/sci/ - Science & Math

>>15711962
give an example

>>15711978
CO2 in the atmosphere vs global temperature. Correlation but not causation.

>>15711962
its on up there but i think its beat out by "its just about finding a balance" or anything relating to that idea

>>15711962
well i means you have an time delay, but you would agree that you pooping in an public tram is not the reason the old lady next door won thelotery. but in some where all correlations are causations if you search long enought for it.

>>15711997
and you also would agree that me insinerating you cause yesterday a dog bit me, is not the way you want me to go alltought i might find a correlation linking this bite to you.

>>15711962
Correlation is not NECESSARILY, or not ALWAYS, causation. You would be surprised to know just how much of our science is based primarily on the former, and not the latter. We have discovered relatively few causative relationships.

>>15711998
or even worse me insinerating you cause tomorrow a dog which owner you cant even stand will bite me, cause my super ai you can comprehend showed me the correlation linked 1 million years from now.

>>15711988
I second this. "Balance" usually just means maintaining the status quo. Balance is only good for you if you're already winning

>>15712001
or evne worse i drown you in glyphosate cause 600 years from now the mayority of your carbon atomes will be mosquitoes

>>15712011
correlation is not causation referse to the time frame you can act think and or improf your live.

>>15712002
That only makes sense if you define the status quo as balanced.

>>15712017
everythink causes everything if you watch at the world as a vector field but then ofc everything is as well your fail.

>>15712023
but if you burn your hand with boling water, is it realy the fail singing on an other continent?

No, the most midwit phrase ever is
>Most midwit phrase ever?

>>15711962
but no correlation is always no causation

>>15712047
fucking recursion man

>>15711978
>>15712000
I always find this concept is used to attack research going against the debater's priors and a lazy way of dismissing out of hand emerging research directions
>IE: THC is associated with smaller brain sizes? C not C muh heckin reddit industry bias etc etc.
What's funny is the basedence they eat up is usually a bunch of loosely correlated sociology

>>15712071
Dismissing emerging research directions is reasonable. The problem isn't that "correlation!=causation" is called too often, it's that it is called only when the study results are inconvenient to the midwit.
Science should be a search for facts, not a tool for journalists to fill pages and advertisers to claim relevance

>>15711962
It is just synonym for "I disagree"

>>15712081
hear, hear.

>>15711962
It's more like "correlation alone is not sufficient evidence"

>>15712081
Correlational studies are just used by idiots to cherry pick data. Suppose some disingenuous scientists asserts "snakebites are correlated with ice cream sales" after doing a (flawed) study. Conservatards pick this up readily and start posting their basedfaces on image boards and linking to the study whenever somebody asserts that, indeed, ice cream does not cause snake bites. They'll post studies which explore the relationship between summer and snakebites and between ice cream and summer, but to no avail, the boundless stupidity of the retards ensures the ice cream snake bite study gets passed around like a crack pipe for years.

>>15712150
So conservatards now are like a meta peer review that mocks disingenuous scientists for being disingenuous? That's way better than when they mocked french fries to promote war. Don't fight it lol.

>>15711962
It's a completely correct phrase. Correlation is not a causation. But a correlation can point the way to finding potential causations.
>>15712155
Conservatards don't do that. They mock all scientists that go against their worldview, regardless of whether they're genuine or disingenuous.

>>15712187
>mock all scientists that go against their worldview, regardless of whether they're genuine or disingenuous
Libtards do that, too, though. The point is that conservatards used to mock french fries to promote war and now they just do the same thing libtards do.

>>15711962
>>15712131
The original Latin "cum hoc ergo propter hoc" translates literally as "with this, therefore because of this." It is a pretty genius tier thing to say, most of the time.

>>15712217
>cum [in that] ho ergo [she a] proper ho
latin genius confirmed

>>15711962
all statistics are wrong because there's no way to accurately collect data tard

>>15711962
it's true. causality has never been proven.

>>15711962
Just because my brother is an idiot doesn't mean he's a nigger...

The same can be said for this very statement in fact: just because that sentence's meaning strongly correlates to a racist viewpoint (use of the word 'nigger' in a derogatory way), it does not necessarily mean that I actually am racist.

Yes it's a midwit thing to say; it's also true when applied properly and within proper context.

>>15711988
>>15712002
brainrot. life is quite literally all about balance with everything we do.

>>15711962
>>Correlation is not Causation
It's a true statement but the problem is it's misused. Midwits love dropping this one to dismiss ideas they don't like out of hand without judging the merits of the correlational study. The fact is that correlations are useful information and you don't need to strictly prove causation in a philosophical sense if you at least have a model that can be empirically verified. Moreover they tend to just say this without considering that there COULD be valid a causal link, since theoretically correlation is necessary but not sufficient for causation, whereas no correlation implies no causation. My other favorite midwit habit is always namedropping
>Occam's razor
And once again misapplying it. Occam's razor is not a proof, you can't use it to say automatically that your (the midwit's) prior belief is the simplest and therefore the true one. Midwits use this tactic to dismiss evidence they don't like on the grounds that it's "unrelated" or "irrelevant" by their own arbitrary whims. Fact is, if there's a valid argument to show why it's relevant then any model not including it is less correct. It's like ignoring the bolt covers on an airframe... small details that have big consequences. Usually they'll also appeal to Occam's razor while alleging you're a "conspiracy theorist" too.

>>15711962
>Actually, not *ALL* ..., and so that is a generalization [and therefore useless evil wrongthink that should be completely ignored].

They miss the point, derail the conversation, and stop listening,
when in fact the rareness of the counterexample/exception raised actually reinforces the strength of the rule of thumb.

Also:
>That's a conspiracy theory!

>>15711962
If I needed to find causation, correlation is where I'd start looking.

>>15712187
Libtards literally try to change scientific definitions.

>>15712516
Only the ones they don't like.

>>15711962
It is a phrase used by truth-deniers. If there is a way to deduce a causal relationship from some basic axioms then the correlation would verify the model you built, that is literally the basis of econometrics.

Oh look there is a really strong relationship between cokes drank and how fat people are - NOOOO YOU CAN'T CONCLUDE ANYTHING

>>15711962
Jesus was blond and had blue/grey eyes, there are copious descriptions of him by contemporaries and they all describe him the same way, fair haired and with pale eyes

>>15712394
Based. You only see absurd counterexamples to correlation and causation.

>>15712223
Post hoc ergo propter hoc is another good one. It means, "After this, therefore because of this."

https://www.tylervigen.com/spurious-correlations

>>15712541
>"After this, therefore because of this."

>>15712543
>Look mom I posted an absurd combination of 2 correlated variables nobody would ever try to form a causal explanation of

Ummmm excuse me correlation is not causation

>>15712551
what do you mean? the data speaks for itself

Correlation is not causation move along nothing to be inferred here

>>15712353
the other big one is Dunning-Kruger. Trust the authorities but never do your own research! 'tis the reddit way

>>15711985
The thing is that no one (smart) believes that CO2 causes global warning because it is correlated with heat temperature. Its because we know that CO2 is a greenhouse gas and will absorb, then release, heat into the atmosphere. Then we can use a correlation as evidence CO2 is causing warming. But a correlation on its own doesn't give you enough evidence. For a strong argument you need to come up with a mechanism or something that comes from physical laws.

>>15712536
Jesus actually had red dreadlocks (black dyed red), was a trans FtM, had black skin and dark hazel eyes.
Think I'm wrong? Try posting passage numbers and the Bible edition next time.

>>15711962
no causation with msnipulation, so they say

>>15712564
>>15712556

kek

>>15712536
>there are copious descriptions of him by contemporaries
There are zero (0) contemporary references to Jesus, period.

>>15712586
None were peer reviewed, that's for sure.

>>15712000
/thread

>>15712682
What's a caustive relationship you can prove

>>15711962
Almost
>Diversity is our strength
>Experts say

>>15712693
Oh, and any who unironically complains about the slippery slope fallacy or whataboutism in a political or economic context

>>15712695
>political or economic context
You can complain about anything you want in either of those contexts. Both are pure garbage.

>>15711985
But it's true. There's no correlation because CO2 is just one possible contributor towards climate increases out of hundreds. 100 years of data don't invalidate billions of years of data.

>>15712710
You're forgetting that the peacock blue line went up so quickly at the end of the graph that you can't see it; therefore it's never happened so quickly on the graph before because otherwise you'd see it.

>>15711962
Yes. The correct statement should be that correlation does not imply causation, but to state correlation is not causation is simply wrong. Sometimes the correlation does have a causal interpretation, such as is the case for Randomized Controlled Trials.

>>15712749
It's still not correct to use the phrase randomly. I think OP is complaining that it's used as a non sequitur too much. (He has too much sand in his pussy, obviously.)

>>15712755
I agree that people will misuse correlation, but I can see where OP is coming from. Some people can't see a correlation and immediately say the phrase, without any care about the context. But you could argue this is still better than blindingly believing in any fancy graph they throw at you, though.

>>15712689
photosynthesis, heating water, disease transmission, chemical reactions, etc. Comparatively very simple things. As things become more complex, the likelihood of proving causation drops drastically. I would shudder to know of the entirety of human knowledge that has causative proof expressed as a percentage. I believe it would be very low.

>>15711962
>Correlation is not Causation
Yes
Retards often think otherwise. See: many scientific papers and /pol/tard opnions

>>15711962
along with dunning kruger and cherry picking yes

>>15711962
Yeah people explaining basic concepts to retards like you is the most midwit thing ever. Kys dumb nigger.

>>15712556
>>15712564
Lmao correlation-deniers destroyed

i've found that the people who post this image as some kind of dunk or own typically strike me as npc midwit types who are very proud of themselves for having just read a malcolm gladwell book

Correlation is not causation