Section 3 is a conversation between two or more people about academic work — a student and a tutor, or two students planning a project. It is where most candidates' scores break, because the answers live inside disagreement: opinions get stated, challenged, softened, and reversed, and the test asks you what someone finally thinks. This lesson installs the tactics for listening to people who change their minds.
Why Section 3 is the hardest
Three properties stack against you:
- Multiple voices. The answer to "What does the student think about the survey data?" is worthless if you attribute the tutor's opinion to the student. Every question silently asks who as well as what.
- Opinions, not facts. Sections 1 and 2 deal in numbers and places. Section 3 deals in attitudes — agreement, doubt, enthusiasm, reluctance — which are expressed in tone and hedges, not keywords.
- The heaviest paraphrase on the test. Section 3 question writers rewrite everything, and — the cruel part — when you do hear a word from the options repeated verbatim in the audio, it is more often a trap than an answer. The correct option is usually the paraphrased one.
The question types match the difficulty: mostly multiple choice and matching, the two formats where a half-understood sentence produces a confident wrong answer.
Tactic 1: pin the voices
In the preview, note who is in the conversation — the questions tell you (What does Martin think..., The tutor suggests...). Then, in the first fifteen seconds of audio, lock each name to a voice: gender, pitch, role (the one asking for help is the student; the one giving advice is the tutor).
For every answer you write, run the one-word check: whose opinion was that? Question writers deliberately place the wrong person's opinion right next to the answer. When the question asks what the student concluded, the tutor's contrary view will be stated in the adjacent sentence, using tempting vocabulary from the options.
Tactic 2: the last opinion standing wins
Academic conversations move by proposal and revision:
"I thought we could hand out questionnaires in the library—" "Hmm, you'll get very few back that way." "You're right. Online, then. I'll set the form up tonight."
If the question asks how they will collect data, the answer is online — the questionnaire-in-the-library plan was born and killed inside four seconds. Never commit to an answer while the topic is still open. Wait for the topic to close — the right, so, okay then, that's settled moment — then write.
The markers of a mind changing: actually, on second thoughts, you're right, fair enough, I hadn't thought of that, mind you, having said that. Every one of these means: the opinion you just heard may already be obsolete.
Tactic 3: hear agreement and disagreement in their real clothes
The test almost never gives you I agree / I disagree. It gives you:
| Sounds like | Actually means |
|---|---|
| That's one way of looking at it... | disagreement, polite |
| I see what you mean, but... | disagreement — everything before but is discarded |
| Exactly what I was thinking. | strong agreement |
| Do you think so? (falling, flat tone) | doubt |
| I suppose so. | weak, reluctant agreement |
| Well, it depends... | refusing to agree |
| You could say that. | lukewarm — not endorsement |
When a matching task asks you to label opinions (positive / negative / mixed), these hedged forms are the answers. Drill the table until I suppose so registers as reluctance without conscious translation.
Tactic 4: use the tutor's structure
Even a free-flowing discussion has a skeleton, and usually one speaker (the tutor, or the more organised student) builds it: Let's start with your methodology... Now, about the second chapter... One last thing before you go. These are the same signposts as Section 2 (lesson 4), just embedded in dialogue. They tell you when a question closes and the next opens — which is exactly when it is safe to commit ink.
The question order still follows the audio. Pen on the live question, dash-and-jump when you lose one (the recovery rule from lesson 4 applies unchanged).
What to train, and how
Section 3 misses are almost never sound misses — the vocabulary is manageable. They are meaning misses (paraphrase, hedges) and attribution misses (right opinion, wrong person). So the training is transcript work:
After every Section 3 you attempt, take the questions you got wrong and find the trap in the transcript. For each miss, write one line: what the audio said, what the correct option said, and which device connected them — synonym, hedge, reversal, or wrong-voice bait. You are building a personal catalogue of the test's tricks, and the catalogue transfers: the same devices repeat test after test.
This is also where dictation (lesson 2) earns interest: a Section 3 dictation session forces you to transcribe the hedges and reversals word by word, which is the fastest way to stop your brain smoothing them over.
Your drill (30 minutes)
- Open Listening 2026-05 Test 5, Section 3. Preview: identify the speakers, underline anchors, note which questions ask for opinions versus decisions.
- Play once, exam conditions. For every answer, run the whose opinion? check before writing.
- Mark it. For each miss, open the transcript and write the one-line trap analysis: audio words → option words → device (synonym / hedge / reversal / wrong voice).
- Replay just the sentences around your misses. Hear the reversal markers you missed live.
- This week, run the same routine on Section 3 of Listening 2026-04 Test 3 and Listening 2026-02 Test 1. Our audio is recreated from real exam recalls with exact transcripts, so every trap you fell for can be located and named. Three tests in, your trap catalogue will already predict most of what test four throws at you.