Flawed research

NBC is promoting a series of misconceptions and misleading anti-gun studies in it's "Health" section. Although they try to avoid presenting details, it's clear to anyone with knowledge of existing research that they are stretching quite a ways to reach conclusions suitable for propaganda use. I'll go through their story and show what they are hiding below the fold.
Having guns in the home triples the risk of suicide and doubles the risk of homicide, researchers reported on Monday.

Their review of 15 studies considered high quality confirms a clear association between gun ownership and violent death in the United States, where more than a third of the population owns firearms.

Their thesis, then, is that owning guns causes violent death. The problem is that the details don't bear that out.

First, let's remember one of the most basic lessons of statistics: correlation is not causation. Even if this research can establish a correlation between gun ownership and suicide or homicide, that doesn't demonstrate that the gun is responsible for any of the outcomes.

They even quote one of the researchers bringing up this point, without noting how it destroys the premise:
"Why having a gun in the home would substantially increase the risk for being murdered by a person who is not a family member is uncertain, and Anglemyer and colleagues do not provide an explanation,” Hemenway wrote.

To make the principle clear, ask yourself how a owning a gun could possibly make someone more likely to be killed by someone else in their own home. There are only a couple possibilities, and the one that stands out is an obvious cause of causality operating in the reverse direction than that assumed by the researchers: people who are at increased risk of being murdered acquire guns to protect themselves from that risk.

But if owning a gun protected you from being a victim of homicide, why the correlation? Simple: most of the studies along these lines follow an established pattern based on an original study by Kellerman, which produced a long-discredited claim that owning a gun made you 43 times more likely to be killed. That study was a cause-of-death study.

In other words, every case in their study started with a dead body, which was then investigated for cause of death. If a gun owner used his firearm to repel a violent criminal attacker, without killing the attacker, it was literally impossible for their experience to show up in the study, because no one had died. Thus, people at high risk of homicide bought guns and only showed up in the study if they actually died despite buying a gun... hence the correlation.

"Firearms cause an estimated 31,000 deaths annually in the United States," they wrote in their report, published in the Annals of Internal Medicine. "Data from the 16-state National Violent Death Reporting System indicate that 51.8 percent of deaths from suicide in 2009 were firearm-related; among homicide victims, 66.5 percent were firearm-related."

Firearms cannot cause anything. They are inanimate objects.

Left unstated in the NBC story, but mentioned in the story they link to, is that murders with firearms are going down rather than up, continuing a 20-year trend in that direction. Suicides with firearms are up, but so are overall suicides, a trend that is driven by things other than firearms availability.

In short, this isn't new research, it's a warmed-up serving of decades-old leftovers in a relabeled package. The stink coming from this so-called "research" reminds me of lutefisk.

If you're still not convinced, read this detailed explanation of the problems in the research. These problems apply to pretty much all of the anti-gun studies out there. Competent studies find no effect for gun control laws, and small but real reductions in crime (More Guns, Less Crime: Understanding Crime and Gun Control Laws, Third Edition (Studies in Law and Economics)) for allowing ordinary people to carry concealed firearms. The most strongly positive effect is found stopping multiple-victim public shootings.

This entry was published Tue Jan 28 07:27:32 CST 2014 by TriggerFinger and last updated 2014-01-28 07:27:32.0. [Tweet]

comments powered by Disqus

Subscribe to Atom Feed

I am not a lawyer, and nothing on this site should be taken as legal advice.

This site is run on custom blog software and is being actively developed. Please be forgiving of errors.

This website is an Amazon affiliate and will receive financial compensation for products purchased from Amazon through links on this site.