If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Slate)   Next time some pseudo-intellectual internet blowhard tries to take away your carefully thought-out arguments with that "correlation does not imply causation" yarn, just send them here because YOU WIN   (slate.com) divider line 226
    More: Cool  
•       •       •

27662 clicks; posted to Main » on 02 Oct 2012 at 5:31 PM (2 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



226 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread

First | « | 1 | 2 | 3 | 4 | 5 | » | Last | Show all
 
2012-10-02 06:52:42 PM  

RexTalionis: I prefer to say "post hoc ergo propter hoc".


I said that on a sports board once to show my skepticism of some stats, and the author said "yes, but it could!"

Sigh.
 
2012-10-02 06:53:21 PM  

Ambivalence: Yah but correlation does NOT equal causation. Just like Opinion does not equal fact. It's a legitimate and commonly used logical fallacy.

And yes it's annoying when someone points it is, but that doesn't make it less true (assuming it is even true. Sometimes you are able to establish causation).


Here are about 100 commonly used logical fallacies.

http://en.wikipedia.org/wiki/List_of_fallacies

I find a high correlation between people repeating your phrase and being complete idiots who have absolutely no understanding of logic or philosophy.

Here's a novel link to improve your life. http://www.theskepticsguide.org/resources/logicalfallacies.aspx However, I would bet you aren't going to read it, because idiots like you don't like to read any actual philosophical or logical literature.


///The true question is, what are you willing to bet on.
 
2012-10-02 06:56:50 PM  

Firethorn: kxs401: I guess I'm just ignorant, because I certainly realize that correlation doesn't PROVE causation, but why doesn't it imply it? If A and B are correlated, possible explanations are that A causes B or B causes A. When we notice that smoking is correlated with lung cancer, why wouldn't we go looking to find causality?

It's logic along the lines of how in science today you get 'theories' instead of 'laws'. Einstein's theory of relativity is more advanced than the 'law' of gravity, but it's still 'only' a theory. When they find something that's 'more true', it'll be replaced/modified.

Anyways, a Correlation doesn't imply causation(imply being a higher standard of proof), since the question becomes what causes what. Let's take being fat and having heart attacks.
'Some' Possibilities:
1. Being fat causes heart Attacks (A causes B)
2. Heart Attacks cause being fat (B causes A)(one could have a case where most heart attacks aren't noticed)
3. Something else casues both(C causes A AND B, but B takes longer)
4. The study was flawed; not a good random sample
5. It's truly a coincidence.

Basically, you're not allowed to assume that A causes B until you have a theory about HOW A causes B and you've tested it.


I think your last sentence is a bit too strong. The theory that eating cyanide causes death was well-established by experiment (or could have been, anyway) long before it was known how eating cyanide causes death.
 
2012-10-02 07:01:20 PM  

DaCaptain19: kxs401: I guess I'm just ignorant, because I certainly realize that correlation doesn't PROVE causation, but why doesn't it imply it? If A and B are correlated, possible explanations are that A causes B or B causes A. When we notice that smoking is correlated with lung cancer, why wouldn't we go looking to find causality?

No, you're not, unless my PhD and 18 years as a professional statistical analyst mean nothing. Correlation absolutely implies causation. If there's no correlation between X and Y it says X has no way of predicting Y thus could NEVER cause Y. Any correlation (unless it's obviously spurious) between X and Y could mean changes in X DO cause changes in Y. It's just not certain based on a correlation, alone.


Do statisticians really use the word "imply" to mean "suggest"? It seems like that would make it difficult for mathematicians to read the statistical literature and vice versa.
 
2012-10-02 07:01:26 PM  

moran: CreamFilling: uber humper: Correlation doesn't equal causation.

Causation does equal causation. Don't forget that.

Everything has a cause. Nothing happens just 'cuz

Maybe, but causation doesn't imply causation, by the same definition of "imply" used in the original. Maybe in mathematics, not in any real life scenario. Which is exactly why that definition of "imply" turns the whole argument worthless.

I know "imply" means something different outside of mathematics, but could you explain why "A causes B" doesn't imply "A causes B" under your definition of "imply"?


Pick any example you want of A causing B, and you'll find it doesn't always hold true. It's mostly an issue of mixing strict logic terminology with traditional common usages. Smoking causes lung cancer? Drunk driving causes accidents? Of course, unless you apply formal rules of logic.
 
2012-10-02 07:01:34 PM  
Nit picking a nit pick, you can all burn.
 
2012-10-02 07:02:02 PM  
I love threads where ignorant people try to sound like experts.

I'd say "don't talk about stuff you're pretty brain can't comprehend" but...hey, it's FARK.

I'll respect any opinion of someone who has my creds or better.

The fact is, ALL causal models are based on covariance/correlation. If two things are uncorrelated we KNOW one doesn't cause the other. If two things ARE correlated (and pass the smell test, i.e. spurious correlation) there may be a causal link one direction or both.

There are other limitations on causality which have nothing to do with statistics. For instance, a "cause" has to precede the "effect". This is a "duh" but there are others.
 
2012-10-02 07:03:51 PM  

uber humper: Correlation doesn't equal causation.

Causation does equal causation. Don't forget that.

Everything has a cause. Nothing happens just 'cuz


1) Combinations of things frequently happen "just 'cuz"
2) There is no cosmic plan; not everything happens for a reason.
 
2012-10-02 07:07:08 PM  

moran: I didn't read TFA, but being a logician I thought it might help to mention what "implies" means in (classical mathematical) logic. "A implies B" does not mean merely that the truth of A suggests that B might also be true. It means that it is impossible for A to be true without B also being true. A single example of A and B where A is true but B is false refutes the implication (and of course it is easy to come up with an example where A and B are correlated but A does not cause B.)


Yes, you are absolutely correct, but when someone says "correlation implies causation", they are NOT using it in a logical sense, as in a connective in propositional calculus. "Correlation" and "causation" are not strict logical prepositions. So the word "implies" is used here in a lay sense, exactly as if they had said "George's silence implies he has something to hide".
 
jvl
2012-10-02 07:07:18 PM  
FTFA: No, correlation does not imply causation, but it sure as hell provides a hint.

Would this be a bad time to point out that hint and imply are synonyms?
 
2012-10-02 07:07:59 PM  
As I tell my students, correlation may not show causation, but it does show association, which can be equally important. And besides, use your common sense. Studying is highly correlated with getting good grades, and I'm pretty sure that getting good grades doesn't cause one to go out and study more.
 
2012-10-02 07:09:50 PM  

HotIgneous Intruder: If the Slate writer sinks, (s)he is not a witch.


What if she/he weighs the same as a duck?
 
2012-10-02 07:10:53 PM  

ThrobblefootSpectre: I've noticed the vast majority of people who use the "correlation does not imply causation" soundbite use it in completely inappropriate context. They use it as a complete non-sequiter in a context where no causation was claimed, apparently in the hope someone thinks it means something.


Not really true. Most people have come to use it because of the prevalence of jumping to conclusions - especially in the media. "Scientistists find that some people with X also have Y" becomes "X causes Y!". The whole correlation != causation came as a push-back to that. Or so it seems to me.
 
2012-10-02 07:11:34 PM  

kxs401: I guess I'm just ignorant, because I certainly realize that correlation doesn't PROVE causation, but why doesn't it imply it? If A and B are correlated, possible explanations are that A causes B or B causes A. When we notice that smoking is correlated with lung cancer, why wouldn't we go looking to find causality?


There's a correlation between correlation and causation, but doesn't imply that correlation implies causation.
 
2012-10-02 07:12:21 PM  

CreamFilling: Smoking causes lung cancer? Drunk driving causes accidents? Of course, unless you apply formal rules of logic.


I would never say any of those things. Rather smoking can "increase the risk of lung cancer." Lung cancer can be caused by a multitude of things or a number of things in conjunction. But a specific cancerous cell has a cause. It doesn't just become cancerous without reason.
 
2012-10-02 07:13:35 PM  

Coolfusis: Oh look, it's another one of the "that thing doesn't exactly mean that!" articles. I'll step aside and let the pedants masturbate furiously over this one.

 
2012-10-02 07:13:50 PM  
Every single logical fallacy only exists if there is a premise that fails to support a conclusion.

Here's a tip, end of argument with "this indicates further attention and observation is needed in xxx", then that becomes your conclusion, and the "XXX" that you're really arguing can still be mentioned and no logical fallacy exist.
 
2012-10-02 07:15:20 PM  

meanmutton: not everything happens for a reason.


It happens because of magic? What's the opposite of reason?
 
2012-10-02 07:16:02 PM  
Correlation actually can imply causation...it's up to the person's perspective. It does not EQUAL causation.
 
ecl
2012-10-02 07:17:10 PM  

rcf1105: As I tell my students, correlation may not show causation, but it does show association, which can be equally important. And besides, use your common sense. Studying is highly correlated with getting good grades, and I'm pretty sure that getting good grades doesn't cause one to go out and study more.


I hope you teach Kindergarten.
 
2012-10-02 07:19:12 PM  
I blame the moronic asswipe who writes xkcd for many popular shortcuts to thinking and unfunny jokes.
 
2012-10-02 07:19:31 PM  
TFA fails to mention self agrandisement, self implication and projection.
 
2012-10-02 07:19:36 PM  
I got your Granger test right here, biotch.
 
2012-10-02 07:20:46 PM  
correlation implies relation, but certainly not causation.

If a coke fiend once tried marijuana in his youth, that doesn't mean that marijuana was the 'gateway' drug... It might simply mean that someone who likes hardcore drugs has no aversion to soft core drugs, is all.
 
2012-10-02 07:20:52 PM  

Donnchadha: kxs401: I guess I'm just ignorant, because I certainly realize that correlation doesn't PROVE causation, but why doesn't it imply it? If A and B are correlated, possible explanations are that A causes B or B causes A. When we notice that smoking is correlated with lung cancer, why wouldn't we go looking to find causality?

You can probably find a correlation between two seemingly random things, however it does not imply that there necessarily has to be a causal link between them. I could probably correlate levels of beer consumptions with hours of football watched, for example. However, just because people who watch more football might also drink more beer means that watching football causes you to drink beer or that drinking beer causes to you watch football. The societal trend of drinking beer while watching football comes from an external social stimulus, not as an inherent property of beer or football.


Dude, have you seen all the beer ads during football games? There's your link. A ton of money sunk into consumer psychology to get you to associate "game time" with "drinking time." Or most any sport for that matter.

There IS an underlying link, and watching lots of football will induce more people to drink during football season. It may not happen for YOU or for any one person in particular because you can't predict specific results for a given person. But you CAN say that out of 1 million viewers, x% will drink more beer during the football season because of the prompts in the advertising. It's not the game of football that does it, it's the social environment that Madison Avenue has spent a lot of money cultivating around the game.

Correlations are very important for tracking back root causes. It's whether modifying one correlate will impact the other as desired that is in question.
 
2012-10-02 07:21:12 PM  
Donnchadha

You don't know that I don't know that 

How do you know that I don't know that you know I know that you don't know that you know that, huh?
 
2012-10-02 07:21:45 PM  
You know who else won?
 
2012-10-02 07:22:14 PM  

meanmutton: This About That: Correlation does imply causation. Correlation does not confirm causation.

Increases in texting in the United States has occurred at the same time that traffic fatalities - both in the absolute sense and per mile driven - has decreased. That doesn't mean that there's any sort of causation.


I think it's more fallicous to automatically assume that "more texting = more accidents". I can think of a few reasons right off the top of my head that more texting can correlate with less accidents (less people driving for one). If I saw a study or something that correlated texting with decreased accidents I would not dismiss it out of hand saying "correlation does not equal causation"
 
2012-10-02 07:23:06 PM  

ecl: rcf1105: As I tell my students, correlation may not show causation, but it does show association, which can be equally important. And besides, use your common sense. Studying is highly correlated with getting good grades, and I'm pretty sure that getting good grades doesn't cause one to go out and study more.

I hope you teach Kindergarten.


I share your hope that we have reasonable people teaching our young children.
 
2012-10-02 07:25:43 PM  

Bhruic: ThrobblefootSpectre: I've noticed the vast majority of people who use the "correlation does not imply causation" soundbite use it in completely inappropriate context. They use it as a complete non-sequiter in a context where no causation was claimed, apparently in the hope someone thinks it means something.

Not really true. Most people have come to use it because of the prevalence of jumping to conclusions - especially in the media. "Scientistists find that some people with X also have Y" becomes "X causes Y!". The whole correlation != causation came as a push-back to that. Or so it seems to me.



To me it seems often times it is. If there's an article about how x was found to be correlated with y, then someone, many people, invariably use "da phrase". But.....no said anything, anything at all, about causation. The correlation alone is an interesting discovery, and worth knowing about and reporting by itself.

They may as well interject "But rutabagas aren't limes!!" Okay, yes, that's pedantically true, but the problems being a) everyone already knows that and b) you are objecting to something that no one said.
 
2012-10-02 07:26:40 PM  

ChuDogg: meanmutton: This About That: Correlation does imply causation. Correlation does not confirm causation.

Increases in texting in the United States has occurred at the same time that traffic fatalities - both in the absolute sense and per mile driven - has decreased. That doesn't mean that there's any sort of causation.

I think it's more fallicous to automatically assume that "more texting = more accidents". I can think of a few reasons right off the top of my head that more texting can correlate with less accidents (less people driving for one). If I saw a study or something that correlated texting with decreased accidents I would not dismiss it out of hand saying "correlation does not equal causation"


fark you!

Put the phone down while driving!
 
2012-10-02 07:27:02 PM  

falcon176: pseudo intellectuals="I will repeat verbatim what someone far smarter than me said and I have nothing else to back it up if you try to question it so I'd better write QED on the end which means I win no matter what in latin or some shiat I don't know and if you reply to me I'll just reply with some general insults that I run through a thesaurus 5 times"


If I change things up by busting out "Pythagorus said so", does that make me the ultimate practitioner of insufferability? BTW I'm sending this from iPhone (further douching it up?) so apologies if I misspelled everyone's favorite dead Greek mathematician's name.
/nearly a Classics minor and definitely an Apple nerd
 
2012-10-02 07:27:02 PM  
So in closing, I like the isle of Manhattan....
smoke on your pipe and PUT THAT IN!
 
2012-10-02 07:28:55 PM  
#slatepitches strikes again...
 
2012-10-02 07:29:52 PM  

DaCaptain19: I love threads where ignorant people try to sound like experts.

I'd say "don't talk about stuff you're pretty brain can't comprehend" but...hey, it's FARK.

I'll respect any opinion of someone who has my creds or better.
....


I love threads where experts try to sound like pompous jackasses. (At least, I assume that's what you were going for.)
 
2012-10-02 07:30:12 PM  

brap: I like the isle of Manhattan....


There truly is no reason in that.

/and here I was arguing that everything has a reason.
 
2012-10-02 07:30:27 PM  
Correlation does not in itself imply causation. It is what people that managed not to black out under the stress of grade-school logic instruction might call "a necessary but not sufficient condition".

The reason the quote/reference is used instead of presenting a full epistemological argument for why a given correlation is insufficient to indicate a causative relationship is twofold:

(1) "Correlation is a necessary, but not sufficient condition to imply a causative relationship in either direction, other necessary factors being control of other variables, lack of selection bias, and basic statistical analysis regarding the meta-statistics of false correlations in the specific case. And obviously only then if one of your base assumptions is the assumption of Empiricism, which underlies most scientific thought," is a speech everyone even remotely educated has heard before and just needs to have referenced to be reminded of. Plus that's a bit of a wall of text.

Basically the same reason people will generally just quote a line of a play or movie instead of retyping the whole script when they're referencing it.

(2) The kind of person that actually falls into the post hoc ergo propter hoc fallacy is waaaaaaay too stupid to even be worth the effort of typing up a full explanation. Literal five-year-olds that I've instructed are already familiar with the concept that you can't just assume shiat based off a hint. No one wants to waste their life trying to shove knowledge into the heads of adults that are litterally dumber than five-year-olds.

kxs401: I guess I'm just ignorant, because I certainly realize that correlation doesn't PROVE causation, but why doesn't it imply it?


*eyes narrow at screen, unsure if serious*

You're the kind of person from which phrases like "evolution is just a theory" originate, aren't you?

//If you are serious, note that "implies" means something somewhat different in the specific, formalized English of logic/rhetoric than it does the vague, colloquial English of hanging at the bar.
 
2012-10-02 07:31:40 PM  

RexTalionis: I prefer to say "post hoc ergo propter hoc".


post hoc ergo propter hoc (after this, therefore because of this) is not the same thing as cum hoc ergo propter hoc (correlation does not imply causation).
 
2012-10-02 07:32:19 PM  

SineSwiper: Lemme see:

1. Fish oil + heart disease = CORRELATION FAIL
2. Lycopene + cancer = CORRELATION FAIL
3. Echinacea + immune system = CORRELATION FAIL

But, hey, keep farking that chicken about how "correlation implies causation". (Meanwhile, the "supplement" industry would like to rake in millions of dollars from you.)


B-b-but it can't hurt, right?
 
2012-10-02 07:33:56 PM  

CreamFilling: moran: CreamFilling: uber humper: Correlation doesn't equal causation.

Causation does equal causation. Don't forget that.

Everything has a cause. Nothing happens just 'cuz

Maybe, but causation doesn't imply causation, by the same definition of "imply" used in the original. Maybe in mathematics, not in any real life scenario. Which is exactly why that definition of "imply" turns the whole argument worthless.

I know "imply" means something different outside of mathematics, but could you explain why "A causes B" doesn't imply "A causes B" under your definition of "imply"?

Pick any example you want of A causing B, and you'll find it doesn't always hold true. It's mostly an issue of mixing strict logic terminology with traditional common usages. Smoking causes lung cancer? Drunk driving causes accidents? Of course, unless you apply formal rules of logic.


It seems to me that "smoking causes cancer" implies "smoking causes cancer" under all three definitions of "imply" on the page http://dictionary.reference.com/browse/imply
 
2012-10-02 07:36:25 PM  

jvl: FTFA: No, correlation does not imply causation, but it sure as hell provides a hint.

Would this be a bad time to point out that hint and imply are synonyms?


Would this be a bad time to point out that Logical term "imply" is not a synonym of hint? It's closer to the standard English meaning of the word "prove" (but not the Logic or math meaning of "prove").

In standard English:

"A implies B" means "If A is true, that suggests that B might also true."

In Logic:

"A implies B" means "If A is true, B must also be true."

"Correlation does not imply causation" is a Logical argument, and is stating "If correlation exists, causation does not necessarily exist."
 
2012-10-02 07:36:50 PM  
I was watching a commercial for a whole-grain cereal a few weeks ago, and the announcer said something along the lines of, "people of a healthy body weight tend to chose whole grain foods."

The implication of course being that eating the whole grain cereal would change your fat, lumpy ass into a healthy body.

I guess it's OK that there are people in this country who accept that correlation unquestioningly. I would prefer it if everyone were smart enough to see the flaw in the logic, but then, who would buy whole-grain cereal with added sugar?
 
ecl
2012-10-02 07:38:46 PM  

lostcat: I was watching a commercial for a whole-grain cereal a few weeks ago, and the announcer said something along the lines of, "people of a healthy body weight tend to chose whole grain foods."

The implication of course being that eating the whole grain cereal would change your fat, lumpy ass into a healthy body.

I guess it's OK that there are people in this country who accept that correlation unquestioningly. I would prefer it if everyone were smart enough to see the flaw in the logic, but then, who would buy whole-grain cereal with added sugar?


Teachers apparently.
 
2012-10-02 07:39:39 PM  
mw2.google.com

^Example of coralation implying causation^
 
2012-10-02 07:44:17 PM  

ecl: lostcat: I was watching a commercial for a whole-grain cereal a few weeks ago, and the announcer said something along the lines of, "people of a healthy body weight tend to chose whole grain foods."

The implication of course being that eating the whole grain cereal would change your fat, lumpy ass into a healthy body.

I guess it's OK that there are people in this country who accept that correlation unquestioningly. I would prefer it if everyone were smart enough to see the flaw in the logic, but then, who would buy whole-grain cereal with added sugar?

Teachers apparently.


LOL r u sayin teachers r fat!
 
2012-10-02 07:48:58 PM  

SineSwiper: Lemme see:

1. Fish oil + heart disease = CORRELATION FAIL
2. Lycopene + cancer = CORRELATION FAIL
3. Echinacea + immune system = CORRELATION FAIL

But, hey, keep farking that chicken about how "correlation implies causation". (Meanwhile, the "supplement" industry would like to rake in millions of dollars from you.)



Great.

Can you point to at least ONE "supplement" that science actually has shown to provide some benefit?

Because if not, I have another correlation for you to examine.
 
2012-10-02 07:49:39 PM  
Author appears to be butthurt by savvy internet scientists exposing the design flaws of studies hyped to be more conclusive than they should be. Correlation does not prove causation. The relationship may be spurious, or due to some third unmeasured variable. On the other hand, a true cause and effect relationship does necessitate correlation.
 
2012-10-02 07:50:21 PM  
I'm not sure that using that argument is a guarantee of success.
 
2012-10-02 07:53:26 PM  
s2.hubimg.com
 
ecl
2012-10-02 07:53:56 PM  

moran: ecl: lostcat: I was watching a commercial for a whole-grain cereal a few weeks ago, and the announcer said something along the lines of, "people of a healthy body weight tend to chose whole grain foods."

The implication of course being that eating the whole grain cereal would change your fat, lumpy ass into a healthy body.

I guess it's OK that there are people in this country who accept that correlation unquestioningly. I would prefer it if everyone were smart enough to see the flaw in the logic, but then, who would buy whole-grain cereal with added sugar?

Teachers apparently.

LOL r u sayin teachers r fat!


That's the conclusion I'd expect someone named moran to jump to.
 
Displayed 50 of 226 comments

First | « | 1 | 2 | 3 | 4 | 5 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »
On Twitter





In Other Media


  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.

Report