rev 8/4/2k (links 12/29/2k)


Gun control advocates will not acknowledge that crime and violence are prevented and deterred by guns being possessed by law-abiding people.  Or, if they do, they minimize the effect by stupid statements such as, "the deterrent effect is unknown" (or "over stated") and "only a very small number of people are killed annually in self protection by private citizens" (as though killing were the only way to protect).

The truth is that guns are used much more by private citizens for protecting the innocent than guns are used in crime.  This is true even for cases of self defense.

It is also true that law-abiding citizens owning guns reduces crime and violence, and allowing such citizens to carry guns out and about (not just at home) also reduces crime and violence.  Gun control advocates screem out against these ideas without even reading the proof, but it is absolutely proven thanks to the great work of John Lott, Jr. and David Mustard.

Their work, attributed mostly to Lott, caused a bit of a stir when published.  Although much of the media coverage has been gun control advocates getting their usual cooperation from the media, there have been a few instances in which Lott has been given an opportunity to speak.  So people with any sense could see that he had facts and his detractors had nothing but emotionalized generalities and lies.

Lott and Mustard's work has generally been characterized incorrectly by people on both sides of the issues as being a study of "concealed carry" laws.  The fact is that their work has been a long series of studies or analyses, and they have not been just about concealed carry laws.

Most of the analyses have been determinations of the extent to which rates of suicide, accidental death, and several different kinds or crime may be predicted by several things that have been suggested as having a causal effect.  One of the factors they considered as possibly having an impact on crime, etc. was the extent to which people were allowed to carry concealed weapons—not just on the basis of the state law, but also the actual practice.  They also considered the impacts of things such as arrest rates, encarceration rates and sentence enhancement laws.

The statistical methods they used in some of the analyses are actually able to determine almost certainly that one thing is a cause of the other, which most types of statistical analyses cannot do, because the one thing always preceeds the other.

Lott and Mustard used in most of the analyses all the data collected by the FBI from law enforcement all across the country for the last 16 years for which the data were available (through 1992).  They did not select out data from all that which was available to make the results show what they wanted it to show as gun controllers do in their "research."  They did most of the analyses using county-level data, and repeated them using state-level data in order to test the effect of using county-level data rather than state-level data.  They found generally that effects are more reliably detectable using county-level data since state-level data tends to average away effects because a state is not a homogenous entity.

A lesser number of the analyses were done using data Lott and Mustard were able to find other than the FBI data.  One of the characteristics of their work is that they tried to find a valid way of checking on the effects of practically anything anybody has ever suggested as possibly having an effect on some type of crime.  Another characteristic is that, to do this, they did some work to find and use data which could reasonably be used as a measure of the things suggested as having a bearing.  The main limitation of their works has been the fact that government agencies have not collected usable data regarding some of the things people have thought might be important.  Lott and Mustard generally did the best they could with what they could get, and they went to great effort to find and get what they could find.

Another characteristic of their analyses is that most of them have covered all areas of the country over the entire period for which data were available and have simultaneously accounted for all factors (including all federal and state laws) that have been billed as importantly related to crime.  So, for example, they took into account when the Brady Law went into effect, when each county started issuing concealed carry permits on a nondiscretionary basis, etc.  They considered time in two different ways so that they could determine whether or not time-since-effective-date was significant (as opposed to an effect being immediate and remaining constant forever).

After completing a great number of analyses and identifying their conclusions, Lott and Mustard published a paper in a criminology journal.  Gun controllers immediately got great press coverage for their criticisms even though they had often not even read the paper.  Lott answered the criticisms and actually re-ran the analyses in ways to accommodate and check them.  He then used two additional years of FBI data to check the earlier analyses, then included it all (with the criticisms and answers) in his book "MORE GUNS, LESS CRIME: Understanding Crime and Gun Control Laws" (U. of Chicago Press).

One thing to keep in mind about the work of Lott and Mustard is that the data they used and the number of different factors considered in their analyses is orders of magnitude greater (probably at least hundreds of times greater) than for any study ever about crimes, violence, or guns.  Besides this, they have run many more analyses.

The original paper is available in .pdf format for free.  The book is cheap ($23 suggested).  Every U.S. citizen should read it.  A politician should not be allowed to vote on a crime bill or gun bill without being able to pass an open-book test on it.  However, the paper and book are a bit technical.  One thing to keep in mind in reading either of them is that Lott frequently refers to a state having a concealed carry law when he really means "a nondiscretionary (shall issue)" concealed carry law.  It obviously makes writing complicated to have to use the more complete term over and over in a discussion about it.

When reading the paper or Lott's book, it will be helpful if you go in with a little understanding about the statistical analyses he and Mustard did and are still doing.  The analyses are based on the idea that something (rate of aggravated assault or murder, for example) is a result of a number of things and could therefore be predicted approximately by totalling up a number of factors involving those things.

For example, one could say that the rate of murder in a county equals some constant value, plus some constant times the poverty rate in the county, plus a constant times the portion of the public (in the county) that have concealed carry permits, plus a constant times the portion of the public that are law enforcement officers, etc., etc.  In their actual analyses, there were many of these different variables.  The statistical analyses they use determine the values (and signs) of all those different constants and, hence, the extent to which a crime rate, gun accident rate, etc. are affected by the various variables (if at all), and the direction of any effect (up or down).

One thing you will find in reading the paper or book is some occasional references to using "dummy" variables.  These are not some kind of trickery.  As already mentioned, data that would have been good to have for the analyses were often not available.  For instance, the numbers of people having concealed carry permits in the many counties (or even in the states) have simply not been recorded by authorities.  So, Lott and Mustard would use a variable for each county and each year, to which they assigned a value of zero for every year before a nondiscretionary concealed carry law went into effect, and a value of one for each year for as long as the law was in effect.  Similarly, they would assign a value that was zero before, then a value equal to the months after the law went into effect (so they could determine if the impact of concealed carry kept increasing after going into effect).

Bookmark for later reading all of the following:

FIGHTING BACK:  Crime, Self-Defense, and the Right to Carry a Handgun by Jeffrey R. Snyder, CATO Institute, Oct 22, 1997 [ALT] [PDF] [alt PDF]

Lives Saved, Injuries Prevented, ... by Dr. Edgar A. Suter of Doctors for Integrity in Policy Research

The Value of Civilian Arms Possession as Deterrent to Crime or Defense Against Crime by Don B. Kates, Jr. ('91) [ALT]

But check out, right now, if you haven't already done so the info at our site about Kleck and Gertz' study on how frequently private citizens in the U.S. use guns in self defense.