Statistics and the Gun Control argument

[SolaScriptura139]

I have noticed that a lot of anti's use statistics in their arguments for gun control. I have also noticed that many people in my classes here use statistics as the basis for their entire arguments. Now I know that statistics are patchy at best, but can anyone provide a good argument as to why statistics are a load of crap, especially as the basis of your whole argument?

 

[SuperNaut]

This is one you are going to have to be very careful with. It is far, far more compelling to speak with data than to just speak. Numbers don't lie, but the people massaging them do. The biggest criticisms of statistics are the criteria utilized for the elimination of outliers, the original questions used to gather the data, and control and sample size.

Remember, everyone has a bias but bias backed up by numbers means something. Bias backed up by flapping gums means nothing. Even if you disagree with the study at least a study was performed. If you disagree and have nothing to back it up you look like a fool.

Tread carefully.

 

[Erebus]

If you fire twice at a target and miss 4' to the left and 4' to the right statistically you hit it. Of coarse it's completely inaccurate but it is mathmatically correct. The average of 4' to the left anf 4' to the right is a bullseye.

You can use statistics as a basis for complete bull****.

 

[mec]

Nothing that will convince a lover of statistics. The old progression goes:
"Lies, Damn Lies, Statistics."

Statisticians argue among themselves about sample sizes and relevance. John Lott took his data from a nation wide county by county data base and dismisses the anti gun Kellerman study because it targeted a single region. Bellisailes apparently made up all of his statistics and the dog ate all his notes. The Bradys and the Violence Policy Center like to claim a huge number of gun related child deaths but include stats on children up to the age of 24 or so.
You can come up with statistics that prove smokers make more productive employees because they take fewer sick days than non smokers.
the only safe bet is to distrust anybody who slings stats even if they agree with you.

 

[Jorg Nysgerrig]

can anyone provide a good argument as to why statistics are a load of crap

Statistics themselves don't mean much. It is all in the interpretation. Statistics can be very valuable and convincing. However, many studies are flawed, resulting in skewed statistics.

However, the classic book on the subject is this: http://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728/sr=8-1/
Any library should have it.

If you scroll down to the "Criticism" section of this article you will find a bit of basic information.

Here's a bit more on statistics.

You won't get far criticizing statistics as a whole or bashing someone for basing their argument on statistics. That said, most people who cite statistics are using them from secondary or even tertiary sources. Your best bet to find flaw with someone's argument that is based entirely on statistics is to find the statistics they are using and critically evaluate them. Are this misrepresenting the statistics? Is their analysis flawed? Has the study been later disproven? Are there other studies that show drastically different results?

Well done statistical evidence is hard to refute. Fortunately for you, outside of serious scholarly journals, the way most statistical evidence is used/presented is not well done.

Besides, providing your own overwhelmingly compelling evidence is always better than merely showing your opponent is wrong.

 

[ETXhiker]

A recent scientific study has shown that 87.42% of all statistics are made up on the spot.

 

[Keith Wheeler]

Properly used statistics are a very powerful tool. Just attacking the statistics won't get you very far, unless you can find error in their math.

The comments about "average" -- that's not statistics, that's basic math. The "average" voltage in your house is zero. It's a sine wave, and the "average" of a sine wave is zero. That doesn't mean that math/statistics are wrong, it means you don't have enough of an education in math to use the right tool for the job. With sinusoidal AC voltage we use "root-mean-square", average is meaningless.

The right tool for the job. Plus a lot of people here are using "statistics" to mean "gathered infromation". Statistics is not demographics. The murder rate is that, the documented historical rate that murders occur in our population.

Now looking at murder rate vs some other factor, say number of firearms or perhaps number of CCW permits, that can be done with the tools of statistics, BUT (and if this sentence means nothing to you, perhaps actually studying statistics would help) correlation does not equal causality. One more time, just because something is statistically correlated, does not mean you've found the (or a) cause for that something.

One thing that statistics as a tool help you do is rule out a hypothesis. Let's say I think that gun crimes are committed by people who watch Spongebob Squarepants. I interview those who have committed gun crime, and find out how many of them watch Spongebob. I'll probably be able to easily demonstrate mathematically that there is no correlation. But, what if there is? That still doesn't prove my hypothesis, and there are a number of alternative concepts that could explain the correlation.

What you need to focus on attacking would be the premise and conclusions.

It's like the Brady Bunch information on the number of "children" killed annually by firearms. Attack their premise -- their definition of "children". (Which I think includes those up to age 25?) Attack their conclusion "more gun laws would mean fewer gun crimes". Attacking their numbers, unless you have enough statistics classes under your belt to prove their math is in error, is grasping at straws.

 

[Juna]

Statistics are often grossly mis-used and mis-interpreted by politicians for their own gain. You need to know a bit about statistics and math to refute the "statistics" they spout off. You need to know the details of the study, any inherent bias, sample size, controls, etc.

One more time, just because something is statistically correlated, does not mean you've found the (or a) cause for that something.

This is very true and frequently ignored by politicians spouting off "statistics". You could do a study showing that people who carry a lighter in their pocket are more likely to get lung cancer, but that doesn't mean the lighter causes lung cancer--it's just correlated in that people who carry a lighter are more likely to smoke. Again, a relationship between two variables does not mean that one causes the other. That's why when politicians spout off stats about guns and violence, it ignores the fact that the gun isn't what's causing the violence--it's the person holding it who causes the violence.

Banning or limiting guns to reduce violence is like banning cigarette lighters to stop lung cancer. People can use any source of heat to light their cigarettes, but it's their behavior (i.e. smoking) that causes the lung cancer. And, people can commit acts of violence with a knife, baseball bat, or their bare hands, but it's the behavior, not the instrument, that is responsible for the violence.

 

[Molon Labe]

I never use statistics when arguing about the right to keep and bear arms. Even if they're accurate. Why? Because it implies that, if the statistics proved the opposite, I would not have a right to keep and bear arms.

So my rule is to never never never bring up statistics when arguing about the right to keep and bear arms.

 

[Molon Labe]

If you are arguing with a very sharp anti-gunner, they will eat you alive if you bring up statistics. Example:

Pro-gunner: "Yes, people should have the right to be armed. According to research, guns in the hands of lawful citizens actually reduce crime."

Anti-gunner: "Hmm. So does that mean people would not have the right to be armed if research proved guns in the hands of lawful citizens increased crime??"

Pro-gunner: "Um, well, um, no, um..."

 

[TX1911fan]

I echo the cautions above in using statistics to prove the RKBA. They don't matter. Also, I would advise that, to the greatest extent possible, you obtain the statistics to be used by the other side in advance. Then, you can verify the source, and also have time to digest what is being said. The source materials for any statistics are vital to understanding the data, and without them, you are at the whim of the interpreter. The classic case is the use of up to 24 year olds in the category of "children" and also the use of suicides and deaths by LEOs in the total number of gun deaths. If an anti simply states that the US had x number of gun deaths last year, and how terrible they all are, and so we need gun control, knowing that 1/3 of those gun deaths were the result of LEOs, and 1/2 of the remaining amount were suicides really takes the wind out of those sails.

All that said, AVOID at all costs arguing that the statistics are actually in favor of RKBA, because you give the impression that you agree that if we had bad statistics you would support gun control.

 

[romma]

Figures Lie and Liars Figure

 

[Carl N. Brown]

The Numerical Reliability of Econometric Software,
McCullough and Vinod, Journal of Economic Literature,
Vol XXXVII, Issue 2, June Quarter 1999.

The Vancouver Stock Exchange started an index at 1,000.000
which was recalculated after each transaction to the nearest
.0001 then truncated to report only the last three decimal
places. Within a few months the stock index had dropped to
520.000 due to constant rounding down. When properly
calculated, the actual stock index was found to be 1098.892
(Toronto Star, 29 Nov 83).

 

[Declaration Day]

I never use statistics when arguing about the right to keep and bear arms. Even if they're accurate. Why? Because it implies that, if the statistics proved the opposite, I would not have a right to keep and bear arms.

Great point, but there's a more important reason why statistics don't matter.

The right to keep and bear arms is a RIGHT, and is (theoretically) unaffected by the irresponsible or criminal acts of others.

 

[Carl N. Brown]

How do statistics prove that gun control would work
against criminals, anyway?

The argument that there are 2.5 million defensive
gun uses versus 430,000 felonies with firerarms is used
against gun control.

Invert the stats: 2.5 million firearms crimes versus 430,000
defensive gun uses would not prove the utility of gun control.

Gun control is legal restrictions on legal sales, purchase and
ownership, and largely affects the law abiding. The impact on
criminals is much less. So even if there were more gun crimes
than gun self defense, the gun laws would impact gun self
defense more than it would impact gun crime. The stats
prove nothing.

What it boils down to is that gun control is about symbolism,
that banning things will change human behavior. Its voodoo.
And voodoo cannot be quantified.

 

[Alex45ACP]

The only statistic you need to know is that the number of people who have been disarmed and massacred by their own governments is exponentially higher than the number who have been killed by common criminals armed with guns.

Even if stricter laws actually did reduce the crime rate, and it could be statistically proven by the antis, (which it can not) I'd still be against them for this reason. People who get all into the crime statistics and whatnot are missing the bigger picture.

 

[StuckInMA]

I've found that the best way to combat statistics is to research and point out the flaws in the very statistics put before you. Especially if you can get the stats from sources on the same side of the argument. Here's an example.

From Mayors Against Illegal Guns
Whereas: 30,000 Americans across the country are killed every year as a result of gun violence, destroying families and communities in big cities and small towns....

From The Brady Campaign
In 2002, 30,242 people in the United States died from firearm-related deaths.
11,829 - of those were murdered.
17,108 - were suicides.
762 - were accidents.
243 - the intent was unknown.

 

[ConstitutionCowboy]

It Isn't About Statistics

It's about the Right to Keep and Bear Arms and honoring the Supreme Law of the Land. It's about defending the nation from invasion from without. It's about defending the nation from subjugation from within. It's about self defense and the defense of your family from crime. It's about being able to feed yourself and family. Bottom line, it is about us being the only truly free people left on earth.

Woody

Look at your rights and freedoms as what would be required to survive and be free as if there were no government. If that doesn't convince you to take a stand and protect your inalienable rights and freedoms, nothing will. If that doesn't convince you to maintain your personal sovereignty, you are already someone else's subject. If you don't secure your rights and freedoms to maintain your personal sovereignty now, it'll be too late to come to me for help when they come for you. I will already be dead because I had to stand alone. B.E.Wood

 

[Art Eatman]

A problem with the use of statistics is whether or not the numbers are appropriate to a particular argument.

First example: It is fact that firearms are used in approximately thirty-thousand fatalities in the U.S., each year. What's not said is that about one-half are actual homicides (including justifiable homicides); the remainder are suicides. Further, it is well-known as fact that almost all suicides will use another method if some preferred method is not available or doesn't work.

Second example: Numbers for accidental deaths via firearms for "children" are often given. What's not said is that the age of these so-called children extends to late teens and (less often) into early twenties. Most of us regard "children" as 14 and under, for whom the accidental death toll from accident is around 100 per year.

Art

 

[antsi]

-----quote----
A problem with the use of statistics is whether or not the numbers are appropriate to a particular argument.
---------------

This is key. The numbers may be correct, but the question with statistics is whether or not the numbers really support the point being made.

In order to make sense of statistics, you have to find out how the numbers were counted and what they're being compared to.

Art's example is a good one: "child gun deaths." Ok, how was 'child' defined? How is 'gun death' defined?

"Child gun death" sounds like a really bad thing that we'd want to prevent, right? Well, what if that "child gun death" was actually a woman defending herself against a 19 year old rapist?

What if someone is saying we need a new law governing storage of guns in homes with kids, because there were XX number of "child gun deaths" last year? By looking at where the numbers come from, you might find out, for instance, that the vast majority of those "child gun deaths" were actually drug gang members in their late teens killing each other off. Not sure how the new gun safety lock law is going to address that.

People throw around figures like "30,000 gun homicides last year." One appropriate response is "compared to what?" If there are 2.5 million defensive uses of firearms per year in the US (as some studies suggest), then it sure looks to me like the benefits of self-defense with guns probably far outweigh the criminal use of guns.

Also, it is not at all clear that all 30,000 of those people would still be alive, even if guns were magically wiped away. About half of them are suicides... what percentage of them would still manage to have killed themselves some other way? Some of those were legitimate self-defense killings... how many innocent lives were thus saved? Even for the murders, how many of the murderers would have figured out some other way to kill their victims?

You don't need a statistics class to interpret numbers and debunk faulty statistical arguments. It is more an exercise in logic than an exercise in math. Misleading statistical arguments rarely hinge on some abstruse point of statistical calculation or theory. They are far more often based on flaws of logic.

 

[Carl N. Brown]

Criminologist Marvin E. Wolfgang once wrote "I hate guns, nasty
brutish things made only to kill people."

He also studied 588 homicides in Philadelphia. Based on the
circumstances of the crimes, he concluded that the presence
or absence of a gun was seldomn a causal or motivating factor,
and that few homicides due to shooting could be avoided
merely if the gun was removed from the equation.

The removal of guns from society might not prevent a significant
number of the quoted 30,000 firearms deaths--murder, suicide or
accident. But a legal ban would probably inhibit the defensive use
of guns by the law-abiding.

Studies by James Wright and Peter Rossi show that criminals fear
being shot by a victim more than they fear the police. Chicago
or New York City or WashingtonDC style "gun control"--de facto
prohibition--removes a significant deterrent to crime and costs
more than it benefits.

 

[Carl N. Brown]

Accidental Child Gun Deaths

For children 0-4 more are killed by drowning in 5 gallon mop buckets
than by firearms accidents.
For children 5-14 more are killed by bicycle accidents than by
firearms accidents.

As for hunting accidents, what guarantee is there that they
would not choose a sport of equal or greater danger?
Check out accidental deaths by swimming versus hunting.

 

[ConstitutionCowboy]

Another point

These statistics -a goodly portion of them, at least - are crimes committed by people who have been through the court system and released back into society. Maybe that is one reason the anti-gun-rights crowd won't even acknowledge who committed these crimes, and that if these criminals were kept in prison, the numbers would be significantly lower. It keeps their numbers up!

Woody

"The Right of the People to move about freely in a secure manner shall not be infringed. Any manner of self defense shall not be restricted, regardless of the mode of travel or where you stop along the way, as it is the right so enumerated at both the beginning and end of any journey." B.E.Wood

 

[svtruth]

What Art said, and

very few people seem to have the moral courage or intellectual integrity to report conclusions based on their analysis. They, instead, have a point of view and martial statistics to support it.
Annual US gun deaths look much different if sliced by CCW, by illegally possessed guns, by criminal on criminal violence, domestic violence etc. A NYC govt spokesman was quoted in ther NYT last summer pointing out that almost all gun violence in NYC was criminal on criminal, "if you are not living a life of crime you are very unlikely to be shot."

 

[Art Eatman]

Then there's the "statstics justification": "X number of children die because of guns; guns should be banned."

Refutation: "Okay, if that's grounds for banning, 2X number of children die from school athletics. Therefore school athletics should be banned."

And, of course, cars and swimming pools and...

Art