Smoking, Pollution and Lung Disease Data
How to interpret data on risk factors for lung disease, tell correlation from causation and evaluate the evidence behind smoking restrictions. Includes the 4-mark exam question, a worked model answer and the examiner trap on saying a correlation proves causation.
Chapters (9)
Exam question [4 marks]
A study followed a large group of adults for many years. Smokers had a higher incidence of lung cancer than non-smokers, and the more cigarettes a person smoked each day, the higher the incidence. A newspaper claimed: "The study proves that smoking causes lung cancer." Evaluate this claim. Explain how evidence like this has led to laws restricting smoking.
Try it on paper first, then check the mark scheme below. The question is original and was written for this lesson.
Mark scheme
| Mark | Mark point | Also accept | Do not accept |
|---|---|---|---|
| 1 | Positive correlation: the more cigarettes smoked, the higher the incidence; but correlation alone does not prove causation | "as smoking increases, lung cancer increases", "link / association but not proof"; both halves needed (correlation described AND not proof): either half alone gets no mark | "smoking causes lung cancer" with no mention of correlation; "the graph goes up" with no link stated |
| 2 | Other variables (e.g. age, air pollution, occupation, family history) may differ between groups and could affect incidence | any sensible named variable; "not all variables controlled" | "other things" with no example or effect; "the sample was small" alone |
| 3 | A causal link is supported by a mechanism: chemicals in smoke (tar) are carcinogens that cause mutations in lung cells | "carcinogens", "mutations / damage to DNA in lung cells", "tar causes cancer cells" | "smoke is harmful" alone; "nicotine causes cancer" |
| 4 | Consistent results from many studies with other variables controlled led governments to pass laws (statutory restrictions), e.g. smoking bans in public places | "repeated studies", "reliable evidence", "ban in enclosed public places", "advertising restrictions"; "repeated studies led to bans in public places" scores even without mention of controlled variables | "the government banned it because it is bad" with no reference to evidence |
Marks 1 and 2 evaluate the claim from the data, mark 3 is the mechanism that supports cause, mark 4 links evidence to statutory restriction. In data questions candidates must quote figures from the data given; this question has no figures, so describe the trend. "Proves" is a trap word: a correlation alone never proves a causal relationship. Pollution is credited as an example of another variable or of a regulated risk factor.
Model answer (4/4)
There is a positive correlation: the more cigarettes smoked each day, the higher the incidence of lung cancer. However, a correlation alone does not prove that smoking causes lung cancer. Other variables, such as age, air pollution or occupation, may differ between the smokers and non-smokers and could explain the difference. A causal link is supported by a mechanism, because tar in smoke contains carcinogens that cause mutations in lung cells. Consistent results from many studies, with other variables controlled, led governments to pass laws restricting smoking, such as banning it in enclosed public places.
Examiner trap
- Saying correlation proves causation
- Not quoting figures in data questions
- Ignoring other variables
Key terms
- risk factor
- correlation
- causal relationship
- incidence
- statutory restriction
- pollution
- smoking
Lesson notes
A written version of the video, chapter by chapter, for revision. Times link to the video.
0:26Risk factors
First, two key terms. A risk factor is something that increases the chance of getting a disease. Smoking is a risk factor for lung disease. So is air pollution. And incidence is how many new cases there are in a population over a set time. Notice that a risk factor raises the chance. It does not guarantee the disease. Some smokers never get lung cancer, and some non-smokers do.
0:54Reading a scatter plot
Now the data. This scatter graph plots cigarettes smoked each day against incidence of lung disease. The points rise. As one increases, so does the other. That is a positive correlation. A correlation is simply a link between two variables. And in a data question, always quote figures. Say what the incidence was at a low value and at a high value, and compare them. Describing a trend without numbers loses marks.
1:25Correlation or causation?
But does a correlation show that one thing causes the other? No. A causal relationship means a change in one variable directly brings about the change in the other. A correlation alone cannot show that. There might be a third factor driving both. Maybe heavy smokers also live in more polluted cities, or work in dusty jobs, or are older, or have a family history of lung disease.
These are other variables, sometimes called confounding variables, and a good answer says that they could affect the result unless they were controlled.
2:00Finding a mechanism
So how do we move from correlation towards cause? First, a mechanism. If scientists can explain how the factor does the harm, the link becomes far more believable. Tobacco smoke contains tar and other chemicals. Some of them are carcinogens, which cause mutations in the DNA of lung cells and can lead to cancer. Smoke also damages the cilia and irritates the airways, so mucus builds up and infections are more likely.
2:29Comparing several studies
Second, repeated evidence. When many studies in different places, with different people, all find the same link, confidence grows. And when researchers control other variables, for example by comparing groups of the same age, the link still holds. Mechanism, repeated results, and controlled variables together make a strong case that smoking is a cause.
2:55Evidence and statutory restrictions
This evidence is why governments acted. A statutory restriction is a law that limits a source of risk. Governments have banned smoking in enclosed public places and restricted tobacco advertising, because the evidence showed harm to health. The same logic applies to pollution. Evidence that polluted air increases lung disease has led to limits on emissions. The key idea is that the law followed the weight of evidence, not a single study.
4:07Examiner trap and recap
The trap. Saying the data proves that smoking causes the disease. Correlation shows a link. You need other variables controlled, and a mechanism, before you can talk about cause. Recap. Risk factors raise the chance. Correlation is a link, not proof. Repeated, controlled evidence and a mechanism lead to laws. Next time, we move on from the lungs to digestion, and see how amylase breaks down starch.
Watch first
This lesson builds on:
- Lung Disease and Gas Exchange · AQA 7402 · 3.3.2
Keep going
More from 3.3 Exchange with the environment
- Gas Exchange: Single Cells and Insects · AQA 7402 · 3.3.2
- Fish Gills and the Counter-current Principle · AQA 7402 · 3.3.2
- Gas Exchange in Leaves: Mesophyll and Stomata · AQA 7402 · 3.3.2
- Gas Exchange vs Water Loss: Xerophytes · AQA 7402 · 3.3.2
All 10 3.3 Exchange with the environment lessons Whole playlist on YouTube
Independent revision resource, not affiliated with or endorsed by AQA. The exam question is original and was written for this lesson; it is not an AQA question. Spotted an error? Email info@medservatory.com with the lesson code AB-060.