epicurusaquinas
1449
42
3
Recently (2d ago) I posted an article on plastics being "potentially" linked to a medical condition. However, even though we see and suspect plastics cause the condition .. it's not a scientific medical fact .. yet.
This post is a little discussion about how correlation of data is NOT the same as causation.
In science, causation is not proven because evidence and data show correlation. Testing and experimentation plus statistical reduction of error establish causation.
In higher temperatures, ice cream sales go up.
In higher temperatures, crime rates go up.
There IS a correlation between ice cream sales and crime rates.
However, just because there IS correlation between ice cream and crime DOES NOT mean ice cream causes crime. See how this works?
Evidence and facts alone MIGHT show correlation.
But science (understanding causation) does not work by getting more and more evidence (facts) alone.
Often people think "Oh look. We have tons and tons of evidence and facts. Obviously this causes that."
No.
Science proves causation by experiments and testing leading to evidence and facts, but it's the statistical alignment WITH low error which establishes causation.
Don't get me wrong. It's ok to notice correlation and say "Well THAT'S suspicious." Well, ofc things get suspicious. That's ok to say. You're not an idiot to see correlation. This is how theories come around.
I just wanted to let people know suspicion is fine, but knowing something is physically caused (law) doesn't depend on correlation alone. To establish causation, at least in science, you need statistically "valid" experiments.
Cheers!!
CuntCrotchula
https://www.tylervigen.com/spurious-correlations
nosimpleway
I read this post. I understand correlation and causation. So the post helped, maybe. Not sure.
itsameeeee
The world is obsessed with how things look
epicurusaquinas
Uhh . I think the world is obsessed with validation of their assumptions. How things work (actually work) is maybe secondary. Maybe tertiary after sex. We need a scientific study to figure that out!!
Br0doSwaggins
Considering how many articles I've read that said microplastics are everywhere in the food and water supply, I've learned to stop worrying and love that I'll either become part plastic or ill die of plastic-related cancer decades from now
epicurusaquinas
Well .. science-backed medicine or not .. it don't look good.
kriswinters
#3 If I saw that duck break a wrought iron fence, I'd be careful not to accuse it of anything either.
epicurusaquinas
Amen
abraxasfury
Did you know that everyone who believes that correlation = causation will eventually die ?
quade
So does everyone who drinks hydrogen dioxide
epicurusaquinas
skylardarkfox
I get the joke, but:
Dihydrogen monoxide (H2O)
Hydrogen dioxide (HO2)
Twrecks123
And crime/heat, ice cream sales/heat have at least something in common (heat). But something like cancer may not. For example, if the cancer rates are going up because people are living longer and surviving other diseases that previously would have killed people long before they got cancer-that’s not in any way connected to cell phones. Even loosely or vaguely.
epicurusaquinas
Are you saying "closely related" things have a better chance of causation?
If you are, I'll say I've been through this countless times. I will say "depends on who thinks they are closely related and why." Honestly it's so complex. I've been surprised by my own suspicions being wrong many times in engineering. From a scientific perspective, proving causality is just not very clear when it will or won't be proven.
Twrecks123
No, I’m saying there’s even less of any sort of vague hint of a connection with the cancer/cell phones than with the heat/ice cream/crime.
epicurusaquinas
That's the same thing I said in different words