DS StatisticsLesson 18Free preview
Stat Correlation vs Causality
This lesson is part of Data Science · All levels
Summary
The principle of 'correlation does not imply causation' reminds us that statistical relationships between two variables do not automatically mean one causes the other. Data mismatch or the presence of a confusing variable can result in false correlations, such as the correlation between ice cream sales and the number of swimming accidents in the summer. Causal claims need to be tested with controlled experimental design.
Key points
- Correlation measures association, not evidence of causation.
- Confounding variables can create false correlations.
- A two-way relationship can be misinterpreted as a causal relationship.
- The evidence of causation is reinforced by randomized controlled experiments.
- A safe conclusion: there is a strong linear relationship, the cause direction needs to be tested.
Full lesson
Code examples and quizzes
Full lessons, code examples, and quizzes are available to subscribers. 5 questions.