Academic Discussion: Engaging With the Other Students
Engaging with the two classmate posts separates band 4 from band 5 on the TOEFL discussion task. Six engagement moves, with examples and a full model.
12 min readMonthly & Premium
Quick answer
To reach band 5 on the TOEFL discussion task you have to engage with what the two classmates actually argued, not just state your own view. Name one, restate their point accurately, then extend it, qualify it, or answer it. Agreement alone is not engagement.
Why strong writers stall at a 4
There is a particular kind of discussion post that is clearly written, correctly punctuated, logically organised, and stuck at band 4 forever. It states a position, gives two reasons, and closes. Nothing is wrong with it. It is simply not a contribution to a discussion, because it would read identically if the two classmate posts had never existed.
The band 5 descriptor asks for a well-elaborated contribution with a strong opinion that engages meaningfully with the discussion. That phrase is not decoration. It is the difference between a post and a reply, and it is the single most reliable band you can gain on this task once your English is already sound.
This lesson is about the mechanics of that engagement: how to read the two posts for the thing worth answering, and six specific moves you can make once you have found it.
How to read the two posts in ninety seconds
Here is a real discussion. Professor Ahmadi asks whether companies should generally adopt a four-day workweek or stick with the five-day model, noting that some companies report happier and equally productive workers, while others worry about longer days, more stress, and gaps in customer service.
Cara supports the four-day week. She cites trials in Iceland and the United Kingdom where productivity stayed steady or improved, says workers used the extra day for medical appointments, child care and rest, which reduced burnout and sick leave, and concedes the model would not work for every industry though office-based companies could benefit.
Marek is sceptical. He argues that many service jobs cannot simply close for an extra day, so the cost falls on retail and restaurant workers who still cover the same hours. He adds that trial results often involved careful management changes that may not transfer when companies simply cut hours without restructuring.
Read those two posts looking for three things, in this order:
1. Where do they actually disagree? Not their conclusions, their reasons. Cara and Marek do not disagree about whether office workers benefit. Cara says office-based companies could clearly benefit; Marek's objection is about service jobs and about restructuring. The real disagreement is narrower than the headline positions suggest, and finding that narrow point is the fastest route to something interesting to say.
2. What has one of them conceded? Cara concedes the model would not work for every industry. A concession is an opening: you can pick it up and either strengthen it or show that it undermines her position more than she thinks.
3. Which claim is doing the most work, and is it safe? Marek's strongest claim is that the trials involved management restructuring, so the results may not transfer. That is a claim about why the evidence works, and it is the hardest thing in either post to answer. If you can answer it, you have a band 5 paragraph.
Six engagement moves, weakest to strongest
Move
Sounds like
What it is worth
1. Bare agreement
I agree with Cara.
Nothing. It names a classmate without touching their argument, so the rubric criterion it is meant to satisfy stays unmet.
2. Restatement
As Cara said, the trials showed productivity stayed steady.
Very little on its own. It proves you read the post but does not move the argument.
3. Extension
Cara is right about the trials, and I would add that the benefit shows up in retention as well as productivity.
Solid. This is the minimum that reads as a genuine contribution.
4. Qualification
Cara is right for office work, but her own point about industries doing the qualifying suggests the policy has to be sector by sector rather than general.
Strong. You have used a classmate against their own conclusion without misrepresenting them.
5. Answering the objection
Marek is right that the trials involved restructuring, but that is a reason to require restructuring, not a reason to reject the schedule.
Strongest single move. You concede the fact and dispute what follows from it.
6. Reframing the disagreement
Cara and Marek are not really disagreeing about whether it works. They are disagreeing about which workers we are talking about.
Top of the scale. It shows you understood both posts well enough to see past their conclusions.
Moves 4 to 6 are what the top of the scale is describing. Moves 1 and 2 are what most band 3 and 4 posts do.
What is happening there
Both classmates are named, which is more than many posts manage, and it is still a 3. Here is why.
She mentioned that productivity did not decrease in the trials, and I think this is true. Restatement plus a verdict. I think this is true adds no information, because the writer offers no reason for thinking it and no consequence that follows from it.
Companies in several countries have tried it and the results were positive. This looks like evidence and is not. No country, no company, no number, no source. A sentence at this level of generality could be written without having read anything.
Marek said that some jobs cannot do this, and he also has a point, because every workplace is different and there is no single rule that fits all of them. This is agreeing with both sides in consecutive sentences. Worse, it flattens Marek's argument: he did not say some jobs cannot do this, he said the cost gets shifted onto service workers who still cover the same hours. The specific mechanism was the valuable part and it has been deleted, and the reason offered in its place, that every workplace is different, is true of every topic ever discussed.
In my view, companies should try it because employees would be happier and less stressed, and happy employees work better. If workers get more rest, they will come back to the office with more energy. The position finally arrives near the end, supported by general claims that neither classmate needed to be read in order to write. The second sentence restates the first rather than developing it.
The post reaches the length it needs and has the appearance of engagement (two names, two summaries) with none of the substance. That is worth noticing on its own: length is not what is holding this at a 3, so adding another hundred words in the same register would not move it. This is the most common band 3 pattern among students who have been told to mention the classmates.
Line by line, what the engagement is doing
Marek's objection is the reason the policy has to be written carefully rather than the reason to abandon it. Position and engagement fused into the opening sentence. The reader knows both what you think and how you will treat the opposition before the second sentence.
Marek's strongest point is not that service jobs are different. It is that the successful trials came with management restructuring. This is move 5, and notice what it does first: it upgrades his argument. Telling a rater which of a classmate's points is strongest proves you understood both, and it is impossible to fake.
I think that is correct, and it is the part of his post supporters usually ignore. A real concession. Conceding the strongest opposing point costs you nothing if you can then show what does and does not follow from it, and it is what separates argument from assertion.
But it is an argument about what a four-day week requires, not about whether it can work. The pivot. The fact is granted, the inference is denied. This one sentence is the highest-value sentence in the post.
A policy that says take Friday off is not the same policy as the one that was tested. The mechanism, in plain language. Short sentence after a long one, which is sentence variety doing structural work.
Cara's concession is where I would push further than she does. Move 4. You are using a classmate's own qualification to narrow their conclusion, which is respectful and more effective than disagreeing with her headline.
A hospital ward and a design studio do not have the same relationship between hours and output. The concrete instance. Two specific workplaces, one contrast, no statistics needed.
So my position is narrower than Cara's and less pessimistic than Marek's. A close that locates your position relative to both classmates. This is what a contribution to a discussion sounds like, and it takes one sentence.
A four-step routine for the engagement paragraph
1
Step 1. Pick one classmate, not both
The directions ask for at least 100 words, and about 140 is what most people comfortably produce in the time. That is a working target, not a ceiling: the band 5 model above runs to 231 words because it develops two classmates properly, and nothing penalises it for that. Engaging properly with one classmate beats naming two and doing nothing with either. Choose the one whose argument is closest to the pressure point in the question. If you have room at the end, add a single clause about the other.
2
Step 2. Restate their strongest version in your own words
Not their whole post, the one claim doing the most work. Use your own phrasing, because copying their sentence looks like padding and does not show comprehension. If you can say which of their points is strongest, say so.
3
Step 3. Concede what is true
Say plainly that the point is correct if it is. Students avoid this because it feels like losing. It is the opposite: a rater reads a genuine concession as confidence, and it sets up the next move.
4
Step 4. Separate the fact from the conclusion
Almost every good objection is a true fact plus a conclusion that does not quite follow. Grant the fact, dispute the conclusion, and say what the fact really implies instead. That sentence pattern (that is an argument about X, not about Y) will serve you on almost any discussion prompt you get.
Does this count as engagement?
Naming a classmate is not the same as engaging with one, and this is where most students who have been told to "mention the other posts" still lose the mark. Each opening below names somebody. Decide whether a rater would credit it as real engagement before reading the reasoning.
Classmate post (Mikhail): Remote work damages the informal learning that junior staff get from overhearing senior colleagues.
Student opening: I agree with Mikhail. Remote work has definitely changed how people learn at work, and I think he makes a good point. In my view, companies should think carefully about this issue.
Would a rater credit this as engagement?
Everything after the name is a restatement. "Remote work has changed how people learn" is Mikhail’s point returned to him in weaker words, and "he makes a good point" is an evaluation with no content. Engagement means doing something to the argument: extending it, limiting it, providing a case it does not cover, or explaining why it holds. A rater who deletes the name here loses nothing from the paragraph, which is the test.Naming is necessary but not sufficient. The name is the cheapest part of engagement and on its own it reads as compliance with an instruction rather than participation in a discussion.Agreement is perfectly valid. The problem is not that this agrees, it is that it agrees without saying anything the classmate had not already said.There is no requirement to disagree. A post that agrees and then strengthens the argument with something new scores just as well as one that objects.
Classmate post (Kaitlyn): Free public transport would cut car use in cities.
Student opening: Kaitlyn assumes the barrier is cost, and for most drivers I do not think it is. The people I know who drive into the city do it because the bus takes fifty minutes and the car takes twenty, and making the bus free does not make it faster. Her policy would work, but on a different group: people who already take the bus and are struggling to afford it.
Would a rater credit this as engagement?
This is engagement at full strength. It names what her argument depends on, that cost is the barrier, and then puts a case against it. It also does the thing that separates a 5 from a 4: it does not simply reject her policy, it relocates it to the group it would actually help. Delete the name here and the paragraph collapses, which is the sign the engagement is real.It does offer evidence, in the form of a concrete travel time comparison. Personal observation is acceptable support in this task; you are not being asked for citations.Bus journey time is not a change of subject, it is the mechanism her claim rests on. Testing the mechanism is exactly what engagement means.Repeating a name adds nothing. What earns the credit is the work done on the argument, and this would score the same with one mention.
Classmate posts: Mikhail argues that university should be free. Kaitlyn argues that free tuition mainly benefits wealthier families who would have attended anyway.
Student opening: Both Mikhail and Kaitlyn make interesting points about this complex issue. As Mikhail says, education is important, and as Kaitlyn notes, there are questions about who benefits. Personally I believe university should be free because education is a right.
Would a rater credit this as engagement?
Kaitlyn’s post is the hard one: it says free tuition mainly helps people who would have gone anyway. The student notes it, then declares that education is a right, which does not answer it at all. Acknowledging an objection and then walking past it is weaker than not raising it, because the reader now knows you saw the problem and had nothing for it. Engagement means the objection has to be handled, even briefly.Both are named, and both are summarised so vaguely that "interesting points" and "education is important" would fit almost any prompt on any topic.Summarise then assert is not the required structure and it is the shape of a band 3 post. The position needs to be built on the engagement, not placed after it.Engaging with two classmates is fine and often stronger. The number is not the problem here.
Engagement judgement
Five questions using the four-day workweek discussion. Cara supports the change and cites trials; Marek is sceptical and points to service jobs and to restructuring.
Which sentence engages most meaningfully with Marek?
It identifies his strongest claim, concedes that it is true, and then separates the fact from the conclusion he drew. That is the highest-value engagement move available.A verdict with no content. It names him and does nothing with what he said.It contradicts him with the evidence he already addressed. Marek anticipated the trials and explained why he doubts they transfer, so this reply shows the post was not read carefully.It sets up a contrast that Marek never denied. Nobody argued workers do not deserve rest, so this is arguing against a position he did not take.
Cara concedes that the model would not work for every industry. What is the most productive way to use that concession?
Extending a classmate concession into a sharper conclusion than they drew is a strong engagement move and it produces a narrower, more defensible position of your own.Ignoring the most interesting sentence in a classmate post is exactly what a post that has not really engaged looks like.A concession is not an inconsistency. Treating it as one misrepresents her and reads as point-scoring rather than argument.One qualification does not overturn her position, and claiming it does overstates your case in a way a rater will notice.
Your post names both classmates, summarises each in one sentence, and then gives your own opinion with one reason. What band does that pattern typically reach?
Summary plus opinion is the standard mid-band pattern. Nothing in it uses the classmate arguments to shape your own, so the post is relevant and supported but not elaborated through the discussion.Naming is not engaging. The descriptor asks for meaningful engagement, and a one-sentence summary that leads nowhere is not that.It does answer the professor and it does take a position, so it is well above the band 2 territory of limited relevance and weak organisation.Balance is not what the top band rewards. It rewards a strong opinion that engages with the discussion.
Which of these misrepresents Marek?
He said the five-day model still works for most industries and questioned whether trial results transfer. He never claimed no workplace could ever benefit, and inflating his claim makes it easier to knock down, which raters notice.Accurate. This is his restructuring argument stated fairly.Accurate. This is his service-sector argument, including the mechanism about who absorbs the hours.Accurate. This is close to his closing sentence and it is the conclusion he actually drew.
Which closing sentence best signals that your post was part of a discussion?
It restates the position while locating it relative to both classmates, and it carries the qualification the argument earned. One sentence, and it closes the discussion rather than just the post.A generic conclusion. Correct and safe, but it could sit at the end of any essay on any topic.It abandons the position in the final sentence, which is a coherence problem as well as a wasted close.It ends by recommending that someone else think about it, which is not a position at all.
Engagement self-audit
Could my post have been written without reading the two classmate posts? If yes, it is capped at a 4.
I named at least one classmate and restated their strongest claim, not their easiest one.
I conceded something that is genuinely true in the opposing argument.
I separated a fact I accept from a conclusion I reject.
I did not agree with both classmates and then avoid taking a side.
My closing sentence positions me relative to the discussion, not just relative to the question.
Nothing I attributed to a classmate is stronger or cruder than what they actually wrote.
One sentence pattern worth memorising
X is right that A, but that is an argument about B, not about C.
That pattern works on nearly every discussion prompt in this format, because almost every objection in an academic argument is a true observation attached to an inference that runs further than the observation supports. It concedes, it demonstrates comprehension, and it defends your position in one move.
Practise filling it in against three different prompts. Once it is automatic you will never write a post that could have been written without reading the discussion, which is the only thing standing between a competent band 4 and a band 5.
Monthly & Premium
12 more sections in this lesson
You have read the opening. The rest covers the method in full, with worked examples and practice questions that explain why each wrong answer is wrong.
All 45 lessons: Reading, Listening, Writing and Speaking
Real exam audio and record-yourself speaking drills
No. One classmate engaged with properly is worth more than two named and summarised. If you have words left at the end, a single clause about the second is enough.
Neither is safer. What matters is doing something with their argument. Agreeing and adding nothing is the weakest possible move, and disagreeing without conceding what is true in their point is the second weakest.
Yes, as long as you engage with the version of the argument they actually made. Restate their strongest claim, say what is right about it, and then explain precisely where you part company.
Then find the point where their reasons conflict, because their conclusions differ so something underneath them must. Reframing that conflict is the strongest engagement move available and it produces a real position of your own.
Around 30 to 45 of your 130 to 160 words. It should be substantial enough to concede and pivot, and short enough to leave room for a developed reason with a concrete example.
Yes. The descriptors are content aware, so a relevant, developed, engaged post with noticeable errors sits at 3 or above, while a flawless post that ignores the discussion is capped at 4. Engagement is usually the cheaper improvement.