Correction and Timing

After a discussion in the comments section of Chiew’s Dogme Diaries, I felt that my thoughts on correction in second language learning were too many to post in another comment, so I am making a first video post as part of my previously-shelved return to blogging. The task is enormous and far too much to speak about in 9 short minutes of video. For every error there seems to be a theory and a whole load of procedural knowledge to add to it, so this is very much the tip of the iceberg!

After a little thought today I have decided to add a reflection template for teachers to use during lessons to use to identify, categorise and prioritise language learners produce. I use the four boxes to note down bits of language I hear or notes on areas of difficulty. In this stage, I decide whether to provide immediate feedback or to wait and exploit the language point in later feedback; if, on the other hand, there’s no immediate need to focus on it – third person ‘s’ or misplaced stress on appreciate for example – I have a record of these still on paper for the future.

22 thoughts on “Correction and Timing

  1. Chris Ożóg says:

    Hi Dale,

    I get the impression that you lean more towards immediate error correction (or a balance of both), which is fine, but one thing I’d add here is when in the lesson you’d consider doing it. Different stages call for different techniques to be applied, which is something teachers should bear in mind. For example, would you

    1) correct learners on multifarious and unrelated errors heard in a warmer?
    2) correct learners on using “will” if you were doing a controlled practice on the present perfect?
    3) correct learners before a language point had emerged, if you prefer your Dogme or deep-end TBL?

    My guess would be a no to the first and second questions (though feel free to immediately correct me if I’m wrong), and I won’t assume anything on the third. After all, if the aim of stage is connected to fluency and this is being done in a meaningful communicative way, then is immediate error correction useful at all, or will it obstruct the building and communicating of the message? If, however, you’re actually practising a language point and they get it wrong, well, that might call for a different, more immediate, approach. Personally, when I was learning Spanish, I remember being corrected immediately (on “ser” and “estar” of all things) while trying to tell a complicated story at the start of a lesson and it infuriated me as that’s not what I was focusing on in that communicative moment (I was about B1 level then).

    Thinking about one of your questions at the end, about product and process, it seems to me that immediate correction would sit more comfortably with the former, while the latter would be a better bed-fellow with delayed. However, it really all depends on how you do it. If you simply correct a learner immediately (providing the correction), you’re really helping no-one in my view; if, however, you try to get the learner to work it out, then perhaps there’s more benefit. Regrettably, in my experience of observing teachers, I tend to find that immediate error correction follows this path and the implicit suggestion is that errors should not be made (see audio-lingualism, etc), while delayed correction suggests the teacher sees error as a process to promote more awareness-raising and noticing, as it pre-supposes more cognitive work to analyse the error and why it is wrong.

    It thus seems to me that any discussion of error-correction necessitates the teacher puts their beliefs about learning on the table in order for the proceeding discussion to really make sense. My general (though by no means my only) approach, unless practising a specific language point, is that if something does not impede the message, then I leave it alone in the moment. After all, I see my role as to help learners do the best they can to communicate (which does not mean being ‘soft’), not necessarily to make them 100% accurate (did you know you could still get the odd past simple wrong, for example, and still get an IELTS 8, a score which means your English is excellent?)

    A final anecdotal musing for you. My girlfriend is a non-native speaker and immediate error correction with her does not work at all. I know it’s very different, particularly as we aren’t in a classroom, but if I were to correct her (which I almost never do), she’d make the same mistake 6 minutes later. I have, however, raised her awareness of the types of errors she makes, such as using present tenses prefixed by time adverbials to talk about the past (“yesterday I tell him..”), by simply telling her about it long after a conversation has taken place. She often then concentrates more on the area identified and starts self-correcting. Make of that what you will.

    Anyway, long may you and Chiew enjoy your blogging wars… Maybe I should’ve blogged this too!


    • dalecoulter says:

      Hi Chris,

      First of all thank you for such an in depth comment! There’s a few areas I want to reflect on with some personal experience and anecdotes too:

      “Personally, when I was learning Spanish, I remember being corrected immediately (on “ser” and “estar” of all things) while trying to tell a complicated story at the start of a lesson and it infuriated me as that’s not what I was focusing on in that communicative moment (I was about B1 level then).”

      Totally agree – it’s infuriating when firstly it’s something you have processed and just don’t have full command over when communicating a complex message. I felt the same in Italian when I was learning some of the more difficult aspects. My general approach is to focus on what impedes their message… consider if it’s something they know or something they need to know to help and then let them come to a natural pause in the message when it’s safe to dive in and help. I have similar experiences of observing teachers in the sense that they see error correction as fixing a broken system instead of improving a functioning one.

      I remember observing a teacher chatting with a learner and during the conversation the learner said “no I don’t vote for him in election last year but he is president” to which the teacher responded “I didn’t vote for him, it’s the past”. The student then parroted this and the teacher praised and moved on. This is the kind of on-the-spot correction that saddens me; it negates the affordance completely.

      “If you simply correct a learner immediately (providing the correction), you’re really helping no-one in my view; if, however, you try to get the learner to work it out, then perhaps there’s more benefit”

      Also back to my own experience of learning Italian and I have to say I’m very similar to your girlfriend in the case; I don’t respond to direct correction and the main reason is that I feel like the person is not giving their attention to what I mean. Again, this depends, especially if they genuinely don’t understand and request clarification, in which case I have to take more care over my language – otherwise I completely ignore it. Instead, getting some feedback on your language and the mistakes you make is something I’ve always responded to. Do students get enough of this – a teacher that listens and studies learner language in order to give helpful one-to-one feedback to raise awareness?

      Great comment and thanks for stopping by. The blogging war is definitely the type where the soldiers have flowers in their guns!


      • Chiew says:

        OK, I’ve oiled the guns but not sure if there’s any ammo left… 😉 I, too, agree with the on-the-spot examples cited by both of you. I, too, got really annoyed when my then girlfriend used to correct me all the time mid-sentence. My reaction was just to ignore her and carried on. I was even more annoyed when this was done in the presence of others. And ++ annoyed when a laughter accompanied it. And +++ annoyed when it was a mistake that I knew was a mistake as soon as the word(s) left my mouth but often, either I didn’t think it was worth re-saying it or I was corrected before I had a chance to do it myself! Doesn’t this happen to you?

        I seldom, if at all, correct in this manner. If it was this immediate, it would be most likely in the gesture form and not verbal, e.g. hand waving back over the shoulder for past, or 3 fingers for third person. Or sometimes, a simple pardon? If the student corrects himself, fine; if not, it’s fine, too. But these are the kind of slip-of-the-tongue type of errors. But there are times when the error or problem is more complicated. To be honest, there are just too many ifs. Like I’ve written before – it all depends. My answer to your questions, Chris, is ‘maybe’. I’m not one for rules. There are too many shades of grey. If it feels right, I will correct it. But at the right time, or what I consider the most appropriate time. On-the-spot can be quite relative, don’t you think? I consider correction at the end of a sentence immediate; in fact, I’d also consider correction at the end of several sentences immediate. Because the context is still fresh in the minds of the students.

        Perhaps, what is more important than when to correct is what to correct. Unless your students are of a very high level, many mistakes are continually being made in the course of a class. As I mentioned in my Part 2 post, what is often neglected is the skill of monitoring. And monitoring is not only to detect the errors but to detect those you think need to be talked about. And, oh, it is also to detect good examples you think should be highlighted.

        You train teachers, right, Chris? I suppose you teach trainees on the methods of correction, maybe even when to correct. But, do you teach them on how to monitor? Or is each left to his/her own to find out by themselves? Or is monitoring supposed to be an innate skill? As fas as I remember, CELTA had input sessions on error correction, but none on monitoring. Or am I the only one to go on about monitoring?

      • dalecoulter says:

        Chiew. If you don’t mind I’m going to jump in on this one as well.

        If I remember clearly, on the CELTA course I did there was feedback available for trainees on monitoring and we were directed towards relevant literature to read up on monitoring. I found this aspect very helpful. Obviously, methods differ from centre to centre even from one tutor to another, which makes it hard to generalise – although on the other hand I think it makes each teacher different; imagine a world of homogenous teachers. Structures are tight and there’s a checklist to get through in the CELTA course, which makes it difficult to slot in potentially beneficial sessions that show trainers monitoring or correcting errors with the students.

        Perhaps in-the-spot correction often gets a bad rep because it is often associated with directly correcting a student by informing them of the error and telling them how they should say; however, the reality, at least amongst many of the teachers I encounter, is quite different indeed. Chiew mentioned lots of non-verbal gestures which encourage what I referred to as accountability for your language. That’s how the real world functions, isn’t it? If your message isn’t clear enough, your interlocutor won’t understand you and request clarification.

        Chris also makes a fine point about the process of awareness raising through error. But I don’t think this is strictly confined to delayed error correction, correct me if I’ve read too much into this. Here’s me laying my cards on the table: I was re-reading Focus on Form in Classroom Second Language Acquisition by Doughty and Williams today in preparation for a seminar I’m giving this weekend and I came across this quote:

        “The proposed advantage of focus on form over the traditional forms-in-isolation type of grammar teaching is the cognitive processing support provided by the “overriding focus… on meaning or communication”. to state this advantage quite simply, the learner’s attention is drawn precisely to a linguistic feature as necessitated by a communicative demand”

        Focus on form is, for anyone reading and not sure (although I imagine Chris et al know it very well), is, in Long’s words “focus on form… overtly draws students’ attention to linguistic elements as they arise incidentally in lessons whose overriding focus in on meaning or communication”

        Of course, the quote was not in reference to error correction; nevertheless, I do feel it has a strong effect on my rationale for error correction. Personally, I give more focus to one-to-one teaching and to pushing each individual in class and I’m lucky enough to have small classes or one-to-ones and this opportunity is available to me. More immediate (on the somewhat polarised scale of on-the-spot to delayed) is the method I lean towards and it’s for the reason that the communicative demand seems more apparent -perhaps a 55-60/45-40 ratio- but by no means does it dominate. I guess this also reflects my leanings on which approaches and methods I prefer as well – it all mixes up to make a big teacher pie.


  2. Chiew says:

    Yea, I know about Celta’s famed checklist; I suppose no matter how a centre programme it’s course, someone will be unhappy. I certainly think the one I attended could have done much better; they went on about correction, but nothing on monitoring.

    Your quote on focus on form (have to admit that I don’t read many books ;)) – that sounds dogmetish, no? 🙂 But, definitely, good teaching is having a big mixed pie and knowing which slice to cut at the right time and the least we could do is to keep trying…

    • Chris Ożóg says:

      Hi again Dale and Chiew,

      First off, some enjoyable and interesting discussion going on here, which always makes my morning coffee all the more pleasant.

      I’m a little confused by the “CELTA checklist”. Do you mean the syllabus? I am a CELTA tutor and I don’t have a checklist, nor have ever seen one, but all courses must follow the syllabus set down by Cambridge, though there is a bit of room to manoeuvre for individual centres and hence you find courses can differ. By the way, both monitoring and error correction feature on the syllabus, so should be at least part of input sessions, or individual sessions, as well as making part of the assessment of TP. And Chiew, I can assure you through lots of experience that monitoring is not an innate skill!

      Speaking of such things, I’ve actually just changed our CELTA observation tasks slightly for our forthcoming part-time course to help the trainees with error and monitoring. Our first observation task is now one in which the trainees must note down learner errors which will then form the basis of an input session on language, level and error the next day. It’s an experiment, but I’m confident it’ll be useful as it should help make the trainees aware of what types of errors are being made at different levels, whether the tutor corrected these (they watch me), why or why not this was the case and how the tutor noted any errors through monitoring or helped individuals with individual errors. We’ll then go into some language analysis of the most common errors. This’ll keep me on my toes… I might blog about how this goes.

      Anyway, back to the original discussion, it’s interesting you brought up Focus on Form, Dale. This is the type of thing I was getting at when I wrote about putting your approach cards on the table. The interesting question for me is the boundary (or is it a Venn Diagram?) between this, emergent language and error correction. One of the simplest ways of helping people understand what emergent language might be is to to focus them on error initially. What I mean is, if you’re listening to your students and they all make errors with “used to + inf/ing”, for example, then that’s a potential area for a focus on form. This is by no means a full definition of emergent language but, for teachers new to Dogme, it’s a nice way in (rather than starting them off on noticing gaps, for example, which they may not have the language awareness for). In fact, am I just repeating the observation task I mentioned above..?

      You also mention paralinguistics as a means of error correction and you are 100% right in my view. There are many non-verbal techniques to make use of rather than simply telling a student they’re wrong and correcting them by providing the correct form which, I’m sorry to say, I have seen thousands of times (sometimes even accompanied by a derisory “noooo”). This really does help no-one at all, in my view, and is a complete waste of time. I’ve never read anything that supports this a means to help learners learn, though this is not to discount immediate correction which, if done appropriately and depending on the individual learner, can be very very effective in helping them (though they should not, of course, suddenly be expected to not make this error again)

      And now I’m going on a bit and so, Chiew, I’ll set you a task to continue the discussion: what is monitoring?



      • Chiew says:

        Yes, by the checklist I meant the syllabus, and I’ve seen significant variance between one centre and another. They generally go on and on about PACS, they have input sessions on “how” to correct errors, but they don’t help – this is a personal opinion – trainees in monitoring. It’s more of a case, you’ll have to monitor the students, listen out for errors, and do a PACS. Ahem, but how do you monitor? Or, for that matter, as you ask, Chris, what is monitoring? It’s many things, isn’t it? Listening out for errors is just one facet of it. You monitor form, lexis, discourse, relevance, interaction, etc, etc. Good and bad usage. Let’s not forget good usage.

      • dalecoulter says:

        On a complete tangent from error correction, you also monitor for task, affective factors, energy levels, seating arrangements, class relations, potential fights and whoopsies (with the little ones)…

      • dalecoulter says:


        Yes, of course, by checklist, I mean the syllabus set down by Cambridge and I can honestly say I don’t any experience with this syllabus, only through seeing what colleagues have taught, so I can’t make a judgement on whether I think it’s effective or not. There are things to be taught and assessed and that’s that.

        Where I do think there’s some great space to move in for trainers is exactly the sort of thing you mentioned with the observations tasks. I’ve tried giving observers from CETLA courses the same sort of task in my lessons and chatting over the language with them afterwards. Unfortunately, that’s about as far as it got but I hope it had some influence on their teaching practice, especially when I taught the classes they used for it. Who knows, but it’s worth a try, right?

        It’s a good point that you raise on the boundary between error-focus on form- emergent language. From my experience, I’ve learnt to recognise the first and the second and can be confident on a day-to-day basis. The third can still often prove to be illusive; I’ve thought back and wondered if my choice was justified many times. To reel off another anecdote, for a conference talk in September on Dogme, I used a picture of a sign that said “what until you’re useful!” to try and get the message across that emergent language differs from slips or errors. Then again, a focus on form can lead to more language emergence as new fields of meaning are made reachable by the exploration of that language. I’m not sure I will ever feel comfortable with a definition of the term and be able to connect to a solid classroom experience.

        To draw from a German lesson I had the other day: we were talking about our housemates and I wanted to tell my classmate that all three of the us could cook, for which I needed a relative clause. I signalled to my teacher that I knew the first part of the sentence but not the relative clause, which she helped me with. My partner also tried to use this and I made a not of it – would this have been emergent language? It definitely wasn’t correction; however, a focus on form would have helped us immensely.

        As for monitoring, I reserve the term for task monitoring – to see if learners are on task. Otherwise, I prefer tuning in and tuning out. Tuning out to leave space between teacher and learner but still at a sufficient distance to listen. Tuning in when I’ve picked up on something and want to involve myself (not necessarily for language – today I tuned in a lot in interviews to act as support for content and to push the class further with their answers and to the edge of their language).

        I’m going to throw the floor back to you, Chris. I won’t ask for a definition of emergent language but would be interested in knowing if you’ve had the realisation of “yes-that is emergent language!” in your experience? And, how did you know it wasn’t error correction?

        Thanks for contributing again to the debate. I’d be interested to know how those observation tasks go.


  3. bealer81 says:

    Hi Dale,

    Welcome back to the blogging world.

    This has been an interesting discussion and I think not just you, but Chris and Chiew should be thanked for continuing the discussion in such great depth.

    I watched your video and it immediately made me question and self analyse the way I have been correcting my students this year.
    As part of my dogme project last year, I spent a lot of time employing the delayed correction method. This was due partly, to the amount of speaking and discussion exercises in class and therefore not wanting to disrupt the flow. Secondly, the follow on discussion about the errors that occured would then throw up a lot more discussion and potential errors.

    This year I have been teaching a few more YL classes and found that my instant reaction was to correct on the spot. In fact I got into the habit of correcting the student, asking them to go back to the beginning and getting them to repeat the sentence until they got it right. I imagine for the student this was very frustrating but it worked. A big thing at this level is past simple verbs. “What did you do at the weekend, Juan?” “I go the the swimming pool” “you went?” “yes, I went to the swimming pool” This has been come such a common feature that the students have now started to correct each other when they speak and I can encourage and praise them for peer correction.

    Last night I was teaching a proficiency class and had decided to do some delayed error correction. They were delivering a 2 minute presentation so on the spot correction would have been stupid and pointless. Afterwards, we corrected the errors, one of which was a classic make/do verb error. The students repeated the task after this correction slot and the same student repeated the same utterance, that had caused the error. She made the same mistake, paused, thought about it, said the same verb, paused, said the same verb again and then with what seemed like an almighty effort, again out loud, self corrected and carried on. Undoubtedly, this was down to the previous delayed error correction. It was great to see this kind of thing in action.

    This has got me to thinking about a very small scale investigation/research project about error correction. In particular, about the effectiveness of on the spot and delayed error correction. I was thinking about a questionnaire for teachers to discover their correction methods before perhaps doing some sort of experiment with a handful of teachers to trial one type of error correction per lesson and getting feedback from both teacher and student.

    Clearly its still in its infancy, but it might be an interesting study and your input would be welcomed. I have a lot of reading to do, but I will start thinking about it and let you know.

    Good to have you back and thanks for the inspiration.


    • dalecoulter says:

      Adam, sorry about the delayed response to your comment. You’re right, a big thank you must go to Chiew and Chris for contributing so in-depth to the discussion. Isn’t it great that we’re lucky enough to have a medium through which we can share and develop our beliefs and practices?

      I would be really interested in the results of that research – something more practical and less theoretical than what one digs up in the journals.


  4. bren brennan (@brenbrennan) says:

    Hi Dale,

    I wrote a short blog ages ago ( about anxiety over which is the best method of error correction as it’s a big area of interest for me personally.

    From my own teaching experience, I’d have to conclude that there definitely isn’t a one-size-fits-all method. Correction has to be individualised to your learners. If you adopt a particular correction method that is a good fit for a certain learner style, even then it MUST be varied. Firstly, because ramming the same method down someone’s throat (immediate or delayed) gets mind-numbingly boring very quickly. Secondly, it becomes ineffective as a student reacts in a Pavlovian way as in your example above about the election/President example.

    I’ve got a Spanish native student at the moment who has a profoundly stubborn & deeply fossilised L1 interference error (using the full infinitive after must: e.g. He must to pay for everything). I have been through every known error correction method to try to resolve this. Every. Single. Method! 🙂
    We are nearly there with it, where in the final stages I only have to slightly adjust my head position in the most subtle ‘gentle’ correction possible… and now he is actually self-correcting (with a knowing smile) sometimes even without my gesturing. It was tough, I won’t deny it. But it’s absolutely sure that if I had stuck to one method, he would’ve NEVER have got to this stage.

    Having said all that, I have the example of my Hungarian partner. When we started living together, she was a strong B2. For the next 2 years, I gave immediate reformulation on the most minuscule of accuracy or fluency errors – EVERYTHING. Don’t blame me! She insisted on me doing it. 🙂
    Now she is completely fluent. Not a strong C2 – I mean incredible. Never ever ever making a mistake and able to play and manipulate the language better than a lot of natives do. However, an interesting and probably quite important point is that after those first 2 years, she started reading the classics in English (Pride and Prej etc.. all the girly stuff), so maybe all that error correction was in vain and actually it was the pleasurable reading that did the trick…. which brings me neatly on to my concluding point which joyfully throws a spanner in the works.

    Last year, I attended a webinar by Stephen Krashen, in which he was quoting John Truscott’s research that concludes that grammar correction in written work is not only ineffective, but actually harmful, too! Truscott’s research paper is here –

    Krashen said that ONLY pleasurable reading works in L2 acquisition.

    Who the f*&k is right in all this???!!!???

    I think no-one is. You’ve just got to adapt and flow with as many methods possible (and have them all ready to hand in your TEFL arsenal) and just have fun with it! 🙂

    Thanks for writing/videoing on Error correction.


    • dalecoulter says:

      Tell me about it! Who is right? Is someone right? Can someone even be right? The only correction I’ve had in the last 3 years in Italian is direct correction – and I mean DIRECT – that has evoked emotions from utter frustration to appreciation. I guess it depends on what sort of mood we’re in as well, right? Anyway, now, the majority of errors I make are lexical.

      Do you think therefore that there may be some people more reactive and reflective on their error correction than others? I certainly think there are. While I don’t think these people are ‘born’ like it, I definitely think there’s a certain element of training one undergoes to be aware of accuracy and how to deal with it when it falls to pieces. Do teachers focus on that enough in class though? What good is an error correction – direct or indirect, immediate or delayed – if the person whom you correct does not have the facilities to process and work on this? I guess it’s the same case for feedback from teaching observations in a way; it requires a reflective mindset. Imagine getting lost down the same road every time you went to work. After a while would you learn? I imagine so. Or maybe the mistaken route then makes up your way to work and over time it just becomes a longer journey. Obviously, someone can put you on the right track, but will you take the advice and use their way the next time?

      Thanks for stopping by,


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s