Error correction involves more decisions than it might first appear — when to step in, when to hold back, and which technique to use in the moment. Get it right and it accelerates learning; overdo it and you undermine the very confidence learners need to develop. This article covers the main principles and the practical techniques that will help you handle it with confidence.
Errors and mistakes: a useful distinction
Teachers often use these terms interchangeably, but they mean different things.
A mistake is a slip – the learner knows the correct form but produces the wrong one in the moment, perhaps because they are nervous or thinking quickly. Given the chance, they would usually correct themselves.
An error reflects a genuine gap in the learner’s current knowledge. They say “I go to the cinema yesterday” not because they have forgotten the past tense, but because they have not yet acquired it reliably. Errors tend to be systematic – the learner will produce the same thing repeatedly, because it reflects where they currently are in their language development.
This distinction matters practically. A slip might just need a gentle prompt for the learner to self-correct. A genuine error may need clearer correction, or even a brief return to the teaching point.
Why do learners make errors?
Understanding the likely cause of an error helps you respond to it more usefully. Most errors come from one of three places.
First language (L1) interference
Learners apply patterns from their mother tongue to English. A Spanish speaker might say “I am agree” because Spanish uses a similar structure with estar de acuerdo. A French speaker might use “actually” to mean “currently” because the French actuellement is a false friend. If you teach the same nationality regularly, you will get to know these patterns quickly, and anticipating them lets you address them before they become habitual.
Overgeneralisation
The learner has grasped a rule in English and applied it more broadly than it actually works. “I goed to the park” is not L1 interference – it’s a learner correctly spotting that English past tense often involves -ed, and applying that rule to an irregular verb. It is actually a sign of progress.
Developmental errors
Some errors simply appear at predictable stages as a learner’s English develops, regardless of their first language. They tell you where a learner is, not what they have misunderstood.
It helps to keep one thing in mind: learners are not just making random mistakes. They are operating in what applied linguists call an interlanguage – a developing language system with its own internal logic, somewhere between their L1 and English. Approaching errors with that in mind, rather than treating them as failures to be eliminated, leads to better correction decisions. As Corder (1967) put it, errors are not something to suppress but evidence of an active, developing language system.
Should you correct everything?
No. Trying to correct every error is one of the most counterproductive things you can do, and it is a very easy trap to fall into.
Constant correction disrupts communication and chips away at confidence. Learners who feel every utterance might be interrupted become reluctant to speak at all, which is the opposite of what you want. There is also good evidence (Lightbown and Spada, How Languages Are Learned) that indiscriminate correction has little effect on learning, while correction that is targeted and timely does.
The better question is always: what is this activity for?
Matching correction to the activity
This is probably the single most useful principle in error correction.
In accuracy-focused activities – drills, controlled practice, form-focused exercises – correction is appropriate and expected. The point is to get the form right, and learners know it. Correct clearly and promptly.
In fluency-focused activities – discussions, role plays, information gaps, communicative tasks – the goal is communication, not accuracy. Interrupting to correct mid-flow damages the activity. The better approach is to monitor quietly, note errors on a piece of paper or in your head, then deal with them afterwards.
The standard technique: at the end of the activity, write a selection of errors on the board, without identifying who said them, and invite the class to spot and correct the problems. It takes five minutes, it’s low-pressure, and it’s often more memorable than in-the-moment correction because learners can see the errors as objects to think about rather than embarrassments to move past quickly.
What should you actually correct?
Even in accuracy work, being selective is more useful than being comprehensive. A few questions worth running through:
Does it impede communication? “I am very boring” (meaning bored) could genuinely mislead a listener. “Yesterday I go to the cinema” almost certainly will not, but it still warrants attention if it’s the lesson’s target language.
Is it related to what you taught? If the lesson focused on the present perfect and a learner uses it incorrectly, address it. An unrelated error from a different area is probably best left for another time.
Is it a class-wide pattern or an individual slip? If several learners are making the same error, it’s worth addressing as a class. An isolated slip may be better handled quietly or individually.
Will correcting it now help? Some errors reflect a stage the learner has not yet moved past. Heavy correction before a learner is ready to absorb the correct form rarely produces lasting change.
Techniques for correcting spoken errors
Having a toolkit of correction techniques, and knowing which to reach for in different situations makes correction feel natural rather than clunky.
Recasting
You repeat back what the learner said, reformulated correctly, without making a fuss.
Learner: “Yesterday I go to the cinema.”
Teacher: “Oh, you went to the cinema. What did you see?”
Recasting is gentle and keeps communication flowing. Its limitation is that learners may not notice they are being corrected – they may think you are just responding to their meaning. It works best for minor errors or for learners who are sensitive to correction.
Indicating the error and eliciting self-correction
Instead of giving the correct form yourself, you signal that something went wrong and give the learner a chance to find it. Self-correction produces better learning than being handed the answer, because the learner has to actively engage with what went wrong. Here are the main techniques:
- Rising intonation – repeat the error with a questioning tone: “Yesterday I go…?” The intonation alone signals “try again” without a word of explanation.
- A short verbal prompt – “Tense?” or just “Yesterday…” with a slight pause is often enough.
- Missing word or article – hold up fingers to represent the words in the sentence, then point to the gap or mime plucking an extra word out of the air. This makes the structure of the sentence visible and locates the error precisely.
- Wrong word order – use your fingers to represent words and mime swapping them around.
- Wrong preposition – repeat the sentence up to the preposition with rising intonation (“I arrived at…?”) and let the learner supply the correct one.
- Contraction missing – press two fingers together to show that two words should merge (“It is” → “it’s”).
- Past tense needed – point back over your shoulder to indicate that something happened in the past.
- Wrong word stress – tap out the correct stress pattern on your hand or the table, or draw stressed and unstressed syllables as large and small circles on the board (● ○ ○ or ○ ● ○).
- Wrong intonation – move your arm or hand in the shape of the intended intonation curve.
It takes a lesson or two for learners to understand what each gesture means, but once they do, it becomes a quick and almost wordless correction system, which is especially valuable during drilling or when you don’t want to slow the lesson down.
Redirecting to the class
If a learner makes an error during open class work, you can open it up: “Does anyone want to help?” or “What do we think – is that right?” This works well in supportive classes and shifts correction away from the teacher-to-student dynamic. Use it carefully – it needs to feel like collaborative problem-solving, not public shaming.
Direct correction
Sometimes you simply want to supply the correct form clearly and move on. There’s nothing wrong with this. It’s often the right choice during accuracy work, or when an error needs addressing without any ambiguity. Keep the tone neutral – the correction is information, not a verdict.
Correcting pronunciation errors
Pronunciation errors deserve a slightly different approach, partly because the correction techniques are different and partly because not all pronunciation errors are equally worth addressing.
The most useful question is: does it affect intelligibility? Wrong word stress often causes more communication breakdown than a grammar error – “CONtent” versus “conTENT” can produce genuine misunderstanding. Other pronunciation variations might simply reflect a learner’s accent without causing any real problem.
For pronunciation, gesture-based and visual techniques are particularly effective because they can show the learner what to do rather than just telling them:
- For word stress, tap the stress pattern, use large and small circles on the board, or use your fingers with one raised higher than the others.
- For individual sounds, point to the relevant position on the phonemic chart, or demonstrate mouth position.
- For linking and connected speech, draw the words on the board with arcs showing where they run together.
The general principle during drills is to give a clear model, signal any problem quickly without lengthy explanation, and give the learner the chance to repeat correctly. Don’t over-dwell on pronunciation correction mid-activity – one or two attempts is usually enough before moving on.
Correcting written errors
Written correction is a different challenge. You have more time to think, but the most common approach – going through writing with a red pen correcting every error – is also one of the least effective. Learners tend to glance at the mark, note that there are corrections, and not engage with them in any depth.
A more effective approach is a correction code: a simple system of symbols or abbreviations in the margin to indicate the type of error, which the learner then uses to find and correct the problem themselves.
A basic code might look like this:
| Symbol | Meaning |
|---|---|
| T | Tense error |
| WW | Wrong word |
| WO | Wrong word order |
| Sp | Spelling |
| Art | Article missing or wrong |
| Prep | Preposition error |
| ? | Unclear – I don’t understand this |
The learner gets the writing back, sees a T in the margin next to a sentence, and has to work out what the tense problem is. This is more cognitively demanding than being handed the correct version, and it tends to produce better long-term retention.
It also helps to be selective. Marking every error in every piece of writing is exhausting for you and overwhelming for the learner. Focusing on two or three significant recurring errors, particularly ones related to what you have recently taught, is more useful than comprehensive marking. And wherever possible, respond to the content of the writing as well as its accuracy. Learners who feel their ideas have genuinely been read are more motivated to improve.
Think about your learners
How correction feels matters as much as what you correct. Some learners actively want to be corrected and feel shortchanged if errors pass without comment. Others, particularly adults, or learners from backgrounds where mistakes were treated as failures, find public correction uncomfortable and may withdraw if they feel singled out.
The general principle is to keep correction low-key and matter-of-fact. Corrections delivered calmly, without fuss in either direction, do not raise the anxiety that blocks learning. As you get to know a class, you will develop a feel for who can be corrected publicly without any problem and who needs a quieter, more individual approach.
There is also a longer-term consideration worth bearing in mind: errors that are never addressed can become permanent features of a learner’s language – what researchers call fossilisation. This is particularly common with errors that don’t cause communication problems, because neither the learner nor the people they speak to ever flag them. Selective, consistent correction over time is the best protection against it.
A quick summary
Match your correction to the activity – more direct during accuracy work, quieter and delayed during fluency work. Correct selectively, not constantly. Build up a toolkit of techniques, including gesture-based ones that let you signal errors without interrupting communication. Use a correction code for written work rather than red-penning everything. And remember that errors are a normal, productive part of language learning – the goal is not a perfect classroom, but learners who keep taking risks and developing.
References
- Corder, S.P. (1967) ‘The significance of learners’ errors’. IRAL, 5(1–4), pp.161–170.
- Edge, J. (1989) Mistakes and Correction. Longman.
- Lightbown, P. and Spada, N. (2013) How Languages Are Learned (4th ed.). Oxford University Press.
- Selinker, L. (1972) ‘Interlanguage’. IRAL, 10(1–4), pp.209–232.
