Corrective Feedback In Eap Speaking Class

Research on Oral Corrective Feedback, especially related to its factors, has become a topic that is still widely discussed in teaching English because various application contexts will produce different conclusions. That is why it is important to study this topic further where this research aims to test a group by applying a single pretest-posttest method by providing oral Corrective Feedback. The participants in this study were a lecturer and 27 students in a Speaking English for Academic Purposes (ESP) class at the State Islamic University in Malang, East Java. The results showed an increase in the mean score from the pretest to the posttest. This means that giving Corrective Feedback provides benefits during the learning process for students in improving. Several factors in Corrective Feedback can explain how students respond to the Corrective Feedback given so that Corrective Feedback can play a significant role in improving the competence of speaking English. It is hoped that by understanding the use of Corrective Feedback, teachers, lecturers and educational practitioners will better understand which types of Corrective Feedback can have a positive impact on improving English speaking competence for English learners as a foreign language.


INTRODUCTION
Abundant studies have discussed the significance of oral corrective feedback to language learning (De Vasconcelos Neto & De Barros Cardoso, 2021;Lochtman, 2002;Mackey et al., 2000;Valezy & Spada, 2006).Although research in oral corrective feedback has been conducted in both laboratory and classroom-based setting, the classroom-based setting is proven to provide more actual evidence.Classroom-based studies are also propitious for acquiring a better understanding of the classroom spoken interaction where learners interact with only one competent target language speaker, the teacher (Spada & Lightbown, 2008).Corrective feedback is considered as "an inherent part of classroom practices in which teachers engage to achieve instructional objectives that include consolidation of learners' L2 knowledge" (Lyster et al., 2013).When corrective feedback is observed in a classroom context, the tendency for finding strong evidence on how the teacher modified the input and on how the feedback can help the learner is higher.
Corrective feedback has both positive and negative impact.However, "corrective feedback does make a great effect on oral accuracy, but the effectiveness for a different level of learner is different" (Chu, 2011).A reserach had further studied how EFL learners emotionally respond to teachers' oral corrective feedback, and it was found that the students demanded error correction because they find it very helpful to help them know how to repair their errors (Martínez-Agudo, 2013).Positive response about error correction is also shown in the study conducted in 2011 whch proved that despite the side effects of corrective feedback (e.g.embarrassment, demotivation), learners appreciated opportunities to receive verbal correction for accuracy enhancement in their English proficiency.It also showed that more proficient students get less demotivated when corrected by their teachers since they have more confidence and more experience in language learning and vice versa.Consequently, more steps in giving feedback must be done for error correction, targeting less proficient learners (Chu, 2011).
Oral corrective feedback is a crucial aspect to be considered when designing learning activities in a speaking class.One of the principles in designing a speaking technique is to supply proper feedback and correction (Brown, 2000).Feedback is necessary, especially in EFL (English as a Foreign Language) class since in most EFL situations, the teacher is the only source that can provide linguistic feedback for students.The teacher is also the proficient language speaker in the classroom that can provide suitable correction at the right moment.Thus, corrective feedback is strongly linked to the learning of speaking in both general English and English for specific purposes (ESP) in the EFL context.
In ESP classes especially for academic purposes, the distinguishing feature of English for academic purposes (EAP) is that it deals with a subject-specific matter that covers knowledge which the average educated native speaker could not reasonably be expected to be familiar with (Tenieshvili, 2019).The goal should be to teach the language that makes the learner can communicate effectively in their work or study areas (Basturkmen, 2010).Therefore, teaching speaking in ESP which focused on academic purposes should consider the students' need on a specific study area.Further, the role of corrective feedback in the EAP context is still as significant as in general English because the speaking micro and macro skills are in the same scope.
There are many attributes related to corrective feedback that can be explored.Most importantly, investigating corrective feedback in speaking classroom need the teacher's ability to identify whether the correction is expressed gently or assertively, supportively or as a condemnation, tactfully or rudely (Chu, 2011;Nassaji & Kartchava, 2017;Sato & Ballinger, 2012;Sheen, 2004;Wang & Li, 2021).Those characteristics of delivering corrective feedback can be used to unveil more profound evidence on the use of feedback type, without solely relying on the type of feedback.Most research on feedback focuses on the feedback type, and only some describe the way the teacher expresses the feedback, which in fact can affect students' response.
In a communicative classroom, the use of intensive corrective feedback is still debatable.
Although communicative practice in context can benefit the students by emphasizing on expressing meaning and understanding, some situations require form-focused corrective feedback (Brown, 2000;Fairbairn & Brown, 2005;Han, 2002).This statement also means that all types of feedback are needed even in the communicative classroom context, but the teacher still needs to decide when to give feedback that can benefit the students.
Still, cognitively to socially-oriented theoretical perspectives suggest that corrective feedback is necessary for moving learners forward in their target language (L2) development and is beneficial for L2 learning (Lyster et al., 2013) .Lyster et al. (2013) also added that in the framework of the cognitive theory, which was popular in the 90s, corrective feedback was seen as facilitative of L2 development by the interactionist model.In line with the interactionist model, the noticing hypothesis also views corrective feedback as valuable to support L2 learning and emphasizes that conscious attention to notice input was essential to proceed L2 learning (Ellis, 2017;Panova & Lyster, 2002).
Moreover, from the perspective of the socio-cultural theory, language learning is not the result of the interaction, but it is a dialogical process in which acquisition occurs in interaction.In this point of view, L2 acquisition is considered as a process in which the learners interact with others.As a result of an interaction, corrective feedback provides learners with assistance.At the same time, they develop from other-regulation to self-regulation which can be applied by providing gradual scaffolding (more implicit corrective feedback over time) as the learners gain more control on the L2 (Sato & Ballinger, 2012).In sum, learners need more proficient target language speaker to help them notice their errors and to correct the errors, and when in the classroom, a teacher must take that role.
Following those theoretical perspectives, corrective feedback used in the contextual language is suggested to be beneficial for learners since oral language production is similar to real-life language use.The support of corrective feedback effectiveness when provided within the context of meaningful and sustained communicative interaction had been proposed by many researchers (Lyster et al., 2013;Mackey, 2002;Sheen, 2004;Spada & Lightbown, 2008).However, debates between these theories and the pedagogical applications are still prevalent among second language researchers, teachers, and educators.
Further, studies proposing types of feedback have also been discussed over time, especially in dealing with some possible errors the learners may produce in a classroom situation.Sheen & Ellis had suggested a taxonomy of oral corrective feedback strategies based on whether it is reformulation or prompt and whether it is implicit or explicit (Lyster et al., 2013).In this distinction, some corrective feedbacks provide both negative and positive evidence such as explicit correction and recast, whereas prompts are seen to provide only negative evidence (Ellis, 2017).Similar categorization of corrective feedback was also proposed by in a reserach where six different corrective feedback types are divided in two broad categories; reformulations and prompts (Lyster & Ranta, 1997).Reformulations include recasts and explicit correction because of both supply learners with target reformulations of their non-target output.Meanwhile prompts include a variety of signals other than reformulations that push learners to self-repair.
Although the distinction of implicit and explicit corrective feedback was seen as pivotal, research that draws on skill-acquisition theory seemed to focus less on the corrective feedback implicitness or explicitness, and it seemed to emphasize more on the opportunities of speaking skill consolidation that can be facilitated by corrective feedback through contextualized practice.
Most used taxonomy of corrective feedback strategies is proposed by Sheen and Ellis (Ellis, 2017;Sheen, 2004).It includes nine corrective feedback types which were categorized based on whether it is implicit or explicit and whether it is input-providing or output-prompting.Other experts have proposed their taxonomy of oral corrective feedback but within the same notion.
The debates on which corrective feedback is more useful for the students will continue.
However, there is still some research in the context of ESP specifically in educational institition classroom where students are young adults but have beginner to intermediate oral English language skill.The need for oral corrective feedback from one group of students to another is, of course, different considering their background.Therefore, this study aims to test the following general hypothesis: there is a difference in the students' pretest and posttest speaking performance in ESP speaking classroom

Research design
This study employed One-Group Pretest-Posttest design by presenting how corrective feedback could affect students' speaking performance.
Two identical tasks were given as pretest and posttest.The task was adapted from IELTS speaking practice in Cambridge test 4-1 and ETS 2002 narrative task, and several changes were applied to suit the tasks for the target students.The tasks were chosen since they were suitable to be used for level check.The level check speaking test aimed to stimulate the test-takers "to respond using expected or predicted forms and formations" (Brown, 2004).The task was used to elicit grammar categories (tense and subject-verb agreement), the discourse structure (vocabulary, coherence, cohesion) and the task achievement (the context).
The scoring scale was used to assess the students' responses on the task and was adapted from the IELTS Band descriptor since it was seen as more suitable for classroom assessment purposes (Luoma, 2004).Before the task was used, a try out was administered to a small number of students by selecting students based on the criteria of a low, moderate, and high achiever.The selection was based on the students' score of the previous semester EAP class.After trying out, the recording taken during the try out was used for task validation using a checklist adapted from O'Sullivan, Weir, and Saville (2002).They developed the checklist to validate speaking task, and they stated that the checklists have "great potential as an evaluative tool and can provide comprehensive insight into various issues" (O'Sullivan, Weir, and Saville, 2002).Based on the validation process, some wordings were revised, but no item was deleted.Finally, the task was used to collect data during pretest and posttest.
An observation checklist was also used to explain the findings.This instrument was complementary to the main instrument that was taken from Communicative Orientation Language Teaching (COLT) observation scheme (Spada and Fröhlich, 1995).The COLT observation scheme Part B was adapted in this study to investigate the classroom communicative features in addition to the investigation of corrective feedback types.Use more L2 The extent to which the information requested and/or exchanged is unpredictable.
The information exchange should be genuine, relatively predictable, and relatively unpredictable.The extent to which the speaker engages in extended discourse or restrict their utterances to a minimal length of one sentence.

Minimal and sustained speech
A correction or other explicit statement which draws attention to the linguistic form of an utterance Correction focus more on-message than form but form is pointed out when necessary

Incorporation of preceding utterances
Correction is available, and types of correction cover: repetition, paraphrase, comment, expansion, clarification request, elaboration request.

Aspect Preferred Criteria Target language
Use more L2 The extent to which the information requested and/or exchanged is unpredictable.
The information exchange should be genuine, relatively predictable, and relatively unpredictable.The extent to which the speaker engages in extended discourse or restrict their utterances to a minimal length of one sentence.

Minimal and sustained speech
A correction or other explicit statement which draws attention to the linguistic form of an utterance Correction focus more on-message than form but form is pointed out when necessary

Incorporation of preceding utterances
Correction is available, and types of correction cover: repetition, paraphrase, comment, expansion, clarification request, elaboration request.

Discourse initiation More initiations are preferred Relative restriction of a linguistic form
The language form preferred is those with unrestricted form rather than limited restriction.It was used in the preliminary study and during data collection.In the data collection, this checklist was used to give additional information, while in the preliminary study, the instrument was used to select the classroom participant from three available classes which were considered as participant candidates.
The data collected on this study were categorized into seven types of corrective feedback that were applied during the treatment, as shown in the table below.

Repetition Clarification requests
Elicitation Paralinguistic signal The communicative classroom was set into activities that required students to produce more genuine utterances, and the focus of collecting the data was on the corrective feedback rather than on the learning materials and methodology.

Research site and participants
The class chosen was EAP class from an Islamic State University, and it was seen as a sample that could yield evidence on the use of oral corrective feedback based on the preliminary study result.
The learners included in the sample ranged between 18-20 years old and had various educational background and included both male and female students.Their speaking proficiency is on B1 level according to the Common European Framework of Reference (CEFR).

Data collection and analysis
The primary data for this study were the scores for speaking performance which were gathered through pretest and posttest.The score was calculated based on the scoring criteria, and the mean scores for pretest and posttest were calculated to see the improvement of students' speaking performance.The mean scores were presented in the chart and descriptive statistic to make it easier to understand.In investigating the result of this study, the null hypothesis and alternate hypothesis were formulated.

Null hypothesis:
There is no significant difference in students' pretest and posttest score.

Alternate hypothesis:
The students had better posttest speaking performance than their pretest speaking performance.
The rejection and acceptance of the hypothesis were measured at .05 level of significance.
The p-value was compared to the critical value .05,and when the p-value was lower than .05(p ≤ .05), the null hypothesis was rejected, and the alternate hypothesis was accepted.On the other hand, the p-value higher than .05(p ≥ .05)accepted the null hypothesis and rejected the alternate hypothesis.This calculation was done using one-sample t-test.
The researcher was a non-participant in the classroom but orchestrated what the lecturer needed to be done with the correction by making sure that the lecturer used all seven designated error correction during the treatment.

Results
This study aimed to reveal how corrective feedback affects students' speaking performance.In measuring the progress of students' speaking performance, pretest and posttest were conducted.Before 12 hours of lessons which were divided into six meetings of two hours of lessons where the treatment was given, a pretest was administered to a total of 27 students.The posttest then was done to measure students' speaking performance after the corrective feedback treatment.The scores of both pretest and posttest were used to see how corrective feedback treatment affect students' speaking performance.

Table 3 Pretest-Posttest Descriptive Statistics
The scores were calculated using compare mean.It was found that the minimum scores from pretest and posttest were different, while the maximum score was still the same.However, the mean showed a significant difference from pretest to posttest.These results led to the need for further analysis to provide evidence on the effect of corrective feedback.Thus, hypothesis testing was done.

Figure 1 Pretest & Post-test Mean Score
The score descriptor used for the speaking task here was taken from the IELTS band score, but the score was converted into the EAP program standard scoring system developed by the Framework of Reference (CEFR), most of the students' speaking comprehension was on B1 meaning that.the students were still on beginner independent language user level.Although the pretest result displayed that the students' level was still in B1, their mean score was improved, from 64.86 in pretest to 67.08 in the posttest.This fact implied that corrective feedback treatment indeed influenced students' speaking performance.
Normality test was done to know whether the data had been normally distributed.The test was using Saphiro-Wilk test after the data collection was finished.The results of the normality test were shown in Table 4 Table 4  The aim of testing the hypothesis was to see whether the null-hypothesis was rejected or accepted.H0: There is no significant difference in students' pretest and posttest score.H1: The students had better posttest speaking performance than their pretest speaking performance.
The calculation results were shown in Table 5.  5 showed that p = .011(p ≤ .05).Thus, the null hypothesis was rejected, and the alternate hypothesis was accepted.It implied that there was a significant difference in the pretest and posttest score.
Since the data had shown improvement on the students' speaking performance after treatment -reflecting through a higher posttest score than pretest -the data from observation showed that the lecturer tried to correct most of the students' significant errors while sometimes ignore minor errors.
The observation result revealed that the purpose was to maintain the communication flow during the lesson.Although communication breakdowns were still found, which resulted in errors, the essence of the communicative classroom was still evident.The correction given that might have been chosen based on lecturer's belief were found to affect students in several ways.However, the lecturer still followed the guideline for giving correction that had been designed by the researcher even though several attributes could influence the choice of correction type and its result on the students, the treatment mainly covered all types of correction effect on students' speaking performance.

Classroom context
Speaking assessment results showed a positive effect of corrective feedback, and it was affected by the fact that the classroom was communicative oriented.The evidence was shown by the result of verbal communication context of classroom communicative features analysis that was analyzed using Communicative Orientation Language Teaching (COLT) part B.
Based on the analysis using COLT, it was found that L2 was mostly used in lecturer students interaction in the classroom although the lecturer sometimes used L1 to explain vocabulary meaning or certain concepts which the students could not understand when it was explained using L2 (Excerpt 1).

.. [continue answering S1 question]
The information exchange during the lesson was mostly unpredictable and genuine as the classroom activities were in the form of presentation and discussion.It was also the reason why the turns were generally minimal and sustained.During the learning process, although the linguistic form was mostly unrestricted, it was sometimes still restricted.This meant that there were several specific language forms expected in the interaction, although the focus was not mainly to produce the correct linguistic form because the classroom interaction was mostly in the form of presentation and discussion.Those findings led to communicative classroom context because more chances were given to students to express their genuine idea to interact with both their friends and the lecturer.The researcher did not interfere with the classroom activities and its communicative context, so the Corrective Feedback In Eap Speaking Class, Dewi Wardah Mazidatur Rohmah, Abdul Halim 6341 lecturer could still maintain the communicative orientation of the class while doing the corrective feedback treatment.

Discussion
Before the data collection, a preliminary study showed problems in the correction, which was related to how the lecturer provided correction and how the students responded to the correction.After the treatment, which emphasized on giving corrective feedback and eliciting students' response was given, the posttest score improved.Although the students were still at B1 level based on CEFR, an increase on the posttest where most students could achieve the minimum target score was seen as a success.Sheen (2004) indicated that corrective feedback was beneficial for students speaking, although different educational settings could affect how each type of feedback contributed to students' performance.
Corrective feedback was proven to be useful to help the learners to notice their error.When they could notice their error, they would learn how to self-correct, which was an essential thing for language learning development.This will help students advance in their language acquisition (Lochtman, 2002;Lyster et al., 2013;Lyster & Saito, 2010).
Although corrective feedback was not the sole factor that could elicit students' response and repair on their utterance which might result in the improvement of students' competence, it was clear that students received advantage from error correction.Moreover, although learners could not perceive all of the feedback they received when the feedback was given at the right time of language development, the learners would be able to perceive a limited amount of feedback which was necessary for their language development (Mackey, 2002(Mackey, , 2004;;MacKey, 2006).Since the students in this study showed eagerness in receiving feedback, this might be the main reason why they could perceive some of the feedback, which later affected their speaking performance.
Overall, this research implied that the balance of form-focused instruction in the communicative classroom setting, which provided a chance for students to develop their target language knowledge was pivotal.Although the communicative interaction was in EAP context, the balance was still as important as in general English classroom because students often could not recognize their error which led to difficulties in producing a correct form of language even though there was no disruption on the communicative interaction.However, since learning a language also required learning the language rule, corrective feedback played an important role to help students got more exposure on the correct language form especially in EAP context where the learners were expected to communicate effectively in their study areas (Basturkmen, 2003(Basturkmen, , 2010)).
Thus, this research supported the theories which emphasized the benefits of corrective feedback.The main theories were cognitive theories based on interaction hypothesis (Long, 1996), output hypothesis, and noticing hypothesis.Interaction hypothesis focused on the negotiation of meaning, which resulted in more emphasis on cognitive factors and corrective feedback.This hypothesis was supported by many studies (Panova & Lyster, 2002;Sheen, 2004;Zhao & Ellis, 2020), which explained how interaction, where the interlocutor provided reformulation of utterance and corrective feedback, helped the communication flow to continue and helped the speaker to reformulate his/her utterance.
To develop speaking skill, learners need to receive correction.In most EFL situations, the teacher is the only proficient language model who can give linguistic feedback for the students.The teacher is also seen as the only speaker in the classroom that can provide suitable correction at the right moment, especially in the EAP classroom.The condition where speakers, in a classroom setting as the learner, were required to stretch their language in order to meet communicative goal (Correia, 2016).would make them produce correct output and as a result, they could learn from their output.
This condition was what the output hypothesis claimed as supportive of students' target language development.
Further, in order to produce correct output, the students also needed to "notice" their error which required them to consciously focus on the specific language form that they were exposed as the input.The input-referred to corrective feedback.Although noticing input did not directly affect the acquisition, it was helpful to lead to language knowledge growth (Hohenst et al., 2008).In summary, this study mainly supported the noticing hypothesis and EAP theory proposed by Basturkmen (2010) which led to the conclusion that interlanguage development needed effort from the learner and needs more target language exposure to help them could communicate effectively in the target language.
Even though these could be facilitated by corrective feedback, more consideration on the classroom context and other factors that could affect how students perceive feedback must be done.Therefore, corrective feedback could serve its role as a support for target language development.

Correction in Communicative Speaking Classroom
Communicative classroom context was preferred in the study of corrective feedback (Sheen, 2004).Thus, this study selected a classroom that showed more features of the communicative classroom by using COLT as the main selection criteria.The role of corrective feedback in ESL and EFL context had been proven to be different and more input-providing feedback were used more by the teacher in EFL context, but it led to low numbers of uptake (Sheen, 2004).Considering that uptakes do not reflect students' success in improving their speaking, this study revealed that still, corrective feedback helped increase students' speaking performance.
Providing more chances for students to use the language during speaking class is, of course, very important.However, although communicative practice in context can benefit the students by emphasizing on expressing meaning and understanding, some situations require form-focused corrective feedback (Fairbairn & Brown, 2005).It also means that all types of feedback are needed even in the communicative classroom context, but the teacher still needs to decide when to give feedback that can benefit the students.Deciding the time to give feedback is also correlated to the type of error occur since different error type may need further corrective feedback.
The classroom context in the current study was communicative and corrective feedback was seen as compulsory by both the lecturer and the students.Further, the finding of this study which showed a high number of recast and other input-providing feedback was in line with what Sheen (2004) stated about the high frequency of recast in communicative EFL classroom context.He said that it reflected the emphasis on fluency over accuracy because the teacher could maintain communication flow through recast.The fact that the students could not fully understand some language concepts when explained in L2 did not mean that the class was far from communicative context.It was the design of classroom activities and how the lecturer interacted to the students, which became the most likely reason why the turns during the classroom interaction were minimal and sustained.
The communicative orientation also could encourage students to improve their speaking by creating an ease atmosphere to increase chances to deliver ideas and could lead to more acceptances toward correction.When less effort was needed for students to speak, they did not feel the pressure to produce perfect language, but they did learn from the correction given as the response for their errors.
It is because communicative classroom environment, which allowed the students to express their ideas freely within the context of the lesson led to a better response on correction and supported the students' learning process.
In a communicative classroom, the focus should be more on keeping the communicative environment during the learning process.As had been found, lecturer in this study used more recast to maintain the communication flow and to boost students' motivation in expressing ideas.It also happened in the study done by (Yoshida, 2008), where the two teachers participated in the study preferred to use recasts because recasts are conducive to maintain a 'supportive classroom environment.'The lecturer in this current study also showed a similar attitude on recast as in research conducted by Yoshida (2008).The lecturer mentioned that the use of recast could make time management more effective, even though prompts could be beneficial, considering it could give learners a chance to work out on linguistic problems.

CONCLUSION
Corrective feedback, although seen, can bring drawback in students' speaking such as pressure and shame, again still proven to have more benefit in supporting students' speaking performance.The increase on pretest to posttest score in this study revealed that pointing forms in a communicative speaking classroom are necessary, especially in the EFL context.However, the teacher still needs to know which corrective feedback type should be given for students' error depending on the situation.A further study on when, how, and why particular feedback should be given to certain types of errors should be conducted in the future to help EFL teachers in choosing best practised corrective feedback in the speaking class.
is happen when one person's position take by other person.T: do you mean someone's position is taken by another person?S: position Miss T: is it active or passive sentence?S: [looked confused] mmmm... T: posisi seseorang diambil oleh, atau mengambil oleh?S: diambil oleh Miss T: so, you use passive then.... [explain about passive in brief] However,students-students' interaction mostly used L1 even though they still tried to use L2 since the lecturer monitor them (Excerpt 2).yang bisa dilakukan dengan itu?S2: oh... ya banyak.Jadi misalnya T: [interrupt] English please S2: Yes Miss.Nah, example is you can ....

Table 1
Checklist for Communicative Classroom Preferred Criteria

Table 2
Types of Corrective Feedback for Treatment Table 4 that p = .001and referred to p ≤ .05.It implied that the data were normally distributed.After the normality testing was done, what was left was hypothesis testing.