Participants

Two teachers participated in the research, one of Portuguese Language (male, 45 years old and 24 years of experience in teaching) and one of Mathematics (female, 41 years old and 11 years of teaching), as the tutoring sessions took place in the classes of these disciplines - which were selected because they have a higher workload than the other disciplines, six weekly classes (Resolution SE-30, of July 10, 2017), which allowed a longer time of interaction between the tutor and the tutee.

As for the participating students, after initial observation and in consultation with teachers, three students were selected as tutors (two girls and one boy, all 13 years old), according to the following criteria: empathy, ease of interaction with peers and teachers, class attendance, emphasis on academic skills, effective participation in classroom activities, responsibility, and expressing interest in participating in the program. Classmates from the same class and age were chosen because, according to the literature, from the point of view of learning and involvement, dyads of students of similar ages are better than dyads formed by students of different ages. It was also decided to have more than one tutor to avoid fatigue arising from the tutoring sessions and also so that the tutored students could exercise their ability to interact with more than one peer, not becoming dependent on a single peer tutor (^{Duran & Vidal, 2007}).

The tutored student selected for the study was appointed by the school and was registered with the São Paulo State Department of Education (Secretaria de Educação do Estado de São Paulo - SEDUC) as a student with intellectual disability (ID), which guaranteed his enrollment in the Resource Room, attended in the extraclass period, twice a week. In the afternoon, he was enrolled in a regular classroom, 8^th grade (corresponding to the 3^rd grade of Middle School in the US), with 25 students, 16 boys and 9 girls. According to the records of the school, the tutee presented delays in his academic development and emotional problems, for which he used medication. The school team identified special educational needs in his school trajectory and presented a medical report that indicated additional psychopedagogical support for his school activities, which justified his selection for the present study.

All participants and their respective guardians consented to the research.

Research environment

The school where the study took place is located in a small city in the hinterlands of the state of São Paulo, Brazil, with 3,246 inhabitants. It is the only state school in the city and attends Elementary School students - final grades and High School students. This school was chosen because of the proximity the researcher had with the institution and because the school has a Resource Room. At the time, the school had 312 students enrolled, 208 at Elementary School and 104 at High School. Of these, ten attended the Resource Room. The school’s staff, at the time of the research, consisted of 24 teachers, three school organization agents, three secretaries and three cleaners. The management team had a substitute principal, a vice- principal and a pedagogical coordinator teacher who attended to both segments.

Data collection instruments

To carry out this study, two types of instruments were used: systematic and non-participant initial and continuous observation scripts and initial and final semi-structured interview scripts.

Data collection procedures

The research design took place in three stages. The first, Elaboration of the peer tutoring program, consisted of five sessions of systematic and non-participant observation carried out in the classroom, prior to the formation of dyads, in order to collect information about the behavior of the tutored student and other peers. The data were registered in the initial observation script, taking into account the following categories: a) relationship between the teacher and the research target student and other students; b) interaction of the tutor with peers and their level of participation in the activities proposed by the teachers; c) participation of other students in the classroom, with emphasis on those indicated by teachers as students with favorable characteristics to be tutors. Also at this stage, the initial semi-structured interviews with participating teachers and students took place.

In the second stage, Implementation of the Peer Tutoring program, careful planning was necessary. This program followed nine essential steps suggested by ^{Topping (2005}): 1) Context - analysis and selection of the SEN learner who should participate in the program, which class, which teachers, which disciplines; 2) Objectives - what is intended to be achieved by implementing the program; 3) Curriculum area - which disciplines will implement the program in class; 4) Selection and training of pairs - through initial observations in the classroom and consultation with teachers; 5) Structure and method of tutoring - choice of the five tutoring models will be adopted; 6) Training of tutors, training of teachers and preparation of tutees - key to the success of the entire program; 7) Sessions - establishment of schedules; 8) Monitoring - through continuous observation for effective monitoring of the dyads, especially the role of the tutor; 9) Assessment - ​​continuous. For data collection at this stage, 30 sessions of non-participant systematic observations in the performance of tutors in the classroom and eight meetings to evaluate the tutoring sessions were carried out.

In the third stage, Evaluation of the Peer Tutoring program, final interviews were carried out with all those involved, in order to analyze and evaluate the influence of the strategy on the participation of the SEN learner, the tutor-tutee relationship, the expectations of tutors and teachers involved in the tutoring action and the impressions, understanding and feelings about the tutoring, after the participation of those involved.

In Table 1, there is an overview of the program’s implementation:

In stage II, the training of tutors took place. ^{Baudrit (2009}) states that tutors need to be well trained and suggests the use of the PPP method: Pause, Prompt, Praise. In the study, the acronym PPP was adopted to refer to the method. The author emphasizes that the method improves the tutor’s psychosocial and interaction skills, as it requires the use of social communicative skills, such as, for example, paying attention or expressing oneself clearly and helping and giving the tutee time to think or ask questions. The three steps of the PPP method instruct the tutor to: wait for the tutee to identify mistakes and have enough time to correct them; intervene, in order to help in the investigation, if the tutee, alone, is not successful in rectifying the mistakes; and encourage self-correcting behaviors, leading the tutee to appropriately use the information provided by the tutor.

Data analysis

For the analysis of the data of this research, the technique of Content Analysis of ^{Bardin (2010}) was adopted. The procedures used to process the data are presented below.

The first of them was the Organization of Analysis, when the pre-analysis took place, through skim reading the material collected throughout the research, at which time it was decided on the documents that made up the corpus of the analysis (the records of the initial observations and records of tutoring meetings and statements from 12 interviews). Finally, the theoretical framework was organized in which the results of the analysis were treated according to the literature consulted throughout the study.

The second procedure was the Coding, when the raw data of the text were transformed into a representation of the content studied in the corpus, also obtaining the characteristics of the written or verbal messages. Three techniques followed: clipping, enumeration and classification and aggregation.

The next procedure was the Categorization, carried out in two stages - inventory and classification -, which allowed defining the ten categories around which the data were discussed, namely: a) Relation of the tutee with tutors and teachers; b) Understanding of the PT program; c) Perceptions of tutors and teachers about the PT program; d) Perceptions about expectations regarding the PT program; e) Perceptions of tutors about the training process for the tutoring action; f) Perceptions of those involved about the participation of the tutored person; g) Duration of tutoring sessions; h) Difficulties in the tutoring action; i) Self-assessment of the tutoring experience; j) Evaluation of the PT program after the end of the research in the perception of teachers and tutors.

The fourth and last procedure - Treatment of results, inference and interpretation of results - made it possible to find the answers to the research questioning, through the statements of the participants, from the answers to the interviews and the statements of the tutee and tutors in the tutoring meetings, with the objective of faithfully comparing the participation of the tutor in the classroom. At this stage, the theoretical frameworks relevant to the investigation were returned. Table 2 below shows how the categories were created.

Results found before implementing the PT program

In the first category, Relationship of the tutee with tutors and teachers, it was observed that the TT had no bond of friendship with any classmate except one who sat behind him, who was the one who helped him to carry out the proposed activities by the teachers. In the initial interview, when reporting how his classmates treated him, the TT answered only: “Well”; the tutors, in turn, about their relationship with the TT before the beginning of the program, answered² (Table 3):

When asked about his relationship with the teachers, TT replied that some helped him and others did not. In a tutoring meeting, he said that the Mathematics teacher was nice, had patience with him, explained well and praised him a lot, but since he could not read or write, the Portuguese teacher thought he did not want to do the activities.

In the second category, Understanding about the Peer Tutoring program, when TT was asked in the initial interview if he understood how his participation in the research would take place, he replied: “Are you going, how to say? You are going to talk, then you are going to talk to the teachers and students”. The continuation of the dialogue was: Who will these students be? “Tutors” (TT). And these peer tutors what will they do? “They is going to sit with me and help me” (TT). ^{Duran and Vidal (2007}) state that tutors are satisfied with the help of a peer to the detriment of the prejudice that holds that students tend to reject the help of their peers and seek help from the teacher (p. 36).

After four days of training the tutors, they reported what they understood about the PT strategy (Table 4):

After training, the teachers were able to understand a significant aspect of the tutoring action, which is the benefit it brings to the classes (Table 5).

^{Duran and Vidal (2007}) emphasize that teachers should take advantage of the ability to cooperate among students and, precisely, see the differences between students as an enriching and helpful element in the educational task (p. 14). The authors claim that many teachers believe that they will lose control of the class and have discipline problems if they use cooperative work, but, on the contrary, the results of cooperative work encourage students to take control of the rules, regulating peers and themselves and that all this change in the classroom should be under the supervision of teachers (p. 36).

In the third category, Perceptions of tutors and teachers about the program, tutors demonstrated that they understood that they were chosen because of their school performance (Table 6).

^{Duran and Vidal (2007}) state that tutors feel that the tutee’s learning will depend on the level of help they will provide and this will cause them to get emotionally involved in the quality of this relationship, of which, for the role they play, is the maximum responsibility (p. 47), which is in line with the T3 statement. It was he who, during the entire process, worked most with TT in Mathematics classes.

As for the teachers, they were very satisfied with having been chosen to participate in the research (Table 7).

Regarding the fourth category, Reports on expectations regarding the Peer Tutoring program, the statements in the initial interviews, after the presentation of everything that would happen during the program, revealed the hopes of the participants placed in a strategy that would be implemented with the objective of evaluating the participation of a student who, until then, had not had good quality participation in the classroom, as shown in Table 8:

The statement of the PLT is in accordance with the definitions presented in the literature. ^{Duran and Vidal (2007}) point out that the PT strategy shows teachers that, in class, they are not alone to carry out their educational work and that students can become allies in this task, as everyone is able to help peers who need additional support to learn. They also emphasize that this moment of interaction between students may even be beneficial to the solution of discipline problems that exist in the classroom, as the focus of learning will be the dialogue on the activities undertaken by the pairs, which will allow learning and metacognition (Duran & Vidal, 2007, p. 65), which is a great opportunity for teachers to understand how their students think.

For tutoring sessions to occur successfully, it is necessary to dedicate quality time to tutors’ training. As pointed out by ^{Mosca and Santiviago (2013}), the experience of training the dyads puts both the tutor and the tutee in an active role in relation to the learning process in academic and affective aspects, and it is precisely these aspects that support the processes of learning.

Therefore, in the fifth category, Perception of tutors about the training process for the tutoring action, the tutors were asked if the training week, prior to the beginning of the program, was enough to prepare them for the tutoring sessions (Table 9).

The tutors were also asked if they understood what role they would play (Table 10):

According to ^{Baudrit (2009}), the tutors’ mission is to help their peers when they have learning problems, but also integration or insertion problems (p. 46).

In the sixth category, Perceptions of those involved about the participation of the tutored, it can be inferred that the tutors understood that, in addition to academic competence, social competence should also be stimulated (Table 11).

Before the end of the training, the researcher carried out the initial interview with the tutors and it can be inferred that they understood that not only academic competence, but social competence should also be stimulated, as shown in Table 11.

The next table presents the statements of tutors and teachers in the initial interviews regarding the participation of TT in the classroom (Table 12).

Results found after implementing the PT program

In the first category, Tutored relationship with tutors and teachers, the tutors confirmed that there was an evolution in the interaction with the TT (Table 13).

This relationship established in a tutoring action promotes social interactions between those involved, as tutors are able to listen, advise and provide information to the TT when he experiences difficulties, which leads him to acquire trust in the tutors (^{Baudrit, 2009}). It was observed that this relationship with the tutors extended to other peers in the classroom, because, unlike the week in which the initial observation was carried out, the TT left his desk to talk with other peers and they also came to his desk to borrow some material or to chat.

At the end of the program, the statements in the final interviews allow us to infer that the relationship of the participants has changed. When TT was asked if receiving help from a classmate made a difference in his daily life at school, his answer was:

It did. Before, I had no friendship with R. (T3) and G. (T1). With I. (T2) I had. Now we are friends and we even went to sleep at another peer’s house. And the next day R. (T3) and I. (T2) and her sister went to eat at my house. This has happened twice already. It had never happened before. I think the three tutors like me a lot. Being friends with them was better than having improved in the activities (TT).

In the fifth category, Perception of tutors on the training process for the tutoring action, based on continuous observations carried out throughout the program, it is believed that the time of one week used for the training of tutors was sufficient for the work carried out in this research. However, it is noteworthy that, in another situation, with more tutees and tutors and, depending on the condition of disability that requires broader training to know how to deal with different characteristics, more time should be reserved for that training.

The next table presents the statements of tutors and teachers in the final interviews (Table 14), which express a change in the participation of the TT, both in academic and social competence, but particularly in social competence:

It is noteworthy that, during the tutoring sessions, the PPP method was used and it is believed that it was essential for the program to run smoothly. The tutors tried to forward the session in the following order: first they explained the activity and waited for the TT’s response; then, when he did not respond or responded unsatisfactorily, they intervened, trying to explain in another way; at the end of the sessions, they always gave feedback, encouraging him to persist. The researcher insisted on this method, because it was a way to prevent the tutors - when explaining the activity, if the TT could not do it - from delivering a ready answer. It is important to inform the TT that the method will be used, so that he is aware that the performance of activities by him is linked to these three steps.

Tutoring meetings were times dedicated to evaluating the tutoring sessions. A total of eight meetings were held. A brief timeline with important facts that marked each one is presented below.

At the last tutoring meeting, an evaluation of the program was carried out. In it the tutors reported that they felt that they themselves had changed, as they realized how difficult it is for a person to want to learn and have difficulty, and how difficult it is to teach. The TT reported:

I’m happy because I’m doing the quizzes. I improved more in Mathematics than in PL. My biggest difficulty is reading and writing. I want to thank my friends [at first he referred to tutors as peers] who helped me. Before, my mother would look at my notebooks and would ask what I learned at school and I didn’t know how to say. Now I explain. And I also liked it because they go to my house (TT).

To complete the study, tutors were asked if they would like to continue with the sessions. The three tutors said yes, but that they would like new tutors to be trained so that they can rest a little (Table 15).

After the end of the program, the assessment of the use of the strategy began, through the final interviews. In the seventh category, Duration of tutoring sessions, participants reported whether the time allocated to the sessions was sufficient or not (Table 16).

The authors consulted do not determine an amount of time that should be allocated to the tutoring action. ^{Duran and Vidal (2007}) leave it up to the teacher to arrange the time in which the sessions will take place, whether inside or outside of school hours, and also cite a PT experience that was carried out within a context of structured format in weekly sessions of 45 minutes, over a period of two and a half to five months, with five pre-initial training sessions (p. 45). ^{Baudrit (2009}, p. 15) cited a successful PT experience in which sessions were spread over eight weeks, at the rate of three sequences of forty-five minutes per week (p. 15). ^{Hoot et al. (2012}) consider that “peer tutoring can occur two to three times per week for 20 minutes” (p. 3). In this program, the dyads worked a total of 100 classes, with daily meetings of 50 or 100 minutes when there were two classes together.

According to the results obtained, it is believed that this time should vary according to each situation: number of students involved, type of disability and the place where the tutoring action will take place. The important thing is that there is careful planning so that this time is significant for the work of the dyads and so that, at the same time, it prevents the tutors from getting tired.

In the eighth category, Difficulties in tutoring action, it can be seen how the implementation of this strategy has to be carefully elaborated. ^{Arguis et al. (2002}, p. 16) warn that tutoring is not the work of a day, it is an interactive work between everyone. There is a danger of drawing a profile so perfect that it is impossible or that it seems so impossible to the tutors that they give up straight away (p. 16). Through continuous observations, four situations were listed that caused fatigue to the tutors: 1) explaining the same exercise three, four or even more times and in different ways, which required patience and skills; 2) working with the TT before they perform the exercise themselves, which reduced their time to perform the activity in the notebook; 3) asking the teacher about any doubts so as not to explain wrongly to the TT, which they would not normally do, as the doubt would be clarified at the time of collective correction, but this situation could not be expected because the orientation to the TT happens before this moment of solution of doubts; 4) being careful when talking to the TT so as not to create any embarrassment. For this reason, it is believed that it is essential to have more tutors so that a better division of the time allocated to tutoring sessions is possible.

In the ninth category, Self-assessment of the tutoring experience, the TT statement demonstrates how benefited he felt through the use of the strategy:

It was cool. It was fun. I got better at other things outside of school, like counting money and texting on WhatsApp. I was just able to add numbers. Now I can subtract, multiply and divide. I managed to improve my reading a little. Before, I couldn’t write any message on WhatsApp. Now I can write a little bit. I text my father and aunt. I couldn’t read the names on the cell. I asked my mother to read it. Now I read almost all of them. (TT)

When asked what he liked most about being tutored, TT replied: “I liked more that I was able to do more things”. When asked what he liked least, TT said: “I couldn’t improve much in Portuguese”. About what he thinks did not work out, TT stated: “It all worked out”. We can infer that he was more confident because he was able to perform activities that he was unable to do before. Regarding this, ^{Fulk and King (2001}) highlight that PT seems to be particularly beneficial in improving the self-esteem of TT who have low social and academic performance. But it was not only the TT that felt the benefits of the strategy in their daily lives, as can be seen from what the tutors declared (Table 17).

The need to help another peer to solve problems that he alone cannot make the tutor organize his own cognitive process to be able to transmit the content more clearly to the TT. The perception of the cognitive benefit that the tutor acquired when performing his role was portrayed in the tutors’ statements (Table 18).

But, when asked about what they liked least, the tutors reported feeling frustrated. In this sense, it is important to demonstrate to tutors, throughout the program, that there is a special concern for them. Table 19 below presents the aspects with which tutors felt most uncomfortable when performing their role.

The tenth category, Evaluation of the Peer Tutoring program after the end of the research in the perception of teachers and tutees, was important, as it allowed us to really understand the impact that the program had on the participants’ lives. When asked if he thought other students deserved this same opportunity, the TT replied: “Yes, because becomes a friend and learn” (TT). And, when asked if he would like to continue receiving this incentive after the end of the research, the TT said: “Yes, of course”. These statements are in line with what ^{Arguis et al. (2002}) state about the PT strategy: it can give rise to improvement from academic performance to interpersonal relationships (p. 79) and the notes of ^{Duran and Vidal (2007}) when they state that PT creates an environment in fundamental classroom for good affective and interpersonal results (p. 17). Teachers were asked if they would continue using this strategy in their classes or if it only served to collaborate with the research. The PLT replied that they were already using the strategy in other classes and that they realized that the student with ID felt more supported to carry out the work and that, for sure, they would continue to adopt that strategy. The MT also said he would continue to use it in his classes.

The data analysis is concluded with the teachers’ statements in the final interview, when the motivation that the tutee and tutors had to carry out the activities using PT was questioned (Table 20).