Skip to main content

View Diary: How Many Of Those are Veterans? Gun Policy, Suicide, and Our National Conversation (185 comments)

Comment Preferences

  •  Two things (0+ / 0-)

    1)  Gun enthusiasts should acknowledge that 30,000+ Americans die every year due to gunshot injuries, and the magnitude of that yearly slaughter is in no way diminished or lessened because some of those gunshot injuries are self-administered.  To say that the tally of Americans who die by gunshot every year is inaccurate because it includes people who kill themselves is factually erroneous and morally repugnant.

    2) Empiric evidence demonstrates that the number of guns in America is directly correlated with the number of gun injuries, both fatal and non-fatal.  It follows that any reduction in the number of guns in America will be accompanied by a reduction in both fatal and non-fatal gunshot injuries.  That is the science of it.

    I am glad your guns have not resulted in any injuries.  But your argument about your guns is like the person who smokes but claims that cigarettes are harmless because they have not gotten cancer.

    The science that tells us cigarettes cause cancer is the same science that tells us more guns are associated with more gunshot injuries.  

    "The fool doth think he is wise: the wise man knows himself to be a fool" - W. Shakespeare

    by Hugh Jim Bissell on Mon Sep 24, 2012 at 12:07:49 PM PDT

    [ Parent ]

    •  Um. (8+ / 0-)

      1. Who's arguing the tally?

      2. This is based on your own research in a previous diary, where it was pointed out that your data set consisted of two consistently increasing variables parameterized over an annual time set.  It is unsurprising that you found a strong correlation given such a toy set up.  It is also incorrect to say that "it follows" that reducing the number of guns would reduce the number of fata and non-fatal gunshot injuries.  That is the science of it.

      Your crude, if somewhat laudable, practice with statistical methods is not epidemiology.  You explore no covariates, your data set is extraordinarily small, and your findings do not support your prohibitionist conclusion.

      •  The study was statistically correct (0+ / 0-)

        The study was performed using publicly available data (so you can check it to see if I got the numbers correct), and I used a common statistics test performed in the usual way.  Go ahead and perform your own Pearson's correlation on the data set and you will arrive at the same results I did.  If you object to the correlation, what statistics test do you think should be done.

        Go ahead and tell me what errors you found in either the data set or the statistics done.  If you cannot identify any errors, than your criticisms of the study are specious.

        The nature of a positive correlation is that if one number goes up, the other number goes up, and if one number goes down, there other number goes down.  That is the science of it.  Anything else is a misinterpretation of a correlation.  The mathematical nature of correlation allows anyone to claim that if the number of guns were to decrease, this would be accompanied by a decrease in gun injuries.

        Go ahead and do a study involving covariates and using a better database (if you can find one).  I look forward to seeing your research.    

        "The fool doth think he is wise: the wise man knows himself to be a fool" - W. Shakespeare

        by Hugh Jim Bissell on Mon Sep 24, 2012 at 01:07:47 PM PDT

        [ Parent ]

        •  I already spelled out my objections. (7+ / 0-)

          Which were reiterated above.  I also pointed out your error, in drawing a conclusion from a correlation that is extremely sensitive to even single point perturbations.  I suspect you're confusing the dependence test represented by Pearson's coefficient with curve fitting (an exercise for you, what is the coefficient for a dataset characterized by y = x^2?).

          •  Your objections are specious (0+ / 0-)

            You objected to my use of a correlation, but correlation such as the one I have done are used by scientists everyday around the world.

            You objected to the fact that no covariates were investigated, where there was no data of covariates to investigate.

            You complain that the dataset was too small, when I used the best and largest dataset available.

            Now you are complaining that the correlation I used is too sensitive to single point perturbations, when, as I have already said, I used the same correlation used by scientists everyday, AND you have not identified any single point perturbation in the data (which you have access to).

            Your criticisms of the research are specious.  The research stands as it is.

            If you can do a better study, I look forward to seeing it.

            "The fool doth think he is wise: the wise man knows himself to be a fool" - W. Shakespeare

            by Hugh Jim Bissell on Mon Sep 24, 2012 at 03:17:32 PM PDT

            [ Parent ]

            •  I did not object to your use of a correlation. (4+ / 0-)
              Recommended by:
              Joieau, oldpunk, PavePusher, Robobagpiper

              That is precisely the tool you would use.  I objected to the quality of your sample and the conclusion you drew from the resulting correlation.

              I don't blame you for the paucity of data, but I do expect you to be judicious in making extraordinary claims in the absence of evidence.

              A single point perturbation is easy.  Your dataset consisted of nine points in each vector.  Perturb one.

              The research does not stand.  The only thing you've shown is that two variables track with one another over a range of nine points.  You would have found a more impressive correlation (R = 0.909, p = 0.00065) between real median household income for the lowest income bracket.

    •  More bad argument. (7+ / 0-)

      (1) Who doesn't "acknowledge" that 30,000 people die of gunshot every year? 100,000 people die of medical errors in hospitals. 47,000 die of no medical care. ~50,000 people a year die of influenza. Cancer, organ failure, motor vehicle accidents, bathtub falls, electrocution, evil boyfriend/husband, drug overdose, house fires, wars... there are lots of ways to die. And every year millions of people do die. That is the nature of life (and death) on planet earth.

      You cannot prevent a single death by confiscating my gun, because my gun hasn't killed anybody and isn't likely ever to kill anybody. That's just a fact that you don't want to acknowledge because it doesn't fit your fallacious argument.

      (2) The number of guns in America still does not correlate with any gun out there that has never been used to commit a crime or kill self/other human being. You might better lobby to make private vehicle ownership illegal because ~30,000 people die in motor vehicle accidents every year.

      You'd get a lot of resistance from people who like their vehicles or claim to need them to get to work and the grocery store, etc. You could claim that car owners "don't care" about the ~30,000 people who die in vehicle accidents every year, but I sincerely doubt they'd be willing to surrender their car because you claim they "don't care."

      There is no "science" in your argument. There is a rather disturbing callousness evident in a totally fallacious argument and I cannot imagine why you'd believe it'll fly. Here at DKos or anywhere else.

Subscribe or Donate to support Daily Kos.

  • Recommended (128)
  • Community (55)
  • 2016 (43)
  • Environment (38)
  • Republicans (34)
  • Elections (34)
  • Bernie Sanders (33)
  • Culture (31)
  • Hillary Clinton (26)
  • Trans-Pacific Partnership (24)
  • Labor (24)
  • Climate Change (22)
  • Education (22)
  • Barack Obama (22)
  • Media (21)
  • GOP (20)
  • Civil Rights (20)
  • Economy (19)
  • Affordable Care Act (18)
  • Texas (18)
  • Click here for the mobile view of the site