Page 26 of 26 FirstFirst ... 16242526
Results 251 to 252 of 252

Thread: One-year-old shot dead by 3-year old in Cleveland home.

  1. #251
    Sage

    Join Date
    Mar 2014
    Location
    Tennessee
    Last Seen
    Today @ 10:57 AM
    Gender
    Lean
    Undisclosed
    Posts
    17,969

    Re: One-year-old shot dead by 3-year old in Cleveland home.

    Quote Originally Posted by Gonzo Rodeo View Post
    No no, what you said was "The point is there isn't just a correlation being noted." The problem is that it is only a correlation. If it's beyond a correlation, it is a cause, which is what you are actually arguing.
    This is a semantic argument. Lot's of stuff we know about health is merely noting 'correlations' between two things. For example, if a new drug is tested on a population of 5,000 individuals and there is a 700 times higher risk of cancer in the drug group versus the control group, researchers would conclude that the drug causes an elevation in the risk of cancer, even when they will (generally) have no idea WHY or HOW the tested drug causes cancer.

    What you're telling me is unless the study can conclusively PROVE the drug CAUSED the cancer risk, that you'd gladly take that drug, and be happy that the drug trials merely showed a 'correlation' but no 'cause.' I highly doubt that. What rational people would do (assuming the drug came to market) is weight that massive increased cancer risk with the benefits of the drug. If it was for pesky headaches, only idiots would take the drug. If the drug cured HIV, maybe some would take it.

    "Successfully" doesn't belong in the same category as "more likely". What these studies prove is that guns provide a clear and easy path for those who wish to end their lives to do so. They do not show that people are more likely to attempt.
    What they show is the ratio - [people dead by suicide]/population goes UP for gun owners.

    The odds are about 25:1 for unsuccessful attempts. That means for every 2 people who killed themselves, 50 people tried. About half of all successful attempts are completed with a gun, so out of 50 suicide attempts (and 2 completions), 1 was the result of a gun. What that tells me is that guns are quite effective as a chosen method... not that they have anything to do with an increased incidence of attempts, i.e. you aren't at any higher risk of committing suicide if you don't already want to commit suicide.
    Attempts isn't the relevant stat. It's DEAD PEOPLE/population. Those who use guns are more likely to be dead after the attempt. About 98% of those who try overdoses are not dead after the attempt. Most of those who try and fail do not later successfully kill themselves. They get treatment, help, the crisis passes and they move on with life. With a gun, there is usually no ability to get treated because they are DEAD.

    It is a dishonest conflation to say increased ability to do the deed is the same as increased desire of doing the deed, which is why it's inappropriate to call it an "increased risk."
    Gun owners are more likely to DIE from suicide than non gun owners. That's what the studies show.

    No, I mean the actual papers. I want access to the papers themselves. Methodology. Raw data. Not a two liner half assed abstract.
    You can use Google and follow links. The cites were there. If you care, you can find the papers. If not, fine. Whatever.


    They aren't at greater risk of suicide; they are more likely to successfully complete their attempt. To claim they are at greater risk of suicide is to imply causation without separating attempts from the total number. I mean, if we assume gun ownership leads to a greater risk of being in the subgroup "completed suicides", then we must also conclude not owning the gun leads to greater risk of being in the subgroup "suicide attempt".... do you believe not owning a gun makes someone more likely to attempt suicide? No, you don't, but if we apply this take on statistics to the 25:1 ratio, this is the only conclusion. So, you see, the reasoning must be flawed somewhere.
    Again - the statistic of interest is #Dead by suicide/population. The ratio is higher for gun owners than for non-gun owners. The ONLY question is WHY. If it's not the gun itself, then something about gun owners other than the gun causes this result.

    If you can find any stat that identifies that cause other than guns, please let me know.

    When Hemenway does a study comparing gun ownership across country lines instead of state lines, I would love to see it.
    The link had many cites to papers other than by Hemenway.

  2. #252
    better late than pregnant
    Gonzo Rodeo's Avatar
    Join Date
    Dec 2011
    Location
    Here
    Last Seen
    Today @ 10:23 AM
    Lean
    Private
    Posts
    3,656

    Re: One-year-old shot dead by 3-year old in Cleveland home.

    Quote Originally Posted by JasperL View Post
    This is a semantic argument. Lot's of stuff we know about health is merely noting 'correlations' between two things. For example, if a new drug is tested on a population of 5,000 individuals and there is a 700 times higher risk of cancer in the drug group versus the control group, researchers would conclude that the drug causes an elevation in the risk of cancer, even when they will (generally) have no idea WHY or HOW the tested drug causes cancer.

    What you're telling me is unless the study can conclusively PROVE the drug CAUSED the cancer risk, that you'd gladly take that drug, and be happy that the drug trials merely showed a 'correlation' but no 'cause.' I highly doubt that. What rational people would do (assuming the drug came to market) is weight that massive increased cancer risk with the benefits of the drug. If it was for pesky headaches, only idiots would take the drug. If the drug cured HIV, maybe some would take it.
    No, what I'm telling you is that there happens to be a correlation between gun ownership and suicide rate. You are continuing to argue for a causative relationship where no evidence of one exists. This correlation can be explained in several different ways, any of which need to be subjected to statistical analysis and experiment to confirm (if possible). For example, people who seriously wish to kill themselves may (very likely) wish to use a gun to accomplish the fact as a reliable and sure method; pills might not work, jumping/cutting may hurt, hanging may take too long, etc. Since gun ownership in the US is high (around 40% of households), it is very likely that those who seriously wish to kill themselves are already in possession of the means with which to do so reliably. 50% of suicides are carried out with a gun, while 40% of households report gun ownership. Those numbers are fairly close. And since none of us have access to the actual numbers and just how much more represented that 40% is within that 50%, we can't say for sure how significant that gap is.

    The only way to control for this variable is to ban private gun ownership and then look at the suicide rate. When a study can control for every variable, it might have some merit... but regression analysis (which this is) cannot control for anything! For example, a child is more likely to drown in a backyard swimming pool than be accidentally (or purposefully) shot by a gun, even though swimming pool ownership is far, far lower than gun ownership.... but do we say that swimming pools cause drownings? There is a correlation between swimming pools and accidental death due to drowning, and the risk of your child dying goes up if you own a swimming pool, but the swimming pool does not cause the risk! All you can definitely say FOR SURE is that if you own a swimming pool, your kid is more likely to drown in a swimming pool than someone else's kid who doesn't own a swimming pool.

    The difference is subtle, but it most certainly exists.

    What they show is the ratio - [people dead by suicide]/population goes UP for gun owners.
    If this were true, then countries with lower private gun ownership rates should see lower suicide rates. This isn't the case, so guns are not a causative agent.

    What is more likely true is that people who wish to kill themselves are more likely to do it with a gun if one is available, but this statement remains unproven and I can't stand by it 100%.

    There is also a whole host of other factors as of yet unaccounted for most likely. For example, in Japan, they smoke like three times as much as Americans but have half the instances of smoking associated lung cancers. They also eat more fish. Is there a correlation? Most certainly. But what does this all say about causation? Well, that's a more complicated story.

    You can use Google and follow links. The cites were there. If you care, you can find the papers. If not, fine. Whatever.
    The link had many cites to papers other than by Hemenway.
    I can't access the papers directly. If you can find a way to do so, please let me know. I would *love* to murder the methodology here in a public forum. Figuratively speaking, of course.
    Last edited by Gonzo Rodeo; 04-21-15 at 05:53 PM.
    "Political speech and writing are largely the defense of the indefensible. . . . Thus political language has to consist largely of euphemism, question-begging and sheer cloudy vagueness."
    ~Orwell, Politics and the English Language

Page 26 of 26 FirstFirst ... 16242526

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •