From: [k--m] at [cv.hp.com] (Keith Marchington)
Newsgroups: talk.politics.guns,alt.conspiracy
Subject: Re: University of Maryland Homicide Study
Date: 24 Mar 1995 16:41:11 GMT
Initial Evaluation of University of Maryland/CDC
Study of State Right to Carry Laws(1)
by Paul H. Blackman, Ph.D.
(March 17, 1995)
This study is being rushed into the public debate before
publication in a "peer reviewed" journal(2) in an effort to
influence decision making. The title is misleading: Since
Florida's homicide rate has been falling dramatically since
adopting right-to-carry legislation, the study looks only at three
counties within the state, at one county in Mississippi, and at
three counties in Oregon.(3)
The study is by the same research group which studied a
handgun ban in Washington,(4) D.C., and pretended they had shown a
dramatic decrease in homicide, even as Washington's homicide rate
first inched upward, declined slightly in response to a mandatory-
penalty provision, and finally skyrocketed to set national records
for big-city homicide rates. That study established the
researchers' anti-gun bona fides for the Centers for Disease
Control and Prevention (CDC), which is thus funding this study. It
uses the same discredited(5) methodology employed in their earlier
study, one which is unable to isolate or test various other factors
which might lead to changes in homicide trends (demographic
changes, sentencing and other legislative changes, trends in drug
trafficking, etc.). Having proven to their own and the CDC's
satisfaction that D.C.'s handgun ban reduced homicide even as the
homicide rate tripled, the same authors now assert that right-to-
carry legislation increases homicide even though the states
adopting it have homicide rates which are defying the dramatic
national murder-rate increase.
The only thing that the methodology used in this research can
show is whether there was a temporary or permanent, sharp or
gradual, change in a measured item -- in this case, homicide, as
all other violent crime is ignored -- at a given point in time;
testing different points of time will often lead to various other
time frames similarly indicating changes, whether there was any
explanation for the change or not. The methodology cannot,
however, explain why a change occurred, or which of a variety of
factors explained it; it is pure post hoc ergo propter hoc even
though there may have been nothing happening to prompt the change.
By averaging homicides or homicide rates for a long period of
time -- nearly 15 years for two Florida counties and over that for
the Mississippi and Oregon counties -- prior to adoption of the
law, impacts of the carry reform are disguised by relatively low
homicide rates in the early '70s and the early '80s; worse, the
authors changed the time frame used for Miami -- adopting a 1983
rather than an 1973 starting point. If they used the same time
frame, it would have appeared that Miami's homicide rate had
declined sharply,(6) using the pre-law averaging method they like
to report. They thus excluded some high homicide rate years which
would make the post-law period seem a decline. The use of long
pre-law time periods can obscure high homicide rates in years
immediately before right-to-carry reform. The study used only
three Florida counties, representing one-fourth of the state's
population, one Mississippi county, representing one-tenth of the
population, and three Oregon counties, representing over 40% of the
state's population and where even their study showed a decline in
homicide. The authors noted a 21% homicide rate decline in Florida
by 1992, the end-point for their research.(7)
The research uses National Center for Health Statistics (NCHS)
data on "homicide" instead of FBI data on "murder and nonnegligent
manslaughter." The major difference is that some civilian
justifiable/self-defense homicides are excluded from FBI data but
self-defense and justifiable homicides by civilians are normally
included in NCHS data. In D.C., the difference was enough so that
applying their methodology to FBI data failed to show the pretended
decline the NCHS data showed, hinting that only non-criminal
homicides were prevented by the handgun ban. Similar use of the
wrong data here could disguise more defensive gun homicides.
More importantly, the study utterly ignores the fact that the
law affects only carrying of handguns in public, not possession.
There were no data reported on homicides involving persons with
carry permits -- presumably because there were no such criminal
homicides. The authors hypothesized that criminals might increase
unlawful carrying where law-abiding people are allowed to carry,
but presented no data or citation to any other study to support the
hypothesis. The study also ignored the location of homicides. In
a previous study of Detroit in which the same authors were
involved,(8) the authors at least acknowledged that one would have
to look at circumstances where carrying was involved in order to
evaluate the change -- and in that study nearly half of the
homicides were indoors, where carrying either with or without a
permit was largely irrelevant.
The authors separated gun-related from non-gun-related
homicides, ignoring the distinction between handguns, subject to
liberalized carry laws, and other firearms, and found greater
increase in gun than non-gun homicide, just as their D.C. study
found a greater decrease in gun than non-gun homicides.
Criminologically, firearms crime leads homicide trends, either
upward or downward, since such fluctuations are normally
indications of crime trends among active criminals, who are more
apt to use firearms. Thus, unsurprisingly, the sharp drop in
Florida's homicide rate since adopting its right-to-carry law was
faster for gun- than for non-gun-related homicides.
Disingenuously, the lead author has asserted that a possible
reason for Portland's decline in homicide is that, while adopting
right-to-carry, it also toughened its waiting period provision.
But Prof. McDowall has, using the same methodology, concluded that
"waiting periods have no influence on either gun homicides or gun
suicides."(9)
Incredibly, the authors suggest that laws against carrying in
public are "easy to enforce and they do not inconvenience most gun
owners." Easy enforcement may be relatively true of laws
regulating licensed firearms manufacturers, importers, dealers, and
distributors, and enforcement of carrying in public may be easier
than enforcement of possession bans in the home. But concealed
carry laws are very difficult to enforce without violating Fourth
Amendment protections against unreasonable search and seizure.(10)
In short, the study ignores that lawful carrying is apparently
involved in none of the criminal homicides reported, it uses
unrepresentative and small segments of three states' populations,
it uses carefully selected time frames, it uses a discredited
methodology which makes it impossible to isolate possible causal
factors for trends, it uses data which counts criminal and self-
defense homicides as equally bad, and it sloughs over the fact that
the homicide trend nationally was increasing while dropping in two
of the three states allegedly studied, and rising minimally in
Mississippi.(11)
----------------------------
(1) David McDowall, Colin Loftin, and Brian Wiersema. Easing
Conceal Firearm Laws: Effects on Homicide in Three States.
Violence Research Group Discussion Paper 15. College Park, Md.:
University of Maryland, January 1995.
(2) The Journal of Criminal Law and Criminology is to publish the
study this summer, in a symposium of "gun control" papers edited by
David McDowall, lead author of the paper.
(3) Indeed, they only wanted to look at one county, Multnomah,
containing Portland, but found too few homicides and so expanded to
three counties, all described to the news media as "Portland."
(4) Colin Loftin, et al. Effects of Restrictive Licensing of
Handguns on Homicide and Suicide in the District of Columbia. New
England Journal of Medicine 325:1615-1620 (1991).
(5) Gary Kleck, Chester L. Britt, and David J. Bordua. The Emperor
Has No Clothes: Using Interrupted Time Series Design to Evaluate
Social Policy Impact. Paper delivered at the annual meeting of the
American Society of Criminology, Phoenix, 1993.
(6) "Except in Miami, we studied the period between January 1973
and December 1992 (240 months). Miami homicides increased sharply
in May 1980, following an influx of refugees from Cuba. Miami's
monthly homicide totals appeared to stabilize by late 1982, and we
thus analyzed the period from January 1983 through December 1992
(120 months)."
(7) Through 1993, the handgun-related homicide rate in Florida had
fallen some 29% in Florida while rising 50% nationally.
(8) Patrick O'Carroll, et al. Preventing Homicide: An Evaluation
of the Efficacy of a Detroit Gun Ordinance. American Journal of
Public Health 81:576-581 (1991).
(9) David McDowall. Preventive Effects of Firearm Regulations on
Injury Mortality. Paper delivered at the annual meeting of the
American Society of Criminology, Phoenix, Arizona, 1993.
(10) Paul Bendis and Steven Balkin. A Look at Gun Control
Enforcement. Journal of Police Science and Administration 7:439-
448 (1979); and J. Star. Why the gun law doesn't work. Chicago
27:128-131+ (February 1978).
(11) FBI Uniform Crime Reports. Crime in the United States, 1987,
1989, 1990, and 1993. Washington, D.C.: U.S. Government Printing
Office, 1988, 1990, 1991, 1994.
=+=+=
This information is presented as a service to the Internet community
by the NRA/ILA. Some useful URLs: http://WWW.NRA.Org,
gopher://GOPHER.NRA.Org, wais://WAIS.NRA.Org, ftp://FTP.NRA.Org,
mailto:[L--TP--C] at [NRA.Org] (Send the word help as the body of a message)
Information can also be obtained by connecting to the NRA-ILA GUN-TALK
BBS at (703) 934-2121.
NRA.org is maintained by Mainstream.com mailto:[i--o] at [mainstream.com]