Correlation vs Causation
Recognise a pattern linking two things across observations. Understand that the pattern alone does not prove one caused the other.
What a learner can do afterwards
- Distinguish a correlation across observations from evidence establishing a cause.
- Consider another factor, the reverse causal direction or chance before accepting a causal claim.
1 · Read
A correlation is a pattern linking two things across several observations. They may tend to rise together, or one may rise as the other falls. Two events happening together once do not establish a pattern. A correlation alone does not tell you what caused it.
Causation means one thing makes a difference to what happens to another. Correlation and causation can occur together. But a correlation does not prove a cause. Another factor could affect both things, or the pattern could happen by chance.
Sunny weather could help explain a pattern between ice cream sales and sunburn. Warm weather may encourage ice cream buying. Time outside can expose skin to the sun's UV light, which can cause sunburn. Heat itself does not cause sunburn. The sales pattern alone does not establish this explanation.
Before naming a cause, ask what else could explain the pattern. Another factor might affect both things. To reverse a causal claim, swap what causes what: 'More cups bring more visitors' becomes 'More visitors lead to more cups being used.' Chance could also explain a pattern. Look at more evidence. A possible explanation is different from a proved cause.
A correlation describes a pattern. More evidence is needed to establish what causes it.
This lesson, its questions and flashcards were written for this dataset. The chart counts and surveys are fictional teaching examples.
This reading was written for this dataset. Its school record is a fictional teaching example.
More to read
Which way might the cause run?
In a made-up school record, more lunch trays are used on days when more children attend. The counts tend to rise together. That is a correlation.
Do extra trays make children attend school? The counts do not show that. More children attending could lead to more trays being used. This possible explanation runs in the other direction.
Causation means one thing makes a difference to another. A pattern alone does not show which way a cause runs or whether another factor explains both counts.
Before accepting a claim about a cause, ask what else could explain the observations and what further evidence would help.
2 · Watch
En anglaisTake it off screen
Where it sits
Learn first
Optional connections
Where this leads
8 questions wait behind this lesson, each with its answer explained. Every answer feeds the sky: stars light as they are learned, and dim when it is time to come back.