Debunking John Lott (was: Re: GUNS: Why here?)

From: Joe Dees (joedees@addall.com)
Date: Mon Sep 25 2000 - 12:55:49 MDT


('binary' encoding is not supported, stored as-is) >Date: Mon, 25 Sep 2000 10:30:53 -0400
>From: "Michael S. Lorrey" <retroman@turbont.net>
>To: extropians@extropy.org
>Subject: Re: GUNS: Why here?
>Reply-To: extropians@extropy.org
>
>Jason Joel Thompson wrote:
>>
>> ----- Original Message -----
>> From: "Samantha Atkins" <samantha@objectent.com>
>>
>> > Statistically many crimes are stopped by an armed
>> > would be victim or witness with no shot fired.
>>
>> Well, since you brought it up, what are the statistics?
>
>Gary Kleck, a rather well respected crime researcher, conducted a study on this
>and found an annual rate of 2.5 million Defensive Gun Uses per year. Even
>anti-gunners admit that there are a minimum of 60,000 defensive gun uses (versus
>9,000 innocent people killed by others with guns, out of a total 32,000 gun
>deaths per year), derived from strictly defined criteria in the FBI database
>(part of which includes not counting DGUs which were found to be illegal by
>defective and onerous state laws and prosecutors). A study done of several dozen
>major studies in this area came up with an average estimate of around 800,000
>DGUs per year. While this is obviously a very broad range, given the
>partisanship of this issue, its to be expected, although Kleck is considered by
>most on both sides of the issue to be credible.
>
>A purely economic study of gun crime vs. gun defense found that so long as
>civilians kill 2,000 criminals a year with private guns defensively, there is a
>break even on the economic impact. If civilians kill 3,000 criminals a year
>defensively, there is somewhere between $1 billion and $30 billion net benefit
>to the economy. This includes not only the fact that les prison space would need
>to be built, but lower insurance rates, lower medical costs, less need for
>police, less need for security systems, etc...
>
>Prof. John Lott, in his book _More Guns, Less Crime_ found a significant impact
>on crime when civilians are more freely allowed to carry concealed weapons,
>while at the same time found almost no impact from the death penalty as it is
>currently administered (he studied ALL FBI crime stats from 1979 to 1995, the
>first and ONLY researcher to do so). The largest anti-gun study of concealed
>weapon use consisted of only five cherry picked counties that happened to have
>statistics which supported the anti-gun position on concealed carry. John Lott
>mentioned this in his book. The author of that study declared Lott's own study
>'flawed' after refusing to even read it.
>
>Mike Lorrey
>
John Lott and his book MORE GUNS, LESS CRIME has become, for the progun faction, the garlic they wave in the face of the dreaded gun control vampires, as if its flawless perfection would exorcise all demons of competing views of safety and security. Not so. Many flaws in methodology and various and sundry unwarranted assumptions and unconfirmed (and unlikely) cause attributions plague the work.

To get an academic view of the work of this Bernard nathanson of the progun lobby, I post the critique of it found on:

http://www.handguncontrol.org/research/progun/lott.asp

                    In 1996, John R. Lott, an Olin Fellow at the University of Chicago School of Law, and David B. Mustard published a
                    controversial study which purportedly showed that states that loosened their carry concealed weapons (CCW) laws experienced a
                    reduction in certain types of crime specifically because they made it easier for citizens to carry concealed weapons. In 1998, Lott
                    published More Guns, Less Crime, a book based on his 1996 study.

                    It is important to note before detailing the major points of criticisms of Lott's book and study that he is unabashedly libertarian and
                    may not be an unbiased source for research on gun-related issues. He has long been a proponent of the "Chicago School" theories
                    of law and economics on subjects ranging from crime to the environment. In the past, he has argued that the benefit of a crime to a
                    criminal can outweigh the harm that a crime inflicts on a society and, according to him, "the worst thing people can expect from
                    dioxin is a bad rash." Two days after the Jonesboro schoolyard shootings, Lott called for arming teachers as the solution to
                    preventing such tragedies. Most recently, Lott has argued that the hiring of more women and minorities in law enforcement has
                    actually increased crime rates.

                    Both Lott's book and his study have been reviewed by academics from a wide range of disciplines from criminology to public
                    health. Many of these scholars found serious, fundamental flaws in Lott's methodology and found his claims to be unsubstantiated.
                    These researchers include Jens Ludwig at Georgetown University; Daniel Black of the University of Kentucky and Daniel Nagin at
                    Carnegie Mellon University; Stephen Teret, Jon Vernick and Daniel Webster, all of Johns Hopkins University; Arthur Kellermann
                    at Emory University; and Douglas Weil at the Center to Prevent Handgun Violence.

                    Unfortunately, while flaws in his research have been widely documented in scientific literature -- and his findings dismissed by
                    numerous, prominent researchers -- the gun lobby has successfully used Dr. Lott's flawed conclusions to persuade several state
                    legislatures to loosen CCW restrictions in the mid-90's.

                    Now, after several years in which the nation as a whole has enjoyed a declining crime rate, there is direct evidence that Lott's
                    conclusions are wrong. A 1999 analysis of crime statistics conducted by the Center to Prevent Handgun Violence (CPHV)
                    demonstrates that allowing people to carry concealed handguns does not mean less crime. The Center found that, as a group,
                    states that rely on permissive concealed weapons laws as a crime fighting strategy had a significantly smaller drop in crime than
                    states which looked to other means to combat crime rather than make it easier to obtain a concealed weapons permit.

                    In the 29 states that have lax CCW laws (where law enforcement must issue CCW licenses to almost all applicants), the crime rate
                    fell 2.1%, from 5397.0 to 5285.1 crimes per 100,000 population from 1996 to 1997. During the same time period, in the 21
                    states and the District of Columbia with strict carry laws or which don't allow the carrying of concealed weapons at all, the crime
                    rate fell 4.4%, from 4810.5 to 4599.9 crimes per 100,000 population. The decline in the crime rate of strict licensing and no-carry
                    states was 2.1 times that of states with lax CCW systems, indicating that there are more effective ways to fight crime than to
                    encourage more people to carry guns.

                    Furthermore, according to the CPHV analysis, violent crime actually rose in 12 of 29 states (41%) which liberalized their CCW
                    laws over the five years beginning in 1992, compared to a similar rise in violent crime in only 4 of 22 states (18%) which did not
                    change their CCW laws. The disparity in the decline is even more obvious for rates of gun violence. From 1992 to 1997 (the last
                    five years for which data exists), the violent crime rate in the strict and no-issue states fell 24.8% while the violent crime rate for
                    states with liberal CCW laws dropped 11.4%. Nationally the violent crime rate fell 19.4%.

                    If allowing more people to carry handguns is supposed to be such an effective crime fighting strategy, why did the crime rate go up
                    in so many "shall issue" states -- particularly when compared to states that employed other strategies to fight crime? The simple
                    answer: allowing thousands of ill-trained citizens to carry guns everywhere they go has no positive effect on the crime rate.
                    Research now substantiates what common sense has always argued: in a society riddled with gun-related crime, reducing the
                    opportunities where guns can be used will actually reduce the rate of gun violence.

                    The following Q&A details other major points of criticism of Lott's book and study:

                    Q: What are the indications that Lott's study of changes in concealed carry laws is flawed?

                    A: Apart from the obvious mistakes (e.g., Lott's inability to accurately identify when states changed their carry laws), several
                    researchers have shown that small changes in the statistical models Lott uses to reach his conclusions result in large changes in his
                    findings - an important indication that his research is fundamentally flawed. Researchers who have reanalyzed Lott's data, for
                    example, found no beneficial impact from changes in carry laws when Florida was not included in the study, or when they restricted
                    their analysis to counties that had populations greater than 100,000 people.

                    Q: Doesn't Lott implicitly acknowledge that his work is fundamentally flawed because he does not account for other
                    factors which could affect both the crime rate and the decision by state legislators to change carry laws?

                    A: Yes. On page 153 of More Guns, Less Crime, Lott writes that "The more serious possibility is that some other factor may
                    have caused both the reduction in crime rates and the passage of the law to occur at the same time." And he goes on to write that
                    "For a critic to attack the paper, the correct approach would have been to state what variables were not included in the analysis."
                    Well, this has been done.

                    Critics of Lott's study have identified a number of factors that affect crime rates, but which Lott failed to address in his research.
                    Examples include changes in how the police go about their business (e.g., implementation of community policing, and
                    crime-mapping techniques used by some police departments), changes in poverty levels, gang activity, maturation of the drug
                    market, and other changes in gun laws.

                    Most important, in a paper published in the Journal of Legal Studies (January 1998), Dan Black and Daniel Nagin used a well
                    known, formal statistical test that proved that Lott failed to include a number of important variables in his study. On the basis of this
                    and other findings, Drs. Black and Nagin, along with Professor Jens Ludwig, concluded that "there is absolutely no credible
                    evidence to support the idea that permissive concealed-carry laws reduce violent crime," and that "it would be a mistake to
                    formulate policy based on the findings from Dr. Lott's study."

                    To this day, John Lott has failed to provide any statistical evidence of his own that counters Black and Nagin's finding that Lott's
                    conclusions are inappropriately attributed to changes in concealed carry laws. Until Lott can do this, it is inappropriate for him to
                    continue to claim that allowing more people to carry concealed handguns causes a drop in violent crime.

                    Q: Is it true that John Lott's findings have been dismissed by Florida State University criminologist Gary Kleck, who
                    is cited as an authority on gun violence research by Lott and whose work is routinely praised by the NRA?

                    A: Yes. Kleck has accepted the Black and Nagin critique, writing in his new book that Lott's thesis "could be challenged, in light
                    of how modest the intervention was. [More] likely, the declines in crime, coinciding with relaxation of carry laws were largely
                    attributable to other factors not controlled for in the Lott and Mustard analysis" (Targeting Guns; p. 372).

                    Q: John Lott claims that changing the law to allow more people to carry concealed handguns causes a fall in violent
                    crime, yet he finds virtually no beneficial effect from changes in handgun carry laws on robbery - the crime most likely
                    to occur between strangers, and in public spaces. Is this finding consistent with his theory -- does it make sense?

                    A: No, it does not make sense. In fact, the finding that changes in concealed carry laws result in a large drop in rape - a crime
                    most often committed within homes by someone who is known to the victim - while showing virtually no beneficial impact on
                    robbery is another indication that Lott's study is fundamentally flawed.

                    In scientific terminology, the basic criticism made of Lott's research is that the statistical model he used to reach his conclusions is
                    "misspecified." This means, in part, that he did not adequately account for other factors which have an impact on crime rates - and
                    which provide an alternate explanations for his findings. When a statistical model is misspecified, it cannot be used as the basis from
                    which to draw conclusions about the impact of policy decisions. One clue that a model is misspecified is if it produces implausible
                    findings.

                    Q: Are there other implausible findings in Lott's research?

                    A: Yes there are. For example, Lott claims that when states ease restrictions on concealed carry laws, criminals do not stop
                    committing crimes - instead, they switch to crimes that decrease the likelihood that they will come in contact with an armed victim.
                    Specifically, he finds that criminals stop committing rape, murder and aggravated assault, and start stealing cars and committing
                    larceny. That is - they substitute auto theft and knocking off coin-operated machines for rape and murder.

                    Does anyone really believe that auto theft is a substitute for rape or for murder? Lott's research also suggests that the presence of
                    elderly black women in the population is associated with higher rates of murder and auto theft despite the fact that these women are
                    neither perpetrators nor victims of these types of crime.

                    Q: I'm confused-isn't it true that there is very little, if any, evidence that changes in carry laws have an impact on
                    actual gun carrying?

                    A: True. One of the biggest mysteries is how changes in carry laws effects changes the risk that a criminal predator will confront
                    an armed victim. Survey research suggests that the percentage of people who report carrying a gun, at least some of the time,
                    exceeds the percentage of people who have a license to carry a gun. And even if the odds did change, why wouldn't criminals
                    respond by becoming more likely to carry a gun themselves, more likely to shoot their victims, and more likely to attack from
                    behind. Certainly, this is more likely to happen than a rapist becoming an auto thief.

                    Q: John Lott's claim is "more guns, less crime," but a substantial portion of that claim is based on his use of two
                    voter exit polls. Can he use these polls to make this claim? And, does the evidence support the claim?

                    A: No. Lott inappropriately uses two voter exit polls to make assertions about changes in the level of gun ownership by adults in
                    14 states, and then compounds his mistake by using this data to make the assertion that "states with the largest increases in gun
                    ownership also have the largest drops in violent crimes."

                    First, according to the Voter News Service (the organization responsible for the 1996 poll) their data is designed to be nationally
                    representative, but does not provide representative data about individual states. Second, according to the Voter News Service it is
                    not possible to compare the 1988 and 1996 exit poll numbers on gun ownership because of differences in how the questions were
                    asked. The fact that making the comparison produces results about changes in the level of gun ownership in America that are wildly
                    out of whack with other survey data support this conclusion.

                    By comparing the two exit polls, and applying a formula that he devised, Lott concludes that the percentage of adults who own a
                    firearm increased by 50% from 1988 to 1996, and the gun ownership among women increased at the fastest pace - yet, the best
                    available evidence on gun ownership (from the General Social Survey) indicates that gun ownership has remained essentially
                    unchanged for men and women since at least 1980. (Changes in Firearm Ownership Among Women, 1980-1994; Journal of
                    Criminal Law and Criminology, Fall 1995) Because Lott has inappropriately misused the exit polls in conducting his study, and so
                    radically miscalculated changes in gun ownership, it is simply impossible for him to make any assertions about the relationship
                    between changes in gun ownership and the crime rate.

                    Q: Is it true that law enforcement has consistently been opposed to weakening carry concealed weapons laws?

                    A: Yes, and for good reason. Law enforcement does not favor allowing more citizens to carry guns as a solution to crime
                    problems. Police feel that an armed citizenry jeopardizes their safety, and public safety in general. When more citizens are armed,
                    police officers have to approach every vehicle stop, every contact with a citizen, as a potential contact with an armed individual.
                    Furthermore, law enforcement has been instrumental in defeating efforts to ease CCW restrictions all across the country, because
                    they know the responsibility involved with carrying a concealed weapon.

                    Proponents of this legislation contend that citizens will be adequately trained to handle firearms responsibly, but this is rarely true.
                    Police departments require officers to go through a great deal of safety and proficiency training before being issued a gun -
                    followed by regular refresher courses and qualifications throughout the officer's career. Citizens armed under the provisions of
                    non-discretionary carry laws are not so highly trained, and frequently not trained at all, thereby further increasing the risk of injury
                    and death with a firearm.

                    Q: Wouldn't allowing more people to carry concealed handguns increase incidents of citizens attacking each other?

                    A: Yes. Contrary to John Lott's assertion that "such fears are unfounded," it appears to be increasingly evident that permit holders
                    commit very serious offenses which, unfortunately, are not isolated events. The Texas Department of Public Safety found that
                    felony and misdemeanor cases involving license holders rose from 431 in 1996 to 666 as of mid-December in 1997, a 54.5
                    percent increase. What makes this particularly disturbing is that Texas has the strictest licensing law of the "shall issue" or
                    non-discretionary states. Texas also has one of the most stringent reporting requirements for CCW infractions. This type of
                    information is either not available from other states or the standard is much more lax for reporting CCW-related events. The
                    information from Texas certainly undermines claims by the gun lobby that only law-abiding citizens will carry.

                    Q: Do gun traffickers benefit when states ease access to concealed-carry permits?

                    A: Yes. The Philadelphia Inquirer (Sunday, January 11, 1998) highlighted the problem in that city of CCW license holders
                    involved in gun trafficking. According to the Inquirer, Pennsylvania CCW licensees, who can avoid background checks and waiting
                    periods, are serving as straw purchasers getting handguns into the illegal markets. In fact, the article concluded CCW licensees are
                    a major source of gun trafficking: according to Stephen T. Haskins, a senior ATF special agent, "We are experiencing more straw
                    purchasing in Philadelphia because they changed the requirement for the permit to carry."

------------------------------------------------------------
Looking for a book? Want a deal? No problem AddALL!
http://www.addall.com compares book price at 41 online stores.



This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:38:57 MDT