YouTube
We have all heard variations of the following "proven truths":
If you buy a gun, it is 43 times more likely to kill you or a family member than an intruder.
40% of firearms transfers are done without a background check.
States with strict gun laws have few "gun deaths"
Members of the media accept all of these "facts" as being true because they come from their chose "experts" - gun control groups - but how are these conclusions reached? What is wrong with their research methods? What happens when you look at the same issues with sound reasoning?
Let's look at the most common logical errors in their "studies":
Mixing Dissimilar Data Sets:
This is one of the most common intentional errors made by gun control advocates. For instance, they habitually combine all deaths involving firearms including criminal homicides, justifiable homicides by citizens and police, accidents and suicides. They then call this the "gun death" figure. No government agency publishes this figure, for a very good reason: Each of these deaths involves a very different cause. Combining them does not result in any useful insight. It does however, provide a means for gun control groups to confuse the public.
This is a very good time to examine the "states with permissive gun laws have more gun deaths" claim. Strictly speaking, this is true. However, it's a bit like asking someone, "Have you stopped beating your wife, answer yes or no?"
Let's break this down:
Depending upon the year, between 63% and 66% of these "gun deaths" are suicides. Gun control advocates argue including these death is legitimate because many of these suicides would not happen if the people who committed suicide did not have access to firearms. Gun rights advocates argue that in the absence of firearms, people determined to commit suicide will simply use other methods.
I must say that my 10 years of EMS experience does allow me to make one unequivocal statement: In the vast majority of suicide attempts, the person does not really want to die - and therefore they choose a method that enables them to survive by calling for help. The do not use firearms, or jump from a high structure, or walk into traffic., or cut a major artery. They take a few pills, or cut their wrists so shallowly that the cuts hardly bleed. They want attention or help, but they do not want to die. This greatly skews the suicide attempt statistics.
There is no dispute that when firearms are available, people will use them to kill themselves, therefore states with more firearms will indeed have more firearms suicides. However, the comparison between state where guns and common and those were they are not, is a bit like comparing the number of residents from the SF Bay Area who jump from the Golden Gate bridge and those from New York who choose the same method. It would be a silly comparison, because the Golden Gate bridge is much more available to those living nearby. People in New York will simply chose another method. It is the same with firearms, if they are not available, people determined to commit suicide will simply chose another method.
So, how do we determine if the absence of firearms prevents suicides? Well, we could look at nations with virtual or actual bans of private firearm and see if people still find ways to kill themselves.
Suicide rates per 100k population:
South Korea (total firearms ban) 24.1
Russia (virtual firearms ban) 17.9
Japan (virtual firearms ban) 15.4
United States (firearms widely available) 12.6
Does this prove that the absence of firearms causes suicides? Of course not! However, it does prove that, when firearms are not available people really want to kill themselves will use other methods. Blaming firearms for suicides is an absurd claim that is easily refuted.
Therefore, 2/3 of their gun death figures are bogus.
So, what happens when you look at gun ownership rates and criminal homicide rates?
Consider these charts. They show the number of firearms sold (number of NICS checks) per homicide. If the anti-gun rights activists are correct and more guns cause more murders, than the bars should be close together. If there is a causal relationship between lawful firearms ownership, the number of firearms sold should be roughly the same in each state - or at least in states with similar homicide rates.
Number of NICS Checks per Homicide
Low Homicide Rate States
All states above have slightly lower than average homicide rates of 3.1 - 4.5 per 100k.
Clearly there is no causal relationship because the firearms purchase rates are very different.
Clearly there is no causal relationship because the firearms purchase rates are very different.
What about states with high murder rates?
Is there a relationship there?
All states above have higher than average homicide rates of 6.0 per 100k or above.
Again, no relationship is seen between legal gun ownership and homicides - even in states with the highest murder rates.
Again, no relationship is seen between legal gun ownership and homicides - even in states with the highest murder rates.
There is, however, one area of consistency:
Most states with lower than average homicide rates have
Most states with lower than average homicide rates have
high rates of legal gun ownership.
All states above have lower than average homicide rates of 1.8 per 100k or below and very high gun ownership rates.
This is why gun control advocates have to make up their own statistics by mixing suicides with homicides. If we only look at homicides, the case for "more guns equals more homicides" simply does not exist.
It is very common for gun control groups to cherry pick data that serves to make their case. An example would be to compare one state with "strict" gun laws and a low violent crime rate and another state with less strict laws and a high crime rate - and then conclude that gun control results in a low violent crime rate. The problem with this is that in any data set, there will be outliers that run counter to the overall trend. This means that virtually anything can be "proven" by carefully selecting what you compare - in this case what states you select.
For instance a recent article compared Hawaii (strict gun control, no concealed carry and low homicide rate) and Louisiana (little gun control, shall issue concealed carry and
 |
| When you do not cherry pick states to compare, restrictive concealed carry laws are associated with higher crime rates. |
What happens when you actually compare all restrictive concealed carry states with states that allow any law abiding citizens who pass background and training requirements to obtain a license to carry? Well, we discover that there is a strong association between restrictive concealed carry laws and higher crime rates. The reverse is also true, shall issue concealed carry laws in 43 states are associated with lower crime rates.
Even in states with higher murder rates than restrictive states, we must beware of the next problem........
Confusing Association With Causation
This is the one error that is also common among gun rights advocates as well as anti-gun rights groups. For instance, just because homicide victims may be more likely to own a gun, does this mean that that owning a gun puts people at risk of being murdered? Well, if that is all the information we have, all we can say is that gun ownership is associated with higher individual rates of homicide.
However, we must always remember that association does not always equal causation. For instance, one explanation for higher rates of gun ownership could be that people at risk for homicide are more likely to buy firearms because of that higher risk. Looking at states with permissive concealed carry and high crime rates, we need to know if the crime rate was higher or lower before the concealed carry law changed. Is concealed carry in that state associated with a drop or an increase in crime? It is very possible that the causal relationship in these cases could be exactly the reverse of what gun control advocates want it to be.
Using Deeply Flawed and/or Outdated Studies
This is a good time to dispense with the "40% of gun transfers are done without a background check" deception. There is indeed a "study" that did find this. It was done over 20 years ago and was highly flawed. So flawed that both Politifact and Factcheck.org have rated the claim as false. So, what are the problems with this study?
1) Small data set: The sample size of gun purchasers was only 300
2) Poor timing: The survey was done in 1994 - the first year in which federal background were required. It looked back two years, which means that for most of the sample period, the current background check law was not in effect!
3) It depended upon buyers memory: After establishing that the person they had called had bought a gun, they asked them if a background check had been conducted. The problems with this are massive.
a) First, as noted above, the background check system was not in force for more than half the time frame sampled.
b) Second, even when sales were done after the new law took effect, while buyers would likely remember filling out the federal form and waiting three days before picking up the firearm, it is doubtful that they would take note of the background check run by the dealer during the waiting period.
c) Third, even in cases where the buyer was aware of the background check at the time, many would no longer remember it
In other words, this "study" clearly massively under counted the number of background checks actually done. In fact, it is amazing that they came up with a 60% figure for background checks.
Yet, anti-gun rights organizations use this study, knowing full well that the numbers from it are completely false. This in spite of the fact that a more recent study - done by pro-gun control researchers is available here. This study puts the percentage of firearms "transfers" done without a background check at 20%. Yet, gun control groups use the 40% figure, over and over, all the while knowing that it is off by at least 100%. What does this say about their level of integrity?
While this study is not as bad as the 1994 study, it too has at least one significant problem: It talks about "acquisitions", not purchases. Why is this a problem? Well, this includes inherited firearms, as well has transfers between family members, Even California does not require a background check under these circumstances. Clearly, the 20% in this study does reflect the actual number of sales without background checks either.
There is a way to know approximately what the real number is - the Colorado experience. Colorado prepared for 420,000 background checks per year, based on the 40% figure. According to this article in the Boulder News, the author of the bill believes that compliance is high - but instead of 420,000 background checks on private sales, Colorado only saw 13,200. That is 96.75% less than the estimate. If the supporters of the Colorado law are right about a high level of compliance, than the real number of private sales done without background checks is 1.3%. Personally, I don't think that compliance is high - but even if it is only 25%, then the number of private sales is still only 5.2%. I wonder why gun control groups don't mention the Colorado experience?
All of the Above
It is now time to tackle the most deceptive claim of all: If you buy a gun, it is 43 times more likely to kill you or a family member than an intruder.
Wow, it sounds like owning a gun is very dangerous....... but is it really? Sadly, this study was published in New England Journal of Medicine - but that isn't proof that it is true. It is proof that doctor are not the people to study gun violence.
This study was done by CDC researchers Arthur Kellermann and Don Reay in 1986 - and even though better studies exist, it is accepted as fact by millions of Americans.
Let's take a hard look at the many things wrong with this "study":
1) Invalid Comparison
Before we begin to look at the many flaws in the study, we must examine the flawed premise of the conclusion. That premise is that a defensive firearm only benefits the owner by killing someone. That is a completely false premise. The purpose of the a defensive firearm is to protect. In a small number of cases, this may indeed involve taking a life, but in many other cases people are protected without shots being fired.
Any valid comparison must include many more cases than the the 200-400 justifiable homicides recorded by the FBI each year - it must take into account Defensive Gun Uses (DGUs). No law enforcement agency tracks these incidents. The lowest estimate of DGUs is 55,000 per year. The highest is over 5 million. The National Self-Defense Survey resulted in an estimate right down the middle: 2.2-2.5 million. All of these figures, even those from respected pro-gun control researchers, greatly exceed the number of firearms related death from all causes.
This invalid comparison not only invalidates the entire conclusion, it reveals the massive bias of the researchers. Ironically in 2013, another CDC study (ordered by Pres. Obama) reached the following conclusions:
- Armed citizens are less likely to be injured by an attacker
- Defensive uses of guns are common
The new CDC study also called for the federal government to begin tracking DGUs. Predictably the Obama administration did nothing.
The bottom line is that the entire premise of the comparison is invalid.
2) Mixed dissimilar data sets
In this study, the researchers committed the error of mixing data sets in two different ways.
a) Suicides included
As we saw above, it is highly questionable to include suicides in a "gun death" figure. In this study, of the 43 gun deaths they count, 37 are suicides.
b) Legal and Illegal ownership included
Researchers made no effort to determine if the firearm was owned legally or illegally. Why is this important? Simple: People who own guns illegally are typically criminals. Criminals are much more likely to be killed by other criminals, much more likely to be careless with their firearms and much more likely to be killed by police in the process of, for instance, a drug raid.
For instance, the vast majority of urban homicides are related to the drug trade. Drug dealers cannot call the police if someone tried to steal their product - so they are usually armed. That means that virtually every drug dealer killed by another drug dealer will own a gun - and the researchers would count them as "gun owners", killed because they owned a gun.
Additionally, consider that people who are selling and using drugs, while keeping their firearms handy, are much more likely to create a situation in which an accident may occur.
Furthermore, criminals who possess firearms illegally are much more likely to use them against a family member or domestic partner. That's why we made it illegal for anyone with a domestic violence history to own firearms.
Again, the premise assumes that the risk to all gun owners legal or illegal is the same - an assumption that is completely absurd.
3) No attempt made to determine if the gun owned was even involved in the fatal event
Yes, you read that right. This "study" counted all firearms deaths in gun owning households. It assumes that the firearm used was a firearm owned by the residents. So, if a man is killed by an armed home invader - with the home invader's gun - and that home has a gun secured in a safe or stored away in the attic - they counted that as a gun related death.
All they cared about were two questions:
- Was someone in the home shot and killed?
- Was there a gun in the home when it happened?
4) No attempt was made to determine of the firearm had been purchased in response to a threat.
Here they make yet another logical error, assuming that that the firearm is the source of the threat, when, in many cases, it is a response to an outside threat. Couple this with no effort to determine if the firearm owned by someone in the home was even used in the homicide and we have another error.
In short, this study from the 1980s has enough holes to drive a Mac truck through - yet supposedly smart people quote it over and over again as fact.
The bottom line is simple: If you are a law abiding person in good mental health, this so called study is completely useless in determining the risks and benefits of owning a firearm.
So, I appeal to all who read this, do your homework. Check out what anyone says, yes even me. Don't assume that they have really done solid research and reached logical conclusions - especially if the "research" comes from an anti-gun rights group. As we have seen, they have a history of deception.
No comments:
Post a Comment