Tuesday, 22 February 2011

Writing feedback: how do you do it?

When it comes to student writing, we all know that feedback and error correction are important.  Sometimes it might just be a vague comment at the end:

Well done, Roberto. A very good piece of work, but next time don't forget to use more adjectives!

This is not very useful.  The student is likely to read the comment and instantly forget about it and wonder why they put all that effort into writing their piece in the first place.

To combat this vagueness, popular advice to teachers in recent years has focused on getting students to think about their own errors, for example by the teacher underlining them and asking the student to go away and work them out and re-correct them.  The problem with this is that it is highly time-consuming and very difficult for the student (if they knew their mistakes perhaps they wouldn't have made them in the first place).  Some teachers use a complicated set of abbreviations in order to point the student to what kind of error they made,

                sp. = spelling error
                lc = use lower case
                cap. = use capital letter
                v. = wrong verb
                adv. = wrong adverb
                ? = I don't understand

                                               ............ and so on.  The trouble with this is that you would have to do a lesson or two teaching them what all the symbols meant.  More importantly, how can you guarantee that the learners would actually bother to look at these pointers and correct the work?  Maybe a few conscientious ones will re-draft at home and hand it in again the following week, but the majority will probably file it away and forget about it.

Imagine if we corrected speaking this way (!), shouting out "VERB!" and "PRONUNCIATION!" as the poor student tried to articulate what he was trying to say.  The quickest way of correcting a speaking error is on the spot, by reformulating what the student said. A rather crude example:

                               -    Student: I go London last week
                               -   Teacher : You went to London last week?
                               -   Student: Yes, I went to London last week and....

The hope is that over time, and once the same errors have been corrected multiple times, the student will eventually assimilate the correct usage. There is no reason why this approach cannot be used for writing skills too.  Myers (1997) in an article for TESL-EJ, detailed how she used reformulation to correct and improve students' sentence-level syntax.  The student writes a first draft, the teacher corrects all the errors  (spelling, grammar etc.) and in addition suggests more 'natural' alternatives.  The student then looks at the corrected draft and re-writes it without the original errors, producing what Myers calls a "clean copy."  Myers then distributed the clean copies amongst the class so they could read each other's work without being impeded by errors, grammatical or otherwise.

Myers used this technique with ESL students in the US studying at university level.  I argue that this technique can be used at any level and for whatever kind of written text.  I have produced some notes explaining the basics:

Reformulation for ESL Writing: A Guide

The advantages:
- Although the text is corrected by the teacher, the 'voice' and meaning remain the student's.
- The student receives feedback that is 100% individualised.  No-one in the class is left out.
- It can be instantaneous, in the same class period (no waiting for the next class for feedback).

Further reading:

Reformulation Revisited (Kusuyama, 2003)


  1. "...(W)e all know that feedback and error correction are important"
    If by "important" you mean it leads to improved accuracy, I'd disagree - there's not a lot of good evidence to show that it's effective. That's not to say we shouldn't do it, of course. Can't beat a good sugar pill. But it doesn't really work.

  2. Thanks for the comment.

    I would say that it has mostly worked for my classes, with improved exam results to boot. Students also rightly expect some sort of feedback on their work, whether it is about their errors or their writing style. Might as well do this in an efficient manner instead of in a haphazard way.

    The 'reformulation' approach is a not an ideal, it is pragmatic, given the often limited amount of time we have with students.

  3. Hi
    Many of us can come up with anecdotal evidence of successful correction leading to improved performance, but the empirical evidence is lacking - at least, I've not seen it, and there are a number of writers, not least of which is Michael Lewis, who reject the popular, seemingly common sense idea that correction leads to improvement.
    Where evidence exists, it actually suggests that correction leads to worse performance, reduced confidence and confusion.
    I have mixed feelings about the idea that, just because students expect it, teachers should provide correction; and if we don't correct, that frees up time to give other, more constructive (and effective) feedback.
    Reformulation isn't especially effective given that students don't always see it as correction, or even repeat the correct formulation, nevermind remember the mistake in the future.
    Again, we can all provide anecdotes, but the evidence doesn't really support what we are doing.

  4. An interesting idea, but I'm not convinced. Are you suggesting that correction codes are a "haphazard way" of marking students' writing? Of course there are many different ways to peel a banana, but I've found in the past that students really appreciate and benefit from correction codes.

    As for submitting just one draft of a writing task then copying it out again after the teacher had made corrections, I can't speak for all learners, but if it were me, I think I might feel a bit affronted, or even patronised, if my teacher expected me to compile a portfolio of written work that I didn't really feel I'd produced by myself. And I certainly wouldn't be able to kid myself that this reflected my 'progress over the course', because the accuracy and appropriacy of the writing in my portfolio would, to my mind, reflect the teacher's proficient writing skills, not mine.

    I'm afraid this is a somewhat slow response to this post (I discovered it a bit late), sorry. But thanks for the food for thought - I've been looking for some inspiration for a new blog post myself! :)

  5. Hi Laura, thanks for the comments! I will reply to this and to your blog post as soon as I can...

  6. Hi, just come across your blog. I like your pages of resources and stuff, nice one.

    I like the idea of reformulation and I have used it before with students of various levels and ages, but I wouldn't want it to be the only method I used. I like to give feedback on writing in a few different ways as I think it helps students think about writing in different ways, covers for the fact that some students may react better to different methods and also so that one way doesn't get too repetitive. So, how about sometimes using code and other times reformulate?

    One method I tried with FCE students was Russell Stannard's screen capture video method, they loved it and said that it really helped them to think about their errors.

  7. Hi thanks for the comments Richard. I think you're right that reformulation shouldn't be the only method used, and after Laura's comments I might become kinder towards a stripped-down version of codes (3 maximum). However, I still maintain it is the most useful in terms of time-saving: The whole thing can be done there and then and neither the student nor teacher have to wait weeks for the final product.

  8. Dear Richard,

    We all want to improve students' writing but if a teacher fixes all the mistakes, even if you get the students to write them again, they won't learn.

    At the same time, you don't want to spend more time reading the essay and fixing the mistakes, which the students' only spend half the time writing, when you know that it won't help them.

    I have been using the codes to let the students know their mistakes, and get them to correct them by themselves first, hoping that they 'realize' their mistakes. It has been working quite well for my students. It takes time to use codes at the beginning. You have to tell them all the codes. However, it only takes 10 min. of your class. Moreover, every year by the time students got the message that I don't put up with those careless mistakes, they start writing carefully.If there are still too many mistakes, particular students need to submit the writing again and again. So, their draft starts to improve. Then, I get them to do unseen impromptu writing task in class. Every year I can see great improvement.

    I do not agree with your comment on "if they knew their mistakes perhaps they wouldn't have made them in the first place". I think they still do make mistakes. Sometimes it fossilizes but it is understandable that native speakers do not question. I feel that using coding is still one of the good ways to improve students' writing.

    I am not a native speaker of English. I often forget to put 's' and 'the' etc. However, I would be careful if I knew that the teacher won't put up with it.

    By Kao

  9. Hi again
    The difficulty we face when assessing the impact of any pedagogical intervention, such as correction, is that our students are exposed to a wide range of inputs, any of which may prove to be the cause of improvement. It's possible that students' writing improves as a result of reading, or grammar instruction. Just because a group of students improved doesn't mean that a particular teaching technique was the cause of that improvement.
    There have been many studies carried out over the years, and, as far as I am aware, there is little good evidence to show that correction, in whatever form it takes, has a positive impact.
    So why do teachers continue to do it? Perhaps many of them are unaware of the research. But I also think that correction is deeply ingrained in the culture of teaching (and learning), based on a simplistic model encapsulated in the sentence, "if they knew their mistakes perhaps they wouldn't have made them in the first place". We know that learning doesn't work that way, but have yet to apply this understanding to the techniques we use to teach.

  10. Kao:

    Thanks for your comment. I think it was aimed at me and not Richard who is actually an advocate of codes! Anyway, really interesting to read your method of using codes, and if, like you say, you see "great improvement" then it's obviously working for you and your classes. I never actually dismissed codes outright. It is a sensible method, but no method is perfect.

    Reformulation is aimed at highlighting sentence-level syntax errors. For other concepts, such as writing style and content, different work is obviously required. Furthermore I would point out that reformulation should not be the central focus of writing classes and only one way among many of trying to improve syntax and vocabulary.


    "It's possible that students' writing improves as a result of reading, or grammar instruction"

    Of course. I don't think I suggested otherwise. Both of these would help enormously.

    Correction and feedback are indeed ingrained in the culture of teaching and there may well be many negative aspects to it. I advocate learner autonomy as much as possible. However, both institutions (through teacher observations, for example) and students demand it to varying degrees. This cannot be escaped from. Even in the absence of that demand, it would be churlish for a teacher not to occasionally point out errors just as would be to never give some kind of positive encouragement.

  11. Hi again Mark, just came across your 25th February comment. Sorry it wasn't published earlier, Blogger designated it as 'spam' for some reason. Anyway I've put it up now. Very interesting comment and I will check out Michael Lewis. About "evidence": there is a lack of evidence for most things that teachers do, is there not? Most academic books on education are pretty inconclusive or parade a series of questionable "theories" such as "VAKT" and "learning styles". I've ordered John Hattie's "Visible Learning", a summary of 800 meta-analyses of educational research in an attempt to point towards the holy grail: evidence based teaching and learning. Ok, his book is about education in general and not specifically ESL, but I look forward to digesting its contents and will put up a review here once I've read it.

  12. Hi There, Nice Blog! This post is really good and blog is really interesting.It gives good details.Thanks for sharing this nice post. esol courses.

  13. A person who wants to learn would want him corrected right away rather than live him wondering were is the mistake. I am also like that I would love you correcting me on hand rather than think that I have solve the mystery and ending up being dismay since I was again wrong. That is what is wrong sometimes with people who teach english or any other subject they all have to let the student do it all since they think of it as learning while in fact if you ask me interacting would be the best way of learning.

  14. While it does depend on the types of students you have and how the class works, I still think that reformulation is a pretty good strategy, and it doesn't hurt to try it out! When I was learning a second language and still a beginner, my professor used the reformulation method for our longer writing assignments. By the time I reached 2nd year, or intermediate, he switched to a no first draft policy (perhaps to challenge us), but having reformulation in my beginner years really helped me.

  15. Thanks for your comment. Yes, I think it is especially useful for beginners.

  16. this is very useful post.English is an international language all over the world.. i always suggest people to go to this web sites for learning english language.
    [url=http://www.professays.com/] custom research paper [/url]

  17. That songs great, we all need to find a more efficient approach to students' writings than stick to the traditional method.

  18. The way of correction is great but doesn't suit all the students. It would be a mistake to use it by default with the assumption it would make every student happy...

  19. Please do more because I really enjoyed reading your article on this.
    order essay

  20. Please do more because I really enjoyed reading your article on this.
    uk writer

  21. Hi, I'm a business english teacher. I teach business writing and I do a lot of writing feedback. This is a helpful post which I'm planning to apply in my next business writing classes. Keep up the good work!

  22. Innovative method. However, the concerned people should be mature enough to handle and give constructive feedback. Also, this method can be helpful for students whose native language is not English or those who have not studied English as a language.

  23. I know that students really appreciate this very individual-student-centred correction. And as an English teacher, I have done this too in the past, but could only manage the time to do it once for each student during the semester. The time involved to complete it for every student for every writing assignment made it impossible for me to do it more frequently.

    I am studying a foreign language at the moment, and I would love it if the teacher were able to do it for me for every piece of homework submitted, but she can't. In place of that I've substituted my own research with Google Translate. After doing my own translation of the English into Spanish, I put the English version through Google translate and check whether its translation is the same as mine. This method gets me quite close to the holy grail of natural foreign language expression. Then I cross-check Google Translate's translation by doing a search on-line. I can see then if what Google Translate has provided is accurate. To check my text this way, I have to do all the work myself, which means that since I put the effort in, I may be able to retain what I learn longer. Well, that's the theory, anyway.

    Thanks for your post! It certainly provides food for thought.

  24. Learn spanish at bazilinga. Because this is one language which can be very beneficial for you while traveling.

  25. Being in the field of languages, I really like some of your article. Please continue with it. Foreign Languages in India LanguageNext