Peer Tutoring Effects on Omani Students ’ English Self-Concept

Based on the social cognitive learning theory (1997), peer learning can be viewed as an effective way of enhancing learning. In this study, peer tutoring, a form of peer learning, was examined. The current study investigated the influence of a peer tutoring program implemented at Sultan Qaboos University on students’ English self-concept. 125 Omani university students participated in the study. The Students English Self-Concept Scale (SESCS) and the Tutorial Programs Factors Scale (TPFS) were utilized to collect the data. The findings demonstrated that peer tutoring has a positive influence on English self-concept. Three tutorial program factors were found to positively predict English self-concept. Implications for EFL teachers, and recommendations for future research were discussed.


Introduction
Self-concept is one of the most influential constructs affecting students' lives.Forming a positive or a negative self-concept can result in remarkable effects on students' learning and school engagement (Alhamodi, 2010;Bong & Skaalvik, 2003;Ribas, 2009).Therefore, examining the importance of this construct and factors related to its development is essential in helping students to be productive individuals who live successful academic and social lives (Bong & Skaalvik, 2003).The current study aims to investigate the levels of English self-concept among Omani university students and the role of a peer tutoring program in its development.

The Definition of Self-Concept and Its Importance
In English learning, developing a positive self-concept will help students significantly improve their academic performance and feel confident in using English as a foreign language.Tang, Zhang, Li, and Zhao (2013) found that positive English pronunciation self-concept increases students' confidence in the English classroom and reduces their anxiety.Language teachers should therefore seek to exploit self-concept as a tool to improve their students' language learning (Liu, 2009).
Although different definitions of self-concept have been documented in the literature, almost all of them share the concept of understanding oneself.Self-concept has been defined as the 'accumulation of knowledge about the self' (p. 2) (Krishnakar & Chengti, 2012).Furthermore, Burns (1982) stated that self-concept can be described as beliefs and evaluations people make about themselves.These beliefs and evaluations indicate what they think about themselves, as well as what they think they can become in the future.
Self-concept can be discussed as a general term related to the overall feeling about the self.On the other hand, some studies have linked it to more specific social, physical, and academic domains (e.g., Harter, 1999;Marčič & Grum, 2011).Bong and Skaalvik (2003) defined academic self-concept as students' 'overall feelings of doing well or poorly in a given subject area ' (p. 71).Hence, it is clear that measurements of self-concept can be associated with a particular subject matter such as mathematics, science, or language.
Research into self-concept has often explored its significant role in affecting behavior.Herbert Marsh is a pioneering researcher in this field who emphasized the important psychological, social and educational outcomes related to varying levels of self-concept (e.g., Craven & Marsh, 2008;Marsh, 2004;Marsh, 1986;Marsh & Martin, 2011).The ultimate goal of the majority of self-concept research is to assist students to work and adapt well when academic demands are imposed on them (Bong & Skaalvik, 2003).Positive associations between self-concept and different school outcomes were found in the literature (e.g.Alhamodi, 2010;Barakat, 2009;Husain & Abdalyama, 2011).More specifically, Bong and Skaalvik (2003) determined that forming positive views about the self encourages students to set challenging, yet achievable goals.A more positive self image also reduces students' test anxiety, pushes them to persist longer in difficult activities, helps them to create plans, and increases motivation (e.g., Bong & Skaalvik, 2003;Ribas, 2009).Consistent with these findings, positive associations between self-concept and different school outcomes were found in literature (e.g., Alhamodi, 2010;Barakat, 2009;Husain & Abdalyama, 2011).
Of all the school outcomes, student academic performance is the one most often studied in relation with self-concept.According to Burns (1982), self-concept can significantly predict academic achievement.Burns found that students with positive self-concept are more satisfied with their relationships with their parents, teachers, and peers, which results in increased confidence when approaching tasks, which in turn leads to high academic attainment.Similarly, other researchers have found positive associations between self-concept and academic achievement (e.g., Ju, Zhang, & Katsiyannis, 2012;Liu, 2009;Marsh & Martin, 2011;Marsh, 2004;Pinxten, Fraine, Damme, & D'Haenens, 2013).Considering this well documented link, students should be encouraged to develop their academic self-concept.

Factors Influencing Students' Self-Concept
Different factors were found to be contributing to the development of self-concept.The most influential factors can be categorized into two main types: Home related factors and school related factors.The home environment is a key element that shapes a child's personality.Many researchers have shown that it is necessary to provide an effective and positive home environment to positively foster child self-concept (e.g.Burns, 1982;Hay, Ashman, & Kraayenoord, 1998;Manning, 2007).Parent involvement in their children's lives has been shown to contribute to the development of a positive or negative self-concept (Ju, 2012;Manning, 2007).Moreover, studies have shown that children's relationships with their siblings can also affect their self-concept (e.g., Barnett & Hunter, 2012;Massey, 1999;Oliva & Arranz, 2005).
In addition to home related factors, the school context has tremendous effects on students' self-concept.In school, children are more concerned with shaping their self-image since they find themselves in an unfamiliar environment that requires new forms of interaction (Burns, 1982).Two main elements can be investigated when studying the effects of a school environment on students' self-concept.The first is teachers, who were found to significantly influence students' self-concept through their teaching practices and relationships with the students (Alrajhi & Aldhafri, 2015;Leflot, Onghena, & Colpin, 2010).A negative school experience caused by teachers leads the students to form a negative self-concept (Komarraju, Musulkin, & Bhattacharya, 2010); on the other hand, if teachers provide a positive school experience, students can be encouraged to form a positive self-concept (Erkman, Caner, Sart, Börkan, & Sahan, 2010).The second element is peers; as revealed by Hay et al. (1998) and Manning (2007), peers can leave a great impact on child self-concept.Also, Egbochuku and Aihie (2009) determined that counselors find it more effective to conduct counseling sessions using peer groups owing to the positive impact on self-concept.These researchers found that peers have a stronger influence over each than adults have over children.Therefore, peer effects should be given careful attention.
In an attempt to exploit the positive effects of peers, peer tutoring programs have started to appear as an instructional method; these combine and exploit aspects of both teaching and peer relationships.It has been found that peer tutoring has many positive effects on the psychological and academic outcomes of children.The current study examines the effects of peer tutoring in a university setting.

Definitions and Types of Peer Tutoring
Topping (2005) defined peer learning as the 'acquisition of knowledge and skills through active helping and supporting among status equals or matched companions.It involves people from similar social grouping who are not professional teachers helping each other to learn and learning themselves by so doing.' (p.631).According to Shabani and Gerdabi (2013), peer tutoring is a cooperative learning method based on the idea of pairing students who share a common goal and 'asymmetrical' relationship (see further definitions of peer tutoring in LeFluer, 2010;Lyttle, 2011;Mayer, 1998).Definitions of peer tutoring, however, differ based on the type of peer tutoring program.Miller, Barbetta, Drevno, Martz, and Heron (n.d) have classified five types of peer tutoring: Classwide peer tutoring, cross-aged tutoring, one-to-one tutoring, small group instruction, and home-based tutoring.Other types have also been identified, such as Peer-Assisted Learning Strategies (PALS), and Cooperative Integrated Reading and Composition (CIRC) (e.g., Boudouris, 2005).
It is argued that the rationale behind peer tutoring programs can be traced back to social cognitive learning theory (Bandura, 1997) which contends that the best way to learn is by imitating others.It is thought that through imitation, one's behavior and cognition are influenced.Bandura also highlights the socialization process as being essential for students.Thus, combining imitation and socialization, peer tutoring helps students to learn at a more suitable pace (Lyttle, 2011;Tella 2013).Tella (2013) emphasized that peer tutoring should be considered a supplementary process, and not a replacement to teacher directed activities.It can thereby function to provide the students with feedback and guidance on concepts that have been previously explained and demonstrated by their teachers.Furthermore, peer tutoring aims to facilitate student learning by having groups of students work independently, with no teaching provided, nor textbook solutions for assigned problems.Ideally, in these independent groups, peer tutors assist junior students by acting as a focus for the group (Schleyer, Langdon, & James, 2005).

The Importance of Peer Tutoring for Students Learning
Peer tutoring programs have been found to have positive effects on several different learning outcomes.Schleyer et al. (2005) found that as a result of the peer tutoring process, students became more confident, developed problem solving and communication skills, and became more responsible for their learning.In addition, Saenz, Fuchs and Fuchs (2005) examined the influence of Peer-Assisted Learning Strategies (PALS) on the reading performance of native Spanish-speaking students with learning disabilities (LD).The findings of this study showed that PALS students outgrew control students on reading performance.Similarly, peer tutoring was referred to as a 'promising intervention' to deal with the social problems in children with ADHD.For example, according to Plumer and Stoner (2005), combining class wide peer tutoring (CWPT) with peer coaching results in positive social behavior in students with ADHD.Hence, it is argued that peer tutoring is an instructional methodology that, if possible, should be included in a teacher's plan.Teachers may, at times, be skeptical of relinquishing so much control over classroom activities; yet, skepticism should not be confused with bias against peer tutoring programs (Miller et al., n.d).Still more positive aspects of peer tutoring include: providing a supportive environment for the students to examine their comprehension of certain difficult concepts introduced in the class session, improving confidence in dealing with complex material, allowing deeper learning, and encouraging students to be more responsible for their learning (Baillie & Grimes, 1999).

The Role of Peer Tutoring in English learning and Self-Concept
It has been suggested by many researchers that peer tutoring programs have positive effects on both EFL and ESL English learners.For instance, Shabani and Gerdabi (2013) found that contrary to individual teacher-guided read aloud, peer-tutored read aloud was effective in vocabulary learning for Iranian EFL students.By fostering peer scaffolding, peer tutoring is also effective in helping learners to overcome task challenges such as lack of ideas, limited content knowledge, difficulties in pronunciation, and presentation skills (Nguyan, 2013).These positive effects all contribute to the formation of positive self-concept in learning English.
Research has also shown the positive effects of peer tutoring on self-concept and attitude.Rosewal et al. (as cited in LeFluer, 2010) found that self-concept and general attitude towards school in students who engaged in a peer tutoring program increased significantly more than the self-concept of students who didn't participate in the program.The researcher stated that after implementing the peer tutoring method in the classroom, the number of students feeling positive about their ability increased.Also, Nelson (1996) showed that parents and teachers of children who experienced a peer tutoring program reported a significant positive change in student behavior, self-perception, and attitude towards their school.
Although it is apparent that self-concept has a significant role in English learning and that using peer tutoring programs is one method for its development, very little research, if any, has been conducted to study the effects of peer tutoring on English language self-concept.Moreover, little has been done to understand what features of these peer tutoring programs help to make them so influential, especially with regards to one-to-one peer tutoring that usually happens outside regular class sessions.Several factors which potentially lead to the success of peer tutoring programs can be examined.Examples of these factors are; the close age between the tutor and the tutee, the sharing of the same social culture, the sharing of learning experiences, and many others.Thus, the current study aims to fill this gap in the literature by examining the effect of a peer tutoring program implemented at Sultan Qaboos University (SQU), the Sultanate of Oman, on English self-concept of students enrolled in the foundation course and who learn English as a foreign language.The study also explores the predictive role of tutoring program related factors in students' English self-concept.This research paper aims to answer the following questions: 1) What are the levels of the students' English self-concept?
2) Can peer tutoring related factors predict students' English self-concept?

Participants and Procedures
The sample consisted of 125 university students who were taking the English foundation course at SQU. Female students represented 43% of the sample.The participants were enrolled in different colleges in the university.The majority were from the colleges of science (28), commerce (21), and arts and social sciences ( 16), but also included other students from the colleges of law, education, medicine, nursing, agriculture and engineering.The participants were informed that participation in the study was voluntary, and that it would not affect their future studies or their grades.The participants were asked not to write any identifying information and were assured of confidentiality.All students who were invited to participate in the study agreed to participate and the data collection was carried out during class sessions by English instructors.All participants were taking sessions (max 3 a day, 30 minutes each) at the Tutorial Center (TC), in the Language Center at SQU.The type of peer tutoring in this center is one-to-one tutoring where senior students studying English majors -from the college of arts and social sciences, or the college of education -work as peer tutors and conduct tutoring sessions for students in the foundation program throughout the year.The peer tutors underwent special training sessions before starting the peer tutoring program.

Instruments
To measure the levels of the participants' English self-concept, the researchers developed The Students' English Self-Concept Scale (SESCS).An alpha coefficient of 0.90 was found for the SESCS.This scale consists of 13 items utilizing a 5-point-Likert scale.An example of a scale item is 'After attending sessions in the TC, I feel more confident in using English.'The second scale developed is the Tutorial Program Factors Scale (TPFS); a 10 item, 5-point-Likert scale which examines students perceptions about factors related to the tutorial center generally, and the peer tutors specifically.An example of one of the factors items is 'I like to talk with the peer tutors because they are close to my age'.An alpha coefficient of 0.85 was obtained for the TPF.The scales were administered in Arabic.

Data Analysis and Results
Data were first screened for outliers and none were identified.There were very few missing data resulting from items left blank.To answer the first research question related to the levels of English self-concept, and the students' perceptions of the peer tutoring program-related factors, descriptive analyses, along with the one-sample t-test, were applied at item and scale levels.
At the item level all (13) items of the English self-concept scale and the (10) items of the factors influencing self-concept were all statistically above the theoretical mean (3.00).Similar results were found at the scale level for both scales of the study.These results indicate that the participants have high levels of English self-concept and perceived the effects of the ten factors to be large.
Next, we analyzed the data to attempt to understand the predictive role of the ten tutoring program-related factors in the levels of students' English self-concept.Linear regression analysis (using the stepwise method) was applied to explore what factors influence students' English self-concept.Prior to the linear regression analysis, we examined the Pearson correlations between the ten tutoring program factors and the mean level of students' English self-concept.Table 1 shows the bivariate correlations.As can be seen from Table 1, all tutoring program factors correlated significantly and positively with the SESC.The highest correlation value was found for 'encouraging the tutee to try using English even if he/she made mistakes'(r = 0.51, p < 0.001) and the lowest correlation value was found for 'sharing the same social culture' (r = 0.30, p = 0.01).The correlations of the other factors with SESC were as follows: 'Providing the opportunity to speak with people who have gone through the same experiences'(r = 0.50, p < 0.001),'the tutors are patient with the tutee' (r = 0.50, p < 0.001), 'using a variety of activities' (r = 0.47, p < 0.001),'small age gap between the tutor and the tutee' (r = 0.45, p < 0.001), 'allow the tutee to speak in Arabic at times' (r = 0.36, p < 0.001), 'informal relationship between the tutor and the tutee' (r = 0.35, p < 0.001), 'understanding the tutees' difficulties' (r = 0.32, p < 0.001) and 'sharing learning experiences with the tutee'(r = 0.31, p < 0.001).
Table 1 also shows that all the tutoring program-related factors correlated positively with each other (except for two correlation values).It is noticed that the Pearson correlations among the ten factors were low to moderate, ranging from (r = 0.11 to r = 0.51) with the exception of the correlation between (item 20 'encouraging the tutee to try using English even if he/she made mistakes' and item 21 'using a variety of activities that allow practicing English', r = 0.60).These levels of correlation among the ten factors are important as a pre-requisite to using these factors to predict SESC.Therefore, co-linearity was not of concern in our regression model.
We proceeded to examine the prediction role of the ten tutoring factors on SESC using the Stepwise multiple regression technique.The stepwise method was preferred over the normal linear regression in order to reach a parsimonious regression model that can explain high levels of variance in the dependent variable (i.e., SESC) with a low number of possible predictors (Tabachnick & Fidell, 2001).The findings of this model are displayed in Table 2.  1) showed that all the ten tutoring factors correlated significantly and positively with the SESC, the regression analysis shows that only three factors appeared to be possible significant predictors of the SESC.These included 'sharing similar experiences in learning English' (β = 0.37, p < 0.001), 'small age gap between the tutor and the tutee' (β = 0.27, p < 0.001), and 'encouraging the tutee to try using English even if he/she made mistakes' (β = 0.261, p < 0.01).The overall model including these three tutoring factors explained 44% of variance in SESC.

Discussion
This paper examined the levels of self-concept in learning English as a foreign language for students enrolled in the foundation course, and attending sessions in a peer tutoring program at SQU.In addition, it explored the effects of this program on the students' English self-concept.The findings support the previous literature, which asserts that peer tutoring facilitates English learning (e.g., Nguyan, 2013;Shabani & Gerdabi, 2013) and contributes to the development of self-concept (e.g., Nelson, 1996;Rosewal et al., as cited in LeFluer, 2010).
First of all, the findings showed that after attending sessions in the tutorial center, the participants showed improvement in their levels of self-concept in learning English.Additionally, they manifested high levels of English self-concept.Although no previous research was found by the researchers that examined the influence of a peer tutoring program on English self-concept, some research has investigated the impact of similar programs in other subject areas, and revealed positive effects (e.g., Nelson, 1996;Rosewal et al., as cited in LeFluer, 2010;Topping, Campbell, Douglas, & Smith, 2003).After attending sessions in the tutorial center, the students felt that their English language levels improved; they also became more confident, and more willing to learn.
Furthermore, the correlations showed that all tutorial program factors (TPFs) significantly and positively correlated with the overall mean of the SESC.This indicates that the aforementioned factors had positive impacts on students' self-concept.The most strongly related factor was 'encouraging the tutee to try using English even if he/she made mistakes.'The students possibly valued this factor most highly because, as commonly accepted, encouragement and motivation are very important in EFL classes (e.g., Awan, 2011;Reza, 2013).Also, Tella (2013) suggested that the ability to motivate the tutee is one of the criteria that teachers should consider when selecting peer tutors.Students may feel more motivated to attempt different tasks, despite their mistakes, when they are encouraged by the peer tutors, since they feel that their tutors are students as well and that they make mistakes just as they do.This is in contrast to student perceptions of teachers, who they may view as faultless figures, whose level of language proficiency is unobtainable.The lowest related factor perceived by the students was 'sharing the same social culture with the tutors.'It can be interpreted that cultural differences are not the most important aspect in learning since students at this particular university are mostly used to being taught by teachers of different nationalities.
Finally, although the findings showed that all (10) tutoring program factors correlated with the SESC, the regression analysis demonstrated that only three factors can possibly predict the SESC.These factors were: 'sharing similar experiences in learning English, 'small age gap between the tutor and tutee', and 'encouraging the tutee to try using English even if he/she made mistakes'.It is obvious that these three factors are qualities exhibited by the peer tutors in the center.The close tutor-tutee relationship that is characterized by sharing the same experience, being of a similar age and offering encouragement, is very important in enabling the students to form a positive self-concept related to their English learning.Therefore, tutorial center staff and school teachers should pay attention to these factors when they choose peer tutors to ensure maximally positive effects.
These factors facilitate the establishment of a more caring and non-threatening atmosphere.One of the tutor roles is to encourage the students to communicate freely in a non-threatening environment (Schleyer et al., 2005).Because these factors account for only 44% of the variance in the SESC, other factors-such as parent and teacher involvement -are likely to be playing a role along with peer interaction in developing self-concept (eg., Ju, 2012;Leflot et al., 2010;Manning, 2007).

Conclusion
The findings of this research supports the notion that peer tutoring has a significantly positive impact on English self-concept.Utilizing this information, English teachers in schools should be encouraged to employ the peer tutoring process inside and/or outside the classroom.Peer tutoring programs in Oman are becoming commonplace in universities and colleges; yet, they are still very rare in schools.Although many teachers in Oman adopt a collaborative learning approach by having students work in groups and pairs, this approach is different from peer tutoring.Often students working collaboratively to execute a task are grouped in one grade level inside the classroom.It is recommended that school teachers start to implement peer tutoring programs which are similar to ones found in higher education institutes.Teachers can select experienced and skilled students, possibly from a higher grade, to tutor low achieving peers.
In addition, it should be noted that peer tutoring programs do not only benefit the tutees, but the tutors as well.
Many studies have proven the positive effects of peer tutoring on the tutor's self-concept (see for example, Flores & Duran, 2013;Gisbert & Font, 2008).It is therefore necessary that future research be conducted on the same peer tutoring program to investigate the influence of peer tutoring for both tutors and tutees, as well as examining the influence of this program on other outcomes such as achievement.Self-concept can be examined as a mediator in the relationship between peer tutoring and academic achievement in English language learning.

Limitations
One limitation of the current study is the measurement of self-concept.The improvement reported for participants' English self-concept was based on their perceptions of peer tutoring effects, and perceptions of their self-concept.However, there was no test for self-concept levels prior to attending peer tutoring sessions, and at the end of it.Future research, may utilize a better research design (for example -experimental) to capture the effects of tutoring.A further limitation was that no achievement scores were obtained to examine the actual level of English performance.

Table 1 .
Correlation coefficient between self-concept and tutoring related factors SA = small age gap between the tutors and the tutees; UTD = Understanding tutees' difficulties; IR = Informal relationship; LE = going through similar English learning experience; SSC = Having the same social culture; SLE = Sharing learning experience with the tutee; E = Encouraging the tutee to try using English; VA = Using a variety of activities; P = Tutors are patient with the tutee; SP = Giving the tutee the chance to speak in Arabic at times.Note: All coefficients are significant at p < 0.01.

Table 2 .
Multiple regression models: Predicting students' English self-concept