Skip to main content

View Diary: Stephen Colbert rips apart debunked austerity economics paper (62 comments)

Comment Preferences

  •  The weird thing about R&R is that (17+ / 0-)

    hey reversd the obvious dietion of causality. Logically countries that are in recession have higher debt to GDP ratios. Recession =>lower tax revenues=>higher debt.
    By what logi would it work the oher way around?

    •  It's the classic ... (9+ / 0-)

      Correlation does not imply Causation mistake that people interpreting scientific research make all the time.

      When you see data set A and data set B that correlate well, there are 4 possibilities:
      1) A implies B
      2) B implies A
      3) A and B are both implied by another data set C
      4) A and B are independent, but random chance has allowed these sets to statistically correlate

      Determining which possibility is correct is where the real science is done. This path is fraught with the peril of confirmation bias.

      •  The phrase "correlation does not imply causation" (6+ / 0-)

        is wrong. Implying causation is precisely what correlation does.

        The phrase should read, "Correlation does not prove causation", or equal causation etc.

        While it's true that correlation does not prove causation, it is frequently one of the best indicators of it.

        So the point really comes down to the word "imply." One of Webster's definitions of imply is "to contain potentially." And this is precisely what correlation suggests: a potentiality.

        I wouldn't be so picky except that this phrase, which has been popularized over the last 20 years or so, is frequently used to shut down rational investigations into, among other things, the cause of climate change, the causes of various health effects from pollution, and other important fields of study like the causative effects of right wing economic policies.

        •  How About... (3+ / 0-)
          Recommended by:
          david78209, fumie, oceanview

          " not infer causation from correlation"?  That is more to the point to me, and it is the other way of saying it does not imply causation.  

          In and of itself, correlation does not allow one to infer (i.e., it does not imply) causation.


          A speaker or writer implies, a hearer or reader infers; implications are incorporated in statements, while inferences are deduced from statements. Imply means "suggest indirectly that something is true," while infer means "conclude or deduce something is true"; furthermore, to imply is to suggest or throw out a suggestion, while to infer is to include or take in a suggestion.
          So, yes, the two phrases, though related, are not exactly equivalent, and you are right.
        •  Sorry, but ... (3+ / 0-)

          from a Logic 101 point-of-view, the statement "Correlation does not imply causation" is correct.

          Remember that, in Logic, "A implies B" means that if A is true, then B must also be true. If A is not true, we know nothing about B, and the truth status of B does not affect the truth status of A.

          The problem here is that science uses the Logic definition of "imply", not Webster's layman definition. If you are going to be pedantic, you must also be precise. Your complaint doesn't meet that standard.

          And in your "correction" you want to change "imply" to "prove". But you should know that Science doesn't "prove" anything. That's the realm of Mathematics. We can't even "prove" the Theory of Gravity, but experiment and experience has shown us it's a darn good bet.

          I agree that correlation is a good indicator of possible causation, but as I said before, the real science happens when you have to determine where the actual causative relationship is, which is where confirmation biases show up.

          •  In statistics class... (0+ / 0-)

            They frequently use the correlation of ice cream sales and shark attacks.

            It seems that as ice cream sales increase on the beach, so do shark attacks.

            Hmmm...  We must therefor conclude that sharks must like ice cream and since they cannot get it for themselves, they must get the next best thing... Swimmers who have just eaten ice cream.

            That or maybe ice cream sales go up in the summer when it is warmer and more people are at the beach. The fact that is also when there are more swimmers during that time must be a complete coincidence.

            "Perhaps the sentiments contained in the following pages, are not YET sufficiently fashionable to procure them general favour..."

            by Buckeye Nut Schell on Wed Apr 24, 2013 at 04:21:52 PM PDT

            [ Parent ]

            •  Did you even read the comment? (1+ / 0-)
              Recommended by:
              Buckeye Nut Schell

              I said there a 4 possibilities when two data sets indicate correlation.

              One of those was that it was random chance.

              Please read before snarking next time.

              •  Actually, I wasn't trying to snark at you... (1+ / 0-)
                Recommended by:

                I was just trying to snark for the fun of snarking in general.  Kind of trying to support your position rather than tear it down.

                Upon re-reading my comment, I can see how it was taken that way.  

                I didn't previously uprate your comment due to trouble I was having with my browser not allowing me to.

                Sorry I wasn't more clear.

                "Perhaps the sentiments contained in the following pages, are not YET sufficiently fashionable to procure them general favour..."

                by Buckeye Nut Schell on Thu Apr 25, 2013 at 01:28:32 PM PDT

                [ Parent ]

          •  Ah, I wondered if this comment might attract (1+ / 0-)
            Recommended by:

            an mathematician. FWIW, I am fully aware of the mathematical definition of imply (and proof). I stumbled on to it again when looking for the definition to use in my comment. I thought about making my comment a lot longer by explaining why the mathematical definition is irrelevant since this expression is rarely used in a strict mathematical context. Then I thought, meh, why bother.

            I also, FWIW, have a huge problem with certain scientific disciplines abusing the English language. This is a peeve of mine going way back. I see it as unnecessarily esoteric and a hindrance to communication. Especially when mixed into layman discussions such as is the case here.

            The problem with pulling out the "it's the Logic definition" here is that this is not a philosophy class. In this context, the one that matters, and using standard English, the expression is entirely misleading.

            So instead of spending the rest of my life educating people what logical implication means, I have chosen to reject that terminology altogether. As it deserves to be rejected.

            Academic esotericism is just another form of elitism - I have little patience for it.

            And a dolphin a fish that happens to be a mammal.

            •  Wow. (0+ / 0-)

              That is truly an amazing amount of condescension there.

              It must be great to be able to reject certain terminologies out-of-hand, because that terminology seems to disturb all those patterns you have in your beautiful mind.

              There's too much fail in this whole comment to break down line by line, but I'll just go after the last, since you obviously threw it up as bait.

              And a dolphin [is] a fish that happens to be a mammal.
              And by the same definition, a sea turtle is a fish that happens to be a reptile. Or are you going to be pedantic about your definition of "fish"?

              Shared definitions are important, my friend, whether to like it or not.

              •  My apologies (0+ / 0-)

                I didn't mean to be condescending. I have a dry sardonic wit that rarely works on the internet. I have to try harder to remember that.

                The fish part was a joke. The annoyance you detected was really directed at the logic authorities and their abuse of the language. They're not the only ones.

                Ironically, I first started thinking about this when I was on the other side of the arguemnt. I was explaining to a commenter in one of the physics forms I hang out in the difference between the scientific meaning of the word "theory" and its non-scientific definition. The commenter stood corrected and we moved on.

                But the issue bothered me and later, I realized that the scientific definition of theory was an abuse of language. It lessens communication instead of enhancing it.

                The war on language is something I've been studying for many years. It began with my observation that the right wing media machine attacks ideas by attacking language.

                Recently, I explained to a wingnut, in person, that not only is "welfare" not a bad word, but that in the constitution, promoting the general welfare is one of the charges of the federal government.

                So shocking an idea was this that the guy didn't believe me.

                Neoliberal is another little gem. It completely confuses people. It sounds so benign - new liberalism. In fact, it's nothing more than a more radical form of laissez faire. Of course, part of that confusion is that the common vernacular of "liberal" is quite different from it's historic origin. But regardless, it was very effective terminology at making a bunch of self-delared liberals miss a giant asteroid that was falling on our heads. Like, "Don't worry, it's not an asteroid, it's a ..., hmm, oh yeah, a space-puffy. You won't feel a thing.

                Language matters. And the people who control it have a lot of power. There's a long history of this I don't have time to go in to. But I didn't want you to think I was just spouting off on the internet. I was spouting off with conviction. :)


Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site