IELTS Speaking Part 3 is the Discussion: 4 to 5 minutes in which the examiner returns to your Part 2 topic and asks about it, in the official wording, “in a more general and abstract way and, where appropriate, in greater depth“.
That sentence is the whole design. Part 2 asked about your bike; Part 3 asks whether cities should be built around cars. The topic is the same and the register is not.
|
Want to FINALLY Pass the IELTS Academic Test And Reach Your Goals? Get Your FREE Lume Account Today to Unlock… ✅ 10,000+ PRACTICE QUESTIONS with detailed answers ✅ FREE MOCK TESTS with fast & accurate AI scoring ✅ Get BETTER SCORES with a PERSONALIZED study plan (Limited seats available!) |
| Part 1 | Part 2 | Part 3 | |
|---|---|---|---|
| Subject | You | One specific instance | Society and general principles |
| Length | 4-5 min | 3-4 min | 4-5 min |
| Answer length | 20-30 sec | 2 min uninterrupted | 40-60 sec |
| Register | Conversational | Narrative | Abstract and evaluative |
| Interaction | Question and answer | Monologue | Discussion, with follow-ups |
| Function | Typical wording | Answer shape |
|---|---|---|
| Opinion | Do you think… / Should… / To what extent… | Position, reason, concession |
| Comparison | What are the differences between… / Which is more… | Both sides, then which matters more and why |
| Cause and effect | Why do you think… / What has caused… | Main cause, secondary cause, mechanism |
| Speculation | How might… in future / What would happen if… | Conditional, likelihood, and what it depends on |
Recognising the function tells you the grammar to reach for. Speculation questions need conditionals and modals; comparison questions need comparative structures; cause questions need the language of mechanism.
| Question | Function | Band 7+ answer |
|---|---|---|
| Do you think people change jobs more often than they used to? | Opinion | “Almost certainly, though I would separate two things. Changing employers is clearly more common — the idea of staying somewhere for thirty years has largely gone. Changing occupation is different, and I am less sure that has increased as much, because retraining is still expensive and slow. So I would say mobility has gone up within a field more than between fields.” |
| What are the advantages of working from home? | Comparison | “The obvious one is the commute, which for a lot of people is an hour a day recovered. Beyond that it suits concentrated work — nobody interrupts you. The trade-off is everything informal: you lose the conversation on the way out of a meeting where half the useful information actually gets exchanged. On balance I think it works well for experienced staff and badly for new ones, who learn by overhearing.” |
| Why do some people stay in jobs they dislike? | Cause | “Several reasons stack up. The immediate one is financial — a mortgage makes risk expensive. Then there is the fact that skills become specific over time, so after ten years your options narrow rather than widen. And I suspect there is a psychological element too: leaving means admitting the last five years were a mistake, and people would often rather not.” |
| How might work change in the next twenty years? | Speculation | “I would expect the routine analytical work to go first, which is the opposite of what people predicted twenty years ago — the assumption then was that manual work was most exposed. If that happens, the jobs that survive will be the ones combining judgement with physical presence, which is an odd category. Though I would say confident predictions about work have a poor track record, so I hold that loosely.” |
| Question | Function | Band 7+ answer |
|---|---|---|
| Should cities restrict private cars in the centre? | Opinion | “I think so, but the sequencing matters enormously. Restricting cars before the alternative exists just penalises people who have no choice, and it tends to fall hardest on those who can least afford it. Where cities have built the public transport first and restricted afterwards, it has been broadly accepted. Where they have done it in the other order it has produced a political backlash that then undoes the whole thing.” |
| What are the differences between living in a city and a small town? | Comparison | “The usual answer is choice against community, and I think that is roughly right but incomplete. A city gives you options — work, food, people who share obscure interests. A small town gives you a social network you did not choose, which is a cost when you are twenty and an asset when you have children or get ill. What is often missed is that cities are also cheaper to be poor in, in the sense that you can live without a car.” |
| Why are some cities more pleasant to walk in than others? | Cause | “Mainly the age of the street layout, I would say. Cities laid out before cars have narrow streets and short blocks, which is exactly what makes walking interesting — you get a decision every thirty seconds. Cities designed around driving have long blocks and wide roads, and those are dull on foot regardless of how much money is spent on paving afterwards. Climate matters too, but less than layout.” |
| Will people still own cars in fifty years? | Speculation | “In dense cities, probably not in the numbers they do now — the economics of leaving an expensive asset parked for twenty-three hours a day are already poor. Outside cities I would expect ownership to persist, simply because shared models need density to work. So my guess is divergence rather than a single trend, which is usually the safer prediction.” |
| Question | Function | Band 7+ answer |
|---|---|---|
| Should university education be free? | Opinion | “I can see the argument both ways and I am genuinely uncertain. Free tuition widens access, which is the point. But it is also regressive in one specific sense — it transfers money to people who will go on to earn more than average, funded by taxpayers who mostly will not. My tentative position is that free access matters more at the point of entry than the funding elegance does, so I would accept the inefficiency.” |
| What is the difference between learning online and in a classroom? | Comparison | “Online is better for information transfer and worse for everything else. If the goal is to learn a defined body of material, a recorded lecture you can pause beats a live one. What a classroom provides is accountability and the incidental learning — the question someone else asks that you would not have thought of. So the format that wins depends on whether the subject is knowledge or judgement.” |
| Why do some students perform better than others with the same teaching? | Cause | “Home circumstances dominate, from everything I have read — whether there is a quiet place to work matters more than most things a school controls. Prior knowledge compounds too: a student who arrives already reading fluently gains more from each lesson, so gaps widen rather than close. Teaching quality matters, but it is acting on very unequal starting positions.” |
| How might schools be different in the future? | Speculation | “I would expect assessment to change before teaching does. If written work can be produced automatically, then coursework as a measure of anything collapses, and the response will probably be more supervised examination rather than less. That is not the direction educationalists wanted, but it may be where the incentives lead.” |
| Question | Function | Band 7+ answer |
|---|---|---|
| Has technology made people less social? | Opinion | “I think it has changed the shape rather than the amount. People I know have more contact with more people and less sustained contact with any of them, which is a real change but not straightforwardly a loss. What I would say is that it has made solitude harder to achieve than company, which is a reversal of how it used to be, and I am not sure that is well understood yet.” |
| What are the differences between how older and younger people use technology? | Comparison | “The stereotype is competence, and I think the real difference is trust. Younger users assume a platform is manipulating them and act accordingly; older users are more likely to take what they see at face value. That produces different failure modes rather than one group being better — the sceptical user dismisses things that are true, and the trusting one accepts things that are not.” |
| Strategy | Phrase | Then |
|---|---|---|
| Answer both sides | “It depends which way you look at it…” | Give one argument each way and say which you find stronger |
| Narrow the question | “It varies a lot by country, so let me take…” | Answer the narrower version fully |
| Say what it depends on | “That would hinge on whether…” | Name the condition, then answer under it |
| Admit uncertainty and continue | “I am genuinely uncertain, but my instinct is…” | Give the instinct and the reason for it |
| Use someone else’s view | “The argument people make is… and I find it partly convincing because…” | Evaluate rather than assert |
Every one of these produces 40 seconds of assessable language. “I do not know” produces none, and that is the only reason it costs you marks.
| Function | Phrases |
|---|---|
| Hedge a claim | I would say… / broadly speaking… / on the whole… / I hold that loosely |
| Concede then continue | Admittedly… but… / there is something in that, though… |
| Distinguish two things | I would separate two things… / that depends what we mean by… |
| Name a trade-off | the trade-off is… / it cuts both ways… / at the cost of… |
| Rank causes | the main driver is… / secondary to that… / less than people assume |
| Speculate carefully | I would expect… / it is plausible that… / that would depend on… |
| Problem | Why | Fix |
|---|---|---|
| Part 1-length answers | Part 3 expects development, and 20 seconds does not show it | Position, reason, concession — three moves minimum |
| Personal anecdotes | Part 3 is general and abstract; a story about your cousin misses the register | Use an example briefly, then generalise from it |
| Absolute claims | Reduces the range of structures and is easy to challenge | Hedge, then commit — “broadly, yes, though…” |
| Memorised opinion essays | Recognisable, and rarely fits the exact question | Prepare functions and phrases, not positions |
| Repeating the question back | Wastes several seconds each time | Start with your position |
| Losing the thread mid-answer | Damages coherence more than a short answer does | Decide your two points before you start the second sentence |
The Discussion: four to five minutes in which the examiner develops your Part 2 topic in a more general and abstract way, and where appropriate in greater depth. It is the final part of a Speaking test lasting 11 to 14 minutes in total.
Roughly 40 to 60 seconds, or four to six sentences — noticeably longer than Part 1 and much shorter than the Part 2 long turn. Part 3 rewards development, so a 20-second answer here reads as underdeveloped even when it is accurate.
Say what it depends on, argue both sides, or narrow the question to a case you can address. All of those produce assessable language, which is the only thing being marked. “I do not know” is the one answer that scores nothing.
No. All four criteria — fluency and coherence, lexical resource, grammatical range and accuracy, pronunciation — assess your English. The examiner does not evaluate whether your view is correct, popular or well-informed.
Briefly, and then generalise from them. Part 3 is set at the level of society and general principles, so an answer that stays inside your own experience misses the register the part is testing.
There is nothing to disagree with — the examiner asks questions rather than stating views. If a question contains an assumption you reject, saying so and explaining why is a strong answer, because it demands more complex language than agreement does.
|
Want to FINALLY Pass the IELTS Academic Test And Reach Your Goals? Get Your FREE Lume Account Today to Unlock… ✅ 10,000+ PRACTICE QUESTIONS with detailed answers ✅ FREE MOCK TESTS with fast & accurate AI scoring ✅ Get BETTER SCORES with a PERSONALIZED study plan (Limited seats available!) |
Share
Lume Test
Editor
Lume Test is a leading platform offering practice questions and mock tests for major English exams like PTE, IELTS, TOEIC, Duolingo English Test, and more. We provide up-to-date, comprehensive materials designed to help learners improve their language skills and achieve high scores. With expert-crafted tests that closely mirror real exam formats, Lume Test ensures you're fully prepared for success. Your goals are our mission!
