Episode 2: Facts about gun violence

If you haven't read the Introduction or Episode 1 you should do so before reading below.
_______________________________________________

"There are three kinds of lies: lies, damned lies, and statistics."

- No one really knows who wrote this though some say it was Mark Twain

I'm going to try and keep the snarky commentary to a minimum and deal in facts. Regular readers know this will be a struggle.

What we know about gun violence : we can count it fairly accurately ^[1]

What we know about the causes of gun violence: virtually nothing

According to the Center for Disease Control (CDC), in 2011, 32,163 people lost their lives from firearms. Of that total, 19,766 lost their lives by suicide by firearm. It's a great irony that in a country where the doctor cannot assist suicide a gunsmith can.^[2] 11,101 people died from homicide by firearm.^[3] The trend in firearm deaths is increasing.

To put that total firearm death number in perspective, the CDC reports that 34,677 died in automobile accidents in 2011. That number has be declining fairly steadily since 1979. The causes of motor vehicle deaths are well understood as are the actions required to make automobiles safer.

According to the CDC, for most gun deaths we do not know whether they were committed by a long gun (rifle or shotgun of some type) or a hand gun. Of the 11,078 firearm deaths in 2010, we know that 899 were from handguns and 576 were from long guns. For the vast majority of firearm deaths the type of weapon is not recorded in the CDC data.

When you look at the FBI Uniform Crime Report (UCR) data for 2010 the death data is incomplete - Florida and Illinois are not included - though it reports more deaths (12,996). Also, the UCR provides data about what types of firearms are used (24% of the firearms type is unknown vs the 87% unknown in the CDC).  The FBI reports 731 deaths from long guns and, of those, 358 from rifles.

This brings us to two important points. First, different data sources provide different data when measuring the exact same thing. The CDC data comes from Death Certificates. UCR data is self reported by law enforcement agencies in the states. Second, the data does not tell a story. The data is used in storytelling and the same data can be used to tell different stories.

Both sides of the gun control discussion attempt to paint the limited data in the best light. Below I will pitch the above data (or data taken from the same source)^[4] in several different ways. You guess which "side" of the gun debate I'm on in each one.

1. Only 3% of gun deaths in 2011 were known to be caused by long guns and not all of those were assault weapons. An assault weapons ban wouldn't matter.
2. According to the CDC, where the gun type was known, 39% of gun deaths were caused by long guns. An assault weapons ban is critical to reducing almost half of all gun related deaths.
3. Only 1/3 of gun deaths are homicides. 90% of those are just gang bangers killing each other. Banning legal weapons won't matter. ^[5]

The astute reader will observe that each of the above statements contains fact followed by a causal statement that is in no way supported by the data. More on the causal point in a moment. I've seen variations of those claims in many places (one of the NRA's is at the bottom of this page). It's all factually correct and totally bogus at the same time.

Another thing I find interesting about the use of weapon-type data in statements about gun violence is what I'll refer to as "peaceful equivalency". The statements assume that the peaceful equivalent use of the weapons are the same. They are not. Clearly, blunt objects, feet, hands, and knives have a different peaceful equivalent use than firearms. And some firearms, say the Winchester Model 70 have a different peaceful equivalent use than the M-4.

A final observation: According to the FBI only 358 people were murdered with rifles. What is the acceptable threshold for the number of people murdered by rifles? By handguns? Spears? When it's data it's very easy to forget that behind the data are dead humans.

So much for data. let's talk about the causal factors.

[THIS SPACE INTENTIONALLY LEFT BLANK]^[6]

We know precious little about the causes of gun violence. That's not because we don't know how to determine the causes. It's because there's little independent research conducted and, notably, little government research.

The reason there's no government research related to the causal factors of injuries from gun violence is because the gun lobby asked for it to be that way. No lie.

Since 1996, the gun lobby has successfully lobbied Congress to specifically prohibit the Center for Disease Control from recommending any form of gun control even if their research demonstrates that such control would reduce deaths from gun injury. The CDC is the government agency specifically tasked with studying and recommending policy for reducing death from injury.

So, one can find CDC recommendations for preventing injuries to children who ride in cars but not for preventing injuries to children who live in homes with guns. They're not doing the research because they're prohibited by law from making those recommendations. So, they focus their research efforts elsewhere which was clearly the intent of the law.^[7]

I understand why it is so, I just think it's stupid for you and I to allow it.

The situation where the causes of gun violence are not known is an unacceptable situation.

This is a place where we need to do more.

Much more.^[8]

_______________________________________________

A CASE STUDY: NRA GRAPHIC RELATED TO SEMI-AUTOMATIC FIREARMS

EVEN THOUGH THERE ARE NUMBERS AND FACTS BELOW, THE MESSAGE WITHHOLDS IMPORTANT INFORMATION AND MAKES UP SOME STUFF TO MAKE IT MORE COMPELLING. IT IS PROVIDED AS AN EXAMPLE ONLY. ^[9]

This graphic appeared on the NRA's Facebook Page on January 7th, 2013.  

The data presented below is from the Preliminary FBI Uniform Crime Report for 2011. The graphic doesn't mention that for 20% of 2011 firearms deaths (1,684 deaths) reported in the UCR we don't know what type of firearm used. The UCR does not report on "Knives" it reports "Knives or Cutting Instruments", The UCR does not report on "Clubs or Hammers". This refers to deaths caused by Blunt Objects. So far, the omissions and changes to data do not appear to be overly misleading; the data is restated in a slightly misleading way to make the message more effective.

Based upon my research, two elements are pure fabrications: 1) "Facts Gun Control Advocates Don't Want You to Know" and 2) "But Obama wants to ban semi-automatic rifles?"

I could not find any evidence that "Gun Control Advocates" were attempting to suppress the FBI UCR data.^[10] While effective at creating an us vs. them mentality, this information is clearly "Facts the NRA wants you to know". At the time this graphic was published the President had made no proposals about a ban on semi-automatic rifles. About the closest I could find to verify this claim was the mention of a renewed Assault Weapons Ban though such a ban would not be a "ban on semi-automatic rifles" it would be a "ban on some semi-automatic rifles". And again, no such proposal had been made at the time of the publication of this graphic. [Update: Two years later still no proposal despite dire warnings by the NRA. Update: Almost 4 years later, still no proposal. NRA still shrieking.] The language is strikingly familiar to language used both before and after the 2008 election and may have been used to specifically strike that cord. But that's just my opinion.

When you see facts such as these, from any source, do some legwork to understand them before you get too worked up. Sadly, social media has made it all too easy for fabrications to be distributed rapidly as fact.

Also, if you have found graphics from those supporting gun control (the other side of this discussion) that have similarly misleading information please put a link in the comments below. I haven't been able to find one to debunk but I'm sure they're out there.

Read the next post in the series


Notes:
1 - The 2011 CDC is here: http://www.cdc.gov/nchs/data/nvsr/nvsr61/nvsr61_06.pdf Hoyert, Donna L. and Jiaquan Xu. 2012. ‘Deaths: Preliminary Data for 2011 - Selected Causes.’ National Vital Statistics Reports (NVSS); Vol 61, No. 6, pp.40-42. Hyattsville, MD: US Department of Health and Human Services, Centers for Disease Control Prevention, Division of Vital Statistics. 10 October. I have also use the data from the preliminary 2011 FBI Uniform Crime Report
2 - Damn, only a few sentences in and snark has been exposed. I will try harder to resist.
3 - The difference in the sums of the amounts is comprised of unintentional deaths and indeterminate deaths.
4 - You can get your own statistics at: http://wonder.cdc.gov
5 - This was a comment to a NY Times article. That 90% claim is amusing. It's not data that's available anywhere (at least not that I've been able to find after searching)
6 - I would have put the sound of crickets here but I hate websites with inane sounds. Yes, this is snarky. Snarky but poignant.
7 - In fact, a task force convened by the CDC to study previous studies regarding the various aspects of firearm violence couldn't find enough studies to study the study data.
8 - If you really want to be shocked, check out restrictions that have been placed on the ATF by the gun lobby. Basically back in 2007 Representative Todd Tiarht wrote some rules, allowed the NRA to mark them up and them put them into an amendment attached to another bill that limits the ATF's ability to collect and share information about gun registrations, limited inspections of gun shops, eliminated the requirement for gun shops to keep an inventory, among other things. Now he's complaining that the problem with guns in this country is that the ATF doesn't enforce the laws.
9 - I chose the NRA example for two reasons: 1) They've got a big megaphone and they need to be held accountable, and 2) the pro-control side seems to do it less frequently. If you've got a good piece of pro-control skewing of facts, I'd like to see it so I can do a similar analysis.
10 - Ironic given that pro-gun lobby is trying to actively suppress firearms research. Damn, another snarky comment.