Correlation Does Not Equal Causation

Correlation does not equal causation. What does that mean? Well, correlation is a statistical relationship between two things, but the key part of the definition would be, causal or not. So, what is causation? Causation is the relationship between cause and effect. Meaning one thing has an effect on the other. So, if reports of cancer and milk consumption have both increased at a similar steady rate over the past 10 years, that’s correlation, but not necessarily causation. They look like statistical brothers, but they’re probably not related.


Let’s say your friend tells you something that seems factually correct; they started sticking blueberries under their armpits, and now they lift more than ever. First things first, the two are not definitely linked. There could be any number of reasons why they’re now lifting more, and it doesn’t mean that the blueberries under the armpits is THE reason, it’s likely that’s just a coincidence. And until proper studies were conducted, you can’t say for sure that it’s the blueberries that did it.


Secondly, just because it happened to them, it doesn’t make it scientific fact. Anecdotal evidence is nothing more than that; anecdotal. It is specific to an individual, and their individual set of circumstances, it is unproven and untested, and it’s very likely that it wasn’t conducted under a controlled scenario. So, if someone else were to do it, or even this person again, they may very well get a wildly different result. And just because it happened to them, doesn’t mean it will happen to you. You’re two different people, probably with totally different genetics. Things will affect you differently, in differing ways.


Also, it being anecdotal and individual means that you are having to trust your friend, and trust that they’re giving you all the pertinent information. Maybe what they forgot to tell you, was that they’ve been on a periodisation programme with their workouts, or they’ve improved their form, or they’ve started eating sufficient protein. They might not even know the real cause themselves, so they’ve made an assumption based on what they think is new in their life. That has to be it, right, the blueberry thing? It’s the only logical conclusion.


Obviously not. Just because two things happen in tandem, it does not necessarily mean that they are linked, that one has caused the other. You can’t jump to those conclusions without research, and you need to be wary of these news stories, where they claim X is the reason for Y. They’re usually using correlation, and not causation. And it’s a huge difference. Correlation does not automatically equal causation. I’ll be back again next week.


Dan Miller

Body Fuel Personal Fitness Trainer