Understanding the statistics

Clayton Cramer's link to my last post about the gun issue prompted a few additional thoughts.

I'd like to think that sooner or later, people who have common sense would perceive that it's a blatantly racist claim to say that black people are unable to resist being turned into criminals by guns. Instead, the idea is promoted that opposition to gun control is racism! How I wish such demagoguery didn't work, but it does.

I'm also continuing to wonder whether statisticians have (in their various "studies") been lumping law abiding gun owners with criminals who own illegal guns. If they have, aren't the data inherently unreliable? Isn't it like lumping legal drug users in with illegal drug users?

When my mom had cancer, she was prescribed certain narcotics that some street users would probably be willing to kill each other over under the right circumstances. But wouldn't it be laughable for a statistician to make pronouncements about what "drug crimes" my mother (or any other legal drug user) would have been "likely" to commit based on statistics gleaned from the entire morphine-using population?

I think it would. That's because it is manifestly incorrect to extrapolate statistics gleaned from a criminal population and then project those statistics onto a law abiding population.

To do so is wrong logically, and wrong statistically. I'm reminded of one of the criticisms of the Kinsey study (that it applied statistics gleaned from the male prison population to the general male population). This is not to say that the data gleaned from studying criminals is not valuable; only that it should be presented in the proper context.

The oft-quoted Kellerman study has also been criticized for precisely this reason:

....the population is composed of a minority subgroup which has a high risk of homicide, and a relatively high gun ownership rate. This subgroup is composed of 'career criminals', gang members and others who have a repeated history of criminal activity. The majority subgroup has a low risk of homicide and a lower gun ownership rate. This majority is the general law-abiding public.
Nevertheless, the argument is made that guns cause these law-abiding people to commit crime even though the crimes are committed by criminals using guns that were already illegal. (In Philadelphia, 80-85 percent of shootings occur when criminals shoot other criminals with already-illegal guns.)

It's one thing to make this argument, but not with data claimed drawn from a non-representative sample.

Of course, it has to be remembered that there is a growing movement in favor of abolishing prison -- even for violent crime. While that's not entirely mainstream, like a lot of activism, it manages to trickles down in the form of projects like this, which lobby for ever-lighter sentences, even foir serious crimes. Editorial-style news "stories" routinely preach about the need for "forgiveness." The result is that criminals are increasingly not being locked up for violent crimes.

It's easy to see why those who advocate freedom for violent criminals would want gun control. They are intelligent enough to realize that they too have to live in the world they want to create. So, in a certain sense, their use of misleading statistics becomes understandable.

(Not excusable, though.)

posted by Eric on 04.15.07 at 10:00 PM





TrackBack

TrackBack URL for this entry:
http://classicalvalues.com/cgi-bin/pings.cgi/4880






Comments

And if imprisonment is eliminated, how long before the death rate among suspects prior to or during apprehension skyrockets?

When there is no way to keep a violent person segregated from society, the ultimate segregation will be enforced.

Alan Kellogg   ·  April 16, 2007 02:32 AM

Post a comment

You may use basic HTML for formatting.





Remember Me?

(you may use HTML tags for style)



April 2007
Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30          

ANCIENT (AND MODERN)
WORLD-WIDE CALENDAR


Search the Site


E-mail




Classics To Go

Classical Values PDA Link



Archives




Recent Entries



Links



Site Credits