Whose numbers?

Health care confusion
One more on health care

2013-01-07 gun control-2

A friend sent us this article (The Gun Challenge) with the following comment: “some stats, maybe it is not so difficult”
I fell off the chair laughing as I was reading the article:

“Australia is an excellent example. ……….
….The laws have worked. The American Journal of Law and Economics reported in 2010 that firearm homicides in Australia dropped 59 percent between 1995 and 2006.”

What made me laugh was the expression “firearm homicides”.
The first thing you learn growing up in a communist country is how to read the news. The friend who sent us the mail was a teenager when she left communism which may explain why she cannot read the New York Truth (Pravda) properly.

The Australians did a MASSIVE confiscation of guns after 1996 and since it is an ‘Island’ nation, replacing the guns was not easy. Naturally, the  firearm homicides did go down, but look at this graph from the Australian government:


Can you see the 59% decline in the overall homicide rates?  Neither can I.

There are other graphs on the site slicing the data in different ways but there is none to justify the number quoted in the NYT.

I do not know about you, but if I really must choose the way I die, I would probably prefer being shot to being beaten to pulp by a baseball bat or hacked to pieces by a machete, but the real question is: what difference does it make how a person dies to the ultimate effect?
If we take away pink baseball bats from people and only leave them with blue ones, rate of murders committed with pink baseball bats will indeed go down.

Unfortunately, the arguments on this subject are about this silly. The whole debate is grossly dishonest; the information is distorted beyond recognition and the response is vehemently partisan.

The dishonesty

There are three kinds of lies: lies, damned lies and statistics.
There are studies everywhere but there is not much consensus.

This study about “Knife-crime: the reality and its implications”  would deserve a study on its own with its clueless platitudes and non-answers.

Reading abstracts like this it is impossible not to notice the controls they had to apply to the samples to make them work. One would think that it is the simplest thing in the world. Before and after, one state to another. What’s there to control for?

“Our study extends previous work by using recent data, looking across both regions and all 50 states, disaggregating victims by age, and adjusting for potential confounders, including poverty, urbanization, unemployment, alcohol consumption, aggravated assault, forcible rape, and robbery”

Got it! We had to control for the ‘potential confounders’.

It is not enough that the studies are dishonest, the New York Truth  article distorts it even further.
These are from the conclusions of the study quoted:

“ It does not appear that the Australian experience with gun buybacks is fully replicable in the United States.”
“For Australia, a difficulty with determining the effect of the law was that gun deaths were falling in the early 1990s”

All the while there is evidence that other types of crimes increased.
This turns into incontrovertible truth in the NYT.

The Gullibility

The reaction of our friend was typical. I can picture how the mail was created – a quick cut and paste from a reliable source – as for the readers of the New York Truth that publication is as authoritative as it can get. Most people invest only a limited effort into determining what do they consider a reliable source mostly by what makes sense to them.

Once we determined what we trust, we tend to rely on them unquestioningly. Only a few ask how much sense the numbers they see actually make. Even fewer can notice the subtle hints such as the “firearms homicides” indicating manipulated data.
Finding good information these days is not that hard. Anybody, who wants to find out the facts can, yet very few do. We must wonder why.
The logic of the pro-gun argument seems to be much stronger but they do not have the emotional vehemence of the gun control advocates.
The source of gullibility is lack of knowledge and wishful thinking. If the wish is strong enough, then knowledge, information and logic is not needed at all. People who really want to believe will believe anything while actively shutting out any contrary information as noise.

The silliness

A few days ago I got an e-mail with the following link: “England Considering Ban On Kitchen Knives”  Note that the article is quoting the Wall Street Journal coming to a conclusion opposite to that of the NYT.  I still could not believe my eyes so I googled the “causes of knife-violence”  to find enough sources to convince me that it is not a joke. Since guns are heavily controlled, there was a sharp rise in crimes committed with knives. Who could have possibly predicted that?? Nevertheless, it is a problem they need to deal with, so we have the perfect liberal response: Let’s ban knives! We need to ask, what’s next? Scissors? Hammers? Fists? Shouldn’t we also consider chain-saws and automobiles?

After 9-11 one must seriously wonder how is it possible that box-cutters are still sold in stores…………. The left is clearly not vigilant enough.

Making sense of it.

The US ranks Number one in the world in household gun ownership. It ranks 28th in homicide by firearm.
Mexico has very strict gun control laws and more than three times the American rate of “firearm homicides”
When ‘gun crime’ goes down other types go up.

I could keep quoting facts ad nauseum, studies ad infinitum but what decides in the end is which of those facts and studies make sense to us.

I am in the middle of reading “The Better Angels Of Our Nature”, a systematic research into the fact and the causes of the historic decline of violence in our world.
The long term trends  are down. Even in Australia, regardless of gun regulations. In front of this background of general downward trend, there are variations.
I don’t have this book (More guns, less crime) yet, but I am planning to get it. I did watch, however, some clips of the author being interviewed by the liberal media hosts with white foam coming out of their mouth. The general attitude is best expressed by this sentence from a book review: “This is too important a topic for such an idiotic contrarian stance.”
If a topic is important, nobody is allowed to disagree with the left. The kind of vehement, violent aggression coming from the gun-control advocates should make everybody take a step back to calmly rethink the issue. The left refuses to have any rational discussion. They just don’t want to let another perfectly good crisis to go to waste.

I fully share the sentiment expressed on this T shirt:

it is not about guns

I never owned a gun in my life and I see no reason to buy one now, but the notion that guns create a safer world makes perfect sense to me.
All you need to consider is that the majority of people in this world are decent, and would use the gun only if it is justified. Protecting themselves and others is justified use. I do not want to live in a world where only criminals have guns – as is the case today in many gun controlled areas.
I want to be a gun parasite, someone who can afford not to have a gun knowing that I am surrounded by decent people who do.

I keep claiming that leftists do not like people. They look down on them, they do not trust them, their basic assumption is that people are bad and they need to be controlled. This is just another example. The left honestly believes that people disagreeing with them are bad people, because the very essence of their existence is the belief that they are good and what makes them good is the set of ideas they hold. This set of ideas function as religious dogma, completely unrelated to reality. The proof they use to justify their beliefs is about as tortured as the modern miracles of Christianity.

Our friend, like any other liberal leftist is only interested in justifying her believes with numbers, not in actually understanding what they mean, imply or suggest.
To be fair to her, I must admit that the other side of the argument is not much better. Few would really look into the numbers or read the books referenced above.

The real causes

The debate is so loudly focused on guns that little room is left to consider the actual causes. Who do I blame? Ken Kesey.
When he wrote his book, “One flew over the Cuckoo’s nest” it was meant as an allegory but society took it literally and started a fundamental change in how we deal with mental illness. The book is excellent, so is the Milos Forman movie, but that does not make the problem they created any better.

Did you ask yourself why the left wouldn’t even talk about THAT problem? Most of the mass murders in recent history were committed by deranged people who should have been in a mental hospital. Why are we only talking about guns and never the people using them? ……. Unless, of course, they have any kind of conservative affiliation………

If they do not, it is assumed and claimed anyway until the evidence makes it impossible to maintain the allegation. Strangely, if mass murderers have any political affiliation, it is overwhelmingly on the left. Maybe the answer should be to ban the Democratic Party of the USA. Just a thought. It could improve the world in more ways than I can count.

The answer

In the end, my answer to our friend is simple:

No, my friend, it is a little more difficult.

Gun control is NOT the answer, no matter how much the ideology driven “scientists” distort the data, no matter how much propaganda the liberal media is spewing about it. The answer is more guns in the hands of decent people, better law enforcement and most importantly a reconsideration of our attitude toward mental illness. You do not even need science to come to this conclusion, just some common sense and a little faith in the goodness of your fellow citizens.

Feel free to disagree, but please try to use some better ‘evidence’ to make your case than what can be found in the rag called the New York Truth (Pravda)

I am generally sceptical about numbers indicating spectacular results because they make little sense and often turn out to be bogus.

If you are serious about trying to understand the problem, you should read serious publications about it, such as this CATO study, the book I referenced above or even this congressional study that would be the closest to supporting your ideas.

Even if you do not have the time to look into a question seriously, sensationalist claims can be checked very easily and quickly. It took me less than 5 minutes to find the graph above from a completely reliable source: Australian Government Statistics.

There is no excuse for using propaganda in a reasonable discussion.

6 replies on “Whose numbers?”

  1. Bogi says:
    Sorry to burst your bubble there but the Australian government stats are correct there. On the same page that you found the overall homicide rates, they have a chart representing the firearm homicide rates. As historically firearm homicides made up no more than about 40 deaths per year a 59% decrease in them would come to about 24 less deaths. There is a larger decrease than that on the overall homicide chart. That suggests that the overall homicide rates have dropped more than it is expected to drop because of the decrease in firearm deaths alone.
    Expecting a 59% drop on overall homicides rates would equate to about wanting a 180 drop on the chart you quoted. I am hoping you get the idea here.
    Also, the 59% rate in firearm deaths is correctly quoted according to the 3rd graph here. (http://www.aic.gov.au/statistics/homicide.html)
    • zorkthehun says:
      Boglárka, virágszálam!
      You shouldn’t worry about my bubble; it is much tougher than you think. I did not question the correctness of the Australian statistics, only the NYT’s use of it. How someone is killed is irrelevant. Should we applaud a policy that simply shifts the tool? Let me ask a hypothetical question to illustrate:
      you have a society with a rate of 100 homicides per 100K population. 50% is committed by guns, the other 50% by knives. Should we applaud a policy that cuts the gun deaths in half while also doubling the knife deaths? (Just to help you with the math, that would be 125 homicides 🙂 ) What if the 50 gun deaths are simply replaced by knife deaths? Would that improve that society in any way? That is what I am talking about when I point out that there is no statistically significant effect on the overall crime rate. (Which, btw, is explicitly stated in the study.) A change in the overall crime rate is the only thing that matters. ….and I did not even start addressing the issues that were pointed out in the linked Youtube video. What if the reduction of homicides also results in vastly increased numbers of aggravated assaults, rapes and robberies? What if gun control brings the opposite effect?
      This last one is not an entirely hypothetical question. Read this article:
      “Figures showed the number of crimes involving handguns had more than doubled since the post-Dunblane massacre ban on the weapons, from 2,636 in 1997-1998 to 5,871.” (in 2003)

      Can you say that gun-control created a better world in the UK?
      From what I understand, this is also the point of the book, “More guns, less crime”
      I don’t “expect” anything; I am questioning the accuracy of and the motivation behind the sensationalist propaganda baloney of the NYT.
      All of this to answer our friend, to show her that the issue is NOT as simple as she hopes and wishes it to be.

      (((and I appreciate your comment more than you know 🙂 )))

      • Bogi says:
        Yay, you replied!
        Right, so the Ozzie stats not producing a statistically significant result does not mean that knife deaths have gone up. They didn’t. It means that in light of the previous years’ fluctuations we can not state that the drop that was experienced is statistically significant.

        I really doubt that it would ever happen that knife deaths would go up because someone doesn’t have access to a gun. (I.e.: your hypothetical scenario would be unlikely). If you think about it logically, it kind of makes sense: the number of homicides tends to be more or less constant over time, x of those are committed by guns, y by other means. No access to guns means some of those x murders will be then committed by knives and other means but it is unlikely that all x or more would be done so, partly because you need to be a) pretty determined to kill someone with a knife/fist/whatever b)real badass. Not everyone is. Shooting someone is relatively easier than butchering them with a knife. So some of those murders will co uncommitted. So in summary, yes, knife crime is likely to go up but unlikely to match or exceed the combined number (unless people get really angry that they don’t have a gun and slash the throats of two people instead of one :)). The net rate is likely to drop.

        I must admit I only kind of skipped through the youtube video but all I saw was a bunch of rednecks lamenting about how their guns were taken and it most certainly did not seem like a reliable study.

        As for the UK, yes, indeed, there can be a side effect like that when banning guns. Gangs (the article states that this is where most of the gun crime happened) are part of urban youth culture and I can very much imagine that a ban on guns makes them into a status symbol and hence desirable to have. Gangs tend to fight other gangs, though, so I am guessing most of that crime is committed sort of “internally”. To us by us in gangworld, haha. Also, the UK didn’t get anywhere near the rates of domestic gun owners before the ban so the gang-effect rise on gun crimes can be such a large one because it is essentially easier to double a small number than a large one. I wouldn’t think that it is a linear curve, though.

        I’d also like to address that other study you quoted- the one with the controlling for confounders, etc. It is a really good thing they did that! No study can call itself a “study” without doing that basic statistical technique. It is done to make results more reliable and it is not some voodoo that is to ensure you get the number you want. In fact, it is quite the opposite. It usually weakens any claims you have and once you control for confounders you can find that you are without results. (A confounder is when another effect is present in a study and you’d be likely to mistakenly draw conclusions based on it rather than the actual effect. A simple example: study on lung cancer finds that there is a 80% correlation between lung cancer and carrying matches. One (stupid person) could conclude that carrying matches causes lung cancer. Whereas it is confounding smoking which is the important variable we want to be looking at). The study states they found no causality only a correlation between ownership and homicide, which is as much as Statisticians can ever hope to say.

        When I said “expect” I meant it in a mathematical sense. As in “when you add two and three you’d expect it to be five”.

        Any case, I agree with you, a very important thing would be addressing the way the US deals with mental health issues as well as thinking about domestic gun ownership. I also agree that the Australian model will not for for the US because a)there’s too much money in guns b)the right to protect one’s self and (insert exaggerated American accent here) property is deeply entrenched in the American psyche (protect yourself from them injuns/lawless hobos/buffaloes/grizzly bears).

        Phew, gotta dash, hope I got it all in.

        • zorkthehun says:
          With all due respect to your chosen profession, Ms. Statistician, I am not stranger to the dark art myself, but over the years I learned to appreciate its dangers, which is how I came to the conclusion that only integrity matters, something I’m afraid the NYT sorely lacks.
          Remember, the point of my post was simply to state that the matter is not as simple as the gun control advocates claim it is.
          My problem with most leftist ideas are a total disregard for consequences, the lack of understanding a simple fact: there are no solutions, only trade-offs and we have to be very careful not to make things worse by ignoring them.
          Even if homicides do actually go down in absolute numbers in a statistically significant way, other types of crimes can go up. How do you measure the value of one victimization over another?
          It is not simply the case of exchanging one type of murder for another, but changing the whole landscape of crime, where you have to ask what is the desired outcome and what is likely to bring it about.
          “I call your homicide and raise you three rapes and an aggravated assault”
          Would that be a good deal?
          This is again (and of course) not a hypothetical point. It is exactly what happened in both England and Australia. Burglaries and robberies both went up in both places. (but I am lazy to look for the reference ). The knife example was just that: an example to illustrate the point.
          Last two links, I promise.
          The first article came to my mailbox yesterday.
          This one, although clearly partisan, seems to be very well researched and referenced.
          DEFINITELY better than anything I came across on the gun-control side
          But the CATO study is the one to read:
          Yesterday I met a guy, who said that after looking into the issue starting from a liberal position got so convinced by the evidence, that he actually bought a gun. I am not there yet. 🙂
          Let me finish with two truisms:
          When guns are outlawed, only outlaws will have guns.
          The best defense against a bad man with a gun is a good man with a gun.
          You do not need to know any statistics, you don’t need to know the evidence to see the logic of either. They are painfully obvious.
          What kind of brain damage prevents the liberal mind from being able to understand them?
  2. Oresztesz says:
    The most important thing I learned from my university statistics course is how unreliable any suveys and statistics are. So many levels to manipulate, starting with the questions in case of surveys: what questions, wording, order, open ended or Y/N, possible answers, then the interviewers: who they decide to talk to, how they look like, then the subjects, in what they think, what they think they should say, the list is endless. Even with statistics, that -should- be straightforward, it is a question what is included, what is omitted, criteria for one set and the other. And don’t even get me started on interpretation, partial analysis, omitting results, coincidence confused with causation, what time scale you are using (how much you zoom in or out on the historical results) what data you compare, and I could go on had my wife not called me that I’m late for my appointment. Our teacher was great, he made us make and interpret surveys that would “prove” completely contradicting results from the same sample. Take statistics for what they are. Cute but it’s the tip of an iceberg of information that you should investigate to your satisfaction before you make any decisions yourself.
    • zorkthehun says:
      You don’t have to tell me. It was exactly this realization that led me to my motto that led to the argument which prompted the mail that got posted on October 29th
      Truth is in the mind of the beholder
      Reality is just an excuse
      Only integrity matters

      We could say statistics is only an excuse. I firmly believe that only integrity matters, which I interpret as honesty toward yourself and your subject alike. I think most gun control advocates want to believe their own touchy-feely BS so badly, that they happily accept sleazy statistics from propaganda organs such as the New York Times.
      The question to which I still seek an answer is why? Do they gain anything beyond the smug self-satisfaction about their own moral superiority? Can that cheap satisfaction compensate for the harm the advocated policies cause? If I ever find the answer, you’ll be first to know.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.