Washington CeaseFire (or maybe CeaseFire Washington, I’ve seen it both ways) has just spent $50,000 buying bus ads touting various anti-gun messages and, as we’ve mentioned, they’re oh so full of fail. One of their most heinous efforts, though, regurgitates the classic Kellerman study figure that a gun in the home is 43 times more likely – oops, sorry he has since recalculated the figure to be only 22 times more likely – to be used to kill someone in the home than to kill an intruder. But with the “correction” RF knows me too well. There’s no way I can let them get away with that bogus number . . .
First let’s put the 43 times number to bed. GunCite.com has a magnificent deconstruction here; basically they use Dr. Kellerman’s own numbers and methodology to show that in homes without a firearm you are ninety-nine times more likely to suffer a non-firearm related killing than you are to kill an intruder without a firearm . . .
But, back to the problems with the new and improved 22 times number.
#1) Dr. Kellerman initially failed to state how he determined that the gun used belonged in the home or had been brought into it by someone else. Four years later he wrote a letter to the NEJM with a correction stating that in follow-up interviews it was determined that in 63% of the cases the gun was organic to the household. So, 0.63 times 22 means you’re only 13.6 times more likely to be ki . . . etc. (I think I’ll start using the acronym TMLTBKBAGIYHTTKAI . . . or, better yet we’ll just use TML (times more likely). So, that has reduced the 22 TML figure down to 13.6.
#2) Dr. Kellerman failed to account for other risk factors like drug use, criminals or criminal activity in the home. According to his study’s Table 3, 53% of the case study households contained at least one adult who had been arrested. So 0.47 times 13.64 leaves 6.4 TML.
#3) Most of the killings didn’t actually occur “in the home,” Oops! According to Kellerman’s own figures, only 23.9% of the homicides happened in the home of the victim. So, 0.239 times 6.4 gives us 1.5321812 so let’s call it 1.5 TML, shall we?
#4) Kellerman’s study includes suicides committed with guns. The problem with that is he did no research to determine if the “victim” acquired the weapon solely in order to commit suicide or if they used it as a “method of opportunity.” And despite the antis claims to the contrary, numerous studies have shown that suicide rates are independent of method. In other words, taking away guns may reduce the gun suicide rate, but non-gun suicides will increase enough to offset this.
#5) And the biggest problem of all: Dr. Kellerman seems to believe that killing someone is the only way to use a gun defensively (one wonders, then, what he thinks of police departments who routinely arrest people instead of killing them). But according to Dr. Kleck’s study Targeting Guns (as cited on page 19 of Gun Facts ver. 6.0) in less than 0.1% of DGUs is the attacker killed. Indeed, in 92% of DGUs the victim merely brandishes the weapon or fires a warning shot.
So, putting all this together, what do we have (besides the fact that 22 TML is a completely bogus number)? After my swipe at number crunching we have reduced TML to 1.5, but what does that mean in real life?
Well, according to the CDC‘s numbers, over the 11 years from 1999 to 2009 we averaged 11,800 firearm-related murders annually. If we carry that number through 2010 and look at the census numbers, that means the your chance of being murdered with a gun is 11,800/309,000,000 or 0.0038%. So even if we accept Dr. Kellerman’s premise that guns are only useful when they kill someone, having a gun in your home raises your odds of being killed from 0.0038% to 0.0059%.
But again, what does that mean? It means that keeping a gun in the home raises your chance of being killed with a gun from 1 in 26,186 to 1 in 17,004 whereas your chance of dying from something other than gunshot? 1 in 205.
I think I’ll take my chances with the guns.