Abstract
Much research work on teacher feedback has concentrated on the perceptions of students and teachers on feedback, but few studies have addressed the extent to which students respond to their teachers’ written feedback, particularly at the tertiary level. This study analysed the extent to which students made appropriate revisions based on the feedback they received. Forty-one sets of drafts and final papers written by first-year undergraduates were compared. The analysis focused on the main components of the grading criteria for the assignment: language and style, rhetorical structure, and format. The findings showed that the students paid more attention to feedback on the rhetorical structure of their writing. There was no statistically significant difference in the students’ revisions of language/style and format. The results suggest that the students were more concerned with macro issues concerning the clarity of their thesis/topic statements and the logical development of ideas, than with the mechanical aspects of writing. This study serves as a useful guide to teachers when providing feedback, and also serves to encourage further research involving different groups of students in different contexts.
Introduction
Feedback generally refers to information provided by others regarding one’s performance. Such a definition is rather wide, and can include both grades and verbal/written comments. In this article, we focus on feedback in the form of written teacher comments. As Butler (1988) pointed out in her influential study, it is narrative feedback, rather than grades, that enhances students’ learning and their level of interest in the task or topic. Recent studies suggest that feedback can be a powerful tool to engage students in the learning process. For example, Hattie and Timperley (2007) believe that when feedback to students is targeted at the appropriate level, it can help to bridge the gap between their current understanding and the desired outcome. Further, Boud and Molloy (2012: 698) promote the idea of ‘sustainable assessment’, where students are perceived as the driving force in obtaining feedback for their progress.
Student Response to Feedback
Extensive studies carried out on the effectiveness of feedback (Leong et al., 2003; Silver and Lee, 2007; Stern and Solomon, 2006) found that formative feedback was ranked highly by both teachers and students. A review by Li and De Luca (2014) of assessment feedback selected from more than 300 journals from 2000 to 2011 revealed that much research concentrated on students’ perspectives of feedback. Among studies on students’ perceptions of teacher feedback, though, little is revealed of students’ actual follow-up on their teachers’ feedback. The small-scale study by Plater (2008) is an exception. The study, involving teacher feedback that was complemented by peer marking and the opportunity for the paper to be resubmitted for grading, found that certain aspects of feedback were not ‘followed up’. Plater’s study of the ‘follow up’ of feedback opens a space for research on what feedback students attend to and what they neglect.
In language pedagogy, the opportunity for students to draft and revise their paper allows them to explore ideas and refine their thinking process and linguistic and communicative expressions (McGarrell and Verbeem, 2007). Hawe and Parr (2013) contend that the process of drafting and re-drafting of student papers is both useful and critical to successful final papers.
At the tertiary level, feedback tends to be focused on the subject matter, as various disciplines provide teaching and understanding of content. However, in ESL and writing courses, feedback is usually given on rhetorical structure (relating to argumentation), language (relating to communication), and format (relating to understanding of genre). These focus areas are considered equally important, as they reflect the necessary writing skills expected of undergraduates. However, research interest in these areas – particularly comparative studies of students’ drafts and final versions at the tertiary level – is limited. The dearth of research in these areas represents a gap in our understanding of how students respond to feedback on their writing.
Models of Feedback
The urgency of closing the gap between teachers’ feedback and student understanding raises questions about how feedback could be more student-oriented (Black and Wiliam, 1998; Flores et al., 2015; Scott, 2014) than they currently are. One such feedback model suggests that feedback is ‘an instructional act within the theory of formative assessment’ (Parr and Timperley, 2010). Another model presents feedback as a vehicle for reflection and learning, otherwise known as feed forward (Quinton and Smallbone, 2010). A more recent model identified two categories of feedback – the degree of explicitness in the feedback, and the ‘amount of rationale provided’, where the student is not only told where the problem lies, but why there is a problem (Mahboob and Devrim, 2011: 112). An extension of coherent feedback, as this model is known, is cohesive feedback. Here, evaluation of writing is focused on purpose and structure of the text, development of meaning across paragraphs, and grammar and expression (Mahboob, 2015: 406). Together, coherent and cohesive feedback depicts effective feedback. Yet another model, proposed by Hattie and Timperley (2007: 87) and derived from their comprehensive review of feedback studies, involves students (and teachers) asking three questions: Where am I going? (feed up), How am I going? (feedback), and Where to next? (feed forward). Each question, in turn, applies to four levels: task, process, self-regulation, and self. The present study utilized this model by Hattie and Timperley as the categories proposed are clearly described and are suitable for application to our data.
However, in line with the purposes of the present study, two of the levels related to feedback (Where am I going?) – feedback on self-regulation and feedback on the self – were omitted because feedback on self-regulation was not our focus, and feedback on the self was, as observed by Hattie and Timperley (2007: 90), ‘least effective’. The other two levels – feedback on task (FT) and feedback on process (FP) – were more pertinent to our analysis. FT is concerned with the extent that students succeed in attaining the task, or goal, such as language accuracy and adherence to stylistic conventions (e.g. using a consistent citation format, maintaining an academic style of writing). On the other hand, FP is concerned with students responding to ‘information relative to a task or performance goal, often in relation to some expected standard, to prior performance, and/or to success or failure on a specific part of the task’ (Hattie and Timperley, 2007: 89). Feedback at this level is intended to help students develop a deeper understanding of the strategies and skills in handling processes underlying a particular task, with the further goal to enable them to transfer these strategies and skills to other tasks.
The Present Study
The objective of our study was to examine the type and extent of students’ revisions following teacher feedback on their essay drafts. The writing task was designed for a discipline-specific writing class. Our analysis focused on the key components of the grading criteria: (1) language and style, (2) rhetorical structure, and (3) format. It is hoped that the findings will guide teachers to provide feedback in ways that help students attend to aspects of writing that they may have overlooked in the past, or in ways that encourage students to respond to feedback as a crucial part of the learning process.
Methodology
This research was a corpus-based, exploratory study on the extent to which students responded to the teacher’s written feedback. Forty-one sets of student writing (with each set comprising a draft and a final version) were examined.
Context of Study
The samples of writing were obtained from an academic writing course for first-year History students at a university in Singapore. Of the 41 students, 27 were Singaporean Chinese, nine were Singaporean Malays and five were Singaporean Indians. All of them underwent at least 12 years of education, with English as the medium of instruction, and their respective mother tongues (Chinese, Malay, and Tamil) as a second language. The course aimed to equip History undergraduates with the skills to write essays about historical texts and to sharpen their argumentation and language skills. Topics in the course included critical reading, development of ideas, and the appropriate use of language, tone, and style. Academic literacy skills were also emphasized. Students were introduced to information search skills, acknowledgment of sources, and the use of EndNote, a citation programme which was available from the university’s library.
The corpus in this study comprised student papers for an assignment that involved a draft (900 words) and a final version (1,200 words). The assignment task, requiring the students to interpret a speech of their choice (from a list of six speeches), was given at the start of the semester. Drafts were submitted in the tenth week of the 13-week semester. Written feedback was provided, and the final versions were then submitted at the end of the semester. The assessment criteria focused on language and style, argumentation, and documentation of sources using the Chicago citation style. All three assessment components were assigned equal weighting. Students were informed of these criteria at the start of the semester.
Data Collection
Approval to conduct this study was granted by the university’s Institutional Review Board. Permission was obtained from students in the writing course for their drafts and final versions to be used in the study. Students were assured of anonymity in the report of the study’s results.
Each researcher compared the final versions of all 41 samples against the drafts, focusing on the revisions made. Responses to teacher markings (e.g. brief comments) and teacher corrections (e.g. answers to grammatical errors), and students’ self-corrections were all counted as instances of revision. The draft and final versions were examined in three aspects, as outlined in the assessment criteria: language and style (LS), rhetorical structure (RS), and format (Fm). Each aspect was examined for the presence of various components as listed in Tables 1–3; representative screenshots – Figures 1–3 – are included for each category.
Evaluation Criteria for Language and Style (LS).
Evaluation Criteria for Rhetorical Structure (RS).
Evaluation Criteria for Format (Fm).
Screenshot of teacher’s feedback on language and style.
Screenshot of teacher’s feedback on rhetorical structure; the feedback here is accompanied by the student’s notes in pencil.
Screenshot of teacher’s feedback on the lack of bibliographic information.
As domain knowledge was not an assessment criterion, accuracy of facts was not considered. This was because the course was conducted to teach History students the mechanics involved in writing essays, rather than to teach them history itself.
Data Analysis
Data analysis took place over four weeks. For each category of analysis, a scoring system of 0 to 5 (Table 4) was used to determine the type and extent of revision affecting the quality of the writing.
Scoring Guide for Extent of Revision.
A pilot run of the scoring was carried out over a week to identify issues with the category descriptions and the scoring guide. One of the problems was that a revision could be counted as an RS revision by one researcher but as an Fm revision by another. In one sample, for instance, the student provided ample evidence in the final draft, after being instructed to do so in the feedback. Two investigators awarded the student five points in the Fm category while the third awarded the student five points in the RS category. Eventually it was decided that extra evidence should be placed under RS, because the structure of the essay was strengthened by the inclusion of additional evidence. The categories were then refined and the scoring guide clarified. Following the pilot, the actual scoring was done independently. Internal consistency of scoring was measured using Cronbach’s alpha, which was 0.97.
The final scores were then subjected to a one-way analysis of variance (ANOVA) test to determine any significant difference among the three categories. The Tukey honest significant difference (HSD) post-hoc test was used for statistically significant ANOVA results. The significance level for all tests was α=0.05.
Results and Discussion
The overall mean score for all samples was 3.335 (s.d. 1.265). Among the three categories, the mean score for RS was the highest; the RS score was also the only one which was above the overall mean. The mean score for each category is presented in Table 5.
Mean Scores of Students’ Responses to Feedback on Language and Style, Rhetorical Structure, and Format.
The one-way ANOVA test revealed that at least one of the means was different from the rest (p=0.016). The Tukey HSD test showed significant differences in two comparisons: RS vs. LS (p=0.040) and RS vs. Fm (p=0.028). No significant difference was detected between LS and Fm.
The results illustrate students’ responses along two levels of feedback as outlined in Hattie and Timperley’s (2007) framework – i.e. feedback about the process (FP), and feedback about the task (FT). The results suggest that the students responded better to FP, providing some evidence of the careful processing of the information in their writing to achieve a desired outcome. Where FT is concerned, though, the ability of the students to distinguish correct from incorrect answers – and, indeed, to build more surface knowledge about such errors – appears to be compromised by the brevity of the feedback they received. We see, then, feedback at the process level appearing to be more effective than at the task level. We now begin our discussion with FP and the students’ performance in RS.
Rhetorical Structure
The emphasis paid by students to feedback on the rhetorical organization of their writing is reflected not only in the mean score for RS, but also in the number of students who received scores in the highest band (4.00–4.99). Of the 41 students, 23 received scores in this band. This is in contrast to the number of students in the same band for the LS and Fm categories – 11 and 13, respectively.
The relatively high RS scores may be attributed to the students’ responses to feedback in two areas – clarity of thesis and topic statements, and logical development of ideas. Of the two areas, issues concerning thesis/topic statements were more pervasive but easier to correct. Interestingly, the feedback itself on such issues was often briefly worded, but the students nevertheless made careful changes in their revisions. A case in point is (1a), which shows a student’s original thesis statement:
(1a) Draft version: In this way, his [Kennedy’s] speech showed his position as holding the mantle of a global leadership, trying to maintain USA’s position in the Cold War while at the same time direct international issues towards peace instead of constant conflict.
As it was not clear whether the student was merely drawing a conclusion based on preliminary information or stating the argument that would be addressed in the rest of the writing, the teacher responded by asking ‘Is this [the] thesis?’ in the margin (Figure 4).
Screenshot of teacher’s feedback on the thesis statement and on editing the opening paragraph.
In the final version, the thesis statement was amended, as follows:
(1b) Final version: Therefore, my essay aims to show the importance of Kennedy’s speech and how it showed his position as one that bore the mantle of a global leadership, trying to maintain USA’s position in the Cold War while at the same time direct international issues towards peace instead of constant conflict.
The thesis revision involved two changes. The first is the explicit indication that the sentence is a thesis statement (‘my essay aims to show’). The second concerns the focus selected by the student. As opposed to the original statement, the revised version is specific about the rhetorical objective – it is centred on the importance and implications of Kennedy’s speech. As seen in Figure 4, the teacher also instructed the student to edit the final line of the first paragraph, and this was done by the student (Figure 5).
Screenshot of student’s edit of the final line of the first paragraph.
In general, corrections involving the logical development of ideas were also fairly well done. There were numerous samples with careful revisions in the re-ordering or deletion of sentences and paragraphs, provision of appropriate supporting evidence, and signposting, as seen in (2a–b). In the opening paragraph of the draft version (2a), the teacher wrote that the writing still needed a strong thesis statement.
(2a) Draft version: Sojourner Truth’s advocacy of equal women’s rights, especially black women’s, reflected the lack of women’s agency in the dominant masculine society of 19th century America. In her speech, “Ain’t I A Woman?”, delivered at the Ohio Women’s Rights Convention in 1851, Sojourner Truth pushed for the advancement of women’s rights by drawing references to her lived experience as a slave. As such, Sojourner Truth’s experiences as a black slave had a significant contribution to her preaching of women’s rights, alongside her Christian beliefs.
In response, the student completely re-worked the paragraph by providing contextual information about the suffrage movement in 19th-century America before ending the paragraph with a clear thesis statement:
(2b) Final version: In the history of feminist movement in 19th century America, one name that stood out was Sojourner Truth. Born into slavery and brought up as a slave, Sojourner Truth employed her lived experiences as a slave into her speeches championing for the rights of women, especially those of black women. This is especially evident in perhaps what was her most well–known speech “Ain’t I A Woman?”. Though there exists various accounts of this particular speech, one could surmise that Sojourner Truth employed her experiences as a slave to discredit any fallacies that supported the lack of women’s rights. Yet, to fully understand the significance of her speech in the suffrage movement in the 19th century and particularly her strategic usage of her experiences, one must look at the intersectionality or race and gender politics and the dilemmas that existed within gender politics, for Sojourner Truth was not only a woman but also of black descent. The suffrage movement in the 19th century America identified the white middle-class women as default ‘Women’ and abolitionist activities were focused on black men. As such, a black woman in the 19th century was generally identified as more than an animal rather than a human, thus Sojourner Truth’s speech was of importance not only because she was a black woman but one who lived in that particular context and stood up against normative beliefs and practices. As such, this essay will seek to understand the circumstances in which black women faced in the 19th century through Sojourner Truth’s speech as well as to understand the significance of the employment of her lived experiences in the speech.
Such revisions can help to improve ‘task performance and self-efficacy, which in turn provides resources for more effective and innovative information and strategy searching’ (Hattie and Timperley, 2007: 93).
Notwithstanding this generally encouraging response to feedback, the analysis also revealed superficial amendments in a few samples. While such exceptions are in the minority – forming about a fifth of the total number of students – they nevertheless present an area of concern as the corrections were lacking in both coverage and adequacy. In an extreme example, only the topic sentence in one of the paragraphs was changed to make it more specific; very little in terms of paragraphing was amended.
As another case in point, consider the draft (3a) below. The teacher remarked that the ‘Beth Fisher’ sentence appeared to have been inserted randomly in the paragraph, thus compromising the writing’s coherence.
(3a) Draft version: The Cold War is defined as the period of intensified hostility and tensions between the two mutually exclusive ideologies, the democratic United States and the communist Soviet Union. This gave rise to the heightened arms race. Hence, the Reagan administration legitimized the appropriation of the hardline approach as the Soviet Union was seen as the primary source of the threat to the national interests of the United States. I argue that the hardliner approach of the Reagan administration had largely aimed to undermine the communist Soviet Union.
In the final version (3b), the only major change involved the shifting up of the ‘Beth Fisher’ sentence. Little else was done to improve the logical development of the writing within the paragraph. It remains unclear how Reagan’s strong anti-communist sentiments led to the demise of the Cold War, since it is entirely possible for such sentiments to have exacerbated the situation instead.
(3b) Final version: The Cold War is defined as the ideological clash between the democratic US and communist USSR, which gave rise to the heightened arms race between the two superpowers. Reagan was one such US president who was recognized for his strong anti-communist sentiments.
Such superficial corrections, albeit in the minority, are a worrying sign. Ideally, teacher feedback should serve to prompt students to work out solutions and alternatives on their own (Bitchener, 2008). The feedback should thus be explicit and indirect, but not exhaustive. The consequence of this is that students who are either unaware of this, or lack the discipline to look beyond the segments singled out by the teacher, may lose out in the learning process.
The larger issue at hand is perhaps the myopic approach that some students adopted in the revision process. There is a hint of this in their responses to feedback in the two areas highlighted in this section. As noted above, the feedback concerning thesis/topic statements was briefly worded, but the students were able to make appropriate changes in their revisions. Where the logical development of ideas is concerned, though, only superficial changes were made in a number of samples. The reason for this difference in responses is the scope involved in the revisions. In the former, the rewording of the thesis or topic statements is typically restricted to only a sentence. While students still need to take into account what is written in the rest of the essay or paragraph, the focus is on making a single sentence clearer and more specific. By contrast, changes to the logical development of ideas involve more holistic considerations. A myopic approach to revision is unlikely to improve the quality of the writing by much.
Language, Style, and Formatting
We turn next to the students’ responses to FT, focusing on mechanical aspects of writing, such as grammar and citation style. Although such issues appear uncomplicated and fairly easy to amend, the scores for LS and Fm were significantly lower than the RS score. The analysis revealed two probable reasons for the lower scores.
Lack of Knowledge
The first concerns the lack of knowledge of grammatical errors. As in the feedback on RS, the feedback on LS was briefly worded and non-exhaustive. Only representative errors in grammar and expression were singled out, with an accompanying comment or two at the end of the essay (see Figure 6).
Screenshot of general comments at the end of a student paper.
Examples of such general comments are given in (4–6):
(4) expr + lang
(5) Be careful → grammar!
(6) Grammar careful!
In the vast majority of the samples, the students merely changed the highlighted errors, but there was little evidence of similar revisions elsewhere in the writing. While this could be attributed to the myopic revision approach of some students, we are convinced that the problem is a more fundamental one – the students did not fully understand what the error was. In the following example, awkward phrasing and grammatical errors are still evident in the final version, despite the teacher’s advice in the draft that some parts of the writing were ‘not the best way to write a paper’:
(7) Final version: The first reason why it is not easy to achieve women’s rights or to acknowledge that women’s right is humans rights as how easy the Clinton puts it, is that culture and traditions in the family plays a crucial role. In her speech, Clinton addressed that it is a violation to women’s rights when they are not allowed to plan and choose who they want to marry. However, she failed to acknowledge and understand that not all women have the freedom to make decisions.
The quality of the writing is inconsistent not merely in the phrases used (‘women’s rights’ ~ ‘women’s right’) but also in the marked contrast between the first sentence and the rest of the extract. While the first sentence has a number of grammatical errors (‘humans rights’, ‘as how easy the Clinton puts it’, ‘culture and traditions in the family plays a crucial role’), the rest of the extract contains only a minor error concerning the relative pronoun ‘who’.
In the Singaporean context, where English is used alongside a number of other languages in daily life, some students may not be aware of the grammatical errors in their use of written Standard English. For instance, in a study involving Singaporean trainee teachers who had attended a degree course on Singaporean English, Schaetzel, Lim, and Low (2008) found that their respondents performed worse in a post-course survey (as compared to the pre-course survey) on the features of Standard English and the local colloquial variety. They further found that grammatical errors were dominant in both the pre- and post-course scores. Indeed, if even trainee teachers have problems with such issues, it stands to reason that students may also face similar challenges.
A case in point concerns the comma splice, an error frequently found in the corpus. This error is exemplified in (8–9) below:
(8) Final version: This brings out the first point of research, why did China block the broadcast and controlled who watches the speech?
(Teacher’s feedback on draft: ‘take care → expr. + gr. slips’)
(9) Final version: Firstly, prior to their declaration of war, America practiced isolationism, they were unwilling to interfere with world affairs.
(Teacher’s feedback on draft: ‘Be careful with gr. slips + errors’)
This error – and others in the corpus, such as subject-verb agreement errors – may well be due to interference from Mandarin Chinese, a language that is widely used in Singapore by the ethnic Chinese majority. The examples (8–9) were written by Singaporean Chinese students. (The reader will recall that two-thirds of the History students in this study were Singaporean Chinese.) Unlike English, comma splices are pervasive in the Chinese language; the comma frequently functions as the clause boundary, performing ‘the same role as [the] period in English in certain context[s]’ (Xu et al., 2013: 809).
Another common error in the students’ essays is the inconsistent use of the definite article. As the English article system is not fully matched in either Mandarin Chinese or Malay (another commonly used language in Singapore), difficulties in the proper use of the English definite article have been reported among ESL learners (e.g. Wong and Quek, 2007). Although English is used as the first language in education and most public domains, these difficulties remain for many of the undergraduates involved in this study. In one instance, the teacher inserted a definite article in ‘(the) Soviet Union’ as a prompter to the student to be more careful about similar slips in the rest of the essay. In the final version, however, no changes were made in three places (underlined) that required the definite article, even though it was used in the very first line (‘the US and the USSR’):
(10) Final version: From his [Reagan’s] speech, it could be inferred that the US and the USSR were not dealing with just a mere lack of understanding but it was more of a conflicting religious views. Many secularists and historians have neglected the aspect of religion in
The inconsistent use of the definite article before nouns, not merely in (10) but in many other samples in the corpus, suggests a lack of understanding of the broader principle concerning the use of determiners.
Lack of Familiarity
In the area of formatting, the students’ apparent lack of familiarity with existing conventions may also help to explain the lower score in Fm. The feedback in this category was focused on the Chicago citation style that the students were expected to use. Unlike grammatical issues, all students were instructed on the format of the Chicago style as part of the course. They were also introduced to EndNote, a software package to help them manage their bibliographies and citation styles.
Despite this, three samples did not follow the Chicago style, despite explicit feedback in this respect. A large number of the other samples displayed inadequacies. These included missing entries, incorrect ordering of entries, and inappropriate use of block letters. The example (11) below illustrates some of the common problems.
(11) Final version: Obenshain, Kate. DIVIDER-IN-CHIEF, THE FRAUD OF HOPE AND CHANGE. United Statesof America: Regnery Publishing,Inc., 2012.
The block letters in the title and the place of publication are inconsistent with the Chicago style; they reflect a lack of familiarity with the conventions of the style. However, the other problem in (11) – concerning the missing spaces in ‘United Statesof America’ and ‘Publishing,Inc’ – reflects simply the carelessness of the student.
Problems with citations likely stem from the students’ grappling with the norms of academic writing, in which the proper listing of references is mandatory. As the students were freshmen, many were probably still familiarizing themselves with this genre of writing, and it is more than likely that some found the process of incorporating scholarly sources in the writing confusing and burdensome.
Recommendations and Conclusion
As an investigation of the revisions by students based on the feedback they received, this study is helpful in highlighting what they perceive to be important, and what teachers can do to improve inadequacies in other areas. The focus on the FT and FP levels of Hattie and Timperley’s (2007) framework has been valuable in highlighting the extent of students’ response to teacher feedback in specific ways. As we have seen, the findings of this study are mixed. The finding that students responded best to feedback on RS – on the arrangement and revision of information (i.e. FP) – is heartening. It signals a level of maturity, with the students displaying a tacit understanding of the importance of coherence and logical development in their written work. This should continue to be reinforced in the writing classroom and in teacher feedback.
On the other hand, the issue concerning superficial corrections needs more careful attention in at least two areas. First, where RS is concerned, the feedback should serve to signal not only areas for improvement, but, more importantly, the role of the teacher as first a reader, and only then, as an evaluator (Leki, 1990; Sperling, 1996). Such a role should be discernible from the feedback, and so heighten the students’ awareness of how written assignments are processed by the teacher. This is particularly crucial in addressing RS problems. Issues on coherence and logical development are unlike grammatical errors; whereas the latter are governed by rules, the former are judged by the reader’s sense of whether the various points in a text form an integrated whole. Teacher feedback, with a clear emphasis on how a particular piece of writing may be perceived by a reader, can go a long way to help students move from being mere writers to careful readers of their own work, and so detect RS problems on their own. They become ‘more discerning, more intuitive, more analytical, and generally more able to create, independently, productions of high quality on demand’ (Sadler, 2013: 62).
An understandable point of concern is the implications this has for the balance of power in the writing-feedback exchange, since the views of the reader can always be challenged or refuted by the writer of the text. Some teachers may be uncomfortable with this. We are of the opinion, though, that at the tertiary level, students should be encouraged to defend their work if need be, and discuss contentious issues openly with their teachers. While this might seem disconcerting to some, it is through dialogues of this nature that both the teacher and student receive valuable input. The teacher will understand more about what the student intended to convey, and so be in a better position to offer advice on how it can be achieved more effectively. The student will also understand how the writing can be perceived differently by various readers (including the teacher), and so learn from the experience to avoid problems in future writing tasks.
Second, where grammatical corrections are concerned, the current feedback, as in (4–6), does not appear adequate. It would have been difficult for the students in this study to respond to such feedback without knowing where they had gone wrong. The use of labelled or coded corrective feedback may be the way forward. Although some scholars have argued against the use of corrective feedback (e.g. Truscott, 2007), studies have provided some evidence that students receiving corrective feedback do improve in accuracy over time (Bitchener, 2008). While the feedback should not be exhaustive, it should nevertheless be explicit, particularly on specific errors in grammar. Providing labelled feedback (e.g. ‘subject-verb error’) or coded feedback (e.g. ‘art’ for ‘article’) is one way to prompt students to find out more about a particular error and how it can be corrected (Sampson, 2012). We concede that views on the use of labelled/coded feedback are still mixed, but based on the analysis in this study, such feedback is felt to be necessary, given how the majority of students continued to commit grammatical errors even in their final version.
Students’ response to FT involves formatting issues as well. The problems with citation styles appear easy to rectify. They have less to do with teacher feedback than the use of available software programs to help students format their bibliographic entries uniformly. The inconsistencies noted in the analysis suggest that the majority of the students created their bibliographies manually, but there is little reason to do so when resources are freely available to ease the process. More crucially, teacher feedback should continue to highlight an often ignored aspect in students’ written work – its appearance and adherence to conventions. This issue is hardly raised in the literature, but it is something that writing teachers encounter on a regular basis. While content and language accuracy are important, essays that are carelessly styled or formatted, as in (11), can leave a negative impression. Indeed, careful attention in this area goes some way to train students to write and present their work in an appropriate style.
In view of the small corpus, the findings of this study are naturally tentative. Much further work remains to be done. It is hoped that the study will serve as an encouragement for similar action research involving a larger corpus of writing samples from other disciplines.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.