Au, H.Y.C. & Bardakçı, M. (2020). An analysis of the effect of peer and teacher feedback on EFL learners’ oral performances and speaking self-efficacy levels. International Online Journal of Education and Teaching (IOJET), 7(4). 1453-1468. http://iojet.org/index.php/IOJET/article/view/895 Received: 12.05.2020 Received in revised form: 25.06.2020 Accepted: 22.07.2020 AN ANALYSIS OF THE EFFECT OF PEER AND TEACHER FEEDBACK ON EFL LEARNERS’ ORAL PERFORMANCES AND SPEAKING SELF-EFFICACY LEVELS Research article Hong Yu Connie Au Gaziantep University connieau@gantep.edu.tr Mehmet Bardakçı Gaziantep University mbardakci@gantep.edu.tr Hong Yu Connie Au is an instructor at the School of Foreign Languages, Gaziantep University. Mehmet Bardakçı is an associate professor of English Language Teaching at Gaziantep University. Copyright by Informascope. Material published and so copyrighted may not be published elsewhere without the written permission of IOJET. mailto:connieau@gantep.edu.tr mailto:mbardakci@gantep.edu.tr http://orcid.org/0000-0003-4574-9763 http://orcid.org/0000-0002-0071-0059 International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1453 AN ANALYSIS OF THE EFFECT OF PEER AND TEACHER FEEDBACK ON EFL LEARNERS’ ORAL PERFORMANCES AND SPEAKING SELF-EFFICACY LEVELS Hong Yu Connie Au connieau@gantep.edu.tr Mehmet Bardakçı mbardakci@gantep.edu.tr Abstract This quasi-experimental study aims to determine EFL learners’ speaking self-efficacy levels and to examine the changes in oral performances after receiving feedback in three module courses. The relationship between speaking self-efficacy and oral performances after receiving feedback was also studied. Quantitative method was applied in this study. The study was conducted among thirty-three EFL prep-school participants in a state university in Turkey. The research began with the implementation of a speaking self-efficacy scale (pre- test) and a speaking test. This process was followed by three oral treatments following teacher and peer feedback and the same speaking test. Repeated treatments and the same speaking test were conducted in the other two modules, as well as a final speaking self- efficacy scale (post-test). The results of the speaking self-efficacy scale revealed that students generally possess high speaking self-efficacy level. The findings also showed that students demonstrated significant changes in their oral performances following feedback treatment in both groups. In the final speaking test, participants who received teacher and peer feedback improved by 212.36% and 161.20% respectively compared to the first speaking test. However, no significant correlation between the two variables was observed. This may suggest that the sample size was insufficient to observe such a relationship. Keywords: peer feedback, teacher feedback, oral performances, speaking self-efficacy 1. Introduction Speaking is a crucial skill in learning English as a foreign language as well as an essential skill for communicating with people in daily life. At the same time, speaking commonly has been viewed as “the most demanding of the four skills” (Bailey & Savage, 1994, p.7). In Turkish EFL classrooms, learners generally cannot speak well due to speaking anxiety and low self-efficacy (Gürsoy & Karaca, 2018). Self-efficacy has been a crucial factor that assisting or hindering language learning progress as well as learners’ preferences for certain learning activities (Bandura, 1977, 1986, 1997; Pajares, 1996; Zimmerman, 2000). In other words, self-efficacy may be regarded as a consistent predictor of students’ motivation and learning strategies. For instance, learners with a higher sense of self-efficacy are more likely to achieve their goals while avoiding negative emotions after making mistakes (Yanar & Bümen, 2012). Without adequate self- efficacy, learners may not be able to attempt more challenging tasks and demonstrate their abilities. Moreover, a higher sense of self-efficacy may lead to increased motivation, which mailto:connieau@gantep.edu.tr mailto:mbardakci@gantep.edu.tr Au & Bardakçı 1454 aids students in focusing on their learning tasks, performing better, and ultimately achieving their learning goals. In addition to self-efficacy, another factor affecting oral performance is feedback. Feedback refers to information provided by an agent such as teacher, peer, or oneself regarding the aspects of one’s performance or understanding (Hattie & Timperley, 2007). It plays an important function in learner output in that it may be viewed as a result of a performance (Sheppard, Flexer, Hiebert, Marion, Mayfield & Weston, 1996). Thus, feedback may significantly impact the processes of learning and achievement (Hattie & Timperley, 2007). Moreover, based on its instructional purpose, feedback should provide information specifically related to a given task or process which fills the gap between what is understood and what is aimed to be understood (Sadler, 1989). Feedback is also given in response to learners’ errors, and these learners may accept, modify, or even reject the feedback. Moreover, feedback may enhance learners’ sense of self-efficacy by enabling them to self- reflect and concentrate on their peers’ judgment of their language capabilities. In particular, positive judgment, praises, and feedback affect the way learners evaluate themselves. It also enhances learners’ willingness to learn and their self-confidence, and thus their learning motivation (Yang & Wu, 2013). In other words, accepting feedback can be viewed as the most effective factors in enhancing self-efficacy and achievement (Yang & Wu, 2013). 2. Literature Review 2.1. Self-Efficacy Many studies have examined learners’ beliefs, also known as self-efficacy (Bandura, 1977, 1986, 1997; Pengajaran, 2018; Zimmerman, 2000). Bandura (1977) introduced the idea of self-efficacy as a key element in learners’ success. Self-efficacy was defined as a person’s beliefs concerning his or her completion of a task and his or her competency level in performing the task (Bandura, 1977). In other words, self-efficacy concerns how learners think about their capabilities to organize and complete a learning task to reach a goal (Zimmerman, 2000). Self-efficacy beliefs are an important aspect of human motivation and behavior, and they directly influence certain actions (Pengajaran, 2018). It also focuses on performance capabilities rather than on personal qualities. Bandura (1977) also suggested that self-efficacious students are usually more hard-working and engage more readily and persistently throughout the learning process. Unlike inefficacious students, self-efficacious students do not doubt their abilities and have more stable emotional reactions when they encounter difficulties (Zimmerman, 2000). That is, individuals who accomplish tasks successfully usually have higher self-efficacy (Ocak & Olur, 2018). Those who have a lower sense of self-efficacy for accomplishing a task may avoid it, while those who believe they are capable often participate readily (Schunk, 1991). Finally, in terms of choosing activities, self- efficacious students usually prefer more challenging tasks. 2.2. Speaking Self-Efficacy and Language Learning Despite the significance of self-efficacy to the learning process, only few have investigated its relationship to students’ speaking ability (Alawiyah, 2018; Asakereh & Dehghannezland, 2015; Dasmo & Sundari, 2014; Liu, 2013). One researcher who has investigated this relationship is Liu (2013), who assessed the impact of an “English Bar” on college students’ speaking self-efficacy. The results demonstrated that students who had often spoken English at the bar possessed higher levels of self-efficacy than those who had rarely or never visited the bar. He also observed four positive effects of speaking English at the bar: (1) students were free to choose their partners to reduce their anxiety level, (2) students with inadequate speaking skills were motivated by International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1455 foreign teachers and their partners, (3) students’ self-efficacy was enhanced while they observed “similar others” who were proficient speakers, and (4) students worked harder as they realized that they were making some improvement in English. In another study concerning speaking skills achievement as it relates to speaking self- efficacy is Asakereh and Dehghannezland (2015). Asakereh and Dehghannezland (2015) investigated the relationship among student satisfaction with speaking classes, speaking self- efficacy beliefs, and speaking skills achievement. The results showed a significant positive correlation among the variables and that students with higher speaking self-efficacy were more likely to achieve higher scores in speaking skills. In a slightly different context, Alawiyah (2018) examined the relationship between speaking self-efficacy and EFL student-teachers’ speaking achievement. The results demonstrated that student-teachers’ self-efficacy levels significantly influenced their speaking achievement. 2.3. Feedback Learners should regularly compare their learning progress with their learning goals in order to develop their linguistic knowledge (Zarei, 2018). Thus, evaluation is essential to any learning and teaching process. Feedback is one of the common classroom evaluation, and there are different forms of instructional feedback for evaluating the knowledge, skills, and performances of learners. Teacher feedback, peer feedback, and self-evaluation are some examples. Evaluation aids learners in identifying their strengths and weaknesses, ultimately enhancing their achievement by highlighting progress rather than deficiency (Zarei, 2018). Moreover, in regard to feedback in speaking, speaking is a highly subjective activity, and the administration of speaking feedback is particularly difficult. Feedback is usually given immediately following the presentation by the teacher and or peers. Afterward, during the feedback process, students may revise their presentation based on the input they have received from their teacher and or peers. The significance of conducting research on feedback and evaluation in speaking is that it may increase both teachers’ and learners’ awareness of effective speaking criteria, improve learners’ speaking ability by enabling them to self-reflect on their performance, and foster more positive attitudes towards speaking. According to Marzano, Pickering, and Pollock (2001, p.187), effective feedback should adhere to the following set of criteria: 1. It should provide students with some explanation of why their productions are correct or incorrect. 2. It should be given immediately. 3. It should be specific to a set of criteria. 4. It should involve students in providing their own feedback based on feedback given by the teacher. 5. It should enable students to learn from their mistakes, make necessary changes, and achieve higher levels of performance. 2.4. Feedback on Speaking Performance A limited number of studies regarding feedback in an EFL setting have aimed to determine whether certain feedback types are more or less effective in improving students’ speaking performance (Lynch & Maclean, 2003; Smith & King, 2004). Au & Bardakçı 1456 One research that has examined the impact of feedback on the oral performance of English for special purposes students is that of Lynch and Maclean (2003). Their study was conducted in the Netherlands among a group of advanced students and involved a 20-minute speaking cycle. Students received tutors’ feedback on their spoken performance in written form and on an individual basis. The results indicated improvement in oral performances that had received feedback highlighting students’ weaknesses. The result is surprising, as students were aware of their language problems identified in teacher feedback as well as problems not identified in feedback. The most striking result was that students were aware of changes in their language use. Another investigation that has examined students’ feedback sensitivity and the efficacy of feedback interventions is that of Smith & King (2004). The results indicated that students with higher sensitivity towards feedback have better speaking behaviours. In particular, feedback was made in low intensity with less direct, personal way of criticism. 2.5. Teacher versus Peer Feedback in the Domain of Speaking Despite teacher and peer feedback are commonly studied, there is a significant lack of research comparing the effects of teacher and peer feedback on students’ speaking performance, both internationally and in a Turkish setting. Only one study has compared these feedback types in the domain of speaking (Murillo-Zamorano & Montanero, 2017). Murillo-Zamorano & Montanero’s study (2017), which involved thirty-two Economics and Business students in a Spanish university, has compared the impact of peer and teacher feedback on the oral presentation. The research tested whether oral presentation skills, with some support instruments such as videos and rubrics, provided improved following teacher and peer feedback. The results indicated that the peer feedback group had more improvement than that of the teacher feedback group in the post-test. Based on the findings, the results suggested that peer assessment could be somehow effective in enhancing oral presentation skills. However, the improvements were not maintained in the peer feedback group in the follow-up re-test. The results implied that a single session of peer feedback with rubric might not be sufficient enough to generalize any improvements in the said competency. As mentioned above, there has been a lack of research comparing teacher and peer feedback on speaking in international EFL context as well as in Turkey. The most striking fact is that the above study is the only one which compares two feedback types in speaking; moreover, it was not in EFL context. Hence, it is necessary for further research at this level. This necessity motivated the research of the present study, which investigated the effects of teacher and peer feedback regarding to speaking. More specifically, this study aimed to answer the following research questions: 1. Are there any changes in the level of students in speaking self-efficacy? 2. Is there a statistically significant difference in terms of oral performances after treatment? 3. Is there a correlation between speaking self-efficacy level and oral performance following feedback? International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1457 3.Methods 3.1. Research Design This quasi-experimental study employed a repeated measures design involving pre-test, treatment, and post-test of speaking performance as well as a speaking self-efficacy scale. In this way, the design utilized a quantitative approach. The duration was twenty-four weeks and spanned the course of three academic modules within the preparatory program (A1 Elementary, A2 Pre-Intermediate, and B1 Intermediate modules). For the repeated treatments, two treatment conditions were established: (1) peer feedback with TOEFL independent speaking rubric and, (2) immediate teacher feedback with the use of TOEFL independent speaking rubric. All of the oral presentations were conducted under these conditions. Participants were divided into two groups: Group A (peer feedback) and Group B (teacher feedback). Each student from Group A received peer feedback, while each student from Group B received immediate teacher feedback. A comparison was made based on the two types of feedback in terms of the content and performances of students’ pre-test and post-test assessments. 3.2. Participants For participant selection, a convenience sampling method was used. A1 students were selected as they typically are insufficient in speaking ability and can better demonstrate improvement in speaking tasks over a certain period of time. During the A1 module, Group A was comprised of sixteen (66.67%) males and eight (33.33%) females, while Group B consisted of seventeen (70.83%) males and seven (29.17%) females. By the end of the A1 module, six students from Group A and five students from Group B had dropped out or failed to pass to the A2 level. During the A2 module, the study continued with eighteen participants in Group A and nineteen participants in Group B. By the end of the A2 module, three students from Group A failed to pass to the B1 level. During the B1 module, the study continues with fifteen students from Group A and nineteen students from Group B. By the end of the B1 module, only one student from Group B quit the study due to the sickness. 3.3. Data Collection 1. Foreign Language Speaking Self-Efficacy Scale A five-point Likert scale was taken from the article, “The Scale Development Study on Foreign Language Speaking Self-Efficacy Perception” by Ocak and Olur (2018) (Appendix A & B). Only 24 items remained, and these were grouped according to three factors. The Cronbach’s Alpha value was found to be .944 in the study of Ocak and Olur (2018), which indicated the internal consistency of the scale. While in this study, the Cronbachs’s Alpha value was found to be. 943 in the pre-test and .944 in the post test. 2. Speaking Test with TOEFL Independent Speaking Rubric It was used to determine the improvement of speaking proficiency in different modules. The speaking test was designed by the researcher based on the syllabus of the School of Foreign Languages. The participants were asked to talk about a holiday that they have taken recently or sometime in the past. The same speaking test was conducted four times at the beginning of A1, at the end of A1, A2 and B1 module. Participants’ responses were evaluated according to the criteria for independent items listed on the TOEFL independent speaking rubric (Appendix C). It employed a four-point analytical scale, and the scores were derived from evaluators’ holistic considerations of a general description based on three dimensions: delivery, language use, and topic development (ETS, 2008). For each category, score bands Au & Bardakçı 1458 and a set of descriptors of student performance were listed and could be used systematically to assign scores to an individual student’s performance. In this study, in order to gain a clear understanding of students’ improvement, four points were assigned within each category, producing a total of twelve points. In addition, the average of two instructors’ scores functioned as the final grade. 3.4. Procedure This research was conducted for approximately twenty-four weeks during the first three modules of the 2018-2019 Academic Year. Prior to the study’s commencement, the consent of both the departmental administration and students was obtained (Appendix D & E). Before obtaining this consent, students were made aware of the study’s purpose, structure, and duration as well as its evaluation methods regarding oral presentation and preparation for giving the presentation. Moreover, participants in Group A were enlightened regarding the use of the analytical speaking rubric for evaluating their peers’ speaking performance throughout the term. The following steps were involved in data collection: Step 1: Administration of the foreign language speaking self-efficacy scale (Pre-test) A foreign language speaking self-efficacy scale (pre-test) was administered on September 27th, 2018, to determine students’ levels of self-efficacy toward speaking in English within the department. Step 2: Administration of the first speaking test Participants in both groups undertook the first speaking test on October 3rd, 2018, to demonstrate their oral performances. The oral performances were rated based on three dimensions: delivery, topic development and language use of TOEFL Independent Speaking Rubric. Two raters individually assessed the participants’ performance and the average of their two scores was considered as the final grade. In addition, prior to the treatments, participants from Group A received a training session regarding peer feedback method for a week. The participants from peer feedback group listened to five responses of the same TOEFL independent speaking task and then compare their rating to the TOEFL speaking grading report. Step 3: Administration of the first treatment (Oral Presentation) in the A1 Module Following the training session for Group A, both groups gave oral presentations on October 11th, 2018. Each student gave a short presentation lasting for approximately 2-3 minutes on a topic assigned by the researcher in accordance to their level and syllabus of the School of Foreign Languages (Appendix F). All the presentations were recorded in case of grading adjustment. Participants from Group A (peer feedback) were divided into groups of three or four based on random grouping to assess their peers. Following the presentation, participants in Group A in accordance with their groups were asked to give oral feedback and comment about their peers’ performance by completing a peer evaluation form (Appendix G). On the other hand, participants from Group B were assessed by the researcher with the use of TOEFL independent speaking rubrics (Appendix C) and a teacher evaluation form (Appendix H), who provided immediate teacher feedback. No teacher feedback was given to Group A participants. At the end of the presentations, the researcher collected the peer evaluation forms. All presentations were recorded in case of different scoring of two raters. The participants of both groups were then given a new topic for the next presentation for the second treatment, which was based on relevant materials from their course syllabi. All participants were given a week to prepare their presentations. International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1459 Step 4: Administration of the second treatment (Oral Presentation) in the A1 Module Following the initial treatment, the participants were given one week to prepare for the second presentation based on feedback from the first session. Their second oral presentations were given on October 22nd, 2018, with conditions identical to those of the first presentation. All presentations were again recorded and evaluated utilizing the same feedback methods (peer feedback in Group A and teacher feedback in Group B) and instruments employed during the first treatment. Step 5: Administration of the third treatment (Oral Presentation) in the A1 Module The participants of both groups were asked to prepare for their final oral presentation, which would take place on November 2nd, 2018. They were again given one week to prepare based on feedback from the first and second treatments, However, unlike with the previous treatments, participants themselves chose their speaking topics by selecting one out of three topics offered by the researchers (Appendix F). Again, all presentations were recorded and evaluated utilizing the same feedback methods and instruments as those of the first two treatments. Step 6: Administration of the second speaking test at the end of the A1 Module Following the three oral presentations, the same speaking test was administered at the end of the A1 module on November12th, 2018. The participants were asked to talk about a holiday which was the same topic as in their first speaking test. Again, two raters judged the learners’ oral performances individually with the use of TOEFL Independent Speaking Rubric. The average of their mean scores was considered as the final grade. The test results were used to determine whether there had been steady improvement in students’ speaking performance. Step 7: Repeated treatments in the A2 Module and the third speaking test at the end of the A2 Module During the A2 module, the three treatments of the A1 module (oral presentations with feedback) were repeated, as was the speaking test component. Participants were asked to give three oral presentations accompanied by a teacher and peer feedback. The presentation topics were designed by the researcher according to the syllabus of foreign language departments and the level (Appendix F). The repeated treatments were conducted on December 3rd, 17th, and 31st, while the A2 speaking test was conducted at the end of the module on January 10th, 2019. Step 8: Repeated treatments in the B1 Module During the B1 module, the three treatments and a speaking test were repeated. Participants were asked to make three oral presentations accompanied by teacher and peer feedback. The presentation topics were designed by the researcher according to the syllabus of the foreign languages department and the level (Appendix F). The repeated treatments were conducted on February 26th, March 7th, March 19th, 2019. Step 9: Administration of the fourth speaking test at the end of the B1 Module At the end of the B1 module, participants in both groups undertook the final speaking test. The B1 speaking test was conducted at the end of the B1 module on March 24th, 2019. The results of the final speaking test were used to assess the improvement of speaking performances comparing to the beginning of the study. Step 10: Facilitation of the same foreign language speaking self-efficacy scale (Post-test) Au & Bardakçı 1460 At the end of the study, participants in both groups were given the same foreign language speaking self-efficacy scale (post-test). The results of the foreign language speaking self- efficacy scale were used to determine whether there had been any significant differences in pre-and post-test performance. 3.5. Data Analysis Quantitative methods were employed for analyzing the data collected from the four speaking tests, and the foreign language speaking self-efficacy scale. Both descriptive and inferential statistics were utilized. The inferential statistics were computed via SPSS Version 21.0 software, and the means, as well as standard deviations, were calculated. Parametric tests were used in this study as the data in this study followed a normal distribution. In addition, parametric tests were employed here as they have more statistical power and are likely to detect a significant effect. A number of independent-samples t-tests were applied to compare the results between two groups, and paired-samples t-tests were used to determine whether any differences within groups in the pre-and post-test scores were statistically significant. A repeated measure analysis of variance (ANOVA) was utilized to compare the oral performances of peer assessment with rubrics and teacher immediate feedback profiles. In addition, a Pearson Bivariate Correlation was conducted to see whether there was any correlation between the two variables. 4.Research Findings 4.1. Results for Research Question #1 Are there any changes in the level of students in speaking self-efficacy? Research question 1 intended to examine the self-efficacy level of students at the School of Foreign Languages at Gaziantep University before and following treatment. The speaking self-efficacy scale was analyzed twice, one in the beginning and one at the end of the study. Table 1 displays the mean and standard deviation of the data collected from the foreign language speaking self-efficacy scale. Means of responses of all participants were calculated. Table 1. Descriptive Statistics and Reliability for Foreign Languages Speaking Self-Efficacy Scale N Mean SD Minimum Maximum Range Alpha pre-test 33 61,5758 17,12095 24 103 79 ,943 post-test 33 88,3030 15,85774 53 120 67 ,944 Group A pre-test 15 65,4667 15,77007 43 103 Group A post-test 15 95,9333 14,92106 71 120 Group B pre-test 18 58,3333 17,95747 24 92 Group B post-test 18 81,9444 13,99538 53 103 The pre-test means score of the whole sample was 61.58 and the standard deviation was 17.12. The Foreign Languages Speaking Self-Efficacy Scale (pre-test) was proved to be reliable with the Cronbach’s Alpha value of .943. The mid-point of the scale is 60 which is the cut-point separating low and high efficacious students. The results revealed that students International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1461 in the School of Foreign Languages possess a high speaking self-efficacy level in the pre-test. A post-test was conducted with the same foreign language speaking self-efficacy scale at the end of the year after all treatments to examine the changes in the level of participants’ speaking self-efficacy. Table 1 indicates the post-test mean score was 88.30, and the standard deviation was 15.86. The range was 67 with a minimum of 53 and a maximum of 120. The results revealed that students in the School of Foreign Languages also possess a high speaking self-efficacy level in the post-test. The Foreign Languages Speaking Self-Efficacy Scale (post-test) was proved to be reliable with the Cronbach’s Alpha value of .944. Comparing the pre-test and post-test total mean scores, an increase from 61.58 to 88.30 in students’ speaking self-efficacy level was observed. The results of a paired-samples t-test was analyzed to reveal the difference in the levels of speaking self-efficacy between the participants of within group A and group B. Table 2. Summary for Paired-Samples t-test for Speaking Self-Efficacy Scale Sources of variance N Mean Difference SD T df R square change P Group A (pre & post- test) 15 -30.46667 10.6962 11.032 14 .758 .000 Within Groups Group B (pre & post- test) 18 -23.61111 15.24752 6.5700 17 .569 .000 Table 1 and 2 displays the mean scores of group A (peer feedback group) increase from 65.47 (SD=15.77) in pre-test to 95.93 (SD=14.92) in post-test with a correlation of .758. The difference in mean in group A was -30.47 (SD=10.70), t(14)=11.032, p=.000. The mean scores of group B (teacher feedback group) increase from 58.33(SD=17.96) in pre-test to 81.94 (SD=13.99) in post-test with a correlation of .569. The mean difference was -23.61 (SD=15.25), t(17)= 15.24752, p=.000. The analysis from Table 1 and Table 2 confirm that group A has a higher speaking self-efficacy level than that of group B in post-test. Moreover, according to the table, there are significant differences within students’ mean scores in both groups on the Foreign Languages Speaking Self-Efficacy (p<.001). The results suggested that the difference between the two scores is statistically significant. The result of another independent-samples t-test (two-tailed) was analyzed for identifying the possible means differences between group A and group B of the same speaking self- efficacy scale and the significant changes between pre-test and post-test. Au & Bardakçı 1462 Table 3. Summary of Independent-samples t-test for Speaking Self-Efficacy Scale Source of Variance Mean Difference SD difference F T df P Between Pre- test 7.13333 5.94482 1.200 1.200 31 .239 Groups Post test 13.98889 5.04122 2.775 2.775 31 .009 Table 3 indicates there was no significant difference found between group A (M=65.47, SD=15.77) and group B (M= 58.33, SD= 17.96) in the pre-test, t (31) =1.200, p=.239. While, for post-test, there was a statistical difference in the mean of the scores in post-test between group A (M=95.93, SD=14.92) and group B (M=81.94, SD= 13.99), t (31) =2.775, P=.009. 4.2. Results for Second Research Question #2 Is there a statistically significant difference in terms of oral performances after treatment? This research question determined the students’ oral performances after receiving peer and teacher feedback. The speaking test was analyzed four times at the beginning of A1 and end of A1, A2 and B1 module. Table 4 represents the results of a repeated measures ANOVA test of speaking tests in different modules. Table 4. Descriptive Statistics for four speaking tests Test Group N Mean SD Speaking Test 1 A 15 3.8667 1.06010 B 18 2.9444 1.10997 Total 33 3.3636 1.16775 Speaking Test 2 A 15 4.2667 1.38701 B 18 3.4444 .70479 Total 33 3.8182 1.13067 Speaking Test3 A 15 8.6667 1.63299 B 18 7.1389 2.11302 Total 33 7.8333 2.03357 Speaking Test 4 A 15 10.1000 1.47842 B 18 8.4444 1.70543 Total 33 9.1970 1.78946 For the first speaking test, the mean scores were 3.87 (SD=1.06) in group A and 2.94 (SD=1.11) in group B. The mean scores of the second speaking test were 4.27 (SD=1.39) in group A and 3.44 (SD=.70) in group B. For the third and fourth speaking test, the mean scores of group A and group B were 8.67 (SD=1.63), 7.14 (SD=2.11) and 10.10 (SD=1.48), 8.44 (SD= 1.71) respectively. The results show that there was a steady increase in both groups. International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1463 The results of the repeated measures ANOVA test for four speaking tests were given in Table 5: Table 5. Summary of repeated measures ANOVA test for four speaking tests. Group N Mean Difference SD T Df R P Source of Variance A 15 -6.23333 1.27988 18.862 14 .533 .000 (Within Groups) B 18 -5.50000 1.85504 12.579 17 .185 .000 As Table 5 illustrates, the mean scores for speaking pre-test of group A (peer feedback group) increase from 3.87 (SD=1.06) in pre-test to 10.10 (SD=1.48) in post-test with a correlation of.533. The difference in mean scores of speaking pre-test in group A was -6.23 (SD=1.28). The mean scores of group B (teacher feedback group) increase from 2.94 (SD=1.11) in pre-test to 8.44 (SD=1.71) in post-test with a correlation of .158. The mean difference was -5.50 (SD=1.86). Comparing the first speaking test score and the final speaking test score, those who participated in the peer assessment made gains of about 161.21% in the post-test mean scores. Those who received teacher feedback improved by 212.36%. The result analysis confirms that group B with teacher feedback has greater improvement in speaking proficiency than that of group A in post-test. Moreover, according to the table, there are significant differences between the two profiles students’ mean scores on the speaking tests (p<.001). The results suggested that the difference between two scores within Group A (M=-6.23, SD = 1.28) and Group B (M=-5.50, SD=1.86) are statistically significant (p<.001). The results of another independent-samples t-test were analyzed to identify the mean scores of each speaking test and to examine the differences between groups in oral performances. Table 6. Summary of Independent-samples t-test for four speaking tests Source of Variance Test Mean Difference SD difference F T Df P Between 1 .92222 .38027 .340 2.425 31 .021 Groups 2 .82222 .37347 8.905 2.202 31 .035 3 1.52778 .66817 .035 2.287 31 .029 4 1.65556 .56177 .949 2.947 31 .006 The results in Table 6 indicated there was a statistical difference in the mean of the scores between group A (M=3.87, SD=1.06) and group B (M=2.94, SD=1.11) in the first speaking test, t (31) =2.425, p=.021. There was also a statistical difference in the mean scores of the second speaking test between group A (M=4.27, SD=1.39) and group B (M=3.44, SD=.70), t (31) =2.202, P=.035. While for the third speaking test, there was also a significant difference in the mean scores between group A (M=8.67, SD=1.63) and group B (M= 7.14, SD=2.11), t (31) =2.287, P=.029. For the final speaking test, a significant difference in the mean scores Au & Bardakçı 1464 between group A (M=10.10, SD=1.48) and group B (M=8.44, SD= 1.71) was found, t (31) =2.947, P=.006. Overall, there is a statistically significant result in oral performances in speaking tests (p<.05). 4.3. Results for Third Research Question #3 Is there a correlation between reported speaking self-efficacy level and oral performances following feedback? This research question aimed to examine the possible relationship between the students’ speaking self-efficacy beliefs and their performances in speaking tests. Table 7. Correlation between the scores on speaking self-efficacy and oral performances Speaking performance Speaking self- efficacy Speaking performance Pearson Correlation 1 .235 Sig. (2-tailed) .189 Speaking self-efficacy Pearson Correlation .235 1 Sig. (2-tailed) .189 The results of a Pearson Bivariate Correlation in Table 7 revealed that there is not a significant relationship between participants’ speaking self-efficacy level and their oral performance (r=.235, n=33, p=.189). Thus, there is not sufficient evidence to state that this correlation exists in the population. 5. Conclusion and Discussion The primary purpose of this study was the exploration of the relationship between speaking self-efficacy level of learners and the effects of oral presentation after receiving teacher and peer feedback at the School of Foreign Languages of Gaziantep University. The results of the Foreign Language Speaking Self-Efficacy Scale revealed that students at the School of Foreign Languages of Gaziantep University possessed a high level of speaking self-efficacy level. The study implies that feedback is essential in developing a learner’s speaking self-efficacy level, which will enrich their positive learning experiences and will develop their speaking skill. The findings also suggest that enhanced speaking self-efficacy can encourage learners in their speaking process and feel efficaciously to perform better. These results were echoed by similar findings in Asakereh, and Dehghnnezhad (2015), which identified a positive relationship between speaking skills achievement and speaking self- efficacy levels. They observed that higher speaking self-efficacious learners are more likely to perform better in speaking. Similarly, a study conducted by Leeming’s study (2017) also revealed that students grew in speaking self-efficacy over the course despite the growth rate varies when they were given a chance to practice. The findings of this study also contributed a clearer understanding of vicarious experiences proposed by Bandura (1997) which suggests that positive experience help to enhance greater self-efficacy, and as a result leading to greater effort and more positive learning experiences. Strengthened self-efficacy can also lead to higher motivation, persistence, and their feelings of self-confidence (Bandura, 1984). Mills (2014) suggests that developing the self-efficacy beliefs of language learners can help International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1465 them feel more competent and capable in their ability to complete a learning task. Ultimately, speaking self-efficacy beliefs was a strong predictor of speaking skill achievement. Secondly, the results of a repeated measures ANOVA test and an independent-samples t- test have also been employed to see the oral performances between and within groups after oral treatments are different. The findings showed that learners in both groups had demonstrated a significant improvement in their oral performances. In addition, students who received teacher feedback have demonstrated more significant improvement than those who received peer assessment. The findings also suggest that a higher number of oral treatments provide learners plenty of opportunities to glimpse at their performances over time and hence develop and enhance their speaking ability. Obtaining feedback also gives learners the opportunity to identify their own mistakes from the teacher or peers’ perspective and hence to regulate ways to approach the task or complete the tasks accordingly and effectively. Feedback can be used as a strategy to achieve learning goals. Considering the results in the first speaking test and the final speaking test, students in peer feedback group improved by 161.21% while students in teacher feedback group improved by 212.36%. These results suggested that teacher feedback group had more significant improvement than that of peer feedback group. These findings were also echoed by those of Murillo-Zamorano and Montanero (2017), who observed that both peer assessment and teacher feedback improved students’ oral performance, however, unlike the findings of our study, students received peer feedback improved more than those who received teacher feedback in the post-test. Similarly, the result of our study was also similar to that of Konold, Miller and Konold (2004). They have found that teacher feedback has enhanced learning and students’ performances and helped learners to focus on what had to be done for improvement. Finally, the result of a Pearson Bivariate correlation analysis was analyzed to determine the existence of a relationship between speaking self-efficacy level and the effects of different feedback on oral performances. The findings showed that there was no relationship between the two variables. The results of this study could not represent the results of all universities in Turkey as the collected data can be different if participants had been selected from different universities. English programs, prep-school syllabi, and background information may cause significant changes in the results. This research was significant, firstly, because it reveals a difference in the effect of feedback type on speaking self-efficacy. Secondly, no similar study concerning speaking self- efficacy level and the effect of peer and teacher feedback on oral performances have been conducted in higher education in the Turkish context. More studies can be conducted to have a better and clear picture of the relationship between these two variables among EFL learners. In addition, no similar studies have been conducted in the EFL context, thus this study can fill in the gap in the EFL field. Future studies may also aid educators in determining the types of feedback form affecting learners and enable them to speak more accurately and fluently. Future studies may also aid educators with the ways of enhancing learners’ beliefs in their speaking abilities. Yet, there are a number of limitations in this study. First, the generalizability of the results is limited by the number of participants (N=33). The number of participants is too small, and hence it limited the scope of sphericity. A larger number of students is needed to prove the compatibility of this study. Future studies might employ a larger scale of the sample consisting of preparatory schools (both public and private universities) in different parts of Turkey for better and more accurate results. Students of different departments can be chosen to determine whether there is a difference in the level of speaking self-efficacy and oral performances. Secondly, only two experimental groups are involved. The reliability of data is Au & Bardakçı 1466 impacted by the lack of a control group. Also, the participants were chosen based on convenience sampling, and this might have had an impact on the results. The sample size was small, with only thirty-three participants; moreover, it was a convenience sample with two experimental groups and, hence cannot be a generalization for the School of Foreign Languages. This might have had slight effects on the current results. In future studies, the sampling procedure could be developed by employing a random sampling method instead of convenience sampling. Another limitation was the time of the study. The study was carried out only for three modules which might have limited the time for students to develop their speaking skills. In addition, the training time for peer feedback group in how to give peer feedback was short in which might have affected the way participants gave peer feedback. In this case, participants may not have exerted much effort in giving peer feedback accurately and fairly. Had it been extended over a longer period of time, more statistically significant results might have been obtained regarding differences and improvements in students’ oral performances. In light of research findings, the following recommendations and suggestions should be considered: EFL teachers should direct the teaching and learning processes to strengthen learners’ efficacy on the subject (Bandura, 1984, 1997). They should employ various speaking activities such as role-play, dialogue practices and projects in class to enhance students’ confidence, fluency, and accuracy in their speaking. Teachers and peers can come together to give feedback after speaking activities. Using praises and giving constructive feedback can help to increase students’ confidence and enrich the positive learning experience. In this way, learners possess positive attitudes in speaking and perform better in daily conversation as well as speaking tests and exams. Moreover, it is impressive to observe students’ significant improvement in oral performances within such a short period of time. Students were more aware of their types of errors they had made in order to avoid repeating them on future presentation and, thus, improve their oral performance. This situation may suggest that more opportunities should be given to students to enhance both speaking self- efficacy and oral performances. A higher number and frequency of oral treatments may be able to improve students’ oral performances within a shorter period of time. For the feedback, teachers should give more detailed and comprehensive feedback to the students. Teachers can point out general speaking errors by students in class and ask students to exchange their ideas. To enhance the effectiveness of peer feedback, more careful training and structuring are required. Teachers should explain peer feedback expectations to the learners as well as guide them throughout the process in order to minimize potential consistencies associated with subjectivity. Moreover, peer feedback group should be trained adequately and should be encouraged to give feedback with fair judgments. Peers can write down the errors on board and make others aware of and spot out the mistakes they have made during presentations. These procedures not only inform students of their abilities and progress in learning but also motivate them to perform well continuously and make them challenge with each other. Also, had it been extended over a longer period of time, more statistically significant results might have been obtained regarding differences and improvements in students’ oral performances. 6. Conflict of Interest The authors declare that there is no conflict of interest. 7. Ethics Committee Approval The authors confirm that the study does not need ethics committee approval according to the research integrity rules in their country. International Online Journal of Education and Teaching (IOJET) 2020, 7(4), 1453-1468. 1467 References Alawiyah, T. (2018). Speaking self-efficacy and EFL student teachers’ speaking achievement. Journal Pendidikan dan Pengajaran, 5(1), 87-96. Asakereh, A., & Dehghannezhad, M. (2015). Student satisfaction with EFL speaking classes: relating speaking self-efficacy and skills achievement. Issues in Educational Research, 25(4), 345-363. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review. 84(2), 191-215. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1997). Self-efficacy: The exercise of control. New York. Freeman. Bailey, K. M., & Savage, L. (1994). New ways in teaching speaking. Alexandria, VA. Teachers of English to Speakers of Other Languages. Cinkara, E., & Tılfarlıoğlu, F. Y. (2009). Self-efficacy in EFL: Differences among proficiency groups and relationships with success. Novitas-ROYAL, 3(2), 129-142. Dasmo, D., & Sundari, H. (2014). The effect of speaking self-efficacy and gender in speaking activities. Bahasa & Sastra, 14(2), 205-217. Gürsoy, E., & Karaca, N. (2018). The effect of speaking anxiety on speaking self-efficacy of children in EFF context. International Journal of Language Academy, 6(3), 194-210. Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research. 77(1), 81-112. Konold, K. E., Millers, S. P., & Konold, K. B. (2004). Using teacher feedback to enhance student learning. Teaching Exceptional Children. 36(6), 64-69. Leeming, P. (2017). A longitudinal investigation into English speaking self-efficacy in a Japanese language classroom. Asia-Pacific Journal of Second and Foreign Language Education, 2(12), 7-35. Liu, M. X. (2013). English bar as a venue to boost students’ speaking self-efficacy at the tertiary level. English Language and Teaching, 6(12), 27-37 Lynch, T., & Maclean, J. (2003). Effects of feedback on performance: A study of advanced learners on an ESP speaking course. Edinburgh Working Papers In Applied Linguistics, 12. Marzano, R. J., Pickering, D. J., & Pollock, J. E. (2001). Classroom instruction that works: research-based strategies for increasing student achievement. ASCD, 187. Mills, N. (2014). Self-efficacy in second language acquisition. In S. Mercer & M. Williams (Eds.), Multiple perspectives on the self in SLA (p.6-19). Bristol: Multilingual Matters. Murillo-Zamorano, L.R. & Montanero, M. (2017). Oral presentations in Higher Education: a comparison of the impact of peer and teacher feedback. Routledge,3, 79-87. Ocak, G., & Olur, B. (2018). The scale development study on foreign language speaking self- efficacy perception. European Journal of Foreign Language Teaching, 3(1), 50-60. Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66(4), 543-578. Pengajaran, E. P. D. (2018). Speaking self-efficacy and EFL student teachers’ speaking achievement. Research Gates, 5(1), 87-96. Sadler, D. R. (1989). Formative assessment and the design of instructional systems. Instructional Science, 18, 119-144. Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psychologist, 26, 207-231. Au & Bardakçı 1468 Shepard, L. A., Flexer, R. J., Hiebert, E. J., Marion, S.F., Mayfield, V., & Weston, T. J. (1996). Effects of introducing classroom performance assessments on student learning. Educational Measurement Issues and Practice, 15, 7-18. Smith, C., & King, P. (2004). Student feedback sensitivity and the efficacy of feedback Iinterventions in public speaking performance improvement. Communication Education, 53(3), 203-216. Yanar, B. H., & Bümen, N. T. (2012). İngilizce ile ilgili öz yeterlilik inancı ölçeğinin geliştirilmesi. Kastamonu Eğitim Dergisi, 20(1), 97-110. Yang, K. H. & Wu, Y. H. (2013). Effects of feedback types on the student’s self-efficacy. International Journal of e-Education, 3(3), 202-205 Zarei, A. A., & Sepehri, S. (2018). Speaking self-efficacy: Through audio-journals. Scholars’ Press. Zimmerman, B.J. (2000). Self-efficacy: An essential motive to learn. Contemporary Educational Psychology, 25, 82-91.