In one line: A correlation is a relationship between two measured variables. It describes; it does not explain.
In 2012 a doctor compared countries and found a striking pattern: the more chocolate a country eats per person, the more Nobel prizes it has won.
The two variables rise together across countries. That is a correlation, and a strong one.
Positive correlation
- Both variables rise together
- More chocolate, more prizes
- More revision, higher marks
Negative correlation
- One rises as the other falls
- More absences, lower marks
- More sleep, fewer errors
A correlation has a direction, positive or negative, and a strength: how closely the two variables keep in step.
Neither tells you why they are related. Nobody thinks that a bar of chocolate wins prizes.
Remember: Even a strong correlation is only a description.
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
In one line: A causal claim says that changing one variable produces a change in another.
Compare two statements about sleep.
"Students who sleep less make more errors" is a correlation. "Keeping students awake makes them commit more errors" is causation, and it is a much stronger claim.
What a causal claim has to show
They vary together
When the cause changes, the effect changes. A correlation shows this much.
The cause comes first
The loss of sleep has to come before the errors, not after them.
Nothing else explains it
Other explanations, such as illness or stress, have been ruled out.
Together · In order · Nothing else
A correlation meets the first condition only. That is why it is where a causal question starts and not where it ends.
Exam tip: Watch the verbs in a study's conclusion.
"Improves", "leads to" and "reduces" are causal claims. Check whether the design earned them.
Know your predicted grade
Take timed mock exams and get detailed feedback on every answer. See exactly where you're losing marks.
Key idea: A correlation is what causation looks like in the data. So is a great deal else.
Three other explanations for a correlation
A third variable
Something unmeasured produces both.
Wealthy countries can afford chocolate and universities.
Reverse causation
The supposed effect is really the cause.
"Teenagers who sleep badly use their phones more at night." Does the phone spoil the sleep, or do the sleepless reach for the phone?
Chance
With enough variables, some will line up by accident.
A researcher who compares fifty things will find a few strong matches that mean nothing.
Third variable · Reverse · Chance
A real study: the marshmallow test: Mischel offered four-year-olds one marshmallow now, or two if they waited. A follow-up found that those who had waited longer had higher test scores as teenagers: a correlation.
The repeat study
Watts and colleagues (2018) repeated the study with about 900 children from more varied homes. The link was about half the size, and most of what was left disappeared when family background and early ability were taken into account.
What looked like an effect of self-control was largely a third variable.
None of this makes correlations worthless. The link between smoking and lung cancer was first seen as a correlation. It showed where to look, and later research established the cause.
What can show a cause: An experiment sets one variable and controls the others, so the three conditions can be met.
Where an experiment is impossible, following people over time at least shows which change came first.
How this is asked: Causality is one of the four concepts offered in the 15-mark question on an unseen study.
A study that draws a causal conclusion from a correlation is one of the commonest things to be given.
The pattern
- Identify the design: variables measured, none set by the researcher.
- Quote the causal verb in the conclusion.
- Offer a third variable and a reversed direction that fit this study.
- Apply a second concept and show how it bears on the first.
- Conclude with what the study can claim and what would test the cause.
The trap: dismissing the study: "It is only a correlation" is half a point.
Say which other explanations are open, and what the correlation is still good for.
Discuss the following study with reference to two or more of the following concepts: bias, causality, measurement and/or responsibility.
Researchers surveyed 1,200 teenagers in one city. Each teenager estimated how many hours a week they spend outdoors and completed a ten-item well-being questionnaire. Teenagers who reported more time outdoors had higher well-being scores. The researchers concluded that spending time outdoors improves well-being and recommended that schools teach one lesson a day outside.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.