🔁 1. Ôn tập – 10 phút
🔁 Ôn thẻ từ vựng đến hạn
🧠 Nhớ lại buổi trước (không xem tài liệu)
- Chuỗi từ vựng trong Task 2 (buổi 20): cho ví dụ.
- “association” và “cause” khác nhau thế nào?
- Kể 3 cụm thể hiện mức chắc chắn trong bài giảng.
Trả lời to hoặc viết ra giấy, rồi mới mở lại buổi trước để kiểm tra.
🎧 2. Đầu vào – 25 phút
🎧 Bài nghe
A lecturer introduces the difference between correlation and causation in health research (giả định)
Nghe tối đa 2 lần trước khi xem lời thoại.
📄 Xem lời thoại (sau khi làm bài)
- Lecturer
- Today's topic is one of the most important ideas in public health research: the difference between finding that two things are linked and showing that one actually causes the other.
- Lecturer
- Let me start with an example. Imagine a large, long-term observational study of office workers in a hypothetical city. Researchers find that people who walk to work have fewer heart problems than those who drive. It's tempting to conclude that walking protects the heart. And it may well do. But an observational study like this can only show an association.
- Lecturer
- Why? The first problem is what we call confounding factors – other differences between the groups that might explain the result. People who walk to work may, on average, live closer to the city centre, have higher incomes, or simply be healthier to begin with. Any of these could account for some of the difference.
- Lecturer
- The second problem is reverse causation. Perhaps it isn't that walking makes people healthy, but that people who already have heart problems are less able to walk long distances, and so they drive. In that case the direction of the relationship is the opposite of what we assumed.
- Lecturer
- So how do researchers get closer to cause and effect? The strongest design is generally considered to be the randomised controlled trial, in which participants are assigned to groups by chance. Because the assignment is random, the groups should be similar in every respect except the one being tested. The difficulty is that many questions can't ethically or practically be studied this way – you can't randomly assign people to smoke for twenty years, for example.
- Lecturer
- Where trials aren't possible, researchers look for several kinds of evidence pointing in the same direction. Is there a plausible biological mechanism? Is the effect consistent across different populations? Does the risk increase with the level of exposure – what's called a dose–response relationship? None of these alone proves causation, but together they can make it highly likely.
- Lecturer
- Finally, a note on how findings are reported. A headline might say 'Walking cuts heart disease', when the study itself only found an association in one group of people. As readers – and as future researchers – it's worth asking three questions: what kind of study was it, how large was the sample, and what else might explain the result? Next week, we'll apply these questions to some recent health claims.
Giọng đọc là giọng máy của trình duyệt; giọng và tốc độ khác nhau tùy thiết bị. Bài thi thật dùng giọng người với nhiều giọng địa phương.
Câu 1–10: Complete the notes. ONE WORD ONLY.
- 1. Example: a large, long-term study of office workers.
- 2. Such studies can only show an .
- 3. Problem 1: confounding factors, e.g. distance from centre, higher or better initial health.
- 4. Problem 2: causation – people with heart problems may drive.
- 5. Strongest design: randomised controlled .
- 6. Participants assigned to groups by .
- 7. Some questions cannot be studied this way for or practical reasons.
- 8. Other evidence: a plausible biological .
- 9. Risk rising with exposure = a dose– relationship.
- 10. Questions to ask: type of study, size of the , other explanations.
✅ Câu 11–12: chọn A, B hoặc C
11. What does the lecturer say about the claim that walking protects the heart?
12. According to the lecturer, several kinds of evidence together can…
🧩 3. Ngôn ngữ trọng tâm – 10 phút: cụm danh từ phức (complex noun phrases)
| Đơn giản | Cụm danh từ phức (học thuật) |
|---|---|
| a study | a large, long-term observational study of office workers |
| a risk | a significantly higher risk of heart disease among drivers |
| a link | a weak but consistent link between walking and heart health |
| a problem | one of the most common problems in health research |
👉 Thứ tự tính từ trước danh từ: ý kiến/kích thước → thời gian → loại (a large, long-term, observational study). Thông tin dài đặt sau danh từ bằng of / among / between / in.
Xây cụm danh từ phức
- The researchers found a weak but consistent link sleep and concentration.
- There was a significantly higher risk of injury night-shift workers.
- It was a small trial of 200 volunteers.
- This is one of the most common mistakes health reporting.
🎧 4. Thực hành – 20 phút
- Nghe lại, mở lời thoại: gạch 6 cụm danh từ phức; viết lại mỗi cụm thành câu ngắn có động từ (10 phút).
- Phân loại mức chắc chắn: tìm 2 câu “đã biết”, 2 câu “có thể”, 2 câu “không thể khẳng định” (5 phút).
- Tóm tắt 1 phút theo cấu trúc: example → two problems → strongest design → other evidence → advice (5 phút).
💼 Dùng ngay – mô tả chính xác một quan sát an toàn – sức khỏe (thêm 5–10 phút)
Viết 3 câu khách quan theo mẫu, tránh kết luận nguyên nhân khi chưa có bằng chứng:
- During the visit on…, we observed that… (what, where, how many).
- This was associated with… / This may have contributed to…, although further checks are needed.
- No evidence was found that… / It is not yet clear whether…
Ví dụ: During the visit on 20 November, we observed that two of the five fire exits on the ground floor were partially blocked by boxes. This may have contributed to the slow evacuation recorded in the last drill, although further checks are needed.
📚 Từ vựng của buổi
- correlation – tương quan
Correlation does not prove causation.
- causation – quan hệ nhân quả
Showing causation requires stronger evidence.
- confounding factor – yếu tố gây nhiễu
Income may be a confounding factor.
- cohort – nhóm theo dõi (nghiên cứu)
The cohort was followed for ten years.
- randomised trial – thử nghiệm ngẫu nhiên
A randomised trial assigns people by chance.
- observational study – nghiên cứu quan sát
An observational study can show an association.
- association – mối liên hệ
There was an association between walking and health.
- risk factor – yếu tố nguy cơ
Smoking is a major risk factor.
- sample size – cỡ mẫu
The sample size was too small.
- reverse causation – nhân quả ngược
Reverse causation could explain the result.
✅ 5. Tự kiểm tra – 10 phút
✅ Tự kiểm tra
Nghe lại bản ghi của mình và tích những điểm bạn đã làm được. Ô tích chỉ lưu trên máy này.