This paper needs a good fisking

[bbhack]

https://www.nber.org/system/files/worki ... w30190.pdf

As I have time, I'm going to go through this paper and research and evaluate their various claims. The premise is that increased carrying (what they call RTC - right to carry) in its various forms increases violent crime. Through combinations of data picking and sorting and interpreting, I don't doubt they have made the massaged data show what they want to show.

I think a question we have to come to grips with is "is bad data better than no data?". The COVID statistics we have are complete garbage - in my opinion good for absolutely nothing. Is the UCR corrupted? If it is, is it worth analyzing?

How do you make the point that urban open-air insane asylums should never be compared with areas where people are functional and rational? That's something that perhaps makes the whole exercise pointless, but I would like to not do that, and see how their methods are otherwise good or flawed.

We analyze a sample of 47 major US cities to illuminate the mechanisms that lead Right-to-Carry
concealed handgun laws to increase crime. The altered behavior of permit holders, career
criminals, and the police combine to generate 29 and 32 percent increases in firearm violent crime
and firearm robbery respectively. The increasing firearm violence is facilitated by a massive 35
percent increase in gun theft (p=0.06), with further crime stimulus flowing from diminished police
effectiveness, as reflected in a 13 percent decline in violent crime clearance rates (p=0.03). Any
crime-inhibiting benefits from increased gun carrying are swamped by the crime-stimulating
impacts.

[srothstein]

Well, here is one thing they did not take into account:

From the FBI UCR page:

National Incident-Based Reporting System 

The National Incident-Based Reporting System, or NIBRS, implemented to improve the overall quality of crime data collected by law enforcement. It captures details on each single crime incident—as well as on separate offenses within the same incident. In 2021, the historic Summary Reporting System (SRS) data collection, which collected more limited information than the more robust NIBRS, was phased out.

This means that there should be both an increase in crime and a decrease in crime solving because of the more thorough data collection. The UCR where agencies reported things like the number of each crime and the number cleared is gone now and the NIBRS data is much more specific. It also allows for reporting more than one crime per incident. For example, under UCR if there was a robbery committed by two people, it would show as one crime. Under NIBRS, it would show as multiple crimes (at least two criminals) and could show as more if they had multiple charges, such as unlawfully carrying or assaulting the victim too. Summarizing their data into just two years shows how they did not properly account for this confounding change in data collection methodology.

Table I also clearly shows they are misreading their data. Guns did not increase crime rates. The worst they can claim is that it reduced the reduction in crime rates. It is interesting that it reduced clearance rates though. I wonder if they did a comparison between clearance rates and attempts versus completed crimes. They both show up the same in the UCR stats.

The use of the end data for all this also hides the effect of the defunding police movement on police effectiveness. A very interesting study would be to judge how much proactive policing takes place in cities like Atlanta and Chicago now compared to about 5 or 10 years ago. That would definitely affect the clearance rates too.

That is from a quick glance at the article. I might study it more later to see exactly what I can find wrong with it.

[bbhack]

Well, here is one thing they did not take into account:

From the FBI UCR page:

National Incident-Based Reporting System 

The National Incident-Based Reporting System, or NIBRS, implemented to improve the overall quality of crime data collected by law enforcement. It captures details on each single crime incident—as well as on separate offenses within the same incident. In 2021, the historic Summary Reporting System (SRS) data collection, which collected more limited information than the more robust NIBRS, was phased out.

This means that there should be both an increase in crime and a decrease in crime solving because of the more thorough data collection. The UCR where agencies reported things like the number of each crime and the number cleared is gone now and the NIBRS data is much more specific. It also allows for reporting more than one crime per incident. For example, under UCR if there was a robbery committed by two people, it would show as one crime. Under NIBRS, it would show as multiple crimes (at least two criminals) and could show as more if they had multiple charges, such as unlawfully carrying or assaulting the victim too. Summarizing their data into just two years shows how they did not properly account for this confounding change in data collection methodology.
...

Good catch. I can't find information about how NIBRS was phased in. It's unlikely that reporting jurisdictions did it all at once. It's been in study for years.

[JustSomeOldGuy]

https://www.nber.org/system/files/worki ... w30190.pdf

We analyze a sample of 47 major US cities to illuminate the mechanisms that lead Right-to-Carry
concealed handgun laws to increase crime. The altered behavior of permit holders, career
criminals, and the police combine to generate 29 and 32 percent increases in firearm violent crime
and firearm robbery respectively...

1) I don't accept that the '47 major cities' are either the majority of, or representative of, the entire population(s) under the new laws, and

2) the three population segments mentioned (LTC, career hoods, cops) mentioned were ALREADY carrying, so there's no change there, at least not in terms of volume.

[The Annoyed Man]

Here’s how what passes for their logic works:

A. Clouds are puffy and white.

B. Sheep are puffy and white.

C. Therefore, clouds are sheep.

[OneGun]

Here’s how what passes for their logic works:

A. Clouds are puffy and white.

B. Sheep are puffy and white.

C. Therefore, clouds are sheep.

This is the same moronic logic that leads to:

If you ban guns, there will be no more gun violence!

[Paladin]

History can add several things. After Texas CHL became the law there was a rapid decreased "crime" in that the crime of "Unlawful Carrying of a Weapon". People who were already carrying got licensed and subsequently legalized their carrying of weapons.

I've seen clear evidence that weapons bans led to increases in crime (ie District of Columbia), because like NY's 1911 Sullivan's law, the root of gun control is disarming law abiding people while empowering certain organized criminal groups. So the result is gun control laws don't decrease crime:

Overwhelming Majority Of Studies Find That Gun Control Policies Don’t Work

[Boxerrider]

I believe your efforts are worthwhile. In addition to the statistical corrections, your evaluation should point out that the logical fallacies utilized in this publication are favorites of other anti-civil rights groups as well.

[Boxerrider]

https://www.nber.org/system/files/worki ... w30190.pdf "The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research."

What if a person's viewpoint was not that of a law professor at Stanford or Duke, but that of a working, middle-class, property owner in a rural part of a red state?
Might that person look at this same data and deduce that the voters in those major U. S. cities routinely elected people who failed to effectively control "career criminals"?
Could this person form the opinion that, since those voters failed to place effective people in positions of authority in their city, then those voters should not have the privilege of participating in state and federal elections?

I am certain that both parties completely respect everyone's constitutional rights - they just believe in a little regulation for the sake of public safety.

[srothstein]

The basic concept of the study is sound. If you compare crime rates for before and after an event (such as adopting licensed carry) and control for other factors, then a difference in the rates is more or less directly attributable to the event you looked for. The way to attack a study of this type is not to point out logical fallacies but to show problems in the data collection or other not accounted for variables that might have affected the results. You may also show flaws in the data collection or study methodology.

IMO, it is not unreasonable for a study to show that more people carrying firearms and defending themselves would result in an increased crime rate per the UCR statistics. After all, you must remember that if someone attempts to rob you, that is reported as a crime. If you shoot him in self-defense, that is also reported as a crime. It is illegal to shoot people, so it gets reported as a crime. It is usually a crime that is easily cleared by exception (declined to prosecute) or by arrest and trial (regardless of trial outcome). Thus, more people carrying guns equals more rime reported in the FBI statistics. Another increase we could reasonably expect is the increase in gun thefts. Again, more people buying guns for the first time indicates they are more likely to not have the "proper" safe storage and the guns will be stolen. Add to that the lack of competence of new gun owners and the number of accidental shootings and deaths will increase too.

Please note that it is common to refer to these statistics as Uniform Crime Reporting (UCR) statistics, but it is not an accurate name. As I mentioned in the first post, these statistics are now correctly called the National Incident Based Reporting System (NIBRS) statistics. They are incidents and not necessarily crimes.

One of the problems I see with the study, and they attempt to cut it off ahead of time with a section answering the complaint, is that they compared the beginning and end of the period only. The problem with this methodology is that the cities did not all adopt right to carry (RTC) at the same time. A better study would have included the specific studies for each city by year over the entire time period, marking when each city adopted the RTC and seeing if the effect occurred immediately or with the same delay in each city. If there were no effect in the studied cities until much later, then it is an indicator that some other variable caused the change and not the RTC adoption. If some of the cities had the effect BEFORE they adopted RTC, it is an even stronger indication that there is some other confounding factor that was missed.

I would argue that another problem with their study is that CITIES do not adopt RTC. It is done at the state level almost exclusively. Thus, by picking just the cities and not the states, you have biased the data and conclusion by ignoring a possible factor in crime: population density. In addition to this, I would argue that the police in smaller cities and rural areas have mush better relations with the population there, not falling victim to the defund the police and charging the police for everything that happens. This affects both the crime rate and the clearance rate and should be better accounted for in the study.

I have some friends who might be interested in reviewing this study and their data. I will see if I can talk to to them about it. The problem I know of is that many schools, like Texas State, put pressure on their faculty to stay away from the gun issue completely. That includes research for and against guns. It is just too politically charged for many schools and people. And my friends are at the bottom of the totem pole academically, trying to make a living and eventually get tenure.

[K.Mooneyham]

Those ivory tower elitists cook data like that so crooked gun-grabbing politicians, particularly Democrats, can quote that stuff when pushing gun control laws. They make the data show what they want it to show, like the "climate change" types do. Real science and math are dead, killed by crooked leftist "academicians" (propagandists). I'm no hero and I'm no tough guy, but if those over-educated bums want to disarm me, let them come on out here to the sticks and try. I'm just sick and tired of their garbage.

[Paladin]

Those ivory tower elitists cook data like that so crooked gun-grabbing politicians, particularly Democrats, can quote that stuff when pushing gun control laws. They make the data show what they want it to show, like the "climate change" types do. Real science and math are dead, killed by crooked leftist "academicians" (propagandists). I'm no hero and I'm no tough guy, but if those over-educated bums want to disarm me, let them come on out here to the sticks and try. I'm just sick and tired of their garbage.

This is pretty much the story... all propaganda, all the time. And we are tired of it.

Data shows there’s no climate catastrophe looming – climatologist Dr J Christy debunks the narrative:


Dr J Christy uses plain language in his discussion of how the politicians and other "climate change" modelers have been shown the real data over many decades that demonstrates the climate models predicting rapid warming are overwhelmingly wrong by a factor of at least 2. Modelers have refused to correct their models... which demonstrates that their "science" is simply propaganda.

[puma guy]

Here’s how what passes for their logic works:

A. Clouds are puffy and white.

B. Sheep are puffy and white.

C. Therefore, clouds are sheep.

'zactly!