/sci/ - Science & Math

It is getting embarrassing guys. 40 years and not one novel testable prediction.

What this should say is, correlation between A and B does not mean A causes B.

However when there is correlation there is a causal relationship in there somewhere. What is not shown by a correlation is the nature of the causal relationship.

Eg everyone who drinks water dies eventually. Therefore drinking water kills you. The causal relationship here is that being a mortal human being causes you to eventually die, and it also causes you to drink water.

In this case

A is correlated to B but the relationship is one of a common cause. Common Cause is a common issue - it is also called Confounding Factors.

X->A
X->B

Being in prison is correlated with having committed a crime. However being in prison does not cause you to have committed a crime in the past. The causation operates in the opposite direction. Committing a crime may cause you to be put in prison.

So: Correlation does imply causation (or a logical relationship). But it does not specify the type of causative relationship.

1. You will get cold.
2. You will miss out on sleep.
3. Mosquitos will bite you.
4. Clouds will go in front of whatever you are trying to see at precisely the wrong moment.

If you look at the study the coffee drinkers actually died younger. However a few nifty statistical tricks later, bingo! after correcting for other factors coffee drinkers theoretically live longer.

Coming soon: a lucrative "consulting assignment" to do some research for a coffee company. Thus science is corrupted.