Marking students’ written work: principles and practice

© David Nott, University of Lancaster

Table of contents

• 1. Introduction
• 2. The task
• 3. Feedback
• 4. The student and the tutor
• 5. The marking process
• 6. Continuous writing, year 1 (sample of course work assignment)
• 7. Written work based on the year abroad (assessment criteria)
• 8. English to French translation, year 4 (sample of unseen classwork assignment)
• 9. Conclusion
• Bibliography

1. Introduction

1.1 Scope of this article

This article is intended to be a practical guide for teachers and tutors of Modern Foreign Languages (MFL) and English as a Foreign Language (EFL) when marking students’ written work, including continuous writing in the Target Language (TL) (free or guided composition and structured exercises such as résumés), language-centred exercises in the TL, translation from and into the TL, and essays (in L1 or in the TL) for socio-cultural studies courses. Assessment of oral work, and feedback on oral performance, are not covered.

Marking is considered here as a means of giving feedback to students (see 3) in the context of formative assessment, semi-formative assessment, and non-assessed work. Marking of work done under test or examination conditions for summative assessment (for final or end-of-course examinations) where markers’ symbols and comments are not seen by students, and the work is not given back before the end of the course, is not covered.

1.2 Definitions

Semi-formative assessment: written tasks closely modelled on those in the final examination and marked according to assessment criteria set out in the course description. The way the practice piece is marked (how these criteria are applied, the symbols used, the comments provided) provides guidance to students as to how the corresponding task in the final examination will be assessed.

Formative assessment: work (such as course work assignments) which counts towards the student’s mark or grade for the course, and which is returned to the student within a few days, with feedback in terms of symbols, comments and a mark or grade.

Marking of non-assessed work: this follows the same principles as formative assessment, in terms of symbols and comments, and speedy return of the marked work; it may not always appropriate to give a mark or grade.

Direct and indirect correction: direct correction consists of writing the correct letter(s), word, phrase or construction directly on the student’s script; indirect correction consists of indicating that there is an error, without writing the correct version on the script. For example, errors can be: coded; circled; underlined; underlined and coded; underlined + description of error; counted in the margin, but neither marked nor coded (see Guénette 2007, p.7). The present article assumes that tutors will wish, when marking, to use indirect correction wherever practicable, and reserve direct correction for phrases requiring reformulation on syntactic or stylistic grounds.

2. The task

No system of marking, however meticulously conceived and rigorously applied, can be of practical use to the student unless the task itself is appropriate. Marking has the best chance of playing a formative role when it is applied to a task directly related to the preceding teaching and learning process. For marking to be meaningful, the parameters of the task need to be clear and applicable to all students: depending on the task, this involves setting word or time limits, and possibly requiring all students to cover a preset number of steps or points. Failure to set clear and fair parameters can, in extreme cases, mean that the tutor is faced with a set of scripts which are so disparate in length and range that they cannot be marked fairly and consistently.

Not all written tasks are suitable for marking: short in-class tests and quizzes as follow-up to a teaching and/or learning sequence, a mini-lecture or a structured class discussion call for immediate closure: students check their own or each others’ answers and the tutor does not normally see the work. Providing a wide spectrum of tasks involving different degrees of intensity of marking and carried out, as appropriate by the tutor, the student, or another student, helps to emphasise the formative role of marking.

2.1 Collaborative writing tasks

As well as being suitable for oral tasks such as pair or group presentations or exposés in class, working in collaboration can be extended to written tasks. Although the system of school and university degree examinations places a high premium on individual performance, in the world of employment, many intellectual, managerial and manual tasks are prepared and/or executed in groups. (In both environments, of course, end-of-task evaluation and appraisal necessarily focus on the individual.)

The common objection to setting group tasks for assessment, namely that it enables weaker students to piggy-back on the skills of stronger students, has weight, and has to be addressed:

• in a degree scheme, only a minority of marks should derive from group assessments, and a cap can, if need be, be set on these;
• working on a task in pairs or groups, in the absence of the tutor, or with occasional guidance, puts the students at the centre of the learning process, and as such can be beneficial both to stronger and to weaker students;
• each student can be given editorial responsibility for a separate section of the written piece, enabling an individual mark to be given to each student;
• students can be encouraged to proof-read each other’s section: playing the part of tutor in this way can help students to be more effective when re-reading and redrafting their other pieces of written work.

2.2 Collaborative marking tasks

If the practice of collaboration between tutor and class, and between student and student is well-established, it may be worth providing opportunities for students to practise collaborative marking. Hagège (2005, p.87) writes of the attractions of this activity for pre-adolescent children: ‘Plus les enfants sont jeunes, plus les corrections qu’ils se suggèrent les uns aux autres à l’exemple et sur l’initiative du maître sont marquées d’un esprit de coopération, et moins elles sont agressives ou dépréciatives, si l’on en croit le témoignage des instituteurs’. With post-adolescent students, the process of collaborative marking can be more structured and reflective, giving students practical experience in applying assessment criteria.

2.3 Assessment criteria

Criteria for assessment normally cover a number of areas of performance (content, understanding, organisation, expression, accuracy, etc) and set benchmarks for the level of performance achieved within each area. (Although this article is a guide to marking, not a discussion of assessment criteria, an example of assessment criteria for written work based on the year abroad is given in 7.) A scheme of assessment based on clear performance criteria, drawn up by the FL teaching team, gives credibility to individual tutors, the team, the department and the institution. It can be particularly beneficial if the criteria, and the scheme of assessment, apply to each FL taught in the institution. If criteria are based on a realistic set of expectations as to students’ performance, and are made clear to all students before marking takes place, they can help to make marking fair to all students, and to make feedback (for example in the form of comments at the end of a piece of work) understandable and acceptable. Marking carried out without reference to explicit criteria risks being unreliable, and is certain to appear to students as arbitrary and unfair: if it is seen by tutors and students as no more than a set of mechanical, arithmetical procedures, together with a number of conventional signs and symbols, it is unlikely to play its full part in the learning process.

3. Feedback

The principal purpose of marking is to provide students with feedback on their performance: marking thus stands at the sharp end, at the point of implementation, of formative assessment. When marking students’ work, it is important to interpret and implement the assessment criteria in a form which suits the needs of the student, and not solely those of the system. How is feedback provided through the marking process?

3.1 Preliminary feedback

All tutors remind their students to check their work before handing it in; few, perhaps, find out how individual students interpret and implement this instruction. How many students make a series of systematic checks for verb endings, adjective agreements, tenses, and so on? Providing early feedback, at intermediate stages in the learning process (such as making comments and suggestions on a draft plan or specimen page, or giving marks in an in-class test), helps bridge the gulf between teaching/learning and assessment, and provides a model for critical review of work in progress which the student can gradually learn to implement independently of the tutor. Leki (1990, p.64) refers to results of a survey by Freedman (1987) showing that ‘writing teachers considered successful agree that intervening during the writing process helps student writers improve’. As Leki points out, ‘multiple drafting with comments on intermediate drafts is an obvious first step away from making each assignment into an isolated miniature test’ (1990, p.64).

3.2 Marking as feedback

Even though the first thing (and in some cases the only thing) they look at is the overall mark or grade, students, as a general rule, welcome feedback: ‘our students expect feedback from their teachers and generally feel that it helps them’ (Guénette 2007, p.11). Written comments on and at the end of a piece of work are an essential part of formative assessment. In the process of awarding points and converting points to marks and grades, the tutor is applying the criteria and the marking scheme fairly to all students’ work; only through written comments on a piece of work can the tutor put each student’s individual needs, at this stage of the course, first.

Commenting on content, organisation and style: comments at the end should refer first of all to features deserving praise and/or encouragement; comments on negative features should be cast in terms of what remedial action can be taken: for example pointing out how the material could have been more effectively organised, or indicating which stylistic features detract from the overall impression and convincingness of the piece.

Commenting on points of language: ‘Students should be left under no illusion that an error is simply a thing to be avoided, like an obstacle on the road; rather, it is a symptom of a fault that has to be diagnosed and adjusted in the vehicle itself – in the learner’s interlanguage’ (Nott 2000, p.242). Simply indicating in the text or in the margin an error that arises from the student’s incomplete grasp of a point of grammar or usage is unlikely to make any difference to future performance. Furthermore, feedback has to be provided in a form and at a level suited to the individual student: ‘Han’s (2001) longitudinal case study reveals that individual differences are an important variable in whether corrective feedback is successful or not. As a result, Han calls for fine-tuning the feedback to the learner’s interlanguage problems in order to directly target the causes of errors’ (Guénette 2007, p.11). Surveys referred to by Leki (1990, pp.62-63) show that some students prefer to have all errors pointed out, while others respond best to supportive comments, which suggests that tutors should provide both kinds of feedback on each script, but not in a one-size-fits-all formula. Diagnosis of errors at the end of a piece of work should be constructive, and focus on no more than two or three areas, showing what the student appears not to have understood and suggesting what needs to be done, including explicit references to grammar manuals and exercises. Making comments and recommendations on each and every shortcoming in linguistic performance will only discourage the student from taking any remedial action at all, and in any case uses up tutor time which would be better spent making a small number of specific recommendations. However, it can be worth reminding students that their work is not being produced in a void, nor for some transcendent reader, but for the class tutor and, for summative assessment tasks, a second marker and/or an external examiner. These markers will be carrying out the very process which students should simulate, at the final stage of checking their work, by reading the piece through as a whole and assessing whether it makes sense and is clearly expressed. (Students, in their blogs, and tutors in their academic papers, are alike accustomed to producing texts which will be read by no more than a handful of readers.)

3.3 Beyond feedback

Reading a batch of scripts will reveal a wide range of areas for improvement, some of which can probably be ascribed to the current state of individual students’ interlanguage. But if a particular point of grammar, style or organisation crops up in several scripts, this should be taken by the tutor as an indication that remedial action is needed on a wider scale – for example by using a few minutes of the current or the following week’s session to micro-teach one or two points to the whole group.

4. The student and the tutor

Marking has perhaps too often been regarded as synonymous with correcting (cf., in French, ‘corriger des copies’), as if the tutor’s task was to provide the student with a complete ‘correct’ version, written in between the lines. However, ‘[t]he role of the teacher is not merely to point out students’ errors, but to collaborate with students in their own investigation of the source in their interlanguage of certain errors’ (Nott 2000, p.242). In itself, the act of marking ‘corrects’ nothing: errors can be corrected (in the sense of not being repeated later in the same or a similar context) only by the students themselves.

4.1 The tutor as coach and as assessor

For marking to be acceptable to the student, let alone effective in improving written performance, it has to take place in a context of mutual trust and understanding between tutor and student. Following Knoblauch and Brannon (1981), Leki (1990, p.63) observes that ‘[a] great deal of research has focused only on comments teachers make on students’ papers without considering the ongoing dialogue between student and teacher’ and that ‘the problem is not the annotation but the entire teaching environment’. What kinds of spoken comments does the tutor make to the class, and to individuals, on students’ spoken and written performance? What goals does the tutor set for the group and for individuals? Is the general atmosphere of the classroom one of collaboration (between tutor and students, and between students) towards achieving specific goals? Following Cowan (1977), Leki (1990, p.59) points out that ‘[t]he role of the writing teacher is schizophrenic, split into three incompatible personas: teacher as real reader (i.e., audience), teacher as coach, and teacher as evaluator’. In practice, ‘teachers must continue to live in this contradiction of trying to be both collaborator and judge’ (Leki 1990, p.60). The tutor should be open with students about this multiple role: it would seem fair to suggest that marking is most effective when seen by students and tutor as part of an ongoing dialogue between them. In other words, students’ response to the tutor as assessor/judge reflects their response to the tutor as coach. One further point: the dialogue between tutor and student is more likely to be open and open-ended if everything the tutor writes on a student’s script is in pencil and not in (red) ink: linguistic judgments, like linguistic phenomena, are relative and provisional.

4.2 Calibration

It is important for tutors to be self-aware, and self-critical, in respect of their attitude to marking in general, and to errors made by students. Tutors’ intuitions as to what constitutes linguistic acceptability can differ widely, and be deeply ingrained; for example, many studies show that non-native speaker correctors tend to be more severe in grading than native speakers, and more likely to regard stylistic variations as errors. What does the tutor expect students to gain from the task and from the process of assessment and feedback? How can tutors be sure that their evaluation of students’ performance of the written task are in line with the published criteria for assessment? Referring to a number of studies dating from 1989 to 2001, Hamp-Lyons (2003) states that ‘[t]he evidence from such studies suggests that, left to their own judgments, raters cannot agree on the absolute quality or the relative quality of essays, nor can they agree on the specific qualities in essays that make them good, worse, or worst’ (p.178); among the specific findings referred to by Hamp-Lyons are that ‘raters are influenced by their own cultural contexts and learning/teaching experiences perhaps as much as by the variation in quality of student essays’, and that ‘even the most experienced and skilled raters act as individuals, using their own values, even in situations with good and extensive rater training and well-defined criteria’ (pp.178-79). Tutors need to be aware that criteria for performance (in terms of fluency, or communicative competence, for example) are always open to interpretation; applying these criteria is not a straightforward, mechanical matter.

These research studies underline the necessity for consultation and co-ordination among all FL tutors before a new academic year gets under way. When several classes at the same level are taught by different tutors, these tutors should also meet regularly as a team to discuss and calibrate their marking practices, for example by comparing their marking levels and mark spread in respect of a particular course work assignment, and carrying out marking exercises such as the one described in 6. In some cases, it makes sense for students’ written work to be marked by someone other than their course tutor, not only for summative assessment, but also, from time to time, for formative assessment: two tutors could agree to mark a set of scripts produced by each other’s students, or to mark a set of scripts jointly. This should be standard practice when a new tutor is being mentored by a more experienced colleague, but it is also needed in cases where two experienced colleagues take differing views of aspects of students’ performance (see 6).

5. The marking process

FL teaching is not solely or even primarily concerned with the avoidance of error; the tutor’s principal role is to provide situations, opportunities and motivation for all students to stretch their linguistic and sociocultural range and proficiency. In so doing, students will inevitably make errors, only some of which (because they are made repeatedly, or involve frequently-used constructions, or are vital to the communication of meaning) should be highlighted by the tutor for the student to concentrate on (see 3.2), and in some cases made the object of further class explanation, demonstration and practice (see 3.3). Tutors should not expect to see and mark everything a student writes: over-marking can be bad for the student (as well as for the tutor). Marking is not the same as proof-reading: when marking, the tutor should always consider, for each student, how much feedback, and expressed in which forms, this student can be reasonably expected to cope with (see 3.2). Marking exists for the benefit of the student; for the tutor, it is a unique opportunity for giving confidential one-to-one feedback.

While consistency in applying criteria is essential in all marking, the procedure should be varied to suit the nature of the task, the stage reached in the course, and students’ proficiency. Possible variations include, separately or in combination:

• marking a piece of work to indicate where there are errors, without showing, in words or symbols, the nature of the error;
• setting students to hand in a revised version of their work for assessment;
• setting the rewriting task to be done, individually or in pairs, in class or before the next class;
• awarding the final mark for the piece wholly or partly on the basis of the revised version;
• setting students to review each other’s work in pairs, using symbols to indicate the type of error, or simply underlining the error;
• awarding (additional) marks both for the number of one’s own errors correctly identified and corrected, and for the number of errors spotted in one’s partner’s work.

In the case of longer pieces of work, such as reports or projects associated with a period of residence abroad, what proportion of the text should be marked, and what proportion should simply be read for assessment purposes? If the purpose of all marking as feedback (see 3.2) is to provide a student with a manageable number of comments and suggestions, it is not at all certain that a trawl through several thousand words will provide a significant increase in the number of useable suggestions, compared to marking just 10-20% of the piece – just as a sample of millions of informants does not necessarily provide more statistically significant information than a sample of thousands.

5.1 Symbols (‘indirect correction’)

For marking to operate effectively as feedback, the language, conventions and symbols used by the tutor must be clear, concise and capable of being acted on by the student. Much time and effort can be saved if a set of symbols is used: the tutor does not need to write in the correct word, form or phrase each time (though it can sometimes be helpful to provide a more suitable or more idiomatic version of a phrase), and the tutor can see at a glance whether a student’s errors tend to fall into particular categories. For students, the symbols serve as a prompt to think through for themselves the process of checking their work for gender, agreement, tense, and so on, instead of passively seeing the correct forms without doing anything to process them mentally.

In order to avoid the use of symbols becoming an unthinking chore for tutor and student alike, it is worth taking a few minutes to explain the symbols to a class, and how the students are expected to make use of them. Once explained, these symbols should be used consistently. Here are some symbols that can be used between the lines or in the margins of written work in the TL. This list is neither exhaustive nor definitive; tutors may have their own preferences for alternative symbols.

Symbols in the margin or between the lines showing the nature of the error:

Symbols in the text indicating the seriousness of the error:

___   ___

/      /

Symbols in the margin:

5.2 The importance of ticks

We all know, from our experience as pupils and students, how much it meant to us to find ticks in the text or in the margin of our work. These memories may tempt us to use ticks as no more than a means of general encouragement, whereas their value in advanced studies is as a clear, precise (and, where appropriate, numerical) guide, to tutor and student alike, as to the actual level of proficiency achieved in carrying out a particular task, such as continuous writing in the TL, translation, socio-cultural essays in the TL or L1.

For continuous writing tasks, ticks for linguistic proficiency can be given in the text, and ticks for quality of content (ideas, argument, organisation, etc) can be given in the margin. How can one ensure consistency and fairness when marking a batch of scripts? Consistency requires the marker to apply the stated criteria in the same way from year to year, judging each script in isolation, against ‘absolute’ standards; fairness requires the marker to judge each script on its merits, in relation to the other scripts in the batch. One procedure could be to put scripts first of all in provisional order of merit on the basis of overall impression, and then to compare each script more closely with its immediate neighbours. The overall impression given by a piece of writing is a realistic element in assessment, as it corresponds to one aspect of how one responds to a piece of writing produced for professional, leisure or entertainment purposes. But impression alone is a poor guide to the quality of a piece, and should always be moderated by more objective factors in arriving at a mark or grade. The more open-ended the continuous writing task, the less one can rely on counting the number of ticks awarded when putting scripts in a provisional order of merit, but counting ticks can serve as a useful check on what has been achieved (see also below).

In the case of socio-cultural essays (in the TL or in L1), if ticks are awarded systematically, the total number of ticks can be the initial criterion for placing scripts in provisional order of merit. Deductions from the total number of ticks can be made for certain shortcomings in the essay as a whole: too many or too few words, repetitions, obscurities of expression, errors of fact, etc. Overall impression can then be brought into play as a moderating factor. This dual procedure provides an invaluable check for the tutor on how much has actually been achieved in an essay: all assessment, whether formative or summative, should reward solid attainment, not mere facility.

5.3 A possible ticks-to-marks scale for essays

The following ticks-to-marks scale has been found useful as a benchmark for both fairness and consistency in marking socio-cultural essays (in English and in French):

It must be emphasised that the number of ticks cannot form the sole criterion for awarding a grade or mark. However, knowing the actual number of positive, relevant points made in an essay provides helpful evidence when comparing the overall merit, particularly at grade boundaries, of two or more essays with a similar number of ticks or overall impression.

For translations from and into the TL, systematically awarding a tick for each example of felicitous translation (lexical items, appropriate syntax, stylistic features, etc) helps the tutor to take a positive approach to assessing translations (see 8). These ticks should form part of the points to marks/grades scheme for the exercise. A simple formula is to count one positive point for each tick; the total number of ticks is then used to offset the number of points deducted for errors of various types and degrees of importance.

Some traditional practices in marking translations, particularly those from L1 to the TL, come down heavily on certain errors of verb morphology (a ‘three strikes and you’re out’ mindset, reminiscent of traditional practices in French primary schools). It could be argued that this approach is not appropriate, either for formative assessment or for summative assessment; in any case (see 2.3), it is important for criteria to be consistent between the two kinds of assessment. The linguistic and other skills called upon for the task of translation are deep, wide and numerous: not for nothing was the rendering into Greek verse of a passage of English long regarded as the pinnacle of Classical studies. Today, the production, in professional life, of a useable translation or summary requires proficiency in areas such as a wide vocabulary, or sensitivity to how the system of tenses in the TL operates, and students deserve guidance and feedback on many of these skills. Knowing basic verb forms, however, is the responsibility of each student, not of the tutor, and written feedback on a script should, if the case arises, address this unequivocally. Criteria for assessment of written translation should give due weight to a range of factors, so that the expression of criteria in terms of points, marks and grades does not give an unbalanced or distorted picture of what has been achieved in a given script.

5.4 A possible points-to-marks scale for translations

The following points-to-marks scale was used for assessing the sample English to French translation done under examination conditions (see 8):

The above scale, where each percentage mark corresponds to two points, is included solely for the purpose of illustration: it should not be the sole basis for deciding whether a script is to be classified as I, II 1, II 2, etc. But some form of points-to-marks scale is an essential element in ensuring fairness among all scripts in a set. The above scale could, for example, be made more severe (one percentage mark = one point) for course work assignments done under uncontrolled conditions. For translations from the TL into English, the bar for achieving 70% would need to be set higher, at +5 or even +10.

6. Continuous writing, year 1 (sample of course work assignment)

Task: (Rédigez vos premières impressions de la vie et du travail à l’université : installation dans votre nouvelle vie, réactions de votre famille, le campus, le travail, les camarades, les professeurs etc. Vous pouvez, si vous voulez, rédiger votre travail sous la forme d’une lettre, ou d’un article pour un magazine ; dans ce cas, précisez pour qui vous écrivez.) (Nott 1993, p.22)

Marked script (adapted):

Download script [pdf, 579kb]

Comments on script, and mark awarded: (Bien conçu. Assez bien composé, malgré quelques points qui sont répétés. Les phrases sont souvent bien formulées, et faciles à lire. Vocabulaire assez étendu, et approprié. Grammaire : attention aux articles (les, des, etc) et aux accords – et à la construction des verbes + préposition et/ou pronom. 57%)

A student essay similar to the above example formed the basis of a marking exercise involving nine first-year French language class tutors (five full-time academics and four experienced part-time language tutors); the marks awarded to the piece ranged from 44 to 59, with a mean mark of 51 (Standard Deviation= 4.2). At that time, an overall mark of 45+ in a subject meant that the student could proceed to study the subject at Part II as a Major (Single or Combined), while an overall mark of 60+ in a subject corresponded to the highest grade obtainable for first-year work.

7. Written work based on the year abroad (assessment criteria)

Guidelines to tutors:

8. English to French translation, year 4 (sample of unseen classwork assignment)

Semi-formative assessment: unseen translation into French, done under test conditions, without notes or dictionary; marked scripts returned to the students.

Task: Traduisez le texte ci-dessous en français.

English text:

He turned over in bed. His calf was still hurting him and he tried flexing his toes and straightening his leg. As soon as the alarm rang, he put out his hand to turn it off, got up and dressed. Then he sat on the large bed, staring at the wall for at least five minutes. It was a pity it had to end like that, he thought, before their relationship had had a chance to succeed. If only we hadn’t had that argument. He had only said that he would see Mary as well when he reached Leeds. Then it occurred to him that the post might bring fresh hope. He hurried downstairs but there was noting but a few bills and advertising pamphlets. ‘Never mind, by working hard I shall forget her’, he determined to himself, and forgetting he was on a diet, he proceeded to eat a hearty breakfast. Drinking his tea from a chipped cup he went through the bills, wondering how he was going to pay for all he had brought from the local grocer. (J. J. Anstruther, The Boy at Large)

Marked script (reproduced with permission):

Download script [pdf, 52.3kb]

Comments on script: (Assez bonne maîtrise du système des temps et de la construction des phrases ; quelques problèmes de vocabulaire (y compris pour des mots courants) ; peu d’erreurs de morphologie (mais : descendit ; buvant) ; à noter : travailleur ; garder ; tout ce qui/tout ce que.)

Totals of positive and negative points: +7, -25; mark awarded: 61

For translations, especially into the TL, it can be helpful to the tutor to draft beforehand a working copy of the TL version, adding acceptable alternatives (including some from students’ scripts). When returning the work to the students, a final ‘suggested version’ of the TL text can be distributed to students; the tutor should point out that s/he has had far longer to think about the translation than the students have had, and that in any case the tutor’s version is only one of a number of possible renderings.

9. Conclusion

It would be helpful to know that the preceding discussion and examples, drawn from practical experience, were supported by research findings. It would seem, however, that we are still a long way from understanding how, or even whether, written feedback can help to improve students’ subsequent TL performance. Introducing her review of research studies on corrective feedback carried out since the early 1980s, Guénette (2007, p.2) declares that she is ‘not much further ahead’ than when she first started teaching: ‘Should teachers provide corrective feedback on form or should they not? The debate still rages between proponents of both options because research so far has not been able to prove, beyond a reasonable doubt, that providing corrective feedback is a decisive factor in the attainment of language fluency and accuracy’. Truscott (2007), evaluating research into how error correction affects learners’ ability to use the TL in writing or speaking for communicative purposes, concludes that if correction has any benefits, these are very small. In their meta-analysis of research on the effects of oral and written corrective feedback, Russell and Spada (2006, p156) come to more positive conclusions (‘the findings of this meta-analysis support a beneficial role for [corrective feedback] overall’, but the basis of their research criteria is challenged in Truscott (2007).

Research is expected to be based on defined, controlled and repeatable experiments, which means that research findings give rise to only modest claims regarding transferability. In the case of research into corrective feedback, this caution reflects the difficulty of replicating in a research project the ongoing reality of the classroom situation. Research studies are often longitudinal downstream from the point of marking, in the sense that they evaluate students’ subsequent performance. However, unless they are also longitudinal upstream, so as to include how the students were taught in the previous weeks or months, they are unlikely to identify those features (if any) of corrective feedback which improve students’ performance: the number and nature of the variables involved is too great. The problem is highlighted in two surveys of research into corrective feedback: ‘We need more research, especially in L2 writing, to look not only at teachers’ written responses but at combinations of classroom settings, course goals, and grading procedures in order to discover what forms our responses can most profitably take’ (Leki 1990, p.66); ‘The success or failure of corrective feedback will depend on the classroom context, the type of errors students make, their proficiency level, the type of writing they are asked to do, and a collection of other variables that are as of yet unknown’ (Guénette 2007, p.12).

Meanwhile, today’s class cannot wait for tomorrow’s research findings: heads of department and students alike expect tutors to mark written work, and students have a right to receive fair and consistent corrective feedback throughout their course as a guide to the quality of their performance. This article is intended to show that FL tutors, before marking their students’ work, have to acquire the art of coaching them in how to learn, and that FL students have to learn (for themselves, and from their tutor’s example) how to tackle head-on the issue of errors, so that instead of being depressed by the experience of having something marked as ‘wrong’, they are stimulated to do something about it. Hagège (2005, p.36) writes persuasively of how, at adolescence, ‘le goût enfantin pour les manipulations verbales se trouve fortement réduit’, and of the devastating effects of a fear of making a mistake (‘une crainte de la faute’): ‘[la faute], au lieu d’être sereinement assumée comme profitable par la correction qu’elle appelle, est obstinément redoutée comme disqualifiante par le ridicule qu’elle produirait’. Part of the role of the FL tutor is to convince post-adolescent undergraduate students to see errors in their written work not as a fault, a sanction or a failing, but as a guide to their next steps in the learning process. Guénette (2007, p.12) concludes that ‘any type of feedback that does not take the crucial variable of motivation into consideration is perhaps doomed to fail. [...] if the students are not committed to improving their writing skills, they will not improve, no matter what type of corrective feedback is provided’. Russell and Spada (2006, p.155) point out that ‘few studies have investigated the impact of individual learner factors in relation to [corrective feedback]. These include motivation, proficiency level, and age’. Given the intensely personal nature of the FL learning process, it is unlikely that the outcomes of teaching, let alone of marking, can be predicted in the case of any individual student, but this uncertainty is in itself one of the reasons for giving all students the opportunity to derive benefit from having their written work marked.

Bibliography

1. Freedman S. W. (1987) Response to student writing. Urbana, Ill. National Council of Teachers of English.
2. Guénette, D. (2007) Is feedback pedagogically correct? Research design issues in studies of feedback on writing. Journal of Second Language Writing, 16 (1), pp.40-53.
3. Hagège, C. (2005) L’enfant aux deux langues. Paris. Poches Odile Jacob.
4. Hamp-Lyons, L. (2003) Writing teachers as assessors of writing. In: Kroll, B (ed.) Exploring the dynamics of second language writing. Cambridge. Cambridge University Press, pp.162-89.
5. Han, Z. H. (2001) Fine-tuning corrective feedback. Foreign Language Annals, 34, pp.582-95.
6. Knoblauch, C. H. and Brannon, L. (1981) Teacher commentary on student writing: The state of the art. Freshman English News, 10, pp. 1-4.
7. Kroll, B. (ed.) (1990) Second language writing: research insights for the classroom. Cambridge. Cambridge University Press.
8. Kroll, B. (ed.) (2003) Exploring the dynamics of second language writing. Cambridge. Cambridge University Press.
9. Leki, I. (1990) Coaching from the margins: issues in written response. In: Kroll, B. (ed.) Second language writing: research insights for the classroom. Cambridge. Cambridge University Press, pp. 57-68.
10. Norris, J M and Ortega, L (eds.) (2006) Synthesizing research on language learning and teaching. Amsterdam. John Benjamins.
11. Nott, D. (1993) Points de départ. Manuel de l’étudiant. London. Hodder & Stoughton.
12. Nott, D. (2000) What language should we aim to teach, how, and why? In: Guillot, M.-N. and M.-M. Kenning (eds.) Changing landscapes in language and language pedagogy: Text, orality and voice. London. Association for French Language Studies in association with the Centre for Information on Language Teaching and Research, pp. 227-248.
13. Nott, D. (2005) Translation from and into the foreign language. LLAS Good Practice Guide. Available from: http://www.llas.ac.uk/resources/gpg/427 [Accessed 26 February 2008].
14. Russell, J. and Spada, N. (2006) The effectiveness of corrective feedback for the acquisition of L2 grammar. In: J M Norris and L Ortega (eds.) Synthesizing research on language learning and teaching. Amsterdam. John Benjamins, pp.133-164.
15. Truscott, J. (2007) The effect of error correction on learners’ ability to write accurately. Journal of Second Language Writing, 16 (4), pp.255-272.

Referencing this article

Below are the possible formats for citing Good Practice Guide articles. If you are writing for a journal, please check the author instructions for full details before submitting your article.

• MLA style:
  Canning, John. "Disability and Residence Abroad". Southampton, 2004. Subject Centre for Languages, Linguistics and Area Studies Guide to Good Practice. 7 October 2008. http://www.llas.ac.uk/resources/gpg/2241.
• Author (Date) style:
  Canning, J. (2004). "Disability and residence abroad." Subject Centre for Languages, Linguistics and Area Studies Good Practice Guide. Retrieved 7 October 2008, from http://www.llas.ac.uk/resources/gpg/2241.