-Caveat Lector-

Doctoring Gun Data

<http://www.stats.org/newsletters/0103/doctorgun.htm>

If guns are a public health problem, it's best to apply public health
standards to gun research

March 2001

Violent crime is a serious issue. Violence involving the misuse of firearms
is all the more serious
because of the lethal nature of the weapons. Given the social cost to the
community of the medical
consequences of gun violence, it is not unreasonable that doctors should
have a role in investigating the
problem. As a result, a considerable body of research on the subject has
been published in medical
journals. However, once proper medical standards are applied in
interpreting such research, the
evidence often seems less convincing, from a medical standpoint, than it
does from a criminological
point of view.
One recent example resulted in the headline "Gun control law helped cut
crime, study says," in the LA
Times. The San Francisco Chronicle declared, "Handgun limit law succeeding"
(both Feb. 28).
These banners ran in response to a report in the Journal of the American
Medical Association
(JAMA) that investigated the effectiveness of a 1991 California law denying
handgun purchases to
violent "misdemeanants" (as the study termed those who had previously been
found guilty of
misdemeanor offenses). The study's authors compared the arrest rate of
those "misdemeanants" who
had their handgun purchases approved before the law to those who were
denied after it was passed.
They concluded that denial was "associated with a specific decrease in risk
of arrest for new gun
and/or violent crimes." But if guns are looked at as a public health
problem, then the standards that
would be applied when looking at risks for disease need to be used. When
that is done, it may turn out
that those headlines were jumping the gun.
The first issue is whether someone who was allowed to purchase a gun is at
a significantly greater risk
of committing a crime compared to someone who was denied purchase. Doctors
measure such things
using "relative risks." These risks are measured by odds ratios, and the
study found an odds ratio of
1.15:1 for the handgun purchasers. This means that for every 100 people not
allowed to buy a gun but
who nevertheless committed a crime, 115 people who had been allowed to do
so broke the law.
Relative risk is a tricky area in medical science, because it is hard to
filter out other factors that may
affect the result. Epidemiologists are therefore loath to consider small
relative risks as indication of any
strong link between the disease and the possible cause.
A relative risk of 1.15 is very small and would therefore be ignored by
experts. The study's authors
did so, to their credit. What they considered significant, however, was the
relative risk for committing
gun and/or violent crime specifically, which they calculated at about
1.3:1. But even a 30 percent
greater risk like this would be regarded as minuscule by epidemiologists.
In fact, they normally require
a difference of 200-300 percent before concluding that they have evidence
of cause and effect (the
relative risk for lung cancer among smokers, for instance, is roughly
300:1). The relative risk at issue
here does not meet the medical standards for real concern about cause and
effect. Therefore the claim
that it is the handgun purchase that makes the difference cannot be made
with any real certainty.
Furthermore, when an epidemiologist looks at the cause of a disease and
discovers a slightly elevated
risk among, say, left-handed people, he does not stop there. He looks
further and asks whether it
affects all left-handed people or predominantly left-handers with bad
eyesight, for example. Most
left-handers could actually be less susceptible to the disease, but the
overall risk for the group as a
whole could be increased by the disproportionate effect in the sub-set of
myopia sufferers. Was there
such a disproportionate effect in this study?
There certainly seems to be. The researchers split the "risky"
population  the "misdemeanants" who'd
gotten guns and then committed crimes  into about 30 sub-groups, depending
on age, number of
prior convictions and number of prior convictions for gun and/or violent
crime. Of these, only three
groups came out with relative risks enabling the researchers to say with
any confidence that they would
be more likely to commit crimes. They were "misdemeanants" arrested for gun
and/or violent crime
who were either aged 30-34 (a relative risk of 1.64), who'd had four or
more prior convictions of any
sort (a relative risk of 1.8), or who'd been convicted for one gun and/or
violent crime previously
(interestingly, more prior convictions for these crimes meant less of a
chance of arrest). All the other
sub- categories possessed either an approximately equal or lesser chance of
arrest.
These findings seem pretty clear. They point to a conclusion that habitual
minor criminals are more
likely than others to commit gun and/or violent crime. It is also important
to bear in mind that "gun
and/or violent crime" need not necessarily involve a firearm  a drunken bar
fight would qualify just as
much as armed robbery. The role of legal gun purchase in leading to such an
eventuality is far from
clear (in epidemiological terms, there is no "biological pathway" here).
 From the evidence the researchers uncovered, then, it might actually be
possible to argue that the
California law forbids many people handgun purchase for no reason. Because
certain categories of
former offenders are shown to be less likely to commit crime when allowed
access to firearms, the
evidence could be interpreted as demonstrating that allowing such people
access would be a useful
public safety measure. But such a conclusion would be just as shaky as the
conclusion that the law
reduced crime.
If the role of guns in violence is to be examined from an epidemiological
standpoint, the rigorous
standards normally used in that field should be used. Drawing conclusions
from the evidence that
cannot stand up to detailed scrutiny does not necessarily help advance the
cause of public safety.

<A HREF="http://www.ctrl.org/">www.ctrl.org</A>
DECLARATION & DISCLAIMER
==========
CTRL is a discussion & informational exchange list. Proselytizing propagandic
screeds are unwelcomed. Substance�not soap-boxing�please!  These are
sordid matters and 'conspiracy theory'�with its many half-truths, mis-
directions and outright frauds�is used politically by different groups with
major and minor effects spread throughout the spectrum of time and thought.
That being said, CTRLgives no endorsement to the validity of posts, and
always suggests to readers; be wary of what you read. CTRL gives no
credence to Holocaust denial and nazi's need not apply.

Let us please be civil and as always, Caveat Lector.
========================================================================
Archives Available at:
http://peach.ease.lsoft.com/archives/ctrl.html
 <A HREF="http://peach.ease.lsoft.com/archives/ctrl.html">Archives of
[EMAIL PROTECTED]</A>

http:[EMAIL PROTECTED]/
 <A HREF="http:[EMAIL PROTECTED]/">ctrl</A>
========================================================================
To subscribe to Conspiracy Theory Research List[CTRL] send email:
SUBSCRIBE CTRL [to:] [EMAIL PROTECTED]

To UNsubscribe to Conspiracy Theory Research List[CTRL] send email:
SIGNOFF CTRL [to:] [EMAIL PROTECTED]

Om

Reply via email to