being needed, valued, and respected by another person produced a new view of the self as a human being” (p. 43). Other researchers also reported that students acting as tutors had better self-concepts, exhibited improved classroom behavior, and had more positive attitudes toward specific subject matter and toward school in general during and after their service as peer tutors (Bernard, 1990; Cohen et al., 1982; Nazzal, 2002; Robinson et al., 2005). Further, work on student outcomes has found that at-risk, low-income, minority, and early elementary students may experience the greatest benefits from becoming peer tutors (Ginsburg-Block, Rohrbeck, & Fantuzzo, 2006; Hedin; Mills, Dunham, & Alpert, 1988;). The research on peer tutoring made me optimistic about implementing a program involving Christopher as a tutor because he was greatly in need of an intervention that would boost his academic performance, attitude, and self-concept. The sections that follow discuss the research design used in this case study, the data collected, and conclusions drawn from this work.


Armed with a deeper understanding of the effects of retention on students and the optimism that involving Christopher in peer tutoring might help him be more successful in school, I thought carefully about how to design an intervention that would be manageable for both Christopher and me. I considered involving Christopher in peer tutoring in the late fall, a time when the January New York State English Language Arts (ELA) test was rapidly approaching and when preparing for it took precedence over many other subjects and activities at my school. The peer tutoring literature suggested that it was not necessary to involve Christopher in a yearlong program; Robinson et al. (2005) and Rohrbeck, Ginsburg-Block, Fantuzzo, and Miller (2003) reported that the duration of a peer tutoring program was not significantly related to the outcomes for either the tutor or the tutee. Further, Cohen et al. (1982) found that longer peer tutoring programs, ranging from 19 to 36 weeks in duration, actually had smaller effects than programs lasting from 0 to 18 weeks, so I was confident that waiting until after the ELA test would not jeopardize the potential benefits of becoming a peer tutor for Christopher.

Based on conversations with his previous teachers and my own observations from the beginning of the year, I learned that Christopher was strongest in his basic math skills—addition, subtraction, and one-step word problems. In addition, the literature suggested that peer tutoring in math generally conferred more benefits to both the tutor and the tutee than peer tutoring in reading (Cohen et al. 1982; Robinson et al., 2005), so I concluded that it would be best for Christopher to tutor another student in math. In mid-December, I approached Christopher about becoming a peer tutor, explaining that I noticed his improvements in math and thought he might be able to help some of his classmates. I told him that I also noticed how patient he was with other students and how good he was at explaining math problems to Jose, his in-class neighbor. I asked if he would be willing to plan with me before school once a week and then tutor Jose in math in the morning before class two to three times per week. Christopher was excited about the opportunity and accepted my offer to take on a new role.

I chose Jose as the tutee because he was behind Christopher academically and struggled in some of the areas in which Christopher had recently improved. I knew that Christopher would be able to understand having difficulty with the types of problems with which Jose needed help. Vygotsky (1978) wrote about the “zone of proximal development,” which I took into consideration in pairing Christopher with Jose and also in deciding what material Christopher would focus on with Jose. Vygotsky defined the zone of proximal development (ZPD) as “the distance between the actual development level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance, or in collaboration with more capable peers” (p.86). In other words, the ZPD is knowledge that a learner has the ability to learn, but only with the help of a more advanced learner. Keeping this in mind, I aimed to have Christopher work on concepts that he had mastered and that were just above Jose’s current performance level.

To prepare for his tutoring sessions with Jose, Christopher agreed to meet with me for 45 minutes on Tuesday mornings before school. Research suggests that training can positively influence tutor behaviors (Fuchs, Fuchs, Bentz, Phillips, & Hamlett, 1994; Reckrut, 1994), so I explained to Christopher that he would have to plan in advance in order to best teach Jose and that he would need to help Jose feel good about his accomplishments. After Christopher and I decided what he and Jose would focus on during a particular week, Christopher would take the problems or worksheets home to do on his own. He and I then went over the problems together in our morning sessions so that he was well prepared each week before he began working with Jose.

After our first Tuesday morning meeting, Christopher began working with Jose. Once they arrived and unpacked, Christopher and Jose went to a table in the back of the classroom to work while the rest of the class did morning exercises. Christopher tutored Jose two or three times per week, depending on how our schedule was changed, for 12 weeks. This began in early January, after the students took the New York State ELA, and continued until the beginning of April.


My initial focus was on student retention and then broadened to include peer tutoring as an intervention aimed at improving Christopher’s low self-esteem and academic difficulties. I gathered information from my entire class and from Christopher using a number of different tools to understand students’ feelings about retention. To assess the impact of peer tutoring on Christopher’s academic and socioemotional outcomes, I also collected data through interviews, questionnaires, observation during planning sessions, and math tests.

Questionnaires. I began by using questionnaires on retention to collect data on all my students’ thoughts on the subject. The whole class completed a questionnaire to measure the students’ feelings about retention (Appendix A). Retention research indicated that students’ retention fears are high (Anderson et al., 2005) and that being retained is very damaging to a student’s self-esteem (Holmes & Matthews, 1984; Jimerson, 2001a, 2001b; Jimerson et al., 2002; Roderick, 1995). It was important for me to determine whether my students’ responses were in line with this research, because if they did, it seemed more plausible that Christopher’s low self-esteem was at least partly related to being retained multiple times.

My initial data collection also included a questionnaire filled out by Christopher that was aimed at gathering his thoughts about his strengths and weaknesses in school. He filled out this questionnaire in November, before he was approached about becoming a peer tutor, and again after completing his 12 weeks as a peer tutor in April (Appendix B). Both times that Christopher completed this questionnaire, the rest of the class was working on filling out another form so that he did not feel singled out.

Interviews. After getting permission from Christopher’s parents to interview him, I met with him to discuss his school experiences in November. I felt that he would be honest with me during the interviews because he had openly shared his feelings about, and frustrations with, school in the past. I asked him about his previous experiences with teachers and other students, as well as how he thought he performed in class. Although I would not be able to draw a causal relationship, it seemed plausible that many of Christopher’s feelings about school were impacted by retention because he was left back multiple times. I also conducted an interview after Christopher finished peer tutoring to evaluate whether he had experienced any clear changes in his attitudes toward school or in his view of himself as a student after spending time as a tutor (Appendix C).

School performance. I collected data from 10 different math tests administered from October to April. Six of these tests were end-of-unit tests from Everyday Math, the school’s math curriculum, and four of the tests were practice tests for the New York State math test. Five of the tests were given before Christopher began acting as a tutor and five were given after he started tutoring Jose in January. Although the unit tests covered different types of material, I calculated the class average for each test and compared Christopher’s test score with the class average to measure any unique changes he experienced during the school year. Math tests were the primary quantifiable measure of Christopher’s academic performance, but I also remained cognizant of the possibility that Christopher might show changes in other subjects, in his attitude, and in his behavior. I kept records of Christopher’s reading level, which was assessed primarily through running records, along with anecdotal records based on observations of his attitude and behavior in school.

Notes from planning sessions with Christopher. Christopher met with me every Tuesday morning from the 2nd week of January until the 1st week in April. After each of our 12 meetings, I wrote notes about our meeting. I included the amount of time for which we met, Christopher’s comments about how his work with Jose was progressing, and notes about who directed the planning for Christopher’s peer tutoring sessions with Jose.


This section reports the data collected using the tools outlined previously. The data suggest that many factors contributed to Christopher’s academic and socioemotional improvement but that each was highly related to the peer tutoring intervention (the data are interpreted and discussed in the subsequent section).


After reviewing the literature, I found that fear of being held back was extremely high among elementary school students (Anderson et al., 2005), but I was not sure whether it was a major concern for my students because grade retention was perhaps more common among my students than among those students included in previous studies. Although a relatively high percentage of them had been retained, many of my students had done very well in school from prekindergarten on and repeatedly received praise from me and from their other teachers. However, nearly half of my students reported that they worried every day about being held back. Another 18% reported that they worried about being held back either a few times a week or once or twice a month (Figure 1). When asked how students might feel if they were told they would be retained, students gave the following types of responses: “They will feel mad or sad because they thought they did so, so, so hard and now their friends will leave them”; “They would be mad, so so mad”; and “They would feel bad. They would say ‘No! No! No! I tried my best.’”

Figure 1. Students’ retention fears

click to enlarge

I also used questionnaires to explore Christopher’s ideas about himself as a student. On a questionnaire he filled out in October, Christopher reported that his favorite subject was gym because he was “good at it and because the gym teacher asks him to show things to the class.” He reported that his least favorite subject was writing because he wasn’t good at it, and he also listed reading, math, and social studies as subjects he was not good at in school. In an interview with me, he confirmed these feelings and added that he knew he was not good at those subjects because he failed third grade, and his previous third-grade teacher told him he wasn’t good at them.

In April, after he finished tutoring, Christopher filled out the same questionnaire that asked about his strengths and weaknesses. He reported that he was strongest in gym, reading, and computers and that he was not good at music, math, or science. In fact, he said that his least favorite subject was math and explained that it was because he thought that he was not good at division. In a follow-up interview, I asked Christopher to name the subject in which he had improved the most this year, and he said it was math. When I asked why he wrote that math was his least favorite subject, he explained that he knew he got better at addition, subtraction, and multiplication, but he expressed some frustration with not being able to master division—the topic we were working on at the time. Although I often told my students that they would get to practice division frequently in fourth grade, Christopher did not realize was that division is not an area that must be mastered in third grade, according to New York State math standards.

In the same interview, I asked Christopher about how he felt while he was tutoring Jose. He told me that he felt “really good because [he] was helping him get smarter.” He added that I should let other kids tutor because helping people “is an important job and more students can get smarter.” He also explained that he wished he could have tutored Jose for longer because he really wanted him to pass the grade, and he was worried that he hadn’t done enough for Jose.


Five math tests were given from October through December, before Christopher started peer tutoring. Of those tests, two were New York State (NYS) practice tests and three were Everyday Math end-of-unit tests included in the school’s math curriculum. Christopher’s average on those five tests was 41%, 22 points below the class average of 63%. Five additional math tests were given from January through April, during Christopher’s time as a peer tutor. Again, two of these five tests were NYS practice tests and three were end-of-unit tests. Christopher averaged 82% on the tests that he took after becoming a peer tutor, which was 13% points above the class average (Figure 2).

Figure 2. Test scores

click to enlarge

The last test that Christopher took before becoming a peer tutor was on December 2, 2005. The test was a multiple-choice NYS practice test. Christopher scored 44% on that test, 20 points below the class average. The first test that Christopher took after he began tutoring in early January was on January 23, 2006; it was another NYS practice test that followed the same format and covered material similar to the December 2 test. Christopher scored 84% on that test. This was the first test on which Christopher scored above the class average; he scored 6 points above the class average of 78%. On every test that followed—Everyday Math unit tests and the NYS practice test—Christopher scored above the class average. The New York State math test was given in March 2006. At the beginning of the 2006–2007 school year, I learned that Christopher scored a 3 on the test, which meant that he was performing at grade-level standards. This score was a significant improvement from the 1 he had earned during his first year in third grade.

Additional improvements. Researchers have found that students who acted as math tutors also experienced benefits in other school subjects (Greenwood & Terry, 1993; Nazzal, 2002). Christopher’s reading level did not improve more dramatically than other students with similar abilities, but he did pass the state test, scoring a 2, which indicated that he was approaching grade-level standards in English Language Arts. I also noted improvements in his classroom behavior. After he started tutoring Jose, Christopher was more focused during reading time and during independent and small-group math activities. In addition, his science, gym, and music teachers, who had often complained that he had problems with other students while in their classrooms, reported that Christopher’s behavior improved during the second half of the year. During the beginning of the year, I received at least one report per week about Christopher’s disruptive behavior from his science, gym, or music teacher. Teachers had often complained that he talked too much in class, spoke to them in a disrespectful manner, failed to follow directions, or threw materials at other students. Between January and the end of the year, I received only three reports of misbehavior that involved Christopher.


My Tuesday morning planning sessions with Christopher were scheduled from 7:40 to 8:25. After speaking with Christopher’s mother, she agreed to let him come to school early to plan his tutoring sessions with Jose. After every meeting, I sat down for a few minutes to write notes about our interactions. I included information about the amount of time for which we met, Christopher’s comments about how his work with Jose was progressing, and who directed the planning for Christopher’s peer tutoring sessions with Jose.

Christopher arrived late to our first five meetings, so each session lasted for no more than 40 minutes. After 6 weeks, Christopher began showing up on time and even a few minutes early, so we had more than 40 minutes to plan together during the latter seven meetings (Figure 3).

Figure 3. Planning sessions

click to enlarge

During our early meetings, I directed much of the planning. I explained what areas I noticed Jose needed the most help with and talked about which problems or worksheets Christopher might want to use with Jose to target those areas. I also talked with Christopher about how to help Jose with the work we chose for him without doing it for him. As time went on, Christopher began to take the reins during our meetings. One morning, he sat down and immediately said, “Ms. Mesler, I think he gets how to do two-digit addition. He gets everything right on his own. He doesn’t need my help with it anymore.” I began to spend less time suggesting activities for Christopher to do with Jose as Christopher made more suggestions about what he thought he should do with Jose. By the time our planning meetings ended, Christopher still relied on me for approval but seemed to have gained more confidence in his abilities and required less support in deciding what material to use while he helped Jose.


This section interprets the data described previously from the perspective of determining whether becoming a peer tutor was an effective intervention for Christopher. I believe that a confluence of factors contributed to Christopher’s academic and socioemotional improvement, but I also believe that each factor was highly related to the peer tutoring intervention. Throughout this section, I address some alternative hypotheses that I considered as I interpreted the data collected during this action research study.

When I analyzed the test score data, it became evident that Christopher’s math test scores improved more than any other student in the class. He had never before scored above his classmates consistently in any subject, in any grade. From the very first test he took after he began peer tutoring, Christopher scored above the class average, and he never scored below the class average again.

Although it is likely that practicing his basic math skills to prepare for his work with Jose helped Christopher perform better on math tests, I do not believe that a hypothesis that Christopher’s improvement was due to practicing more math problems can fully explain his progress. It seems logical that working with me in planning sessions sharpened Christopher’s math skills, but the celerity of his turnaround after tutoring began suggests that something else was also involved. Christopher had attended two planning sessions before taking his first math test as a peer tutor, which meant that he completed only two extra assignments to prepare for subsequent work with Jose. Further, the extra assignments focused primarily on basic addition and subtraction, which were much less difficult than the problems on the NYS practice test. Christopher made fewer calculation errors on simple addition and subtraction, and he also scored better on multiplication and word problems—two types of problems on which he had not spent any extra time. This further supports the notion that extra math practice does not fully explain Christopher’s progress.

Other alternative hypotheses that I considered as possible explanations for Christopher’s academic improvements were maturation and regression to the mean. By comparing and contrasting Christopher’s progress with the academic performance of the whole class, I determined that Christopher’s increase in math performance was not likely due to maturation, because his improvement far exceeded that of any of his classmates. Additionally, although regression to the mean, which may operate when extremely low- or high-performing students are selected for study, might explain some of Christopher’s progress, it cannot account for Christopher’s advancement well beyond the class mean.

Data from observations and anecdotal records also indicated that Christopher’s self-esteem improved tremendously after becoming a peer tutor. In an interview early in the year, he told me that he knew he was not smart for a variety of reasons. In February, however, it became clear to me that Christopher’s self-concept had changed dramatically when another teacher came to pick up some students to work on problems for the upcoming state math test. I heard him say to another student, “I don’t need extra help in math. I’m smart in math. That’s why I’m tutoring Jose.” In addition, Christopher demonstrated that his confidence was increasing while he was tutoring; he began to take the lead during our planning meetings and did not need as much direction from me.

Christopher also began to hold himself to a higher standard. Although I was initially alarmed by the questionnaire on which he chose math as his worst subject, I later realized that this indicated he was expecting more of himself. Christopher explained that math was his least favorite subject because he was not good at division, which the class was working on at the time. He stated that he knew he was good at addition, subtraction, and multiplication, but that he didn’t like division. I was struck by his response for two reasons. First, I knew that in November, Christopher never would have asserted that he was good at addition, subtraction, and multiplication. Second, division was not a skill my students were expected to master by the end of third grade, and the fact that Christopher pushed himself to learn division led me to believe that he had more confidence in his own abilities and had begun to hold himself to a higher standard than he had previously.

My data suggest that for Christopher, the experience of peer tutoring was a successful academic and socioemotional intervention. The experience of peer tutoring seemed to help fulfill his need for self-esteem, which likely allowed him to perform better in school. His attitude toward school and about himself began to change as he took on his new responsibilities tutoring Jose. In addition to experiencing improved test scores when he started peer tutoring, Christopher became a better student in other areas as well. He showed increased responsibility by coming to our meetings on time or early. His mother explained that by the end of his tutoring, he would push her to leave the house in the morning to make it to our meetings on time, whereas the opposite had been true in January, when we first began meeting. As noted, Christopher also became more focused in class when we had reading and writing workshops, and his behavior improved in other classes as well. I believe that becoming a peer tutor helped improve Christopher’s academic and socioemotional outcomes.

Although it is difficult to make a strong causal argument based on the results from this action research study, I believe that this intervention was the primary reason for Christopher’s improvement. He knew that I had enough confidence in his abilities to ask him to be a peer tutor, and he began to believe more in himself. I believe that these factors, combined with extra math practice and the positive feelings that came from helping another student, translated into better outcomes for Christopher. Overall, I concluded that peer tutoring was a very successful intervention for Christopher, and I hope that he continues to succeed in subsequent grades. Although he did not continue his role as a peer tutor the following year, his fourth-grade teacher reported that he had relatively few behavioral problems in her class. She also reported that he was performing near the class average on academic tasks and passed the NYS standardized tests in reading and math at the end of fourth grade. That was the first time that Christopher passed a grade on his first attempt in 5 years.

Retention research suggests that Christopher’s low self-esteem was typical for retained students (Holmes & Matthews, 1984; Jimerson, 2001a, 2001b; Jimerson et al., 2002; Roderick, 1995). By providing students with the opportunity to feel successful and help others, peer tutoring could potentially be a successful intervention with other students who have been held back and/or who are struggling academically. Christopher began his retained year as a student with poor grades and a negative attitude toward school. I had seen many such students in my school and I believe that students very much like Christopher exist in elementary schools across the country. Peer tutoring may be a useful tool to help other struggling students find success because it offers students the opportunity to both assist others and strengthen their own academic skills. In addition to supporting the student-level effects of retention discussed in the literature, the data from this case study are consistent with research findings that students, especially at-risk minority students, benefit greatly from serving as peer tutors. However, because some work has found that even when students participate in successful interventions during the retained year, the positive effects may fade over time, it will be important to follow Christopher’s progress in fifth grade and beyond.


The extant literature on grade retention and my work with Christopher suggest that in its current form, grade retention largely fails to improve student outcomes. However, my action research suggests that when students are retained, planned interventions can help them succeed. This work highlights the importance of developing student-specific interventions for retained students. If students must be retained, the repeated year should be a time when students receive the targeted assistance they need, as opposed to a mandatory “sentence served” for failing to meet grade-level standards the year before. Although there have been promises from the school chancellor that New York City’s retained third graders will be provided extra tutoring after school, on weekends, and on holidays (Herszenhorn, 2004b)—a proposal that acknowledges the importance of ensuring that students get additional help during the repeated year—no such offer was made to the students at my school. I believe that one of the most pertinent policy recommendations stemming from this work is that it is necessary for school districts practicing grade retention to offer support to students and teachers to better ensure that repeating a grade leads to improved student outcomes rather than simply “more of the same.”

Although there may be various forms of district or school support that would help retained students, this case study suggests that peer tutoring should be considered a feasible classroom- or school-level option when developing individualized programs aimed at raising the academic achievement of retained students. Peer tutoring provided opportunities for Christopher to receive individual positive attention, practice basic math skills, and become involved in an activity that helped build his self-esteem. Although it is difficult to know the relative influence of each of those components on Christopher’s progress, this combination of factors as part of the peer tutoring intervention appears to have been beneficial.

Finally, I recommend that schools develop early intervention plans that focus on both the academic and socioemotional needs of young students. Christopher is just one example of a student whose poor school performance and low self-esteem likely worked in concert to result in his being retained twice in 4 years. Retention researchers have often suggested that early intervention programs may be greatly beneficial to students well before they reach grades that involve high-stakes tests (Jimerson, 2001b; Roderick & Nagaoka, 2005). Although early intervention programs often involve students working with specialists to improve their language and math skills, programs like peer tutoring, which are inexpensive to implement, may also be instrumental in raising and sustaining students’ self-concepts.

There are also recommendations for classroom practice that emerged from my work with Christopher. I must note that the intervention I implemented required significant amounts of time, energy, and monitoring and that this example only highlights the experience of one student, in one class, in one school. Although it would be nearly impossible for one teacher to devote the same amount of time and attention to every struggling student, devoting some part of regular class time to classwide peer tutoring, which research has also shown to be beneficial to students (Greenwood, Delquadri, & Hall, 1989; Greenwood & Terry, 1993), might be less time consuming and allow more students to participate and benefit. In addition, there are some general practices of which teachers should remain cognizant in their interactions with students. First, it is important to be conscious of the significant impact that a student’s self-esteem has on school performance, both academic and behavioral. It can be challenging to highlight the positive qualities of the most difficult students, but having the ability and patience to do so as a teacher can make an important difference in students’ experiences in school. In addition, by taking the time to get to know each student individually and challenging the reputation that sometimes follows struggling students, teachers may impact students’ academic careers well beyond the time they spend in one classroom.

Although the extant literature on grade retention paints a rather grim picture of its ability to improve student outcomes, it is a policy that will most likely endure given the escalated calls for greater accountability and increased standards. It is important that teachers and school leaders take responsibility for ensuring that a students’ retained year is marked by growth and progress. Although this article outlines a single case study, it suggests that there are feasible student-specific interventions that can support struggling students. Teachers should be encouraged to find effective methods to help ensure the success of retained and other academically struggling students.


Alexander, K., Entwisle, D., & Dauber, S. (1994). On the success of failure: A reassessment of the effects of retention in the primary grades. Port Chester, NY: Cambridge University Press.

Allensworth, E. (2004). Ending social promotion in Chicago: Drop out rates in Chicago after implementation of the eighth-grade gate. Chicago: Consortium on Chicago School Research.

Anderson, G. E., Jimerson, S. R., & Whipple, A. D. (2005). Students’ ratings of stressful experiences at home and school: Loss of a parent and grade retention as superlative stressors. Journal of Applied School Psychology, 21, 1–20.

Bernard, B. (1990). The case for peers. Portland, OR: Northwest Regional Educational Laboratory.

Cohen, P. A., Kulik, J. A., & Kulik, C. L. (1982). Educational outcomes of tutoring: A meta-analysis of findings. American Educational Research Journal, 19, 237–248.

Fuchs, L. S., Fuchs, D., Bentz, J., Phillips, N. B., & Hamlett, C. L. (1994). The nature of student interactions during peer tutoring with and without prior training and experience. American Educational Research Journal, 31, 75–103.

Ginsburg-Block, M. D., Rohrbeck, C. A, & Fantuzzo, J. W. (2006). A meta-analytic review of social, self-concept, and behavioral outcomes of peer-assisted learning. Journal of Educational Psychology, 98, 732–749.

Greene, J., & Winters, M. (2006, Spring). Getting ahead by staying behind. Education Next, 65–69.

Greenwood, C. R., Delquadri, J. C., & Hall, R. V. (1989). Longitudinal effects of classwide peer tutoring. Journal of Educational Psychology, 81, 371–383.

Greenwood, C. R., & Terry, B. (1993). Achievement, placement, and services: Middle school benefits of classwide peer tutoring used at the elementary school. School Psychology Review, 22, 497–517.

Hedin, D. (1987). Students as teachers: A tool for improving school climate and productivity. Social Policy, 17(3), 42–47.

Herszenhorn, D. (2004a, January 9). Stricter standards in New York may hold 15,000 in 3rd grade. New York Times. Retrieved April 8, 2006, from http://www.nytimes.com

Herszenhorn, D. (2004b, January 14). Special help for 3rd graders is promised. New York Times. Retrieved April 8, 2006, from http://www.nytimes.com

Holmes, C. T., & Matthews, K. M. (1984). The effects of non-promotion on elementary and junior high pupils: A meta-analysis. Review of Educational Research, 54, 225–236.

Jimerson, S. R. (2001a). A synthesis of grade retention research: Looking backward and moving forward. The California School Psychologist, 6, 47–59.

Jimerson, S. R. (2001b). Meta-analysis of grade retention research: Implications for practice in the 21st century. School Psychology Review, 30, 420–347.

Jimerson, S., Anderson, G. E., & Whipple, A. D. (2002). Winning the battle and losing the war: Examining the relation between grade retention and dropping out of high school. Psychology in the Schools, 39, 441–457.

Maslow, A. H. (1970). Motivation and personality (2nd ed.). New York: Harper & Row.

Mills, R., Dunham, R. G., & Alpert, G. P. (1988). Working with high-risk youth in prevention and early intervention programs: Toward a comprehensive wellness model. Adolescence, 23(91), 643–660.

New York City Department of Education. (n.d.). Assessment and accountability. Retrieved July 26, 2007, from http://schools.nyc.gov/daa/test_info/default.asp#NYSMATH

Roderick, M., & Nagaoka, J. (2005). Retention under Chicago’s high-stakes testing program: Helpful, harmful, or harmless? Educational Evaluation and Policy Analysis. 27, 309–340.

Nazzal, A. (2002). Peer tutoring and at-risk students: An exploratory study. Action in Teacher Education, 24(1), 68–80.

Reckrut, M. D. (1994). Peer and cross-age tutoring: The lessons of research. Journal of Reading, 37, 356–362.

Robinson, D. R., Schofield, J. W., & Steers-Wentzell, K. L (2005). Peer and cross-age tutoring in math: Outcomes and their design implications. Educational Psychology Review 17, 327–362.

Roderick, M. (1995). Grade retention and school dropout: Policy debate and research questions. Phi Delta Kappan Research Bulletin, 15, 1–6.

Roderick, M., & Nagaoka, J. (2005). Retention under Chicago’s high-stakes testing program: Helpful, harmful, or harmless? Educational Evaluation and Policy Analysis, 27, 309–340.

Rohrbeck, C. A., Ginsburg-Block, M. D., Fantuzzo, J. W., & Miller, T. R. (2003). Peer-assisted learning interventions with elementary school students: A meta-analytic review. Journal of Educational Psychology, 95, 240–257.

Vygotsky, L. S. (1978). Mind and society: The development of higher mental processes. Cambridge, MA: Harvard University Press.


 Have you ever repeated a grade?

Yes, grade ______



Why do some kids have to repeat a grade?


What is the biggest reason why a student might be held back?


How do kids feel when they find out they are going to be held back?


How often do you worry about being left back? (circle one)

      every day

  a few times per week


   once or twice a month

      a few times per year


What can kids do so that they won’t need to repeat a grade?


Should kids be held back? Why or why not?


Name ______________________________

Date ___________________

1. Circle the areas that you think you are good at in school.




social studies






2. Circle the areas that you think you are not good at in school









social studies

3. What is your favorite subject in school? _________________

4. Why is that your favorite? ______________________________________


5. What subject do you like the least? _______________

6. Why is that your least favorite? _________________________________



1. How did you feel when you were acting as a peer tutor? Why do you think you felt that way?

2. Do you think that you tutored Jose for a good amount of time? Why or why not?

3. Would you like to act as a peer tutor again? Why or why not?

4. Have you gotten better or worse at any subject in school this year?

5. What do you think caused that change?

Cite This Article as: Teachers College Record Volume 111 Number 8, 2009, p. 1894-1915
https://www.tcrecord.org ID Number: 15503, Date Accessed: 1/26/2022 12:08:42 PM

Purchase Reprint Rights for this article or review
Article Tools
Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Leigh Mesler
    Northwestern University
    E-mail Author
    LEIGH MESLER is a doctoral student in human development and social policy at Northwestern University’s School of Education and Social Policy. Before attending Northwestern, she was an elementary school teacher in the New York City public schools. Her research interests include education policy, teacher learning and change, the social organization of schools, and student achievement in urban schools. Recent work: J. Spillane, L. Gomez, & L. Mesler, “Notes on Reframing the Role of the Organizations in Policy Implementation: Resources for Practice, in Practice,” in D. Plank, G. Sykes, & B. Schneider (Eds.), Handbook of Education Policy. Mahwah, NJ: Lawrence Erlbaum, forthcoming.
Member Center
In Print
This Month's Issue