Quick answer
TOEFL Listening is 29 minutes long and contains 47 items across two adaptive modules. You meet four task types: Choose a Response, Announcements, Conversations, and Academic Talks. Plan for one listen per clip. Your Listening band is reported from 1.0 to 6.0 in half-point steps, aligned to CEFR levels.
What you are actually sitting down to
Listening is the second shortest section of the test by clock time and the densest by item count. You get 29 minutes and 47 questions. That is roughly 37 seconds per item on average, and a large slice of every minute is spent listening rather than answering, so the real time you have to think is smaller than the arithmetic suggests.
That single fact explains most of what goes wrong for test takers. Students who lose marks in Listening usually do not have a comprehension problem. They have a recovery problem. One clip goes badly, they spend the next ninety seconds replaying it in their head, and they miss the opening of the next one. By the time they settle, three items have gone.
So the goal of this first lesson is not vocabulary or accent training. It is orientation. If you know exactly what is coming, in what order, and what each task is asking of you, you stop spending attention on surprise and start spending it on the audio.
The section is built from four task types, and they are not distributed evenly. Short reply items make up the largest single block. Longer passages, the conversations, announcements and academic talks, carry more questions each but appear fewer times. You will see the exact shape of that in a moment.
One more thing to accept early: the audio is not designed to be difficult English. It is designed to be normal English delivered at normal speed with normal reductions. Sentences run together. Speakers say gonna and wanna and swallow the ends of words. Nothing about that is a trick. It is simply what a lecture and a campus conversation actually sound like, and the test would be invalid if it were cleaner.
The four Listening task types
| Task type | What you hear | What the questions ask |
|---|---|---|
| Choose a Response | One person says one short thing to you. A question, a complaint, an offer, a piece of news. Usually one or two sentences. | Which of four replies is the natural next turn in the conversation. One question per clip. |
| Announcement | A single speaker reading a campus or public notice. Financial aid session, hot water shutdown, career fair, library hours. | The purpose of the announcement, plus one specific detail such as a time, a place, a requirement or a reason. |
| Conversation | Two speakers, back and forth, on a practical campus matter. A trip, a paper, a schedule clash, an exhibit. | What the problem is, what each speaker agrees to do, why someone mentions something, and what happens next. |
| Academic Talk | A lecturer speaking on one topic from a real university subject. Psychology, economics, astronomy, philosophy, biology and more. | Main topic, stated facts, an inference the speaker implies but never says, and what the talk will move on to next. |
What each task sounds like and what the questions after it want. Task names come from the shipped test data, not from an outside guide.
Hear it now: a Choose a Response item
Choose the best response.
Choose a Response gives you one line and asks what a normal person says back. Two things are happening in that line: a yes or no question, and a reason that quietly explains why she is asking rather than moving. The reply answers the question and accounts for the seat, which is what a real conversation does.True, on topic, and answering a question nobody asked. This is the commonest distractor in the task: it reuses the setting so it feels connected, while responding to nothing that was actually said.It uses "the row behind" straight from the clip, which is exactly why it is there. Repeated vocabulary is a warning sign in this task, not a confirmation. It also does not answer whether the seat is free.Advice nobody asked for. The task is testing whether you can hear the function of an utterance, and the function here was a request, not a complaint about seating.Hear it now: an announcement
You signed up for the workshop last week. What do you need to do?
Two facts, and the question only works if you hold both. The room changed, and people who already signed up keep their place. Announcements are built this way on purpose: several small facts, and the question picks the pair that interact.Halloway 210 is the old room, and it is named first. First-mentioned numbers and room codes are the easiest thing to write down and the most likely to be superseded two sentences later.The six spaces are for people who have not signed up. You have. This option is correct information aimed at the wrong listener, which is how these distractors are usually built.Half right, which is worse than obviously wrong. The room is right, but the announcement says explicitly there is no need to register again.What the 29 minutes actually feel like
- 1
Minutes 0 to 1: the intro audio
A short spoken introduction sets up the module. Do not treat this as dead time. This is the moment your ear locks on to the voice, the recording level and the pace. Sit up, put both hands where you will write, and listen to it properly. Students who tune out here spend the first real clip catching up.
- 2
The short reply block
You then work through a run of Choose a Response items. These come fast. Each one is a single short utterance followed by four possible replies. There is no reading passage to hide in and no second chance to hear the line, so your whole job is to catch the speaker's intent in one pass: are they asking, complaining, offering, or sharing news? Answer, commit, move on. Lingering on a short item you already half-guessed is the most expensive habit in this section.
- 3
The conversation block
Next come the two-speaker conversations, each followed by a small set of questions. The audio is longer, so the pressure changes: now you have to hold two people's positions in your head at once. Who has the problem, who offers the solution, who agrees to do what. Most conversation questions in the shipped tests are about exactly those three things.
- 4
The announcement block
Announcements are the most predictable material in the section and the easiest place to bank points. There is almost always a purpose question and almost always a detail question. The detail is a time, a room number, a required document, an eligibility rule or a phone extension. Knowing that in advance tells you what to write down while the clip is still running.
- 5
The academic talk, then the second module
The module closes with a lecture on one academic topic, carrying more questions than any other single clip. Then the second module begins and the same shape repeats in a shorter form. Nothing new is introduced. If you have trained the four task types, the second module is the same test at a different difficulty setting.
What the section is really testing
It is tempting to describe Listening as a memory test. It is not. Almost nothing in the shipped question set asks you to recall a long chain of detail. Look at what the questions actually target and a pattern appears very quickly.
Intent, not words. A speaker says "I can't believe how long the line is at the registrar's office." The correct reply is not about the registrar's office. It is "You can do most things online now, you know." The item is testing whether you heard a complaint and recognised that the natural next turn is a solution. Every wrong option in that item contains real, true, topic-relevant information. Truth is not the criterion. Fit is.
Relationships, not facts. Conversation questions ask why a speaker mentions something, what one speaker agrees to do, what the other will probably do next. Those are all questions about how the two turns connect. If you write down only nouns while you listen, you will have the topic and none of the relationships, and you will lose exactly these items.
One implied idea per talk. Academic talks reliably include a question you cannot answer from any single sentence. The talk on the Zeigarnik Effect describes psychological tension that keeps a task active in memory until closure, and then asks what can be inferred about unfinished tasks. The answer, that they occupy the mind until completed, is assembled from the explanation rather than quoted from it. That is the one place in a talk where careful listening pays more than fast note-taking.
Forward direction. Lectures often end with a signposted turn: "Next, let's take a look at some popular apps and inventions that people use today to improve their focus." That last line is a question waiting to happen. Never let your attention drop in the final ten seconds of a talk, which is exactly when most tired test takers relax.
Check your orientation
Three questions on format and approach. Each one maps to something a real test taker gets wrong in their first week of practice.
You have 29 minutes for 47 Listening items. What does that mean for how you handle a clip you did not understand?
Time in Listening is spent listening, not deliberating. Roughly 37 seconds per item on average, most of it audio, means there is no slack to reconstruct a clip you already missed. The recoverable cost is one item. The unrecoverable cost is losing the opening of the next clip while you brood.Reconstruction sounds responsible but it is the single most expensive habit in this section. You cannot replay the audio in your head accurately enough to fix the item, and the attention you spend trying is taken directly from the next clip.Coming back later assumes spare time at the end. With 47 items in 29 minutes there is no spare time, and a clip you have already heard once cannot be re-heard just because you returned to the item.Option length is not a signal. In the shipped test data the long options are frequently the distractors, precisely because extra detail is how a plausible-sounding wrong answer gets built.Why does the first Listening module deserve more of your energy than the second?
Listening runs as two adaptive modules and the first is the router. How you do there determines which difficulty band the second module is drawn from, which is why arriving already warmed up matters so much.Nothing in the test data supports per-item weighting differences between modules. The routing effect is the reason the first module matters, not a points multiplier.Both modules contain academic talks. The first module is not defined by holding the hardest task type.Both modules sit inside the same 29 minute section clock. Neither is untimed.A speaker says: "I left my umbrella at home again, and it looks like it is about to rain." What kind of reply is the test looking for?
The speaker has not asked for anything directly, but the implication is clear: they need an umbrella now. The natural next turn is an offer, which in the shipped item is "I have an extra one you can borrow." Choose a Response is almost always testing implied intent rather than surface content.Related weather information is exactly how the distractors are built. "It rained a lot last week too" is true, relevant sounding, and completely unresponsive to what the speaker needs.Corrections such as "You should check the weather forecast" are grammatically fine and conversationally cold. The test consistently prefers the cooperative reply over the corrective one.Asking for repetition is never the intended answer in this task. The clip is always clear enough to respond to on one listen.Before your first full Listening practice test
- Test your headphones on a clip you have never heard, not on music you know well.
- Have paper and a pen ready before the intro audio starts, not after the first clip.
- Decide in advance what you will write for each task type, so you are not inventing a note system mid-clip.
- Accept in advance that you will miss something. Decide now that you will let it go rather than chase it.
- Do the whole 29 minutes in one sitting. Practising in five-minute chunks trains a stamina you will not have on test day.
- Sit somewhere with normal background noise at least once. A perfectly silent room is not what the test centre gives you.
How to use the rest of this course
The lessons that follow are ordered the way a section is ordered, not the way a textbook is. First scoring, so you know what a band actually means and stop guessing at your level. Then one lesson per task type, each with real audio you can listen to and answer, because reading about Choose a Response and doing Choose a Response are not the same skill.
After the task lessons come the cross-cutting ones: note-taking, accents and connected speech, listening for attitude and function, pacing across the full 29 minutes, and finally the distractor patterns that make wrong answers feel right. If you only have time for four lessons before your exam, do the task lesson for whichever type you fail most, plus the traps lesson, plus note-taking, plus pacing.
Every audio lesson in this course carries clips with transcripts and full reasoning for every option, right and wrong. Read the wrong-answer reasoning even when you got the item right. That is where the pattern lives, and the pattern is what transfers to the clip you have not heard yet.
Common questions
Listening is 29 minutes and contains 47 items. It runs as two adaptive modules inside that single section clock.
There are 47 items in total. In our practice tests these are split as 30 items in the first module and 17 in the second, matching the two-module adaptive design.
Four: Choose a Response, Announcements, Conversations, and Academic Talks. Choose a Response is a short reply item; the other three are longer clips carrying several questions each.
Yes, and you should. What matters is what you write. Notes that capture relationships, such as who has the problem and who offers to do what, beat notes that capture only nouns.
As a band from 1.0 to 6.0 in half-point steps, aligned to CEFR levels. The next lesson in this course covers what each band means and how routing between the two modules affects it.
No. The test plays at normal conversational and lecture speed, with normal reductions and linking. Slowing the audio trains an ear for a version of English you will not hear on test day. Train at full speed and use the transcript afterwards instead.