Liberals Lying about Guns Again

[JohnStOnge]

Either that or they believe so fervently in what they're doing that they are deceiving themselves. Saw this yesterday morning while watching MSNBC:

http://www.msnbc.com/morning-joe/watch/ ... 9275843593" onclick="window.open(this.href);return false;

Wait a while and you'll see the video and the Rattner chart thing starts about 45 seconds in. I reproduced the last two charts at the bottom of this post. The problem with the second chart that shows a correlation between gun ownership rate and gun homicide rate it that he cherry picked a handful of countries. Take a look at the table at http://www.washingtonpost.com/wp-srv/sp ... hip/table/" onclick="window.open(this.href);return false; . It's a gun homicide rate table so not quite the same as gun deaths. But if I just use the data in the table from the eight countries Rattner used in the video I get a highly significant correlation between gun ownership rate and gun homicides. 99.6% confidence level. Consistent with the point he was trying to make.

HOWEVER, if I don't cherry pick and just use data in the table for all 107 countries for which the necessary data are available, I do not get a significant correlation. Also, the correlation coefficient is negative. That means that to the extent that one should lean one way or the other one would lean towards thinking gun homicide rates tend to go DOWN as gun ownership goes UP. You could be about 70 percent confident that's the case.

And if you do some math and look at the TOTAL homicide rate you get a negative correlation that gives you about 84 percent confidence that homicides tend to go DOWN when gun ownership rates go UP.

The other chart is bad because he acted like it showed something and it doesn't. It just shows a downward trend from the beginnning of the period depicted. If all the bars were of one color you could not tell when the law he referenced was passed. There's no obvious change in the downward trend.

[kalm]

Your ability to twist statistics to suit your ideology is truly exceptional!

[JohnStOnge]

Your ability to twist statistics to suit your ideology is truly exceptional!

I hope that's a joke. All I did was find a source of data...the Washington Post...and use all available data. The point is that whoever did that for Rattner or maybe Rattner himself selected a small handful of countries that would give them the result they wanted when they did the correlation.

And the second graph is what it is. You make all those bars blue or all those bars red with no labels and you can't tell when the law was passed. It's just a downward trend from start to finish. There's certainly no dramatic change in the downward trend associated with the point at which the law was passed. Surely you can see that. In fact it looks to me like if you were to fit a trend line you'd see a steeper decline in the area where the red bars are than you would see in the area where the blue bars are.

That's just a case where somebody points to a chart saying it shows something and it just doesn't show it.

[kalm]

Your ability to twist statistics to suit your ideology is truly exceptional!

I hope that's a joke. All I did was find a source of data...the Washington Post...and use all available data. The point is that whoever did that for Rattner or maybe Rattner himself selected a small handful of countries that would give them the result they wanted when they did the correlation.

And the second graph is what it is. You make all those bars blue or all those bars red with no labels and you can't tell when the law was passed. It's just a downward trend from start to finish. There's certainly no dramatic change in the downward trend associated with the point at which the law was passed. Surely you can see that. In fact it looks to me like if you were to fit a trend line you'd see a steeper decline in the area where the red bars are than you would see in the area where the blue bars are.

That's just a case where somebody points to a chart saying it shows something and it just doesn't show it.

So Japan's homicide rates should be through the roof!

[CAA Flagship]

I hope that's a joke. All I did was find a source of data...the Washington Post...and use all available data. The point is that whoever did that for Rattner or maybe Rattner himself selected a small handful of countries that would give them the result they wanted when they did the correlation.

And the second graph is what it is. You make all those bars blue or all those bars red with no labels and you can't tell when the law was passed. It's just a downward trend from start to finish. There's certainly no dramatic change in the downward trend associated with the point at which the law was passed. Surely you can see that. In fact it looks to me like if you were to fit a trend line you'd see a steeper decline in the area where the red bars are than you would see in the area where the blue bars are.

That's just a case where somebody points to a chart saying it shows something and it just doesn't show it.

So Japan's homicide rates should be through the roof!

No. They are too obsessed with porn pixels.

[JohnStOnge]

So Japan's homicide rates should be through the roof!

No, what it means is that there is no clear trend one way or the other. If you had to bet you'd bet there is a weak trend, though, such that countries with fewer guns per capita actually TEND to have higher homicide rates. The main thing is that when you just look at the data that are available instead of cherry picking a few countries that allow for creating the impression you want to create there is no support in the data for the Statement that more guns equal deaths country by country.

And I can say that even though the Washington Post table just includes homicides and not all gun related deaths. That's because you have some countries with very low gun ownership rates and very high homicide rates that exceed the overall total gun death rate in the United States. Note that you can tell from the chart Rattner provided that the total gun deaths rate for the United States, which is ranked #1 in gun ownership rate, was 10 per 100,000 people. The Washington Post Chart has Honduras with 68 gun related and 83 total homicides per 100,000 population even though it is rated in the lower half of the 107 countries for which all necessary data are available a 66th in gun ownership rate a. Ecuador is 94th among the 107 in gun ownership rate but its gun homicide rate exceeds the total gun related death rate of the United States shown on Rattner's chart (12 to 10).

The "more guns equals more deaths" by jurisdiction association just is not there.

I mean c'mon man. Just look for yourself at that Washington Post chart for a while. Just kind of look around at different countries and their gun ownership rates and their gun homicide rates. You can also get a calculator out and calculate their total homicide rates if you want to because the percent of all homicides that are gun homicides is provided. You start looking around at that table and it's OBVIOUS that there is no pattern whereby higher gun ownership rates are generally associated with higher gun homicide rates in particular or homicide rates in general.

It's not like the Washington Post is working for the NRA. The data are what they are.

[houndawg]

....never miss an opportunity for mental masturbation..

[CID1990]

So Japan's homicide rates should be through the roof!

No, what it means is that there is no clear trend one way or the other. If you had to bet you'd bet there is a weak trend, though, such that countries with fewer guns per capita actually TEND to have higher homicide rates. The main thing is that when you just look at the data that are available instead of cherry picking a few countries that allow for creating the impression you want to create there is no support in the data for the Statement that more guns equal deaths country by country.

And I can say that even though the Washington Post table just includes homicides and not all gun related deaths. That's because you have some countries with very low gun ownership rates and very high homicide rates that exceed the overall total gun death rate in the United States. Note that you can tell from the chart Rattner provided that the total gun deaths rate for the United States, which is ranked #1 in gun ownership rate, was 10 per 100,000 people. The Washington Post Chart has Honduras with 68 gun related and 83 total homicides per 100,000 population even though it is rated in the lower half of the 107 countries for which all necessary data are available a 66th in gun ownership rate a. Ecuador is 94th among the 107 in gun ownership rate but its gun homicide rate exceeds the total gun related death rate of the United States shown on Rattner's chart (12 to 10).

The "more guns equals more deaths" by jurisdiction association just is not there.

I mean c'mon man. Just look for yourself at that Washington Post chart for a while. Just kind of look around at different countries and their gun ownership rates and their gun homicide rates. You can also get a calculator out and calculate their total homicide rates if you want to because the percent of all homicides that are gun homicides is provided. You start looking around at that table and it's OBVIOUS that there is no pattern whereby higher gun ownership rates are generally associated with higher gun homicide rates in particular or homicide rates in general.

It's not like the Washington Post is working for the NRA. The data are what they are.

there you go again bringing qualitative analysis to an emotional argument

[JohnStOnge]

Here's one for you:

You can do a little math and determine that Honduras has NON GUN ASSOCIATED homicide rate of 13.9 per 100,000. That exceeds the TOTAL homicide rate of the United States, 4.7 per 100,000, by a factor of about 3.

No doubt guns are A factor. They're effective tools. But they are not one of the primary factors contributing to the problem. Liberals have this "thing" where they act like passing the gun control laws they want would solve the problem. And it won't.

Guns are tools. This thing of focusing on the tool is absurd.

[JohnStOnge]

....never miss an opportunity for mental masturbation..

Do you enjoy being lied to?

[JohnStOnge]

So Japan's homicide rates should be through the roof!

Gun ownership is clearly not the reason why the United States has a higher homicide rate than Japan does. If you just look at United States homicides that did NOT involve guns that rate exceeds the total homicide rate of Japan by about a factor of 3.

[Chizzang]

Johnny,
Me thinks thou doth protest too much...

[houndawg]

....never miss an opportunity for mental masturbation..

Do you enjoy being lied to?

No, but I appreciate the effort to whitewash whatever horseshit you're trying to sell at the moment with a pseudo-scientific veneer of statistics.

Statistics are like bikinis John. What they reveal is interesting, what they don't reveal is vital. The person who said that was a real scientist John and it would do you well to brush up on his work if you're going to keep trying to pass yourself off as a real scientist too.

[CID1990]

Do you enjoy being lied to?

No, but I appreciate the effort to whitewash whatever horseshit you're trying to sell at the moment with a pseudo-scientific veneer of statistics.

Statistics are like bikinis John. What they reveal is interesting, what they don't reveal is vital. The person who said that was a real scientist John and it would do you well to brush up on his work if you're going to keep trying to pass yourself off as a real scientist too.

"My statistics are truer than your statistics"

or

I reject your reality and substitute my own

[kalm]

So Japan's homicide rates should be through the roof!

Gun ownership is clearly not the reason why the United States has a higher homicide rate than Japan does. If you just look at United States homicides that did NOT involve guns that rate exceeds the total homicide rate of Japan by about a factor of 3.

But would US rates go down if we had Japan's laws?

Or perhaps there'd be 10,000 more knife and throwing start related deaths?

[JohnStOnge]

No, but I appreciate the effort to whitewash whatever horseshit you're trying to sell at the moment with a pseudo-scientific veneer of statistics.

Statistics are like bikinis John. What they reveal is interesting, what they don't reveal is vital. The person who said that was a real scientist John and it would do you well to brush up on his work if you're going to keep trying to pass yourself off as a real scientist too.

Dude, I provided a link to the data. Just look at it. There is no way there's evidence in those data to support the idea that there's an association among countries between higher gun ownership rates and higher gun homicide rates or higher overall homicide rates.

The obvious misuse of statistics was on the part of Rattner as he pulled out eight countries with numbers that fit the idea he wanted to promote and ignored all the other ones. Seriously. Are you really defending what he did? Can you REALLY not see how deceptive it is?

[Skjellyfetti]

there you go again bringing qualitative analysis to an emotional argument

quantitative

[CID1990]

there you go again bringing qualitative analysis to an emotional argument

quantitative

winner winner chicken dinner

i was trying for clitz on that one

[ASUG8]

Gun ownership is clearly not the reason why the United States has a higher homicide rate than Japan does. If you just look at United States homicides that did NOT involve guns that rate exceeds the total homicide rate of Japan by about a factor of 3.

But would US rates go down if we had Japan's laws?

Or perhaps there'd be 10,000 more knife and throwing start related deaths?

To rephrase: Would you rather have the US gun culture, or a shit ton of ninjas?

[93henfan]

Why are we even having this argument? France has some of the toughest gun control on the planet and it hasn't stopped multiple AK-47 attacks this year.

Gun control doesn't work, plain and simple. Bad guys will get guns, regardless of the law. Why take the fundamental right of defense away from the good guys?

Gun control is liberal feel-good legislation - see the Brady Bill.

[JohnStOnge]

"My statistics are truer than your statistics"

or

I reject your reality and substitute my own

If there is an attempt to mislead with statistics one should be able to recognize it and describe the problem. For instance, you can easily see that there is a problem with Rattner's correlation between gun ownership rate and gun deaths by country because he just used 8 countries and ignored the data from all the other countries for which data are available.

There's nothing deceptive or misleading about what I did. I used all the data available and linked it so people can look at it themselves. All I did beyond that was calculate correlation coefficients to see if there's sufficient evidence to say there is an association such that gun homicides rates in particular or homicides rates in general tend to be higher in countries with higher gun ownership rates. And there isn't. Nothing tricky or deceptive about that. It is what it is.

And I used the friggin' Washington Post as the data source, for Pete's sake. That's not exactly a "conservative" pro gun source.

[JohnStOnge]

And on the other one look at the chart below. Imagine that there is no legend to tell you when the law went into effect. No red and blue. Let's say all the bars are grey. Suppose someone said that a grey graph like that where you couldn't see the different colors showed that "Gun Control Works," that the Graph shows an effect of a gun control law and you can see from the graph when the gun control law went into effect. Do you think you could tell when it went into effect without the colors? I don't think you could. All you could see without the colors and the legend is that the number was high on the left side and low on the right side and that, while there were some ups and downs from year to year, there was generally a declining trend throughout the period represented by the graph. There's nothing deceptive about just pointing out that the graph below doesn't really show anything.

[JohnStOnge]

Hey, here's a good one for you. It pertains to our two immediate neighbors.

Canada has a gun ownership rate of 30.8 per 100,000 people while Mexico has a gun ownership rate of 15 per 100,000 people.

Oh but wait: Canada has a gun homicide rate of 0.51 per 100,000 people while Mexico has a gun homicide rate of 9.97 per 100,000 people. So...let's see...Canada has over twice the gun ownership rate of Mexico yet it has about 1/20th the gun homicide rate. CLEARLY higher gun ownership means higher gun homicide rate. Right? CLEARLY the reason a country has a high homicide rate is the the rate at which guns are available.

Where's that "tongue in cheek" emoticon?

[93henfan]

Say what you will, but JSO is kicking ass in this thread.

He might not be Mr. Charisma, but what he does, he does well.

[Pwns]

Your ability to twist statistics to suit your ideology is truly exceptional!

Is that all you've got?

JSO is right, especially on the second graph…if you think gun control reduces gun crimes is a good conclusion from the second graph, I have a theory that the creation of DHS in 2002 also reduced gun crime. Look at the graph for yourself! Also, global temperatures are correlated with the number of abortions in the US.

Speaking of which, I thought Roe vs. Wade was the reason cited by liberals for the fall in crime in the 1990s?