Introduction

Revision, a part of process writing, is defined as any type of change made to a written text which can be done at any point of the writing process: brainstorming, drafting or revision (Freedman, 1985). Writers may decide to change their work during the brainstorming, drafting or the revision stages. Sommers (cited in Witte, 1985) defined revision as a series of changes that have a cause-effect relationship in which revision is triggered by a cue and can happen repeatedly throughout the writing process. It is not a linear activity which occurs only after writers have considered their work to be completed and then revise one more time. Instead, students can become aware of any discrepancies in their writing and intervene even if the text has not yet been completed (Allal & Chanquoy, 2004). Revision is possible in written language, in the procedures the writer follows in order to produce the text, or in the cognitive processes that the writer undergoes when revising (Freedman, 1985).

Allal and Chanquoy (2004) classify revision in two essential categories: editing and rewriting. While editing is considered to be any modification that does not change the meaning of the text, rewriting entails the transformation of meaning. Faigley and Witte (cited in Asenavage & Connor, 1994) developed a taxonomy for revision in which various “multidimensional classifications” were included. These classifications are: meaning-preserving and meaning-transformation modifications, the impact on language, the effect of revision on the text, and the specific modifications writers make to revise. These classifications have given insight into the variety of revision processes a writer can undergo (Allal & Chanquoy, 2004).

When provided with feedback, writers are able to reconsider their work, reflect on the meaning of their work and modify their information, if they consider it necessary. Therefore, feedback plays a key role in students’ revision activities and it contributes to the quality of students’ writing (Freedman, 1985; Hyland & Hyland, 2006). Teacher feedback guides students through the revision stage in three different aspects: 1) it aids students to detect and to handle problems they may face while writing, 2) it provides opportunities to practice the writing skill through multiple drafts, and 3) it encourages students to analyze the comments received, to choose which suggestions are useful for them and to aid them in the production of new writings (Freedman, 1985). 

Feedback is considered a source of input that encourages writers to improve their written work and to develop their writing skills (Hyland & Hyland, 2006; Li Wai Shing, 1992). Some commonly used sources of feedback or feedback techniques are oral feedback or writing conferences, peer feedback, written feedback (end notes, side notes, or correction codes) and computer-mediated feedback (e-feedback or computer programs).

Teacher written feedback (TWF), in any of its delivery modes, allows students to benefit from working with a more experienced and knowledgeable person (Goldstein, 2005). Providing teacher-written comments enables students to reflect upon whether what they intended to write was what the reader understood. These comments also give writers ideas for possible ways to mend the mismatch between what they intended to express and what was actually written (Goldstein, 2005).  Moreover, it is permanently available for the writer to refer to when necessary and it gives the teacher the opportunity to expand her/his comments with full explanations of suggestions. However, teacher written feedback does not provide space for meaning negotiation; if the teacher’s feedback is unclear or misunderstood, the writer does not have the opportunity to ask for clarification. Furthermore, writing personalized feedback to every student is time consuming for the teacher (Goldstein, 2005). 

Teacher/student conferencing (TSC), another way of providing feedback to writers, is considered a “conversational dialogue” in which meanings are constantly being negotiated while a strong emphasis is made on the two-way communication (Freedman, 1985; Freedman & Sperling, 1985; Hyland & Hyland, 2006). Students benefit from conferencing because it encourages the development of autonomy and it allows them to construct their revision plan independently (Hyland & Hyland, 2006). This one-on-one dialogue allows the writer to reflect and change the main idea of the composition.  It encourages or discourages changes on drafts and it helps the writer notice any issues that may arise in the written draft (Freedman & Sperling, 1985). Nevertheless, some researchers believe face-to-face conferencing may have certain reservations. For instance, the power relations between teacher and student may have a strong influence on the revision outcomes (Hyland & Hyland, 2006).  Conferencing requires large amounts of time and specific interaction skills which teachers and students may not have.

Several studies have addressed the importance of feedback in the revision process. For instance, Huang (2000) compared the effectiveness of teacher audio-taped feedback (ATF) and teacher written feedback on the drafts of twenty-three English major students at a Taiwanese university. It was found that the teacher  produced more words in her feedback comments with ATF (54,258) on the final draft than with TWF (4,757). Additionally, 83% of the students stated they preferred ATF, 13% preferred TWF, and the remaining 4% preferred both.

Although Huang’s study gives insight into the effectiveness of each feedback technique by analyzing the number of words that was used to give input to the participants, the study gives a limited view of how feedback and revision are related. It would be interesting to record the types of revisions participants produce and how the teacher’s feedback influenced those revisions. On the other hand, the difference between Huang’s study and the current  study can provide information concerning how the program of study, teaching techniques, teaching methodology, and the students’ personal preferences influence the writer’s revision process.

Another study carried out by Conrad and Goldstein in 1991 documented how a Vietnamese student’s revisions were influenced by the teacher’s written feedback and conferencing sessions (cited in Goldstein, 2006). A recent revision of this study revealed that after receiving feedback through teacher/student conferencing sessions, the participant produced revisions such as addition and deletion of information on the written drafts. However, it was found that the participant’s personal beliefs, attitudes and individual factors such as the lack of adequate knowledge to write the analysis and to carry out the discussions that were required had also influenced the revisions. Conrad and Goldstein reported that when written feedback was provided, the participant’s revisions and quality of written work were influenced by the feedback per se and by other factors such as the clarity with which the feedback was provided, the participant’s motivation at the time of revising and the participant’s lack of time to look up information concerning the topic of the assignment.  Goldstein (2006) found that affective factors such as students’ beliefs (any idea that the student believes is true or any information that the student becomes aware of) and perceptions (the way the student perceives feedback suggestions) can have an effect on revisions and the quality of writing. It is my opinion that in addition to students’ beliefs and perceptions, the predisposition a student may have specific feedback technique can also influence the revisions made on drafts.  Therefore, knowing students’ feedback preferences can help writing teachers to improve their input and adapt this input to their students’ needs.

As a contribution to previous EFL writing research, the current case study attempts to describe the relationship between teacher feedback and students’ revisions by revealing the type and number of revisions students make due to teacher feedback. The participants of this study took part in multi-draft activities in which teacher feedback was provided using two different techniques: teacher written feedback and teacher-student conferencing. The participants made as many modifications as they considered necessary based upon the feedback they received. The source of the feedback, the number of revisions and the types of revisions carried out are documented.

This study also documents students’ preferences towards the feedback techniques in an EFL context.  This case study addresses the following research questions:

1.      Which feedback technique produced the most revisions in participants’ final drafts, Teacher Written Feedback or Teacher/Student Conferencing sessions?

2.      What types of revisions did students produce?

3.      What are the students’ preferences towards the feedback techniques? 

Methods

This case study followed a mixed approach of quantitative and qualitative methods. Merriam (1998) considers case study research to be an investigation of a “bounded system” (p. 27) which focuses on a single entity or unit. It aims to describe the complexity and particularity of a single case within a certain context (Merriam, 1998; Stake, 1995).  Both quantitative and qualitative methods were combined to obtain a more reliable understanding of the results of the study (Condelli & Wrigley, 2004).

The study’s quantitative traits allowed the researcher to determine what the students revised as well as how the teacher feedback influenced those revisions by providing statistics concerning the number of revisions and feedback suggestions given. This numerical representation was obtained by coding each revision according to a specific taxonomy, thereby producing a more controlled result (Matveev, 2002). On the other hand, qualitative data gave the researcher an opportunity to explore the students’ preferences and opinions concerning the use of feedback.  This represents the affective side that quantitative data may not be able to show easily.

The data collection instruments used in this study are interviews, questionnaires, observations and the students’ written work. It is important to mention that the results of this study cannot be generalized due to the small number of participants.

Participants

An EFL class consisting of twenty students in a private high school was initially recruited for this study.  The students were informed of the conditions of the study and what their participation would consist of.  Only ten students volunteered to take part in the study. However, as the study progressed three students gradually dropped out due to absences or illness. Thus, only seven students (three females and four males) actually took part in this study. The other ten students of the class did not participate in the study; however, both the students who participated in the study and those that did not participate were treated equally during the classes and carried out all of the class activities. The work of the ten students who did not want to participate in this research was not used.  

As shown in Table 1, the students’ ages ranged from fifteen to seventeen. The data obtained from a background questionnaire (see Appendix A) revealed that four of the participants had previously taken a yearlong academic writing course while the three remaining students had not taken any type of previous writing preparation course. Six participants mentioned they enjoyed writing for a variety of reasons while one mentioned not enjoying writing.

TABLE 1. PARTICIPANTS’ BACKGROUND

The Mexican teacher participant has taught EFL for six years and had previously taught two academic writing courses.

Setting

This study took place in a private Mexican high school. The students were enrolled in an intermediate EFL course as part of their semester schedule. The class met for ninety minutes daily for a total of 105 hours of class time during the fall semester of 2007.

The teacher adopted an integrated skills (listening, reading, speaking and writing) approach for her lessons. Daily activities included the practice of the four skills. Grammar and vocabulary instruction were approached from an inductive perspective where students were encouraged to discover meaning and function through the use of examples and practice.  The writing instruction included activities such as brainstorming, discussions, planning, writing and revising, which were carried out through individual, pair or group work.

The assignments were written in class and varied from letters, emails, descriptions and stories with a length of between 80 and 120 words.  Writing was practiced two times a week and the 90-minute classes allowed for sufficient time for the stages of the process of writing to be completed. For each assignment two to three drafts were written and revised.  A number of feedback strategies such as written, oral or peer feedback were used randomly with each assignment. However, both feedback techniques were used during the writing process of each assignment until the final draft was finished. 

Procedures

Data Collection

The data collection was carried out in two phases near the end of the course.  Each phase lasted two to three days with two weeks between each phase. The end of the course was chosen so as to give the participants the opportunity to be acquainted with the feedback and revision techniques. As the researcher, I was also an observer during both phases and I took notes to record the teacher’s procedures in class. I did not take an active role in any of the phases.

Two writing tasks were used for this study: a letter to an imaginary friend in Phase 1 and a story in Phase 2.  In Phase 1 the teacher focused on using the TWF technique and in Phase 2 attention was given to the T/SC technique. Whether a student had consented to be part of the study or not, all of the students were given the same tasks at all times.  It should also be mentioned that there were other writing tasks besides these two, as well as activities in the other three skills of listening, reading and speaking.

a) Phase 1:On the first day, the teacher conducted a group discussion concerning a specific writing task by eliciting information such as: What would you include in a letter written to a friend that lives in a foreign country? and How would you begin to write?  As the teacher and students carried out the discussion, the teacher wrote the important information on the whiteboard so that the students could refer to it later while doing their writing assignment. Then, the participants were given ten to fifteen minutes to plan and organize their letter.  Finally, all students had 30 minutes to write an 80-120 word letter to an imaginary friend. Once the session was over, the teacher collected all of the drafts for analysis. End comments were used to praise the students’ work and to give organization and content suggestions while a correction code was used for language suggestions. On the second day, the teacher gave each student a copy of the correction code that was used for the language suggestions and led a group discussion to clarify any doubts.

The students were subsequently given their first draft back with TWF and had 30 to 45 minutes to revise their letters and write their final drafts. Although most of the participants finished their final drafts during class time, others were allowed to finish them as homework. The students also answered a background questionnaire (see Appendix A), which consisted of multiple choice and open-ended questions written in the participants’ L1 so that students had  a full understanding of each question and could express their ideas without any problems.  

b) Phase 2:  On the first day of Phase 2 – two weeks after Phase 1 - the teacher followed the same teaching procedure as in Phase 1. Discussions, elicitation, brainstorming and planning were activities done prior to the thirty minutes of class time given to writing an 80-120 word story. The teacher then collected the finished drafts. However, this time on the second day the teacher called each student to her desk to lead personal T/SC sessions and give them her feedback orally.  They also had the opportunity to interact directly with the teacher to clarify any doubts. The amount of time dedicated to each student was approximately five minutes.  During this time the rest of the class was given a revision activity to work on from a previous writing assignment. Once the feedback session was over, each student reviewed their first draft, revised it, wrote their final draft and turned it in. As in Phase I, the teacher received the final drafts, gave written feedback and assessed the writing by giving each story a score.  The students were given the opportunity to clarify any doubts concerning the scores.

Once the class was over, the researcher carried out a semi-structured interview with the teacher in the teachers’ lounge in order to gather information concerning the teacher’s opinions.  A list of questions was used as a flexible guide for the interview depending on the teacher’s responses. The interview was recorded and later transcribed for analysis. 

In addition, at the end of Phase 2 the students answered a ten-item questionnaire (see Appendix B) to obtain their perceptions and preferences regarding the two feedback techniques. 

Data Analysis

The information obtained from the four drafts, the TWF and transcriptions of the T/SC sessions, were analyzed in two steps.

Step 1) Draft one was compared with the final draft for both tasks.  Revisions were coded using Faigley and Witte’s (cited in Asenavage & Connor, 1994) Taxonomy of Revisions (Appendix C). This taxonomy describes surface and deeper, text-based changes and includes a variety of revision categories for each type of change (modification). Therefore, it allowed the researcher to identify the types of changes students had made.

Step 2)After each revision was coded, the suggestions given in the TWF and T/SC sessions were noted and compared with the final drafts of each task. The purpose was to identify which revisions were a product of the teacher’s input. Each revision was marked TF (teacher feedback) when it was the result of the teacher’s feedback and SSR (student self-revision) when it was the result of the student’s own initiative. This allowed the researcher to identify which revision was a product of which type of feedback technique. 

Lastly the teacher interview was analyzed with the purpose of identifying the teacher’s opinions concerning the use of feedback and her preference for feedback.  Also the student’s preference questionnaires were analyzed in order to obtain the students’ perceptions and preferences concerning the feedback techniques.

Results

Question 1Which feedback technique produced the most revisions on participants’ final drafts, TWF or T/SC sessions?The data obtained indicates that participants made a total of 75 revisions on Task 1. As shown in Table 2, the majority of the revisions were a result of TWF rather than of the students’ own initiative.On the other hand, data from Task 2 (T/SC) indicates that the students made a total of 32 changes to their texts (see Table 3). Once again, the input obtained from the teacher during T/SC produced the majority of revisions.

TABLE 2. NUMBER AND PERCENTAGE OF REVISIONS MADE ON TASK 1

TABLE 3. NUMBER AND PERCENTAGE OF REVISIONS MADE ON TASK 2

The information obtained suggests that TWF resulted in more revisions, with a total of 75 changes, than T/SC which produced a total of 32 revisions. 

Question 2: What types of revisions did students produce? Students modified a total of 107 items on which a variety of revision techniques were carried out. As shown in Table 4, participants reorganized information, deleted information, added information, substituted words, deleted words, corrected spelling, paraphrased, changed tense number or modality, and corrected punctuation. Surface changes (change of tense, number or modality, deletion of words, paraphrasing, and punctuation) were the types of revisions students carried out the most.

TABLE 4. TYPES OF CHANGES

An analysis of the first and final drafts shows that the students made more surface modifications (84%) than text-based changes (16%).

Question 3   How do students perceive the feedback techniques?  The data revealed that the seven participants considered both techniques to be useful for various reasons.  The seven participants perceived TWF to be useful because:  1) the students believed they became aware of their mistakes, 2) others felt that it enabled them to remember what they need to improve in their writing, and 3) others stated:

 “It is motivating.”

 “”It allows me to improve writing.”

Additionally, the seven participants considered T/SC useful for various reasons and commented:

 “It allows me to improve my writing.”

 “It is easier for me to understand.”

“I can learn better and clear doubts faster.”

Despite the usefulness of each technique that the students commented upon, they had preferences.  Three students stated they preferred TWF; two students preferred T/SC and only one student preferred a combination of the two techniques.  The remaining participant did not express a preference for either technique. However, due to the small number of participants, more research should be carried out.

Concerning the teacher’s interview after Phase 2, the teacher mentioned she believed that it was easier for students to understand what they needed to improve upon in their writing using the TWF technique. However, she perceived T/SC sessions to be useful because they allowed her to negotiate meaning with the students and to be more specific in her comments. Of the two techniques, her preference was T/SC for feedback purposes.  She also mentioned that she believed that using revision techniques and teaching writing with a process approach was a good way of guiding students with their writing.

The assignments that were carried out throughout the course and the study seemed to contribute to the construction of the instructor’s opinion.  The results of this study were only shared with the teacher after the data analysis had been carried out with the purpose of avoiding any influence on her opinions and preference.

Discussion of Results

The results obtained in this study are consistent with those obtained by Paulus (1999). In this study the first and final drafts of 11 ESL (English as a Second Language) students were compared and the revisions were categorized. Paulus’ (1999) data revealed that 62.5% of the revisions (527 modifications) were surface- level modifications while the remaining revisions were meaning-changing modifications. In the current study, a total of 107 revisions were made on Tasks 1 and 2, of which 84% (90 revisions) were surface changes and 16% (17 revisions) were text-based changes. Both of these studies suggest that even though students made mostly meaning-preserving changes, they were indeed capable of modifying the meaning (text-based changes) of their first draft.

This study contradicts the results of Asenavage and Connor (1994). Their data showed that 60% of the revisions made by their participants were a result of the writer’s self-initiation – self-revision.  Only 35% were a result of teacher feedback. The remaining 5% were revisions made from feedback from the participants’ peers. It is important to mention that Asenavage and Connor (1994) reaffirm the idea of triggering student self-initiated revisions through teacher input. In the current study, students made fewer self-initiated revisions (SSR) on both Tasks 1 and 2. Only 33% (35 modifications) of the revisions made on both tasks were a product of the students’ self-initiated revisions.  Most changes were a result of either the TWF or the T/SC.  The results obtained from both studies indicate that teacher input may play a major role in triggering students’ written analysis and revision processes. However, revisions that were made by the students themselves without any input from other people cannot always be expected especially if the teacher’s teaching methods do not train students to revise on their own. Participants in the current study mentioned they preferred TWF and this technique produced the most revisions. This seems to indicate that students’ preferences can influence the modifications made on final drafts. Therefore, it is important to take into account students’ opinions when teaching and when teachers form their rationale for the use of feedback. Again, due to the small number of participants, more research is needed.

Future research might focus more on finding how TWF or TSC can influence participants’ revision processes.  By considering how these two techniques of feedback may or may not affect students’ written outcomes, we can improve our writing lessons and provide opportunities for students to improve their work.  It would be interesting to investigate teacher-student relationships; for example how the teacher’s power over the students during conferencing sessions can influence participants’ revisions and improvement in writing. Understanding the affective domain of the relationship between feedback and revision may allow writing teachers to improve their techniques and help students feel more comfortable with writing. On the other hand, during the T/SC sessions that were carried out in this study, participants did not take notes of the teacher’s comments. Instead they needed to reply orally regarding their understanding of the received feedback and the teacher’s notes written on the draft.  Further research could focus on how students’ note taking could influence the number and type of revisions that were produced.

In conclusion, the data obtained from the written drafts, the feedback comments and the conferencing transcripts revealed that participants’ were more influenced by written feedback than by conferencing sessions. The students made more surface-level revisions (change of tense, number or modality and punctuation changes) than text-based modifications (reorganized information, deleted information and added information).  The data obtained from the interview with the teacher and preference questionnaires from the students revealed that students preferred written comments over conferencing sessions while the teacher preferred giving oral comments in conferencing sessions. Yet, the number of participants was small and more research needs to be carried out.

Teaching Implications for Instruction and Conclusions

The results obtained in this study suggest that feedback techniques can influence students’ revision outcomes.  Furthermore, it gives an insight into the different techniques that EFL teachers can use to motivate their students to improve their texts and how the teacher’s feedback can influence the revisions of students.  For instance, teacher’s written input can motivate students to analyze their writing and to make any modification they consider appropriate even if the teacher does not necessarily suggest an analysis or modification.  That is, feedback can encourage students to initiate their own corrections.  I believe that one of the main purposes of training students to revise their writing is to lead them to self-analysis and self-improvement in their writing.  On the other hand, rather than focusing on the surface level problems such as punctuation, tense or modality, teacher feedback may focus on the overall intention of the writing and any text-based aspects that may improve the meaning of the text. This refers to making sure that students’ intentions when writing match what they actually wrote.  In other words, if the meaning of the written text is obscured or does not match what the writer intends to communicate, then the teacher may need to focus her or his feedback on text-based aspects.  The teacher should keep in mind the needs of the students when choosing a feedback technique or a combination of feedback techniques.  Taking into account the writers’ feedback preference may help the teacher to make a suitable selection.

This study gives a perspective on how a teacher and seven students perceive feedback techniques. However, the results obtained in this study cannot be generalized to all EFL contexts due to the small number of participants. Therefore, it is of major importance to carry out more research in other EFL contexts that can lead us to provide adequate feedback which can best serve our students’ needs.

Acknowledgements

I would very much like to thank Izela Partida Montalvo, Jose Luis Rodriguez Ramirez and Ruth Roux for their unconditional support in this study. A special thought of appreciation goes to the Mexican students who participated in this study.

References:

Allal, L. & Chanquoy, L. (2004). Revision revisited introduction. In L. Allal, L. Chanquoy & P. Largy (Ed.) Revision and Cognitive Instructional Processes: Studies in Writing. (pp.1-7). Vol. 13. Norwell: Kluwer Academic Publishers. 

Asenavage, K. & Connor, U. (1994). Peer response groups in ESL writing classes: How much impact on revision? Journal of Second Language Writing, 3(3), 257-276. 

Berg, E. C. (1999). The effects of trained peer response on ESL students’ revision types and writing quality. Journal of Second Language Writing, 8(3), 215-241.

Condelli, L. & Wrigley, H.S. (2004). Real world research: Combining qualitative and quantitative research for adult ESL. [Word document] Retrieved  from www.leslla.org/files/resources/RealWorldResearch.doc

Freedman, S. (1985). The Acquisition of Written Language. New Jersey: Ablex Publishing Corporation.  

Freedman, S. & Sperling, M. (1985). Written language acquisition: The role of response and the writing conference. In S.W. Freedman (Ed.) The Acquisition of Written Language. (pp.106-130). New Jersey: Ablex Publishing Corporation.  

Goldstein, L. (2006). Feedback and revision in second language writing: Contextual, teacher and student variables. In K. Hyland  & F. Hyland (Eds.)  Feedback in Second Language Writing: Contexts and Issues. (pp.185-205). New York: Cambridge University Press. 

Goldstein, L. (2005).  Teacher Written Commentary in Second Language Writing Classrooms. Ann Arbor: The University of Michigan Press.

Hyland, K. & Hyland, F. (2006). Contexts and issues in feedback on L2 writing: An  introduction. In K. Hyland  & F. Hyland (Eds.)  Feedback in Second Language Writing: Contexts and Issues. (pp. 1-19) New York: Cambridge University Press. 

Huang, S. (2000). A quantitative analysis of audio-taped and written feedback produced for students’ writing and students’ perceptions of the two methods. Tungai Journal 41, 199-232. 

Li Wai Shing, J. (1992). A process approach to feedback in writing. Perspectives: Hong Kong Journals Online, Vol.4. Retrieved August 10, 2007 from www.sunzi1.lib.hku.hk/hjo/view/10/000038.pdf  .  

Matveev, A. (2002). The advantages of employing quantitative and qualitative methods in intercultural research:  Practical implications from the study of the perceptions of intercultural communication competence by American and Russian managers. Retrieved on March 9, 2009 from http://www.russcomm.ru/eng/rca_biblio/m/matveev01_eng.shtml

Mendonça, C. O. & Johnson, K. E. (1994). Peer review negotiations: Revision activities in ESL writing instruction. TESOL Quarterly, 28(4), 745-769.

Merriam, S. (1998). Qualitative Research and Case Study Applications in Education. San Francisco: Jossey-Bass Inc. Publishers. 

Paulus, T. M. (1999). The effect of peer and teacher feedback on student writing. Journal of Second Language Writing, 8(3), 265-289.

Stake, R. E. (1995). The Art of Case Study Research. Thousand Oaks, CA: Sage Publications. 

Tsui, A.  B.  M. & Ng, M. (2000). Do secondary L2 writers benefit from peer comments? Journal of Second Language Writing, 9(2), 147-170.

Witte,S.  P. (1985) Revising, composing theory and research design. In S. W. Freedman, (Ed.), The Acquisition of Written Language: Response and Revision. (pp. 250-284) Norwood: Ablex Publishing Corporation.