Key Idea: *A correlation shows two things move together; only a controlled experiment can show one causes the other.*
Topic 1.2 at a glance
- Correlation vs causation — A link isn't a cause — a third variable or reverse direction may explain it.
- Establishing cause — Manipulate an IV, control other variables, use random allocation.
- Bidirectional links — Each variable can cause the other, forming a feedback loop.
Correlation≠cause · Establish cause · Bidirectional
IB-style questionExplain[6 marks]
A report says 'students who sleep more get better grades', implying sleep causes grades. Explain why this may not be causal, and how you could test the cause.
🔒 Model answer plan
See the mark-by-mark plan — for / against / judgement, with marking guidance — in study mode.
Why isn't correlation causation? A third variable could drive both, or the arrow could run the other way.
How do you show cause? A true experiment: manipulate the IV, control the rest, allocate randomly.
What is a bidirectional relationship? Two things each cause the other (e.g. stress and poor sleep).
Faced with a 'X is linked to Y' claim, always ask: third variable? reverse causation? Then say what study would test the cause.