Attitudes of second language students towards self-editing their own written texts

Recognizing students’ deliberate e!orts to minimize errors in their written texts is valuable in seeing them as responsible active agents in text creation. "is paper reports on a brief survey of the attitudes towards self-editing of seventy university students using a questionnaire and class discussion. "e context of the study is characterized by its emphasis on evaluating the #nished written product. Findings show that students appreciate the role of self-editing in minimizing errors in their texts and that it helps in eventually producing well-written texts. Conceptualizing writing as discourse and therefore as social practice leads to an understanding of writers as socially-situated actors; repositions the student writer as an active agent in text creation; and is central to student-centred pedagogy. We recommend the recognition of self-editing as a vital element in the writing process and that additional error detection mechanisms namely peers, the lecturer, and the computer, increase student autonomy.


Introduction
One way of recognizing students' deliberate e orts to minimize errors in their academic writing is to understand their attitudes towards self-editing.While we agree that "no learner intentionally writes strings of incoherent text" (Yates and Kenkel 2002: 43), we suspected that an attitudinal problem existed amongst our students, because of the apparent indi erence exhibited in the written texts that they submitted on a writing support course we teach.is rendered their texts very tedious to read.Recurrent errors in our students' work convinced us that students do not self-edit their texts, do not appreciate its value, and deliberately wait for the lecturer to correct their errors for them.In practical terms "self-editing" is the writer's ability to independently or otherwise identify and act on textual inaccuracies and loss of clarity in content, organization and mechanics.Charles (cited in Cresswell, 2000: 235) proposed a "self-monitoring" technique where "students write marginal annotations R W & about problems in their evolving compositions, to which the teacher responds".While we did not dismiss the potential of self-monitoring for improving students' nal dra s, we questioned the second language (L2) writer's ability to act on errors identi ed via self-editing and thought that research was required, so that we could better understand the extent to which student writers take responsibility for textual accuracy, and are able to do so.
A system of assessment that focuses only on the nal written product, as is the case at our university, provides unclear feedback on what students are capable of doing as writers.While students might be expected to be able to improve their written texts by way of acting on the lecturers' feedback, the system requires students to submit, not multiple dra s, but only one nal dra of the essay.In this way, the formative role of teacher feedback (Glover and Brown, 2006) seems non-existent.e expectation of markers is that that single submission is as error-free as possible; and that over the years of their education, students have received su cient feedback on earlier submissions to have learnt what constitutes correctness, and can therefore produce error-free texts.
In this paper, we hypothesize that in a largely product-focused writing context such as ours, students hold speci c attitudes on self-editing that may detract from the quality of their writing, even though they might want to minimize errors that might occur in their work.is generated the research question: what are students' attitudes to self-editing?

Why self-editing is important
We regard self-editing as central to increasing students' facility in meeting their lecturers' expectations.Only the writer, via dra ing and redra ing, reviewing (by self or peers), re-casting, and repeated self-editing, can respond to the entirety of textual detail, ranging from punctuation to word appropriateness to sentence length, cohesiveness, viewpoint, force of argument, pacing, and so on.We also regard self-editing as extremely important in the era of electronic communication because clicking "send" or "print" before attending to possible errors in form, content, and organization can be a source of embarassment for writers or annoyance for their readers.e promoting of self-editing practices therefore has lasting value for today's university student.More importantly, enhancing student self-editing capacity eliminates a culture of over-dependency on teachers, enabling the teacher to assume the role of a facilitator, co-learner or collaborator (Atkinson, 2003) during the writing process.It may also be important in reducing the teachers' workload, so that they can focus on providing feedback that is relevant to students' future work.It is also an important resource for learner-centred pedagogy (Vollmer, 2006) which places the student as an active agent in knowledge creation.Using our experience as teachers of a university writing course, we explore students' attitudes towards self-editing and identify possible reasons for these attitudes before making our recommendations.

eoretical framework
When Krashen (1985) proposed his Monitor hypothesis to explain how L2 users apply their learnt knowledge of L2 rules to monitor (self-edit) their texts, he was of the view that self-editing was sometimes bad practice because of the resultant hesitant speech of L2 users.In contrast, however, we regard self-editing as a vital competency for L2 users in writing, especially, since the academic essay has become an entrenched assessment tool within higher education and must therefore be carefully cra ed to optimise chances of success in assessment exercises.
Increasingly, studies in the eld of L2 writing, especially research on revision and text quality, have come to view the writer as a responsible and active participant in the writing process.Focusing on what the student knows about communication and language, Yates and Kenkel (2002) adopt a learner interlanguage perspective on error correction, which they use to critique error correction procedures in the literature.ey propose that "learners have principles which, if understood by the writing teacher, provide insights which are more useful than a target deviation perspective" (Yates and Kenkel, 2002: 31).Cresswell (2000) studied learner autonomy resulting from training in using the self-monitoring technique whereby students indicate their doubts by annotating their texts so that the lecturer can give feedback on these doubts and the essay itself.Results showed that learners appreciated the degree of independence gained and showed willingness to continue using the technique (Cresswell, 2000: 243).Similarly, Xiang (2004: 245) found self-monitoring particularly bene cial for high-achievers, although this applied only to the organizational aspect of their compositions.Both these ndings show that students are not passive recipients of feedback but can be active participants in the construction and meaning of that feedback (Xiang 2004: 244).Charles (1990: 292) also proposed the self-monitoring technique and claimed that the technique "encourages students to look critically and analytically at their writing and to place themselves in the position of the readers".e above studies questioned the e ectiveness of established ways of giving feedback i.e. from the teacher to the student.Research in this area of L2 writing (Storch 2005, Brender 2002, Sugita 2006, Truscott 1999, Ferris 1996, Carson and Nelson 1996) is on whether and how teachers should correct errors in students' writing, whether teacher feedback improves students' writing prociency, or whether peer-editing helps.Currently, debate rages around the question whether or not teacher's error feedback makes a di erence.Over the past ten years the debate has featured Truscott against Ferris.e former argues that error feedback in L2 writing is counter-productive as it detrimentally a ects learners' writing development and that it has not improved students' writing (Truscott 1999); while the latter argues that error feedback can improve language accuracy over a period of time (Ferris 1996).Williams (2003) suggested using individual conferencing as one way to explain the teacher's feedback to each student but this strategy is less feasible in large-class settings, such as ours.e literature reviewed by Glover and Brown (2006) claimed that in large classes the frequency and quantity of teacher feedback is reduced; consequently, the formative value of such feedback was lacking; and students argued that because written assignments were topic-focused, feedback lacked relevance to future assignments.Under these di culties surrounding the e ectiveness of feedback, additional techniques such as self-editing are needed.
Students' self-editing attitudes have, however, received relatively little attention in L2 research.Polio, Fleck, and Leder (1998) studied ESL students' editing for sentence level errors and Francis (2002) investigated the editing and correction strategies of much younger bilingual children.In both studies learners showed remarkable attentiveness with regard to their texts.Additionally, there are several writing manuals available that mention self-editing and the Internet o ers access to numerous checklists for self-editing purposes.However, we still do not know clearly why errors that would seem to be author-correctable continue to end up in students' texts.e purpose of this paper therefore is to ascertain students' attitudes that may in uence their ability to self-edit their written texts.Once identi ed, these attitudes may provide insights for instructional purposes for teachers of academic writing.

e subjects
Two out of the ve classes taking an optional post-Year One course called ' Advanced Writing Skills' responded to a questionnaire and took part in the class discussion therea er.(Both these activities were intended as consciousness-raisers for a self-editing activity that followed but is not reported here).
e course is housed in a 'study skills' unit for student academic support programmes at our university, and is underpinned by behavioral psychological approaches characterized by genre writing drills.
Altogether there were seventy students (23 males and 47 females) from seven di erent faculties as follows: the Faculty of Social Sciences (36), Humanities (12), Education (8), Science (7), Health Sciences (4), Business (2), and Engineering (1).ese gures show the uneven popularity of the Advanced Writing Skills course across Faculties and disciplines, which can be attributed to the relative importance that the Faculties of Social Sciences, Humanities, and Education attach to essay writing, as reported by the students.e sample included forty-two second-year students, twenty-three third-years, and ve fourth-years.Respondents' ages varied widely: y-two respondents were below twenty-ve indicating that they le senior secondary school less than ve years ago, which indicates that their familiarity with academic writing was more recent than the few who were over thirty.ey were also a very complex multilingual group whose home languages included: Setswana (37 respondents), Kalanga (3), Sebirwa (3), Hindi (1), Ndebele (1), Setswapong (1) and Herero (1).Some spoke a combination of two or more home languages, either English and Setswana ( 14); or Kalanga and Setswana (6); or English and Sekgalagadi (2); or Setswana, Afrikaans and Herero (1).Despite this home language diversity, the sample shared similar school experiences of using English as the medium of instruction.

Methodology
Using, rst, a questionnaire and then a class discussion, the study explored university students' self-editing attitudes in order to see whether students thought that self-editing improves textual quality.e three-item open-ended questionnaire, initially piloted with a di erent group of students taking the Advanced Writing Skills course, introduced students to the idea that writers must remain consciously in control of the writing and editing process (Cresswell, 2000:237) so as to minimize the errors that slip into the nal dra .Again, taking hints from Cresswell (2000), the class discussion was based on students' memories of writing experiences.

Findings
e rst item on the questionnaire was: "Do you think writers are able to selfedit their work themselves?"It sought respondents' perception of writers as autonomous individuals and as members of a writing community capable of self-editing and acting on the textual errors they spotted.All their responses communicated the attitude that self-editing is di cult, ine ective and complex.In the class discussion, they also expressed the view that they had not attained the status of 'writer' , so as to be able to overcome these di culties.(Indeed many self-editing online manuals convey a similar view.)ree such responses expressed the confounding nature of self-editing as follows: "yes, writers are able to self-edit their work but it is quite di cult because we tend to believe that we did everything correctly, thus defending our work"; and "because it is your work, you will understand it your own way, and some mistakes you will not identify"; and "they [writers] can [self-edit] and maybe not, because sometimes when you think you are reading what is written, in fact you are saying what you thought you were writing".
We further sought to know respondents' thoughts regarding the importance of self-editing by asking the question: "What in your opinion is the importance of self-editing to the text writer?"Respondents agreed that self-editing was an important part of writing because: "you [the writer] are the one who knows what you want to say and it will be di cult for another person who does not know what you want to say to do that for you"; "work with lots of mistakes turns o readers"; "I have realized how much I make mistakes when writing a er I had my exercise marked"; "bumping into someone's editing mistakes is irritating".One respondent lamented the omission of self-editing activities in early education: "it is a bit di cult for most to grasp this concept [self-editing] because from elementary school we were taught in such a way that the teacher has to be the identi er of mistakes instead of us communicating through our writing".Another suggested that "time should be made for self-editing in exams and tests like probably a er the test duration has lapsed"; while yet another suggested "maybe we can write our academic papers via a computer as it easily picks up errors".
We sought to see if students ever thought of alternative assistance with their dra s in the form of peer editing, by asking the question: "Do you ever ask a friend to edit your work before you submit it?".Almost half (34 out of 70) the respondents said they did.While they saw the value in peer editing, many saw it as merely a way to: "see if I have any mistakes"; " nd each and every mistake'; "see mistakes I overlooked"; "be corrected by somebody"; "identify errors you the writer cannot see"; and "correct construction of words and spelling".One respondent wanted to "ensure that the work is readable".
Evidently, peer-editors mainly focus on the mechanics, and not on content and organization of ideas.However, this help was not sought by many because, as several respondents put it, "there was no time to show your work to a friend".e thirty-six respondents who said "no" to peer-editing gave reasons that indicated that they doubted if their peers were any better skilled than they were themselves.Peer-editing was also viewed with suspicion re ecting the competitiveness students attach to texts submitted for assessment and for that reason they were worried that peer-editing might result in "plagiarizing my points"; or "copying from me to improve [their] work and get higher marks"; or even "friends making fun of your mistakes"; or worse still "missing submission deadlines".
As expected, students' attitudes to self-editing are divided.On their ability to self-edit, they are unanimous that self-editing is complex, but that despite its complexity, it is important for the writer.However, on the value of asking a friend to assist in the editing, some students say that friends are helpful while others view that help with suspicion.e hypothesis of the study is therefore con rmed: that in a largely product-focused writing context such as ours, students' attitudes towards self-editing are not helping the quality of their writing, even though these students would want to take full responsibility to minimize the errors that occur in their work.

Discussion of ndings
Increasingly, studies in the eld of L2 writing, especially research on revision, have come to view the writer as a responsible and active participant in the writing process (Charles 1990, Cresswell 2000, Xiang 2004).However, as writers, students do not usually position themselves as co-researchers or as creators of new knowledge (McIntosh, 2001), a situation con rmed by their responses to the questionnaire.Responses also implied a strong need to minimize the sense of competition in the learning process. is is because developing academic writing skills in L2 can be theorized as a process of apprenticeship, where learning is viewed as a process of social participation rather than simply as acquisition of knowledge.e teacher's role, too, di ers from that of a disseminator of knowledge.Within such a learning approach to knowledge acquisition, teacher and/or peer feedback may be viewed as part of the process of apprenticing students into legitimate participation.When student writers position themselves as communicators in a discourse community (consisting of their peers, their lecturer, and themselves), they become their own rst readers of the texts they produce.However, evidence from the questionnaire and class discussion showed that mainly because writing tasks are competitively understood by students, audience is perceived as either assessors (their lecturer) or plagiarizers (their peers).
In the class discussion, students argued that feedback from their faculty lecturers was emphatic around correct use of the conventions of linguistic and textual features.is reinforces students' view of self-editing as a complex process shrouded in uncertainties.Students also rightly perceive a model of correctness to exist somewhere; a view advanced by the genre approach to academic literacy.Unfortunately, that ideal model seemed to remain obscure to them, either because their faculty lecturers do not model the genres for them, or the models are not made accessible to students; and also because such ideal models (or "genres") fail to produce autonomous writers because of an over-emphasis on the technical features of genres, rather than on their expressive resources.Additionally, the writing tasks are for purpose of compiling Continuous Assessment scores, and less because students need to practice and develop their discipline-based writing facility.us, the function of academic writing in the di erent faculties is seen to be mainly that of an evaluative tool that determines pass or failure; a view perpetuated by the absence of a real audience beyond the lecturer who chose the topic(s) and by the requirement to submit, not multiple dra s, but only one nal product of the essay (Wright, 2006:90).As a result, students' attitudes to their writing tended to imply a process over which they had little control.
With regard to involving peers in editing one's work, one respondent indicated a reluctance to be critiqued by a friend because "friends may not want to disappoint you". is comment may be attributed to the Tswana cultural philosophy of botho which means compassion and caring.In the students' view, friends are expected to show their "goodness" in assessing what a friend has written; thus one respondent argued that, out of modesty, friends may not do a thorough job of editing because "they don't want to disappoint you", in case they are seen as bad or unsupportive friends.Carson and Nelson (1996) reported similar results among Chinese learners which they attributed to the Asian collectivist culture, saying that more successful peer interactions come from students who share a common language and cultural expectations than from students in heterogeneous cultural groupings.e same study identi ed additional cultural factors that underlie a reluctance to involve peers.One of these factors is "mutual status inequality" which was exempli ed in the current study by one respondent who felt much belittled by peer-editing saying, "Someone who is not my lecturer reading my work!!!I feel as though they judge me".e other factor is "trustworthiness of peers' language pro ciency".For instance one respondent dismissed her peers' linguistic pro ciency as "most of their [peers] English is not very good… almost useless to have them do it [peer-edit]".ese comments indicate a reluctance to accept guidance that comes from elsewhere other than from the lecturer, a problem attributable to classroom culture and power, where the teacher is perceived as the only source of knowledge.Such teacher-fronted perceptions of learning to write are not very helpful in large classes.For instance, at our university, semesterization reduced contact time per week for the Advanced Writing course from three to two hours, and due to heavy marking loads, student's work is returned long a er submission but without any direct contact with the student.O en students do not even collect the marked scripts.ose who do are only curious to see their score but make very little use of the feedback.
Earlier studies on revision cited in Cresswell (2000:236) found that students tended to edit for grammar at the expense of other textual elements such as logic, relevance, and appropriateness of content.Similar results are evident in the respondents' comments, where editing is only associated with their overemphasis on local, surface-level components and ignorance of global structures of texts.In the class discussion there were suggestions that since the computer can edit their work for them, there was little need to worry about errors.However, this is only partly true.For instance, with regard to essay content, the computer cannot supply the description, argumentation, thesis statement, focus, or di erentiate factual and experiential information.e writer must also deal with the logical organization of ideas and arguments, the e ectiveness of the introduction and conclusion, and the sequencing of ideas in order of importance.us the computer can only be a supplement to detecting textual inaccuracies.More importantly, when the peer's primary concern is "to see if I have any mistakes" or "to nd each and every mistake", the attitude conveyed is that the original purpose of academic writing is not a genuine concern to understand something.Instead it is an opportunity for the reader to judge the degree of adherence or divergence to the writing conventions; and for the writer to display awareness of such textual features and; and that the form rather than the message is at the centre of writing. is view was further con rmed in the class discussion: among the main areas of writing mentioned by students during the discussion as requiring improvement were those relating to mechanics (referencing, grammar, punctuation, and spelling) and organization (sequencing of ideas, cohesion and coherence).
During the class discussion the students also revealed that, based on the feedback they were getting on their assignments, their faculty lecturers perceived the Advanced Writing Skills course as essentially remedial and that taking the course would enable them to write better.Such a de cit view of the students who take the course has implications which might be apparent in their written scripts in the Advanced Writing Skills course.Vollmer (2002) argues that a de cit perspective "sees them [L2 writers] as developmentally weak and their texts as riddled with errors".For their part the students said they found writing in English easy although they admitted that they needed help.Linking these statements to their responses to the questionnaire items, it is possible to suggest that greater autonomy via self-editing skills could enhance students' textual control better.Because these respondents did not position themselves as purposeful communicators, they o en failed in their attempt to communicate meaning to their readers, probably as a result of the doubts they hold about their capabilities to do so.Such doubts are the result of recurrent disappointments from earlier assessed work.According to Garcia-Sanchez and de Caso-Fuertes (2005:273) a long history of failure in uences task perseverance, the level of e ort, and the degree of success achieved among other things.
On the basis of the attitudes to self-editing ascertained in this brief study, a strong case can be made regarding the teaching of writing conventions to L2 writers.From an L1 perspective, McIntosh (2001) regarded writing conventions as domesticating and limiting because they discourage subjectivity.However, the important question for L2 writing instruction is "what bene t do students get, as writers, from a genre approach to literacy"?e textual inaccuracies in mechanics, organization, and content show that the learning of academic writing conventions is still needed by L2 writers if their sense of being in control is to be realized.Omitting the teaching of writing conventions marginalizes students within academia and relegates them to the back row of academic literacy.We are convinced that self-editing, initiated by way of direct instruction, requires a good or growing command of the conventions of writing.Without conscious engagement with and exposure to the conventions, students are at risk of failing and of being ever-subject to their lecturers' corrections, rather than developing their own facility.

Recommendations
Ideally, a genre approach to writing suggests that students acquaint themselves with actual examples of a variety of textual types.In this way they get to recognize the di erent linguistic and textual features.When this is followed by learner training in self-editing, students' texts have a basis of correctness to follow.Learners also need to be alerted to the expectations of the target audience.Additional error-detection mechanisms such as peers, the lecturer, and the computer promote a sense of discourse community within which meaning contained in the written text is constructed.Because the computer provides impersonal feedback on the mechanics (spelling, punctuation, and grammar) of writing, it protects the L2 writer against embarrassment and feelings of humiliation over the errors committed.It also saves the writers some e ort, enabling them to focus beyond grammar and spelling on more global textual concerns such as logic, cohesion, word appropriateness, and overall textual organization.

Future research
Analyses in this brief survey of student attitudes to self-editing did not factor in the full range of variables involved.For example, it would be interesting for writing instruction to determine how age, gender, home language, and course and level of university study of the respondent impact on the attitudes to self-editing.ere is also a need to investigate the impact on self-editing of contextual realities, such as large class sizes which reinforce classroom organizational practices that result in product-focused (rather than process-focused) writing and assessment practices.Due to the large number of learners involved, heavy marking loads are a constant burden for the lecturer.Hence, the intervening teacher feedback on the dra s and revisions are o en impossible to give.Under these conditions, students' actual self-editing practices need to be documented and developed.e documenting of actual practices serves as indicators of how students are exhibiting control of the text they produce.

Conclusion
Overall the study has shown that although L2 writers in the research sample see self-editing as complex, they value it in reducing textual inaccuracies.Although a larger study sample would have provided more generalizable results, the ndings of this brief attitudinal study contribute to the debate over how e ectiveness within L2 writing can be developed: that, despite students' attitude that self-editing is complex, self-editing is a vital skill for improving textual quality; and writing instruction that nurtures its development is bene cial for purposes of developing autonomous L2 writers.