Classroom skills · Article

Monitoring in ELT: Keep It Low-Key, but Make It Count

Learn how to monitor English learners without taking over. Practical CELTA/DELTA examples of task checks, evidence collection, feedback and reactive teaching.

By Ahmed Bin Nasser, CELTA Main Course Tutor · · Last updated

Lumina ELT article cover: Monitoring in ELT: Keep It Low-Key, but Make It Count

The teacher who walks around the room isn't necessarily the teacher who knows what's happening in it.

You set up the task. Learners get into pairs. They start talking.

Excellent. Now it's time to monitor.

So you walk around the classroom, stop behind one pair, nod at their answers, move to another pair, correct a sentence, smile, and return to the front.

You've monitored. Haven't you?

Well, you've certainly walked around.

But what did you find out?

Did everyone understand the task? Which answers caused disagreement? Which language problems are worth addressing? Was the activity too demanding for anyone? And if you had to give feedback right now, would you know where to begin?

This is why I like the expression keep your monitoring low-key. It doesn't mean becoming invisible or refusing to help. It means making your presence purposeful without taking over. Let's look at what that involves, from the moment a task starts to the decisions you make during feedback.

First, what is monitoring?

In ELT, monitoring means observing learners as they work so you can assess their progress, understand their needs and respond appropriately. Depending on the task, you may be looking at procedure, participation, task achievement, language use, strategy use or the quality of ideas.

There are two useful questions to separate:

  1. Are learners doing the task they were asked to do?
  2. What does their performance tell me about what they need next?

ELT Concourse describes related purposes as check monitoring and support monitoring. The distinction is helpful, provided we don't turn it into two rigid categories. You may check instructions and notice a language gap during the same visit to a pair.

Good monitoring is also connected to formative assessment. A classroom observation becomes formative when the evidence informs what you, or the learners, do next. As Black and Wiliam (2009) argue, assessment is about using evidence of learning to adapt teaching and learning—not merely collecting evidence for its own sake.

Monitoring begins before you move

Some teachers treat monitoring as a stage that begins after they say, 'Start.' I'd argue that effective monitoring begins earlier, in your planning.

Before the lesson, ask:

  • What is the aim of this task?
  • What would successful performance look or sound like?
  • What problems are learners likely to encounter?
  • Which problems would require immediate intervention?
  • What evidence will I need for feedback?

Consider a listening lesson. Learners are matching five speakers to five attitudes. Your main concern is whether they can infer attitude from the recording, not whether they spell every name correctly.

So when you monitor the peer check, listen for disagreements about which words or intonation patterns led to their decisions. That's more useful than counting how many answers they wrote.

Now compare that with a controlled grammar task. If the next stage depends on learners understanding the contrast between used to and usually, the accuracy of that distinction matters more. You may need to establish whether the class is ready to move on.

Same word: monitoring. Different purpose.

The first thirty seconds: Check the setup

Illustrative scenario: An intermediate speaking lesson.

You've asked learners to choose three workplace problems and agree on the most urgent one. Two pairs are discussing a problem. One pair is copying the instructions. Another has begun listing workplace vocabulary instead of prioritising problems.

Don't correct their grammar yet. You have a more immediate issue: they are not all doing the same task.

Move close enough to establish what's happening. If one pair misunderstood, clarify quietly. If half the class misunderstood, stop briefly, demonstrate the first decision with the group and restart.

That is responsive teaching, not an admission of failure.

A well-timed 20-second reset can save a much longer and more frustrating feedback stage. But don't assume that every task needs multiple ICQs. Clear instructions and an effective demonstration may already have done the job.

Think before you reveal: Two pairs have misunderstood your task. Another six pairs are working correctly. Would you stop the entire class or intervene locally?

Reveal the trainer's reasoning

There is no compulsory answer. If the problem is isolated and you can repair it without disrupting others, a brief local intervention is sensible. If the misunderstanding reveals a flaw in your instructions likely to affect more learners, a short whole-class reset may be more efficient. Diagnose the scale of the problem first.

During the task: Keep it low-key

Once learners have started, give them room to think, communicate and make mistakes. Your presence changes the interaction. A teacher standing directly behind two learners may unintentionally become the third participant—or the person they wait for to approve every answer.

In a face-to-face lesson, change your position deliberately. Stay near enough to listen, but not so close that your body language invites everyone to stop and look at you. With freer speaking, sometimes sitting at the edge of the group works better than looming over them. In a cramped classroom, moving constantly may distract more than it helps.

Online, the question changes. You may need to visit breakout rooms, use the shared document or observe a chat-based task. Announcing every visit or entering with a long explanation can interrupt momentum. Decide what information you need before you enter.

Low-key doesn't mean passive. You're active intellectually, not necessarily verbally.

For many tasks, your monitoring has three layers:

Layer What you notice What you might do
Process Are learners engaged and following the task? Repair a misunderstanding; adjust timing.
Performance What strategies, answers or language are emerging? Gather evidence; identify common patterns.
Support Is a difficulty preventing meaningful progress? Prompt, model or clarify selectively.

The layers overlap. Don't turn the table into another rigid classroom framework.

What exactly should you write down?

Please don't try to transcribe the entire lesson. You aren't writing a novel.

A small, organised record is often more useful. For a speaking task, I recommend something like this:

Task aim: Make polite requests Notes
Useful language “Would it be possible to change my appointment?”
Recurring issue “I want you change…” in three pairs
One pronunciation issue Stress in possible obscures the word in one exchange
Content worth sharing Two very different, sensible solutions to the same problem
Teaching decision Short clarification of indirect requests after the task

These are illustrative notes, not data from an observed lesson.

You don't need to write down every slip, and you shouldn't correct everything you record. Choose what is relevant to the aim, widespread enough to matter, or likely to disrupt communication.

Also record successful language. Learners need evidence of what worked, not a list of everything they did wrong.

Monitoring should change the feedback you give

Here's where monitoring pays off.

Imagine a reading-for-detail activity with eight questions. You circulate during the peer check and notice that learners agree on questions 1–5 and 8, but pairs disagree over 6 and 7.

What happens next?

You could elicit all eight answers, one by one, at exactly the same pace. But you've already learned something useful: six items appear straightforward; two need attention.

Instead, reveal the answer key for the straightforward items, allow learners to confirm it, and devote more feedback time to the two difficult questions. Ask learners to identify the text evidence supporting their answers. If the disagreement relates to an inference, discuss the reasoning; if it comes from a lexical misunderstanding, address that.

Monitoring tells you what to focus on. Feedback lets you respond.

Don't overinterpret: one pair's wrong answer doesn't prove the whole class struggled. Sample across groups and check the scale of the issue. An answer key is also not feedback by itself; the useful part is what learners do with it and what uncertainty is resolved.

A quick decision you can use tomorrow

Illustrative situation: During a grammar peer check, six pairs agree that item 4 is has lived; two pairs choose lived. Your lesson targets the distinction between a continuing situation and a finished past event.

What should you prioritise?

Reveal one sensible feedback route

Show the complete key so nobody is left uncertain, then focus briefly on item 4. Ask whether the person still lives there and what in the context tells us. If the learner's answer suggests a deeper misconception, clarify it. You needn't make every item into a mini-lecture.

Reactive teaching: The lesson is allowed to respond

Sometimes you monitor and discover something you didn't anticipate. Perhaps learners understand the target grammar but cannot pronounce the key phrase intelligibly. Perhaps they lack an essential expression for the speaking task. Or perhaps they are already performing confidently and no longer need the scaffold you prepared.

A professional decision might be to:

  • pause for a short reactive clarification;
  • give an additional model or useful chunk;
  • reduce support and move to a more demanding task;
  • revise timing because learners need longer to process;
  • postpone a non-urgent language point and return to it later.

But a word of caution: teaching reactively isn't chasing every interesting error. If the main aim is reading for gist, a lengthy explanation of a rare phrasal verb may consume time without serving that aim. Decide whether the issue is urgent, shared and relevant.

In a Test–Teach–Test lesson, this responsiveness becomes particularly visible: performance in the initial task should shape what you teach. In other frameworks, the same professional judgement matters, even if there is no formal diagnostic stage.

Monitoring is not the same during every stage

During controlled practice, accuracy on target items is often more important; you may monitor fairly closely to identify misunderstandings. During fluency-oriented speaking, listen for ideas, task achievement and useful language, and avoid turning every conversation into a correction exercise. During reading or listening, monitor procedure and learners' responses without interrupting input unnecessarily. During a lead-in, your purpose may be engagement and relevant background knowledge rather than error correction.

These are tendencies, not rules. A fluency task can still require intervention when communication breaks down; a controlled exercise can be completed independently with little teacher attention.

One question captures the principle: What is this stage supposed to achieve?

The teacher's position is also a teaching decision

Try sketching your classroom on a piece of paper. Mark where learners sit, where you stand and which groups you can realistically hear. Now draw two routes: one for quickly checking task set-up, another for listening more carefully to two or three groups.

You may discover that your usual route is more of a walking habit than a monitoring strategy.

The classroom layout matters; so do learners' preferences and accessibility needs. If teacher presence is starting to dominate the interaction, the related Teacher Talking Time guide offers additional ways to create space for learners. A learner with hearing difficulties or a group working in a noisy space may need you closer. In online lessons, you cannot see every group simultaneously, so your sampling strategy and shared artefacts become especially important.

In-article interactive suggestion: Display an original classroom floor plan. Allow readers to select the teacher's position and reveal what they can hear, what they may miss, and when that position is useful. Do not copy another site's diagrams.

Five habits that make monitoring more useful

1. Know your target. Choose two or three things to listen for, not twenty.

2. Sample the room fairly. Avoid spending the whole activity with one confident pair while quieter learners remain unseen.

3. Distinguish a task problem from a language problem. Clear the procedure before diagnosing linguistic difficulty.

4. Take notes that support decisions. A list of isolated errors is less useful than a pattern, an example and a plan for feedback.

5. Follow through. If you noticed an issue but did nothing with the evidence, ask why. Sometimes postponing it is right; sometimes you've missed the teachable moment.

And please don't confuse consistency with doing exactly the same thing during every activity. Be consistent about observing purposefully and responding to evidence. Let your procedures vary with the task.

A CELTA and DELTA perspective

For CELTA trainees, monitoring needs to be observable through its effects: learners understand tasks, have suitable opportunities to work, receive appropriate support and benefit from subsequent feedback. Cambridge's CELTA syllabus includes monitoring learner behaviour and responding appropriately, incorporating assessment procedures and making planning decisions from assessment evidence.

For DELTA candidates, there's an additional level of analysis. Why did you choose to intervene? Why at that moment? What did the evidence suggest? How did your response serve the aim, and what alternatives were available? Simply writing “I monitored the pairs” tells us almost nothing about the quality of the decision.

This is the shift I want teachers to make: from describing what you did to explaining why it mattered.

Try it in your next lesson

Before one task, write three things you intend to monitor: procedure, one target-language feature, and one feature of successful task performance. During the task, jot down no more than five useful observations. Immediately afterwards, design your feedback around what those notes reveal.

Then ask yourself: What did I notice that I would have missed if I'd simply walked around?

Low-key monitoring can be one of the busiest things your mind does in a lesson. You just don't need to announce it to the whole room.


Quick FAQs

Should I correct mistakes while monitoring? Sometimes, particularly when an error blocks progress or when accuracy is the focus. During fluency tasks, delayed correction may serve the aim better.

Should I monitor every pair? Aim for fair coverage, but don't assume every pair needs equal time. Sample strategically and respond to emerging evidence.

Is monitoring a CELTA assessment skill? Yes. Cambridge's CELTA syllabus includes monitoring learners, responding appropriately and using assessment evidence.

Is monitoring the same as feedback? No. Monitoring gathers and interprets evidence during work; feedback uses evidence to support understanding and improve subsequent performance.

Continue exploring with Lumina: ELT Terminology: Part 1 · ELT Terminology: Part 2 · Scaffolding or Spoon-Feeding? · Test–Teach–Test · Lesson Planning

References and further reading

  • Cambridge English. CELTA syllabus and assessment guidelines. https://www.cambridgeenglish.org/images/21816-celta-syllabus.pdf
  • Black, P., & Wiliam, D. (2009). Developing the theory of formative assessment. Educational Assessment, Evaluation and Accountability, 21, 5–31. https://doi.org/10.1007/s11092-008-9068-5
  • Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. https://doi.org/10.3102/003465430298487

Check your understanding

6 short questions. No account needed, no score, and nothing you choose or write is saved or sent anywhere.

  1. 1 You discover during a peer check that half the class misunderstood an essential instruction. What's most useful first?
    Show the answer (no automatic checking)

    Answer: B. The procedural issue prevents valid task performance; repairing it takes priority.

    By Lumina Trainers

  2. 2 During reading feedback, eight answers need confirming, but learners disagreed only on items 6 and 7. What is an efficient response?
    Show the answer (no automatic checking)

    Answer: C. Everyone gets confirmation while time goes where the evidence indicates uncertainty.

    By Lumina Trainers

  3. 3 A low-key monitor is best described as someone who…
    Show the answer (no automatic checking)

    Answer: C. Low-key monitoring is purposeful observation with proportionate intervention.

    By Lumina Trainers

  4. 4 True or false: Monitoring is formative whenever the teacher writes a long list of errors.
    Show the answer (no automatic checking)

    Answer: B. Evidence becomes formative when used to inform learning and teaching decisions.

    By Lumina Trainers

  5. 5 Match each teaching situation with the most appropriate response.
    • A Consider advancing or enriching task
    • B Repair task instructions
    • C Plan focused language feedback
    Show the answer (no automatic checking)

    Answer: 1–B, 2–C, 3–A. Compare the learning need and the purpose of the response; there is no benefit in using one routine for every situation.

    By Lumina Trainers

  6. 6 After a speaking task, your notes include five isolated errors and two excellent expressions. How would you decide what belongs in feedback?

    Show a suggested answer

    I would begin with the task aim, check whether issues were repeated or obstructed communication, share successful examples, prioritise one or two useful areas for improvement, and allow learners to reuse the refined language. The list should support decisions, not become the feedback automatically.

    Suggested answer by Lumina Trainers

Ahmed Bin NasserCELTA Tutor & Teacher Trainer | Lumina