The Logical Fraud Conflating "Gun Deaths" with Public Safety

Vox recently published an article by German Lopez that makes a logical error. The article conflates an increase in "gun deaths" with a decrease in public safety. The jump is made from an article published last year, in 2018, where the Rand Corporation published a chart based on a number of different and dubious studies on "gun deaths". From the previous article:

But there were some things that could be gleaned from the available evidence. While RAND as a nonpartisan group avoided any sweeping policy conclusions in its analysis, its review does seem to point in a direction, based on my own reading: More permissive gun policies lead to more gun deaths, while more restrictive policies lead to fewer gun deaths. Coupled with other evidence in this area, that supports the idea that more guns lead to more gun deaths. 

The key way the Rand compilation of studies is biased is by the focus on “gun deaths”. The major bone of contention in restrictions on gun ownership and use is *not* gun deaths. It is, first, a philosophical determination of political power;  secondarily, a question of practical effect, primarily about whether other methods will be substituted for guns in crime and suicide.  A major part of that equation is whether guns have significant benefits for defense, which has little to do with how many people are killed.

The claim is made that Rand is "non-partisan".  The unstated assumption, made by Rand, and by Lopez, is that microscale pragmatism is the lens by which governmental decisions should be made. They implicitly reject the argument that political power grows out of the barrel of a gun. That is a partisan assumption. They ignore Constitutional questions. That is a partisan approach. Rand is one of the think tanks that has prospered as the nation came under the sway of the Mediacracy. It has a vested interest in pushing increased regulatory schemes.  It assumes the regulatory state is benevolent.

Even with the assumption of a benevolent state, if you ban guns, and "gun deaths" decrease, but overall homicide and suicide rates are not effected, you have not accomplished anything to reduce homicide or suicide.  You have increased restrictions on ordinary citizens.

Focusing only on "gun deaths" is intellectually dishonest.

I have not read all of the studies. I have read most of the Australian studies and looked at some of the raw data. The articles' contention, that over all homicides and suicides were decreased by the gun laws in Australia, is false.

Most of the studies acknowledge there was little change. Suicides went up for a few years, driven by other methods. Homicide, followed the same trend line it previously had.

The research on Australian gun law is mixed, with significant variation in quality.

I only know of two studies  about firearms accidents and concealed carry law. One study was national, about fatalities. The nationwide study done by Lott and Mustard in 1997, looking at unintentional deaths, was inconclusive. The other study only looked at one state, Arizona, and only from 1988 to 2003 It is unknown why the study, which was done in 2013, cut the date for the Arizona study off at 2003. The study only looked at unintentional firearm injuries, not at unintentional firearm deaths. It included adults over the age of 18 who were not legally capable of obtaining concealed carry licenses. There are many other candidates for minor variations in accident rates.

To claim there is a correlation with increased concealed carry and increased firearms injuries sounds plausible; it is not supported by fact. If it were, we should see a measurable increase in fatal firearms accidents over the last 20 years. We do not.  The number of concealed carry permit holders has increased from 2.7 million in 1999 to over 17.25 million a year ago.

During the same period, fatal firearms accidents have decreased from 824 to 486, with the rate decreasing from .29 per 100,000 population in 1999 to .15 per 100,000 population, in 2018, almost in half.  There may be other factors, but if the legal carry of guns increases firearms accidents, it is an exceedingly small amount, which is beyond measurement.

Concealed carry might decrease accidents. It is easy to create hypotheticals; carrying a defensive firearm may keep it under the control of an adult, for example, rather than leave it at home to be stolen. Carrying a firearm may increase training and awareness of firearms safety. These are unknowable imponderables that are difficult to measure. People with concealed carry permits are extremely law-abiding, several times less likely to commit crimes than police.

Part of the problem with these studies is the bias inherent in the language used to ask the question.

"Gun violence" is a propaganda term used to link five separate phenomena involving guns, all together, and claiming all the problems can be solved by infringements on the Second Amendment.

The proponents of the term link suicides, non-justifiable homicides, justifiable homicides,homicides by police, and firearms accidents all together under the "gun violence" banner.  These are separate phenomena with separate causes and dynamics.

There is no serious reason to believe suicides will decrease in any significant amount if private sales are outlawed by "Universal Background Checks". Most suicides are by older white men who have owned guns for decades. Numerous other methods exist. In Australia, suicides went up after the extreme gun restrictions were put in place. Different methods, particularly hanging and single car fatalities were used.

There has been no reduction in overall homicides when the access to guns was made much more difficult.  One study, in a medical journal, claimed that homicides have increased with "stand your ground" law in Florida. Half of the increase was from recorded justified homicides. The study used justified homicide data from a law enforcement data base, then compared it to total homicides from a medical data base. Any homicide not recorded as justified was assumed to be non-justifed. If you infer problems with conflating different data sets, you are correct.

Two other studies, in Arizona and Texas, showed no effect.

Overall effects are small, involving a few homicides or suicides or accidents out of hundreds or thousands. It is difficult to differentiate differences in reporting from differences in fact. This allows result oriented researchers to pick through the data to find results to reinforce their ideological point. It need not be a conscious decision. Selection bias and the confirmation bias are well known.

There is no strong evidence one way or another, which magnifies bias effects.

One of the problems is, to find evidence, the researchers are reduced to using complex models to eliminate the effect of other relationships. The selection of baselines, timing, and factors to be considered are all subject to selection bias and fiddling.  This is easiest done for studies limited in time and space. The more limited the study in time and space, the more potential for biased results. The most credible studies are those involving national, detailed data, over long time lines.

Studies are notorious for selection bias in international comparisons. They choose some subjective criteria, such as "developed' or "rich" countries while ignoring demographic differences among populations. If you see a study that only compares England and Sweden, but leaves out Brazil and Russia, you know you are seeing bias. All international studies are suspect, because of definitional difficulties. 

It is not hard to understand that academics, especially in the medical field, have biases against gun ownership.

There is a general bias against gun ownership in most of the medical journal studies I have read.

There seems to be less in the criminological journals.

In determining positives and negatives of gun ownership, there are many imponderables and value judgments. The rule of law, and the supremacy of the Constitution, are seldom mentioned by those who want the population disarmed.

If you believe firearms are useful for defense of self, others, and the common polity, there are studies that agree with you. 

If you believe firearms have no legitimate purposes, there are studies promoting your view. No regulation or ban will be considered extreme or irrational.

People familiar with firearms tend to fall into the first camp. Those without firearms knowledge tend to fall into the second. 

These are two competing world views. Power brokers can always find an expert to support the policy decision they wish to make.

Historically, those with arms have ruled those without.

©2019 by Dean Weingarten: Permission to share is granted when this notice and link are included.

Gun Watch