Causation vs correlation
When a correlation is 100% strong, there must be a causation, right? But how do I find out the direction of the causation?
Let's take for example the observation that 100% of people who wear leather jackets are douchebags. Cam we conclude that leather jackets turn people into douchebags or that douchebags prefer to wear leather jackets? Which one is the right explanation?
>>8203297
>douchebags
Define 'douchebag'.
>douchebags prefer to wear leather jackets
This one, as douchebags exist without leather jackets, however this scenario leather jackets don't exist/aren't worn by anyone who isn't a douchebag.
As in, the wouldn't exist if nobody was ever around to buy them/wear them.
No.
>>8203297
The answer is both. There are myriad studies about how clothing, and specifically uniforms, affect human interactions and behavior within a given culture.
>>8203297
Causation requires sequentiality of events. The cause must precede its effect
>>8203297
>100% correlation
What I'd be looking for is faked statistics frankly.
Also no. Things can be perfectly correlated and not have a causal connection.
100% of cars I have seen have been in Poland
Therefore either Poland causes cars (lol we don't) or cars cause Poland.
>>8203324
>This one, as douchebags exist without leather jackets,
leather jackets could cause people to be douchebags along with plenty of other things (for example, if wearing manbuns also causes people to be douchebags someone could be a douchebag without necessarily wearing a leather jacket, even though leather jackets could cause people to be douchebags too).
what the fuck is wrong with leathert jackets
it's cold what the fuck else am I supposed to wear?fuck off
>>8203388
Don't be such a douchebag about it.
>>8203388
Wow, what a douche.
>>8203388
douchebag detected
>>8203297
>When a correlation is 100% strong, there must be a causation, right?
No. Correlation will always be present if there is causation, but may or may not be present if there is not causation.
Every day I have been alive the sun has risen. This is some evidence that I cause the sun to rise (or vice versa), but it's not much evidence compared to the entire body of knowledge we have about how humans and sunrises work.
>But how do I find out the direction of the causation?
Experiment and investigation. If you are sure that there is a causal relationship between me and the sunrise, you could kill me and see if the sun rises the next day. Or you could investigate and try to figure out the mechanism by which one causes the other.
