I’ve seen this referenced a lot, especially with psychology, but can someone explain what exactly it means? How does correlation not imply causation? Sometimes, does correlation ever imply causation?
In: 160
A common example used is ice cream sales go up. So does crime at the same time. Ice cream sales drop, so does crime at the same time. Does that mean that ice cream causes crime or crime makes people buy ice cream? Of course not. That’d be a pretty silly thing. Instead there’s another variable going on – the weather. Hot weather increases the desire for ice cream, while it also means more people are out and about, interacting and on shorter fuses.
The other possibility is that they just happen to correlate by pure happenstance (see the previously mentioned Pirates-Global Warming example) and there’s nothing at all joining the two.
Now, if a there is a causation you there WOULD be a correlation; so seeing a correlation may be a signal to a causation, and worth looking into. But showing a causation takes more than just showing correlation.
You could say “Spain implemented this policy to address crime, and crime dropped shortly after”. OK. That’s a potential causation there. But you have to look at more than that. What happened in other countries in that time? Did crime drop in other countries that did not implement the policy? If so, maybe there’s something else going on, like jump in economic growth around the world. It didn’t? Ok. Have other areas implemented similar policies? Did they see similar results?
Latest Answers