TOEFL Speaking Task 4 Mastery — From 22 to 28+

Table of Contents

- Chapter 1: How TOEFL Speaking Task 4 is Scored (FREE)

- Chapter 2: 3 Model Responses (FREE: Response 1)

- Chapter 3: 20 Model Responses (PAID)

- Chapter 4: 100 Power Phrases (PAID)

- Chapter 5: 30 Common Topics & Templates (PAID)

- Chapter 6: 30 Common Errors (PAID)

- Chapter 7: 20 FAQs (PAID)

- Appendix A-E

---

Chapter 1: How TOEFL Speaking Task 4 is Scored

1.1 The Format — What Actually Happens on Test Day

Let's start with the raw mechanics, because most candidates fail Task 4 not from weak English but from not knowing exactly what the task demands.

Here is the sequence, in order, with no surprises:

1. You hear a lecture excerpt. It runs 60–90 seconds. A professor (voice actor) delivers a short academic mini-lecture. Topics come from intro-level university courses: biology, psychology, economics, environmental science, art history, business, etc.

2. You get 20 seconds of preparation time. A countdown timer appears. No paper is graded, but you take notes on scratch paper.

3. You speak for up to 60 seconds. A clock counts up. You should aim to fill 50–60 seconds.

4. You are scored 0–4 on that single response, then it's converted into your 0–30 Speaking section score.

That's it. One lecture, one question, one minute.

The question itself is almost always phrased one of these ways:

- *"Using the examples from the lecture, explain [concept]."*

- *"Explain how [concept] works, using the examples the professor gives."*

- *"Using the points and examples from the lecture, explain [concept]."*

Notice the pattern: the question always asks you to USE the lecture to EXPLAIN something. You are never asked for your opinion. You are never asked to agree or disagree. You are a summarizer, not a commentator.

(中文：Task 4 的题目永远是"用讲座里的例子解释某个概念"，你只负责复述，不发表任何个人观点。)

1.2 No Reading Material — The Only Pure-Listening Task

This is the single most important structural fact about Task 4, and it's the reason many otherwise-strong candidates plateau at 22–24.

Task 2 (Campus Situation): reading + listening → speaking

Task 3 (Academic Course Topic): reading + listening → speaking

Task 4 (Academic Lecture): listening only → speaking

In Tasks 2 and 3, the reading passage gives you a conceptual anchor. You read a definition, a policy, a theory — then the audio gives you an example. Even if you miss half the audio, the reading rescues you. You know what the topic is. You know the vocabulary. You can guess.

Task 4 gives you nothing. There is no safety net. If you don't catch the main topic in the first 10–15 seconds, you are lost for the entire response. There is no text to glance back at. There is no title on a screen. The professor simply starts talking, and you must build the entire mental scaffold from sound alone.

This is why Task 4 is widely considered the hardest of the four Speaking tasks — not because the language is harder, but because the cognitive load is highest. You're listening, comprehending, selecting, organizing, and note-taking all at once, with zero visual support.

(中文：Task 4 是唯一没有阅读材料的任务。没有"救命稻草"，全靠听力。前 10 秒抓不住主题，整题就废了。)

1.3 Score 0–4, Mapped to 0–30

Each Speaking task is scored 0–4 by a trained rater (and increasingly, by an AI + human hybrid system). Your four task scores are averaged, then converted to the 0–30 scale.

Here's the practical mapping you need to internalize:

Task 4 Score	What It Means	Rough 0–30 Equivalent (per-task average)
**4.0**	Fully intelligible, well-organized, complete summary	28–30
**3.5**	Strong, minor gaps in detail or fluency	26–27
**3.0**	Understandable, but incomplete or repetitive	23–25
**2.5**	Noticeable problems: missing points, choppy delivery	20–22
**2.0**	Hard to follow, major content missing	17–19
**1.5**	Very limited, fragmented	13–16
**1.0**	Minimal intelligible content	8–12
**0**	No response / unintelligible / off-topic	0–7

The 26+ target: To hit 26+ overall in Speaking, you generally need at least two tasks at 3.5–4.0 and the others no lower than 3.0. For Task 4 specifically, you want to be consistently at 3.5 or above.

The jump from 22 → 28 is NOT about vocabulary. It's about completeness and clarity. A 3.0 response is understandable but leaves the listener thinking "something's missing." A 4.0 response makes the listener feel they just heard a clean, complete mini-lecture summary.

1.4 The Three Scoring Criteria — In Detail

ETS officially scores speaking on three dimensions. Understand each one, because you can lose points on any of them independently.

Criterion 1: Delivery

What it measures: How clearly and fluently you speak.

Sub-factors:

- Pronunciation — Are individual sounds and words intelligible? (You do NOT need a native accent. You need to be understood.)

- Intonation & stress — Do you stress the right words? Does your pitch rise and fall naturally, or is it flat?

- Fluency — Do you speak smoothly, or with long pauses, false starts, and "uh… uh…"?

- Pacing — Not too fast (garbled), not too slow (you run out of time).

What kills your Delivery score:

- Long silences (3+ seconds) mid-response

- Repeated filler ("um, um, um")

- Monotone delivery — the rater can't tell what's important

- Mumbling the technical terms

Key insight: Delivery is scored holistically. A few small stumbles are fine if the overall flow is smooth. But if you pause 5 times for 3 seconds each, you're capping yourself at 3.0 no matter how good your content is.

(中文：Delivery 看的是"顺不顺、清不清楚"。不需要美音，但必须流畅、有语调、不能长时间卡壳。)

Criterion 2: Language Use

What it measures: Your grammar and vocabulary as demonstrated in the response.

Sub-factors:

- Grammar — Sentence structure, verb tense, agreement. Are your sentences well-formed?

- Vocabulary — Do you use appropriate academic words? Can you paraphrase the professor's terms, or do you just repeat them awkwardly?

- Cohesion — Do your sentences connect logically? (First… Second… For example…)

What kills your Language Use score:

- Broken sentences that trail off

- Wrong verb tenses (mixing past/present randomly)

- Over-reliance on one structure ("The professor says… The professor says… The professor says…")

- Vocabulary so basic it can't convey the concept

Key insight: You do NOT need fancy words. You need accurate, connected sentences. "The professor explains that the birds use the stars to find their way" beats "The professor, uh, the bird, it use star, uh, for direction" every time.

Criterion 3: Topic Development

What it measures: Did you actually summarize the lecture accurately and completely?

Sub-factors:

- Main topic captured — Did you state what the lecture is about?

- Supporting points included — Did you cover the 2–3 key points?

- Examples included — Did you mention the specific examples the professor gave?

- Accuracy — Did you get the facts right?

- Completeness — Did you cover everything important within 60 seconds?

What kills your Topic Development score:

- Only covering ONE of the two points

- Describing the topic vaguely ("something about animals")

- Getting the example wrong

- Adding your own opinion (which is irrelevant and wastes time)

- Running out of time before the second point

Key insight: This is where 22 vs. 28 is decided. A 22-level candidate speaks clearly but covers only 60% of the content. A 28-level candidate covers 90%+ of the content and organizes it cleanly.

(中文：Topic Development 是拉开差距的关键。说得再流利，如果只讲了一半内容，最多 3.0。必须覆盖主题+2-3个要点+例子。)

1.5 The Structure Template — Your 60-Second Skeleton

Here is the template that reliably produces 3.5–4.0 responses. Memorize the frame, not the words.

> "The professor discusses [MAIN TOPIC]. First, she explains [POINT 1]. For example, [EXAMPLE 1]. Second, she mentions [POINT 2]. For example, [EXAMPLE 2]."

That's the core. Let's break down the timing:

Segment	Time	Content
Opening	0–8s	State the main topic
Point 1	8–28s	First key point + example
Point 2	28–50s	Second key point + example
Closing (optional)	50–60s	Brief wrap-up or third point if time

Why this works:

- It signals structure immediately, which boosts Language Use (cohesion).

- It ensures you hit both points, which boosts Topic Development.

- It's simple enough to execute under pressure, which boosts Delivery.

Important: This is a FRAME, not a script. Do not memorize fixed sentences. The content changes every time. What stays constant is the organizational pattern: topic → point 1 → example → point 2 → example.

(中文：记住的是"框架"不是"模板句"。主题→要点1→例子→要点2→例子。内容每次都变，框架不变。)

1.6 The Golden Rule: Summarize, Don't Opine

This is non-negotiable and it's where nervous candidates sabotage themselves.

You must summarize the lecture. You must NOT give your own opinion.

Signs you're accidentally giving an opinion:

- "I think this is a good example because…"

- "In my country, we also…"

- "This is very interesting because…"

- "The professor is right that…"

These sentences waste precious seconds and do nothing for your score. The rater is listening for whether you captured the lecture, not whether you have thoughts about it.

What to do instead: Every sentence should be traceable to something the professor said. If you can't point to "the professor said this," don't say it.

Exception that isn't an exception: You can use framing language like "The professor explains…" or "According to the lecture…" — this is not opinion, it's attribution. That's exactly what you want.

(中文：绝对不要加个人观点。每句话都要能追溯到"教授说的"。加了观点 = 浪费时间 + 不得分。)

1.7 Note-Taking Strategy — Capture the Skeleton

You have 60–90 seconds of audio and limited working memory. Your notes must be fast, sparse, and structured.

The 3-layer note system:

Layer 1 — Main topic (capture in first 15 seconds):

The professor almost always states the topic in the first 1–2 sentences. Write it as a short phrase.

- Example: "animal navigation — magnetic field"

Layer 2 — Supporting points (2–3 of them):

Listen for signpost words: *First… Second… Another… Also… For example… For instance…*

Write each point as 3–5 words.

- Example: "1) sense magnetic field / 2) use as compass"

Layer 3 — Examples (one per point):

Each point usually comes with a concrete example. Write a keyword or two.

- Example: "birds / sea turtles"

What your notes should look like (real example):

TOPIC: animal navigation — Earth's magnetic field

P1: detect magnetic field
   ex: birds migrate thousands of miles

P2: use as internal compass
   ex: sea turtles return to birth beach

That's it. That's all you need. If you write full sentences, you're too slow and you'll miss the second point.

Note-taking rules:

1. Abbreviate everything. "magnetic" → "mag", "navigation" → "nav".

2. Use symbols. → for "leads to", = for "is", + for "and".

3. Skip articles and small words. "the bird uses the field" → "bird use field".

4. Leave space between points. You'll add details as they come.

5. Never write during the first sentence. Just listen — the topic is coming.

(中文：笔记只记骨架——主题 + 2-3 个要点 + 每个要点的例子。写关键词，不写整句。第一句话先听，别急着写。)

1.8 Top 5 Chinese Candidate Mistakes

After coaching hundreds of Chinese candidates, these five mistakes appear again and again. Fix these and you'll gain 3–5 points almost immediately.

Mistake #1: Missing the main topic

Many candidates start speaking without clearly stating what the lecture is about. They dive straight into a detail. The rater thinks: "What is this about?"

Fix: Your first sentence MUST state the main topic. "The professor discusses how animals navigate using Earth's magnetic field." Done. Now the rater knows the frame.

Mistake #2: Only covering one point

The lecture almost always has two main points (sometimes three). Candidates who only cover one point cap themselves at 3.0 — no matter how fluent they are.

Fix: Before you start speaking, glance at your notes and confirm you have two points written down. If you only have one, you missed half the lecture. Train yourself to listen for "Second…" or "Another way…"

Mistake #3: Memorizing templates

Candidates who memorize fixed sentences ("The professor discusses a topic. First, she explains a point. Second, she mentions another point.") sound robotic and often force content into a template that doesn't fit.

Fix: Memorize the structure (topic → point 1 → example → point 2 → example), but generate the words fresh each time based on the actual lecture.

Mistake #4: Poor notes

Either writing too much (full sentences, can't keep up) or too little (just the topic, no details). Both lead to incomplete responses.

Fix: Practice the 3-layer system above. Aim for 8–12 words total on your scratch paper. That's enough.

Mistake #5: Panic from no reading

The absence of reading material triggers anxiety. Candidates freeze in the first 10 seconds, miss the topic, and never recover.

Fix: Reframe it. No reading means no reading to worry about. You only have one input: the audio. Train with pure-listening practice (no transcript) until the first 15 seconds feel routine. The topic is ALWAYS in the first two sentences. Trust that.

1.9 Putting It Together — What a 28+ Response Looks Like

Before we move to model responses, let's crystallize what separates a 28+ response from a 22 response.

Dimension	22-Level Response	28-Level Response
**Delivery**	Understandable but choppy; some long pauses; flat intonation	Smooth, natural pacing; clear stress on key terms; minimal hesitation
**Language Use**	Simple sentences; some grammar slips; repetitive connectors	Varied but accurate sentences; clean connectors (First/Second/For example)
**Topic Development**	Covers main topic + one point; vague on examples	Covers main topic + two points + both examples; accurate
**Time management**	Runs out of time or finishes in 35s	Fills 50–60s with complete content
**Opinion**	Occasionally adds personal view	Purely summarizes

The 28+ candidate is not a genius. They simply executed the frame completely under pressure.

Here's the mental checklist you should run during your 20-second prep:

1. ✅ What's the main topic? (Write it as a phrase.)

2. ✅ What are the two points? (Number them.)

3. ✅ What's the example for each? (One keyword each.)

4. ✅ First sentence: "The professor discusses ___."

5. ✅ Connectors ready: First… For example… Second… For example…

If you can run this checklist in 20 seconds, you're ready to speak.

(中文：20 秒准备时，脑子里过一遍：主题是什么？两个要点是什么？每个要点的例子是什么？第一句怎么说？——想清楚这四点，开口就稳了。)

1.10 Why This Chapter Matters More Than Any "Tips" Video

Most Task 4 advice online is either too vague ("just relax and speak clearly") or too gimmicky ("use this magic phrase to get 30!"). Neither helps.

What actually moves your score is understanding:

1. The exact format — so nothing surprises you.

2. The three criteria — so you know where points come from.

3. The structure template — so you have a repeatable skeleton.

4. The note-taking system — so you capture the right content.

5. The five mistakes — so you stop losing points you don't need to lose.

Everything in Chapter 2 builds on this foundation. You'll see three real lectures, three sets of notes, three full responses, and a detailed scoring breakdown for each. By the end, you'll be able to reverse-engineer your own responses and self-diagnose exactly why you're stuck at 22, 24, or 26.

The path from 22 to 28+ is not about becoming a better English speaker. It's about becoming a more disciplined summarizer. That's a skill you can drill in a week.

---

Chapter 2: 3 Model Responses with Detailed Feedback

Now we get concrete. Below are three complete Task 4 walkthroughs. Each includes:

- The lecture content (summarized)

- The 20-second prep notes (exactly what to write)

- The full 60-second response (transcribed)

- Delivery notes

- A criterion-by-criterion scoring analysis

Response 1 is free. Responses 2 and 3 are for paid subscribers.

Read Response 1 carefully. Then, before you look at the analysis, try to score it yourself using the three criteria. This active practice is what makes the models stick.

---

Response 1 (FREE): "How Animals Navigate Using Earth's Magnetic Field"

The Lecture (summary, ~120 words)

The professor explains that many animals can navigate long distances using Earth's magnetic field. She describes two ways this works.

First, some animals can detect the magnetic field directly and use it to determine direction. For example, birds such as the Arctic tern migrate thousands of miles each year, and they can sense the magnetic field to stay on course even when landmarks are invisible — for instance, over open ocean.

Second, animals can use the magnetic field as an internal compass that works alongside other senses. The professor gives the example of sea turtles. Baby sea turtles hatch on a beach and immediately enter the ocean, then years later return to the same beach to lay eggs. They use the magnetic field as a map-like compass to find their way back to their birth beach.

The professor emphasizes that this ability is innate — the animals are born with it, not taught.

The 20-Second Prep Notes

Here's exactly what a strong candidate writes during prep:

TOPIC: animals navigate — Earth's magnetic field

P1: detect field → direction
   ex: Arctic tern, migrate 1000s miles, ocean

P2: internal compass + other senses
   ex: sea turtles, return to birth beach

NOTE: innate (born with it)

That's under 30 words. Notice:

- Topic in one line

- Two points, numbered

- One example each, with a keyword

- One extra note (innate) in case there's time

The Full Response (transcript, ~90 words, ~55 seconds)

> "The professor discusses how animals navigate using Earth's magnetic field. First, she explains that some animals can detect the magnetic field directly to find direction. For example, birds like the Arctic tern migrate thousands of miles, and they use the magnetic field to stay on course, even over the open ocean where there are no landmarks. Second, she mentions that animals can use the magnetic field as an internal compass along with other senses. For example, sea turtles return to the beach where they were born to lay their eggs, using the magnetic field to find their way back. The professor notes this ability is innate."

Word count: ~95 words. Time: ~52–56 seconds at natural pace.

Delivery Notes

- Pacing: Steady. No rushing. The speaker pauses briefly after "magnetic field" and "First" to signal structure.

- Stress: Key terms stressed — *magnetic field*, *Arctic tern*, *internal compass*, *sea turtles*, *innate*.

- Intonation: Slight rise on "First…" and "Second…" to mark transitions. Falls at the end of each example.

- Fluency: No long pauses. One small hesitation is fine; the speaker does not restart sentences.

- Clarity: Technical terms (magnetic field, Arctic tern, innate) pronounced clearly.

Scoring Analysis

Delivery: 4.0

The response is smooth, well-paced, and easy to follow. Transitions are marked clearly with intonation. No disruptive pauses. All key terms are intelligible. This is exactly what a 4.0 delivery sounds like — not fast, not slow, just *clear*.

Language Use: 4.0

Sentences are well-formed and varied:

- "The professor discusses how animals navigate…" (complex noun clause)

- "First, she explains that some animals can detect…" (reported speech)

- "For example, birds like the Arctic tern migrate…" (example structure)

- "Second, she mentions that animals can use…" (parallel structure)

Connectors are clean: *First… For example… Second… For example…* No grammar errors. Vocabulary is appropriate: *navigate, detect, migrate, internal compass, innate*.

Topic Development: 4.0

- ✅ Main topic stated in first sentence

- ✅ Point 1 covered with example (Arctic tern)

- ✅ Point 2 covered with example (sea turtles)

- ✅ Extra detail included (innate) — bonus

- ✅ No opinion added

- ✅ Complete within time

This response would score 4.0 and contribute strongly to a 28+ Speaking section.

Why This Response Works

Three things make it a 4.0:

1. The frame is executed perfectly. Topic → Point 1 → Example → Point 2 → Example → Extra detail. The rater never has to wonder "where is this going?"

2. Both points are covered. Many candidates would stop after the Arctic tern example and run out of content. This response covers both points with time to spare.

3. The language is simple but accurate. No fancy vocabulary. No complex grammar that risks errors. Just clean, connected sentences that convey the content.

What a 3.0 Version Would Look Like

For contrast, here's what the same lecture might sound like from a 3.0 candidate:

> "The professor talks about animals and magnetic field. Uh, birds, they use magnetic field to, uh, migrate. Like Arctic tern, they fly many miles. And also turtles. Sea turtles, they go back to beach. So animals use magnetic field. Yeah."

Why this is a 3.0:

- Topic stated vaguely ("animals and magnetic field")

- No clear "First… Second…" structure

- Hesitations ("uh") disrupt delivery

- Second point not explained (why do turtles return?)

- Missing the "innate" detail

- Ends weakly ("Yeah")

Same content, but disorganized and incomplete. That's the difference between 3.0 and 4.0.

(中文：同样的内容，3.0 和 4.0 的差别在于——有没有清晰框架、有没有覆盖两个要点、有没有卡壳。内容不是关键，组织才是。)

---

*Responses 2 and 3 are available to paid subscribers. They cover:*

- Response 2: "Why Some Birds Abandon Their Chicks" — a psychology/biology lecture with a two-point structure and a counterintuitive example.

- Response 3: "How Plants Communicate Through Chemicals" — a botany lecture with a three-point structure, testing whether you can compress three points into 60 seconds.

*Each includes full notes, transcript, delivery analysis, and criterion-by-criterion scoring.*

---

End of free preview. The paid chapters continue with Responses 2 and 3, plus a drilling plan to move from 22 to 28+ in two weeks.

---