Teacher Feedback : Student Perceptions , Teacher Perceptions , and Actual Teacher Performance

This study sets out to investigate a teacher’s and her students’ perceptions of written teacher feedback in a college English as a foreign language (EFL) writing class in China. Essays, questionnaires, and interviews were employed to identify the types of feedback given by the teacher, the perceptions and preferences of students and the perceptions of the teacher. The teacher and her six students were interviewed, and sixty two students completed the questionnaires. The results are that the written teacher feedback covered content, organization, vocabulary, grammar and mechanics, and students reported that they benefited most from feedback on organization, which was focused on by the teacher but not specific enough and in a small amount. Moreover, the preferences of students were not expected by the teacher. Foreign language writing teachers were suggested to communicate more with their students about their feedback practices, and be aware of students’ perceptions and preferences, so that their writing instructions could be more effective.


Introduction
Written teacher feedback plays an important role in improving students' writing skills.Not only does it provide a valuable opportunity for individualized, text-based, contextualized instruction from teachers (Ferris, 2001), but also it is likely to be taken seriously by students than feedback given to groups of students (Hattie, 2012, p. 271).Especially for second or foreign language writing, research has found that both teachers and students hold that teacher feedback on student writing is a critical part of writing instruction and can have a great influence on student writing (Ferris & Hedgcock, 2014, p. 237;Goldestin, 2004;Leki, 1990, p. 58).
Research on teacher feedback has been exploring about its different aspects, such as its description and its effects, and recently the survey from students' perspectives on teacher feedback has emerged as one of the main aspects (Ferris & Hedgcock, 2014, p. 238).It indicates the importance of students' views in investigating about teacher feedback.What do students need and pay attention to influence the impact written teacher feedback have on students.Most studies on students' reactions to teacher feedback find out that it may reduce students' motivation if they do not receive what they expect from teachers (Ferris, 2003).Only if students read and take seriously teacher feedback can teacher feedback be successful in helping students improve their writing (Ferris, 1995).Therefore, many scholars have suggested that how students receive feedback on their writing and their preferences for feedback should be carefully considered, understood and acknowledged by teachers (Ferris, 2012, p. 229;Hedgcock & Lefkowitz, 1996;Goldstein, 2004).

Previous Research on Students' Opinions about and Reactions to Written Teacher Feedback
Research on second language students' opinions about and reactions to written teacher feedback has been mostly done by means of survey.It has been found that learners vary greatly in their response to teacher feedback, and they prefer different types of feedback for different reasons.First, the focus of students' attention on feedback has been found different for different learners.Some learners paid the most attention to almost all aspects of their writing, some paid more attention on form, and some paid more attention on content (Cohen, 1987;Ferris, 1995;Hedgcock & Lefkowitz, 1996;Hyland, 1998;Leki, 1991).Second, students' views about usefulness of teacher feedback are mixed.Most students found teacher feedback useful (Ferris, 1995;Saito, 1994).Some students found feedback on errors more useful than feedback on content (Hedgcock & Lefkowitz, 1996;Saito, 1994).While students in Leki's (1991) study profited little from error corrections, and students in Cohen's (1991) study reported to benefit most from comments on organization.Third, students' preferences for feedback have also been found to vary according to writing contexts.Some studies suggested that students want teachers to comment on form as well as on content (Cohen & Cavalcanti, 1990;Leki, 1991;Hedgcock & Lefkowitz, 1996).Some students preferred more comments on form (Hedgcock & Lefkowitz, 1996).These diverse findings of previous studies have revealed that students' views should not be treated alone, isolated from its context, especially their teachers' beliefs and practices.Hedgcock and Lefkowitz (1996) put forward that students' response pattern could reflect their instructors' practices in the writing classes.Written feedback reinforces the points made through explicit teaching and any research of written teacher feedback should consider the connections and interactions between teachers, students, texts, and writing purposes (Goldstein, 2001, p. 86;Hyland & Hyland, 2006, p. 213).
However, research considering teachers, students and texts together has not been much.Montgomery and Baker (2007) point out the lack of comparing student preferences and perceptions about feedback to actual teacher feedback or teachers' self-evaluations of that feedback.Lee (2008) concludes that there have been few attempts that relate student reactions to actual teacher feedback in specific instructional contexts.Teachers have rarely been consulted as informants in teacher feedback research (Ferris, Brown, Liu, & Stine, 2011).
Only several studies have studied actual teacher feedback, student preferences and perceptions about feedback, and teachers' opinions of their feedback practices at the same time.Cohen (1991) carried out a study on both teachers' and students' perspectives on teacher feedback on compositions to find out the focus of teacher feedback and student attitudes and preferences toward teacher feedback.The study involved 2 teachers and 13 EFL students and 19 Portuguese L1 students from two universities.Teachers did verbal report and finished a questionnaire.All students completed questionnaires, and among them, 3 EFL and 3 Portuguese L1 students provided verbal report.Also, the teachers' comments on three students' texts were analysed and classified according to their focus (i.e., content, organization, vocabulary, grammar and mechanics).In the EFL case study, the teacher reported that they focused on all categories of feedback, and felt that students benefited more from the comments on organization.Most students described their teacher as providing comments mainly on organization, content and grammar.Nearly half of the students preferred more emphasis on content and vocabulary.Montgomery and Baker (2007) invited 13 writing teachers and 98 students from a writing centre of a university to complete the questionnaire adapted from the one used in Cohen (1987) and Ferris (1995), and later showed the results to the teachers.Teacher feedback on some student compositions was also examined.It was found that teachers gave more feedback on local issues than on global issues, which was not the same as what the teachers thought they did.It revealed that students seemed generally satisfied with the amount of feedback they received, which may result from students' belief that addressing local issues is a vital part of writing courses.Lee (2008) made an attempt to involve more contextual factors in her research.She investigated student reactions to their teacher feedback in Hong Kong secondary classrooms through questionnaires, interviews, content analysis and classroom observations over one school year.Fifty-eight students completed questionnaires, and nine of them finished checklists and interviews.Two teachers were observed and interviewed.Results showed that teacher feedback were mainly directed toward language errors, while it was more so in the lower-level class.Students of one teacher responded favourably to their teacher feedback, while most students of the other teacher wanted more comments on ideas.Much more lower-proficiency students did not want their teacher to respond to their errors.
Few attempts have been made in the EFL context of mainland China to examine written teacher feedback from both students' perspective and teachers' perspective.Moreover, the research of student response in EFL is much less than that of research in ESL, and is by no means exhaustive or conclusive.Furthermore, the same questions should be examined with comparable research methodology across different contexts to create a more comprehensive picture of written teacher feedback (Goldstein, 2001, p. 77).
This study contributes to existing feedback research by linking student perceptions of written teacher feedback to the actual written teacher feedback on final drafts in a multi-draft setting and the teacher perceptions in one Chinese college classroom.A survey was done to collect data.The following research questions were addressed in this study: 1) What types of written teacher feedback are given on students' final drafts in a Chinese EFL writing class for English majors?
2) What are the teacher's perceptions of the written teacher feedback?
3) What are students' perceptions of and preferences for the written teacher feedback?

Subjects
One writing teacher and her 62 students from a university in China participated in this study.The teacher was a native English speaker, who held a bachelor degree in English literature and a master degree in English composition in the United States, and had been teaching for five and a half years (one and a half in China).The students participated in the research were all sophomore English majors enrolled in the writing course in which the writing of essays and the receiving written teacher feedback on the final drafts was an integral part.The students were all Chinese native speakers who learned English as a foreign language, and had been studying English for over seven years.All students completed the questionnaires for the study, and 6 of them volunteered to be interviewees.
The course on which the study was based was a writing course for sophomore English majors, which counted towards graduation.The course mainly focused on essay development for different genres including important elements, rhetorical devices, local coherence with transitions, and language use.The class met one and forty minutes every week.In-class activities included instruction on the knowledge about the definition, the components and features of each type of essays, writing strategies (e.g., making an outline, annotating texts), formatting, exercises of using the basics of each genre of essay for group work, peer response on major assignments, and general oral teacher feedback activities.

Essays
The substance of teacher feedback was investigated through text analysis of teacher feedback on students' writing assignments.The feedback analysis in the present study only covered the focus of feedback.The coding scheme for written teacher comments was adapted from Ene and Upton's ( 2014) study.The six students' graded writing assignments were copied after they were graded.In total, twelve graded essays were collected, with two essays from each student.The first essay was descriptive essay and the second one narrative.Table 1 presents the basic information about the essays collected for analysis.Each paper with the teacher's original handwritten commentary was read and coded by two raters.A questionnaire was designed to collect self-report data concerning the feedback students received.It asked the respondents to think of the papers that they had received from their teacher in the semester, and to give their response in the questionnaire.
The questionnaire was based on a hybrid of surveys used in Elwood and Bode (2014), Ferris (1995), Hedgcock and Lefkowitz (1996), Nash (2012), andWang (2014).It included Likert scales and an open-ended item.The questionnaire was comprised of several sets of questions.The first set queried the basic information about teacher feedback.The next set tried to elicit students' perceived usefulness of teacher feedback.The final set asked about students' preferences of feedback.The first item was adapted from that used in the study of Ferris (1995), which was to examine students' perceptions of different types of feedback they had received.This item was a question on a 5-point Likert scale, with 1 being never and 5 always.The following four questions were on a 6-point Likert scale, with 1 being strongly disagree and 6 strongly agree.Even numbered scales were employed to prevent neutral answers and demand respondents to make real choice (Dornyei & Taguchi, 2011, p. 28).The second item was based on that in Hedgcock and Lefkowitz's (1996) study, which focused on students' beliefs about the benefits of microlevel aspects of teacher feedback.Students were to reflect on how helpful they found the teacher's feedback concerning the content, organization, lexical choice, grammatical and mechanical accuracy of their written texts.
The third item was adapted from Wang's (2014) study.In Wang's (2014) study, this item was used to study the perceived usefulness of peer feedback in revision.In the current study, it was adapted to investigate students' perceived usefulness of teacher feedback for improving different aspects of their EFL writing.These two items probed into the extent to which the participants found the feedback useful in a general sense and in developing their writing in content, organization, vocabulary, grammar and mechanics.The fourth item was borrowed from that of Hedgcock and Lefkowitz's (1996) study, and was used to elicit information of students' preferences for various focuses of feedback.The fifth item was adapted from Elwood and Bode (2014), and was to ask students' preference for certain types of written feedback.These two items were about students' preferences of feedback.
The sixth item was borrowed from that used in Nash's ( 2012) study, and was to discover other alternatives that were not covered in the previous part.The items on the questionnaire were all translated into Chinese.The Chinese version was distributed to students.
Student questionnaires were distributed to 62 students a week after they received their second graded drafts during a class break, which was near the end of the semester.In this study, Cronbach's alpha for the whole questionnaire was .93,suggesting that the reliability of the instrument was satisfactory.

Interviews
A semi-structured format was followed.An interview guide with questions was be written in advance.Depending on the development of interviews, the order and the amount of questions may change, and further questions may be added for clarification and exploration.

Teacher interview
The teacher interview questions were based on the studies of Hyland and Hyland (2001) and Shine (2008).Apart from the basic information about the course and the way feedback was given, the core questions asked about the teacher's perceptions of the aspects of writing she focused on, the helpfulness of the feedback she gave and the comments she expected to be wanted by students.The interview was carried out three days after students were interviewed.

Student interview
The participants looked at their final drafts with teacher feedback, and described their reactions to the feedback as well as commented on the feedback within two weeks after they received the final drafts of their second assignments.
The interview schedule was divided into two sections.The first section consisted of a series of structured questions, derived from interview questions of related literature, which were Shine (2008) and Leki (2006).The core questions asked the participants whether receiving teacher feedback was of any use to them and which specific aspects they found helpful or not and the reasons behind that, and their preferences for focuses of teacher feedback.
The second section comprises of a verbal protocol by means of a checklist.The checklist was adapted from the study of Lee (2008).The students were asked to provide their reactions to the feedback-their general evaluation of its usefulness with the help of a checklist.For each of the teachers' comments, students were asked whether they knew how to resolve it (on a scale of 5 to 1, "totally" to "not at all") and their evaluation of its usefulness (on a scale of 5 to 1, "totally" to "not at all").The graded final drafts of the two assignments were used as a visual prompt during the interview.At least 23 feedback points were randomly selected from the essays, which the participants were asked to respond to according to the checklist.The interviews were carried out three days after the collection of questionnaires.The interviews were carried out in Chinese with translated questions.

Types of Written Teacher Feedback
The first research question examined the types of written teacher feedback students received on their final drafts.
Results from different data sources are slightly different from each other.Generally, written teacher feedback under study covered content, organization, vocabulary, grammar and mechanics.First, materials for training teaching assistants to grade students' essays were collected from the teacher.The rules for giving written feedback for the class were written on them.The statements about which aspects of writing should be commented were identified.It was stated that, "the overall goal with feedback is to help students see both macro and micro techniques they can add to revise or edit their own work."End notes of every final drafts were supposed to include "one holistic goal (like organization) and a technical pattern of error, like tense or unclear explanations".Second, response to the first set of questions on the questionnaire was calculated to show students' self-report of the frequency of each type of feedback they had received.The results are shown in Table 2. From the table, it is clear that students reported that they often received feedback on content, organization, vocabulary, grammar, and mechanics, while the teacher feedback they felt that they received the most was on grammar, followed by content, vocabulary, mechanics and organization.Third, feedback on students' final drafts was categorized and calculated in terms of numbers and percentages.A total of 731 feedback points were collected from the 12 essays written by the 6 volunteer interviewees.Table 3 shows that most teacher feedback focused on content and vocabulary.Other types of feedback were much less, with 17% of the feedback on grammar, 14% on organization and 10% on mechanics.Four, the data from the interviews was analyzed to support students' response in the questionnaires.In the interviews, over half of the students reported to receive feedback mainly on grammar, and not much on content.Jutta said, "The teacher focuses less on content than we do, and instead concentrates more on other detailed problems.The teacher did not give me many comments, but the comments she gave me draws my most attention to formatting, punctuation and grammar".
However, the relative amount of feedback on each area was not sure.Except from the training materials, all other sources proved that teacher feedback on organization and mechanics was not much.As for the frequency of feedback on grammar and content, there is some mismatch.Students reported that they had received most feedback on grammar, content and vocabulary in the questionnaire, and more on grammar than on content in the interviews.Whereas, the analysis of teacher feedback on several students' papers uncovered that the number of feedback on content and vocabulary respectively was more than that on grammar.The conflict is mainly on the relative amount of feedback on content and feedback on grammar.It might result from that most feedback on content the teacher gave to students was about clarity or understandability.Comments like "confusing", "unclear", "hard to follow" were often on students' papers, and students might treat these comments as pointing out their grammatical usage, because both feedback on clarity and feedback on grammar are on the micro level, and do not directly deal with the quality of ideas students express.However, it is certain that students did not receive as enough feedback on content as they want.

The Teacher's Perceptions of Feedback
The second research question examined what the teacher thought of her feedback in terms of its focus, usefulness and preferences for students through an interview.
The teacher stated that she focused on different aspects for different genres.For descriptive essays, micro-level things, such as rhetorical devices, descriptive details, dialogues and grammar were focused on.For narrative essays, macro-level things, such as a strong thesis with plot, climax and a change.Content, organization and grammar were implied to be focused on in the essays under examination.
As for whether students would think her feedback useful, the teacher answered that if students read the feedback, they would benefit from it.
The teacher thought that students would most want to receive honest personal reader response to the content of their writing.She explained that students did not receive enough attention from teachers in China, and some students may not like feedback on writing techniques, but honest feedback on the writing content could make students feel real.

Student Attitudes towards Feedback and Their Preferences
The third research question first considered students' perceptions of the usefulness of the teacher feedback.It was answered by data from checklists, questionnaires, and interviews.In general, the students' assessments of the value of teacher feedback were positive.They thought they could learn a lot from the feedback and would improve their writing.
First, a part of the interview employed checklists to ask participants to comment on the usefulness of the feedback they had received on their final drafts.The responses were summarized in Table 4 below.Table 4 shows that out of a total of the 164 feedback points collected form the essays of six students, close to half of them were considered to be totally applicable and useful, and over half of them to some extent could not be applied to revise the essays and not useful for the students.Which is worth pointing out is that the number of feedback considered useful was larger than that of feedback considered applicable (76%>70%, 42%>26%, "4" and "5" on the Likert scale).It is also indicated by the responses on the questionnaires.Next, Table 5 and Table 6 summarize the responses from students on the questionnaires regarding their perceptions of the usefulness of different types of feedback on different areas of writing.In general, students reported that they had learned a lot from all types of teacher feedback on their final drafts, and were able to improve on corresponding areas of writing in the future.Students agreed that they had benefited a lot from feedback on grammar, and would reduce grammar mistakes in their future writing.They agreed less on the benefits of feedback regarding organization, next on vocabulary and content.It is worth pointing out is that according to Table 6, students least agreed on that they would improve on content.Finally, the data from interviews provide more details for the answer.Table 7 represents a summary of details of the part of interview when participants were asked to report what use they made of their written teacher feedback.Data from the interviews revealed that the participants predominantly held that present written teacher feedback made them aware of their problems in grammar, expressions, and organization but might not necessarily help them correct their problems in the future.Apart from Jutta who claimed that the feedback she had received did not help her much in writing, the other five interviewees all mentioned the specific benefits they got from it.
Most interviewees (Helen, Woods, Queenie, Candy and Gina) mentioned that the feedback pointed out where they expressed awkwardly.Woods and Helen said that they had tried themselves best to express in an "English" way.Gina and Candy admitted that Chinese, as their native language, had influence on their English expressions.Whereas, all four of them hadn't known which of their expressions were awkward, and the teacher pointed them out in their writing.Helen said: "I thought that these words were often used by English-speaking people, but in fact they are not.The teacher thought they were awkward, which made me realize that I need to express myself more clearly." Half of the interviewees (Woods, Queenie and Candy) thought that the teacher's corrections in grammar reminded them of their mistakes in grammar, and they would try to avoid the same mistakes next time.Candy said that from the teacher feedback, she would know what to correct, and pay attention to her problematic grammar usages next time she wrote.
Half of the participants (Candy, Helen and Gina) revealed that the teacher feedback reminded that they should pay attention to transitions between paragraphs and connections between them.Gina said: "after receiving teacher feedback on the essays, I am conscious of the transitions and connections among the introduction, the body and the conclusion in my writing." Although the teacher feedback helped them locate their problems in expressions and organization, it was not concrete enough to help them make corrections.Most interviewees expressed their difficulties in revising their essays based on their written teacher feedback to some extent.Woods and Gina said that they did not understand what was wrong with their expressions, even though the feedback pointed specific expressions out.Queenie and Candy said that they did not know how to correct their awkward expressions.Gina expressed that she might write the same awkward sentences next time.

Gina
-Find out weaknesses, and remember them, and make some small changes.
-Pay more attention to the grammar corrected and the unnatural expressions pointed out.
-Find out the problems in grammar, transitions, and vocabulary choice.
-Find out what would not be understood by native speakers (e.g., vocabulary and expressions), and pay more attention to transitions.
-Find out the mistakes pointed out, mainly on grammar.
-Pay more attention to unnatural expressions, clarity, transitions.
The second part of the third research question addressed students' preferences of the teacher feedback.It was answered by data from closed and open questions on the questionnaires, and interviews.The students agreed that teachers should provide feedback on all five areas of writing.They predominantly preferred to receive feedback on organization, vocabulary, content, and grammar.Whereas, mechanics was the least welcome.
First, the means and standard deviations of students' responses on their preferences of feedback types are presented in Table 8 below.Students agreed that teachers should give feedback on organization, vocabulary, content, grammar and mechanics, with organization most agreed and mechanics least agreed.When it comes to their expectations (Table 9), they expected to receive feedback on all aspects, while they strongly expected feedback on organization, content and vocabulary.Second, the open-ended item on the questionnaire asked students to state how the teacher feedback would lead them to "greater improvement".Out of 62 students who handed completed questionnaires, 55 students responded to this item.Over half of them (51%) needed more specific, detailed and clear feedback.Apart from that, four aspects of writing were most often mentioned.They are organization (36%), vocabulary (35%), grammar (22%) and content (15%).For organization, close to half of the students pointed out that they need concrete and detailed feedback so that they would know how to improve, and the other half simply expressed that they need feedback on it.For vocabulary, most students hoped that they could have more detailed feedback, which can provide them the possible appropriate words.Four students directly pointed out that the teacher should not focus that much on mechanics.One student wrote that, "focus more on the depth of the content and give advice on language, such as writing techniques and style instead of focusing on formatting and fixed writing patterns." Third, Table 10 presents what the interviewed students hoped to be commented on in their writing, and further elaborates students' preferences.They generally would like to receive teacher feedback on organization, content, vocabulary, grammar, writing process, expressions and sentence structures, in the order.
Feedback on organization was highly expected by all participants.Most participants (Woods, Queenie, Candy and Gina) wanted to know from the teacher whether their essays were well organized, while Helen was specifically concerned about the connections between the paragraphs.Woods reflected, "I feel that the structure of my essays is very chaotic, and its logic is not very strong.I hope that the teacher can give me guidance on logic step by step, which tells me how I can effectively organize my essay to support the thesis statement." Almost all the interviewees expected the teacher to give feedback on content, but they focused on different aspects of the content.Jutta, Queenie and Candy expressed their worry about whether the content of their writing was against the requirements of the teacher.Jutta said: "when I am writing an essay, sometimes I have no idea of whether it is what the teacher expects us to write, so I hope that she could take a look at it."Whereas, Gina and Helen were more concerned about the clarity of their writing.Helen said that she expected the teacher to point out whether the key point of her writing was prominent.However, one exception was Woods.He thought that feedback on content was not very important, because it would restrict students' writing content.
Most interviewees hoped that the teacher could point out their problems in expressions and word choice.Candy, Woods and Gina expressed that they wanted to express in a more "native" way, or at least not in a way that seemed "strange" in an English native's eyes, meanwhile, Gina also hoped to express herself with more accurate words.
Feedback on grammar was also mentioned by half of the interviewees.Candy said that she was weak in grammar, so she needed feedback on it.Feedback on sentence structures was mentioned by half participants as well.They wanted their writing to seem more interesting and more advanced.And feedback on writing process was brought out by two interviewees.Candy and Helen said that their ideas were "in conflicts" when they were writing the essays, they did not know how to construct the essay, such as where to -Grammar, content, organization, and expressions.
-Vocabulary, expressions, organization and sentence structure.

Discussion
It is interesting to note that the students were more positive in the benefits they could have from the feedback than they could actually apply the feedback to their writing or improve on relevant issues.Feedback on grammar, organization and vocabulary were agreed to benefit them most.It helped them find their problems and know what to improve or avoid.Feedback on grammar could directly point out their errors and help them avoid the same errors in the future, while feedback on organization and vocabulary, which was not specific enough might not reach the same effects.This result is similar to what Ferris (1995) has found on students' feelings of the help from their teachers' feedback, which is that overall students found their teachers' feedback helpful in improving their writing.The result that feedback on grammar was more applicable than other types of feedback is similar to the result of Ene and Upton (2014).
The students slightly agreed that they would improve on the content of their writing.Two students interviewed gave their reasons for that.First, it was because they had thought very hard before finishing writing an essay, and could not change much even the teacher pointed out a big problem.Second, the content of writing each time was different, and it was hard for them to transfer much from one writing to another.The first reason was suggested in the review of Leki (1990), which concluded from previous research, that students did not like their teachers to take over their ideas.The second reason lends support to Ferris's (1995) claim that students may not see much relevance in teacher feedback on their ideas on final drafts to a new assignment with a new topic.
The findings about foreign language students' preferences regrading teacher feedback do support those of Hedgcock and Lefkowitz (1996), in which data of questionnaires indicated that foreign language students were concerned about content, rhetorical soundness as well as about linguistic accuracy.Moreover, the students' preference for more detailed, specific and clear feedback is similar to that of the students in Elwood and Bode's (2014) study.
Compared together, the perceptions of the teacher are slightly different from those of students.For the focus of feedback, the teacher thought she had focused on organization, while the students and the actual feedback showed that feedback on organization was not much.For the types of feedback students preferred, the teacher assumed that the students would like her personal reader response to the content of writing, while the students expected to receive feedback on organization most.The findings of the present study also support what Hedgcock and Lefkowitz (1996) revealed in their study, that students frequently operate on principles and assumptions that may not match those of their teachers.

Pedagogical Implications
The results of the present research have confirmed that students' perceptions and preferences of teacher feedback should be considered as an important part in foreign language writing courses.Obviously, an increased knowledge of students' response toward teacher feedback practices would benefit individual teachers, not to mention students.It can be difficult for teachers to provide feedback that will serve all students, and each student may have different perceptions of what is useful feedback (Hyland, 1998).It is not the intention of the research to suggest that specific categories of feedback, such as feedback on content or organization, should receive equal treatment or priority.The preference for one or another category depends on the educational background of the students and the teacher, the needs, certain writing assignment, and so on.The real issue is that of fit between what the teacher provides and what the students want, and to strengthen the communication between the teacher and students to contribute to the development of students' writing skills.Clear teacher-student agreements on feedback procedures could lead to more productive writing (Cohen & Cavalcanti, 1990).
The teacher could help ensure that the needs of the students are accommodated and their own intentions and priorities of feedback practices explained to students (Leki, 1991).On the one hand, the teacher should be aware that students are likely to have unexpected perceptions and preferences of the feedback.The type feedback that the teacher gives the least or focuses the least may be the most beneficial and highly valued by students.Ignoring students' request for feedback would reduce their motivation in writing (Leki, 1991).On the other hand, the teacher should take some time to explain the overall philosophy of giving feedback to the students, and discuss it with the class (Ferris, 1995), so that a much clearer consensus can be reached between the teacher and the students.The teacher should take actions to get timely response from students and modify instruction and feedback practices, which contributes to effective cooperation and to achieve mutual goals.
What has been identified in this study is the particular type of feedback that participants reported to benefit most from and most preferred while received almost the least from the teacher.It was found that their preferred to receive feedback on organization, leading to a conclusion that the teacher should provide more specific comments on organization which instructs students to effectively arrange their ideas and content so that their essays will be clear and logical.As an extension of this approach, the teacher could elaborate more and demonstrate on how to organize the ideas in class.

Conclusion
The primary purpose of the present study was to examine English major students' perceptions and the teacher's perceptions of written teacher feedback.A teacher and her 62 students enrolled in an intermediate-level English writing course at a China university participated in this study.Multiple instruments such as essays, questionnaires and interviews were used to collect the data.
The study provided important empirical evidence that students' perceptions and the teacher's perceptions of teacher feedback were found to not frequently be the same.
Results from this survey pointed out pedagogical implications for second language writing instruction.It was suggested that instructors should be informed of students' perceptions of their feedback practices in class.Increased awareness may be expected to lead them to take practical steps to explain and discuss their feedback in class, which contributes to effective writing instructions.
Further research is suggested to investigate students' and teachers' perceptions with a larger number of participants.Longitudinal studies are recommended in order to document students' perceptions and teachers' perceptions of feedback more systematically over time.
start from, what to write first, what second and what next.Table 10.Summary of Details from the Interview (sentence structure, and organization.

Table 2 .
Types of feedback received reported by 62 students

Table 3 .
Frequencies of types of feedback in essays analysed

Table 4 .
Results of student checklist

Table 5 .
Students' beliefs about the benefits of different types of teacher feedback

Table 7 .
Summary of Details from the Interview (1)

Table 8 .
Students' beliefs about different types of teacher feedback should be given

Table 9 .
Students' expectations of different types of teacher feedback