3.4 Correlation and causation
Create image of newspapers with following headlines:“Watching funny cat memes sparks increase in biomass power generated in Latvia!”“Increase in American cheese consumption could help Peru generate more solar power!”“As more UFOs are seen in Nevada, movie makers receive record budgets to make their new movies!” Source: Tyler Vigen (2024)It is very easy to jump from two variables being correlated to stating that a change of one variable causes a change in the other variable. However, just because two variables are related does not mean there is a causal link. Some correlation may be entirely coincidental, some may have other underlying explanatory variables, and in some cases, there may be a direct link.
The headlines above have all been created from datasets which show a relationship between the two variables concerned however, hopefully, the silliness of these claims is apparent. That said, the final headline about UFOs could be made to seem believable if the new movies were all about aliens and outer space. See the further reading section for more on this.
Correlation is not causation.
Figure 21 Correlation/causation cartoonCredit: The Open University, M140 Introducing statistics