STUDENTS’ PERCEPTIONS TOWARDS THE EFFECTIVE FEEDBACK PRACTICES IN THE LARGE EFL WRITING CLASS BASED ON STUDENTS’ ENGLISH PROFICIENCY LEVEL

Motivated by my own experience in teaching and learning academic writing in a large class of more than 50 students, I aim to find out the students’ perceptions towards the effective feedback practices in a large EFL writing class in Indonesia. There were 150 students participating by answering the the questionnaire. The data was analyzed based on English proficiency level (the students’ TOEFL scores). The findings showed that based on the students’ English proficiency level, they perceived feedback from their lecturers is effective when it is given in written form, while from their peers, it should be in oral form.


INTRODUCTION
Writing is not only about putting the letters together to form words, then combining them to make sentences, and arranging them to become paragraphs, but it is about choosing the appropriate vocabulary, forming the meaning, and organizing the ideas. According to Sokolik as cited in Eksi (2010) "Writing is the mental work of inventing ideas, thinking about how to express them, and organizing them into statements and paragraphs that will be clearer to the reader" (p. 33). Good writing will be understandable both by the writer and the readers. It is the way to communicate ideas in written form.
As one component of the writing processes, feedback (corrections and comments) is needed to support students' writing development and build their confidence in writing. Specifically, according to Hyland (2003) "feedback helps the writer work out the text's potential and to comprehend the writing context, providing a sense of audience and an understanding of expectations of the communities they are writing for" (p. 177). As English learners who use English as a Foreign Language (EFL), writing means communicating with the readers by making the communication understandable in both ways. Since every language has its own style in writing, the input from the readers is very helpful as a way of communicating the ideas and writing components. Teachers can use feedback as a way of communicating the strength and the weaknesses of their students (Mcgrath, Taylor, & Phycyl, 2011).

THEORETICAL FRAMEWORK
Comparing the feedback from teacher and peers, students prefer to choose teacher feedback as the feedback that they need the most. According to Cresswell (2000) "the advantages of teacher feedback are first, intervention, teacher response to students' questions annotated in the margin of writings, is directed to students' ideas in the developing of essays. Teacher feedback is exactly what student writers need in terms of global content (theme, purpose and readership) and organization (argument structure, main and supporting points). Second, teacher response can be more accurately targeted at the levels of students' language proficiency" (p. 238).
Peer feedback itself is defined by Kroll (2001) as "simply putting students together in groups and then having each students read and react to the strength and weaknesses of each other's papers" (p. 228). To this, research still agreed and argued about peer feedback as the effective practice to be applied in ESL/EFL writing classes. Besides concerning the feedback from teacher and peers, the ways of giving/providing the feedback both from the lecturers or peers also have the very significant difference.
First of all, the feedback can be given in two ways; in oral and written forms. According to Berg and Pilot "written feedback was concentrated mainly on evaluative comments, whereas in the oral feedback students provided arguments and suggestions for text revision" (p. 145-146). After the feedback is known to be effective when it is given in oral or written forms, the lecturers need to know whether their students need to have the feedback directly by showing the errors and the correction (Bitchener, Young, & Cameron, 2005;Ferris & Roberts, 2001). Another thing is whether the feedback is given by showing the errors only without having many corrections (Ellis, 2009).
Besides this, whether the feedback is given in the earlier text or in the last of the final paper, M. Yang et al (2006) indicates about the reason when the feedback is best given by the lecturers/peers as follows: He: if the teacher gives feedback first, the peer would feel the pressure and say nothing for fear of saying something wrong because we all trust the teacher more. But if the peer gives feedback first, he would be much freer to express his opinions. Teacher feedback that comes later could evaluate both the essay and the peer's comments, which, I think, is of great help (p. 194).
As it was debated among the effective feedback for the students, the students in a large class do have more challenges than the students in a standard class. As As Shamim et al. (2007) note, "a large class in a western context might be considered small for both teachers and learners in most teaching-learning contexts in Africa" (p. 12) or even super small in the Ivory Coast (Bamba, 2012).
The study about feedback in the large class is found in the two studies; 1. Miao, Badger, and Zhen (2006), who did the research in a Chinese EFL writing class 2. Lin (2009) who explored how large multi level EFL writing class experiences and interacts with self, peer, and teacher feedback These two studies showed that the research results agreed with the peer feedback for the benefits for the students' writing development but still promote the feedback from teachers for the knowledge of writing. More specific, the students preferred to have the feedback from their teacher directly in the last draft of their writing (Diab, 2005). The reason found in that study is because the students perceived that the feedback from the teacher in the last draft will not result them in being late of submitting the paper/work.

METHODOLOGY
This study used quantitative research design. Purposive sampling method was chosen as the students were in a large class, and had experienced all kinds of feedback practices. There were 150 participants from the undergraduate English study program in one of the private universities in Indonesia. The English proficiency levels of the participants was determined by their TOEFL prediction scores ranging from 200-300, 301-400, 401-500, and 500 and above. More than half (52.67%) of the participants' TOEFL prediction scores ranges from 301 to 400. More than one third (38%) of the participants' TOEFL prediction scores ranges from 401 to 500. There were nine participants (6%) who did not fill out the information about the TOEFL score ranges. The lowest TOEFL prediction scores which ranging from 200 to 300 were possessed by only three participants (2%) while only two participants (1.33%) who hold the highest TOEFL prediction scores of 500 and above.

Figure 1. Pie chart of English proficiency level
This study used closed-ended questions due to the following reasons; firstly, it is easier to make an analysis, statistical count, and comparison among the sample; secondly, it is quicker to be coded and analyzed than open-ended questions (Cohen, Manion, & Morrison, 2011). The questions were divided into two categories: feedback from lecturer and peer. Each lecturer and peer has 13 questions of the kinds of feedback (Written, Oral, Grammar, Spelling, Punctuation, Vocabulary Choice, Style, Organization, Direct, Indirect, Feedback in the first draft, Feedback in the second draft) which the students prefer to get from their peer and teacher. The students were asked to express their opinion to a series of statements by answering the questionnaire using the Likert Scale (Not Helpful at All 1 -2 -3 -4 -5 -6 -7 Very Helpful). The data were analyzed by using the Statistical Package for Social Sciences (SPSS) (Version 20 software).

Perceptions about the effective feedback practices based on English proficiency level.
Based on the English proficiency level, the students from four different TOEFL scores ranging from 200 to 500 and above preferred to get feedback from their lecturers more than from their peers (see Table 2). There is a light different mean score between feedback from lecturers and peers of the students with the TOEFL score ranging from 500 and above. The different is more than 1.00. As shown in Table 3, the students from four different English proficiency level overall perceived that written feedback as the effective feedback from their lecturers. Interestingly, only the students with TOEFL scores ranging from 200 to 300 had a different in the mean and standard deviation scores for oral (M=5.00, SD=3.46) and written (M=7.00, SD=0.00) feedback. The other thing was from the students with TOEFL score ranging from 500 and above, the two students valued feedback in oral (M=6.50, SD=0.71) and in written form (M=6.50, SD=0.71), the same way when the feedback came from their lecturers. The students with TOEFL scores ranging from 200 to 300 seemed to strongly believe that written feedback from their lecturers was more effective than oral feedback. However, the students with the mean scores ranging from 500 and above perceived both written (M=6.50, SD=0.71) and oral feedback (M=6.50, SD=0.71) from the lecturers as the effective feedback. It was also found the same that all the students with different TOEFL scores perceived oral feedback as the effective feedback from their peers not written feedback (Table 4). Similar to the result of feedback from lecturers, students with TOEFL scores ranging on 200 to 300 chose the Likert scale with the highest rating 7 for their choices of the effective feedback.  show that almost all the students preferred their lecturers to show them where the mistakes in their writing were and have their lecturer correct them by writing the correct words or structures (Direct feedback). They were the students with the TOEFL score ranging from 200-300 (M=7.00, SD=0.00), 401-500 (M=6.63, SD=0.84), and 500+ (M=7.00, SD=0.00). Different from the students with those TOEFL score ranges, students with TOEFL score ranging The rating shows that direct feedback is in the first rating, indirect feedback 2 as the second, and indirect feedback 1 as the last one for the students with the TOEFL scores ranging from 200-300, 401-500, and 500+. Conversely, the students with TOEFL scores ranging from 300 to 400 rated them differently. They chose indirect feedback 2 as the first, direct feedback in the second rating, and indirect feedback 1 as the last. It was found the same on indirect feedback 1. They all perceived that when their lecturers as well as their peers only showed where the mistakes were without correcting them (Indirect feedback 1) as the less effective feedback. An interesting finding is on the feedback in the first or second draft from peers (Table 7 and 8). It is interesting because students with the highest and the lowest TOEFL scores only wanted their peers to show where the mistakes were. The students in TOEFL scores 200-300 and 500and above perceived that indirect feedback 1 as the most effective feedback from their peers. Students with TOEFL scores 200-300 and 500 and above also had the same way in rating the three kinds of feedback practices. They preferred indirect feedback 1 as the first choice, indirect feedback 2 as the second, and direct feedback as the last. On the other hand, direct feedback was perceived as the effective feedback from peers by the students with TOEFL scores 301 to 500.  As seen in table 9 and 10, the students with TOEFL scores ranging from 200 to 300 were the only ones who perceived feedback on the first draft as the effective feedback when it was given by their lecturers (M=4.67, SD=4.04). The other three TOEFL score ranges believed that feedback on the final draft was the effective one from their lecturers. Students with TOEFL scores from 200 to 300 perceived feedback from peers in the same preferences with feedback from lecturers (see Table 11 and 12). They chose the first draft as the effective draft to be corrected both by lecturers and peers. Not only did students with TOEFL scores 200-300 perceive feedback on the first draft as the effective feedback practice from their peers, but also students with TOEFL scores 500 and above perceived the same way. They perceived feedback on the final draft as not as effective as on the first draft when it was given by their peers. The other students with the TOEFL score ranging from 301 to 500 had a different perception; they agreed that the final draft was the effective draft when their peers had to correct their mistakes. On the whole feedback practices, it was found the same that all students from four TOEFL score ranges put written feedback as more effective from their lecturers. Oral feedback was assumed as the effective feedback practice when it was given by their peers. However, there were different perceptions on direct and indirect feedback both from lecturers and peers. Students with TOEFL scores 200-300, 401-500, and 500 and above perceived that direct feedback was the most effective feedback given by their lecturers. Only students with TOEFL score 301-400 believed that indirect feedback 2 as the most effective feedback practice given by their lecturers.
An interesting finding was on the feedback practices trend from students with the highest and lowest TOEFL scores. They both agreed that when their peers just showed them where the mistakes were as the most effective way to correct their writing. On the other hand, students with TOEFL scores ranging from 301 to 500 still perceived that direct feedback was the most effective feedback both from their lecturers and peers.
Students with the lowest TOEFL scores preferred to get the feedback from their lecturers and peers in their first draft. The highest TOEFL scoring students also preferred to get Feedback on the first draft but only when it was from their peers. From their lecturers, students with TOEFL scores ranging from 500 and above were the same with students with the TOEFL scores ranging from 301 to 500; they perceived feedback on the final draft as more effective given by their lecturers. Only students with the TOEFL scores ranging from 301 to 500 consistently chose feedback on the final draft both from their lecturers and peers.

CONCLUSION
Since not many studies specifically discussed students' perceptions or preferences toward feedback based on gender and English proficiency level, this study could not use many references to support or to argue the findings. Moreover, the current study tried to find out the students' perceptions toward the types of feedback they expect from their lecturers and peers. Although the results showed from analyzing the data based on participants, gender, and English proficiency level showed that the students preferred to get feedback from their lecturers rather than their peers (Table 13), it does not mean that the results will only focus on feedback from lecturers. That is why this study explored both feedback types from the lecturers and peers. The reason is because both feedback from lecturers and peers work interchangeably in the large writing class (Lin, 2009).
The students with TOEFL scores ranging from 301 to 400 perceived feedback from their lecturers would be effective when their lecturers correct their mistakes by showing them where errors are and giving them clues on how to correct them in written form. They expect the feedback to be given in the final draft of their writing. Feedback from their peers would be effective when their peers speak to them in the final draft of their writing.
The students with TOEFL scores ranging from 401 to 500 perceived feedback from their lecturers as effective when their lecturers correct their errors by showing them where they are and giving them clues on how to correct them in written form. They expect the feedback is given in the final draft of their writing. Feedback from their peers would be effective when their peers speak to them in the final draft of their writing.
The students with TOEFL scores ranging from 500 and above perceived feedback from their lecturers as effective when their lecturers showed them where the mistakes are and giving them clues on how to correct them in written form. They expect the feedback is given in the second draft of their writing. Feedback from their peers is considered to be effective when their peers only show them where the mistakes are in the first draft of their writing by speaking to them.