Seems like a flawed anti-gun study as they don't mention what is considered violent crime and is any committed by legal gun owners.

http://news.stanford.edu/2017/06/21/violent-crime-increases-right-carry-states/

Donahue has spent his entire career attempting to prove that guns are bad. It is virtually impossible that he would ever produce any "research" that would support any other conclusion. That doesn't necessarily mean that his work is bogus or his conclusions erroneous. But probably (at least according to my rigorous data analysis and model projections).

We'll have to wait for Dr. Lott to fisk the methodological flaws of the study, as very few people could knowledgeably comment on that aspect.

We could comment on the "commonsense" aspects of what was and wasn't included as data in the study, but how many of us want to pay $5 for the privilege of reading the paper? I don't want to give them any money. I'll write the authors and ask for free copy... often they send them if personally requested.

I can almost guarantee that at some point in the paper they qualify that they have not demonstrated "causality", but only an "association", meaning that RTC laws do not cause an increase in whatever violent crimes they measured. I'm sure Lott will point out that by manipulating data and "models" that one can demonstrate virtually any association one wants (e.g. there's an association between larger shoe size and increased reading ability... i.e. 8-year-olds read better than 4-year-olds, but the shoe size isn't the cause of the increased ability.).

I'm extremely suspicious when anyone claims that every single RTC state had a greater crime increase than every single state that didn't have RTC. Not that it couldn't be true, just seems statistically improbable. I'd also like to see what constitutes RTC and not RTC. Every single state has RTC and they all issue to at least some degree, except for our beloved Hawaii of course. That's gonna be some " nuanced" "defining", aka "data manipulation". How would they explain one state having a 2% population rate of CCW having the same increase in crime as a state that had a 14% rate of population CCW? That seems "counterintuitive" if the actual CCW has something to do causally with the increase in crime, which is what they "imply". Otherwise you are left with the mere fact of "passing a RTC law" as being the causal factor for increased crime, not the consequences of the law in terms of what percentage of the population then legally carries.

I'd like to see their definition of "defensive gun use", as I suspect from the statistic they cite that they 1. mean the discharge of a firearm, and/or 2. they have not included attempted or suspected violent crimes that were not initiated (i.e. the would-be attacker was stopped by the brandishing of a weapon by the potential victim prior to the actual onset of the attack (read their language to see what I'm talking about: "in 99.2 percent of the violent attacks in the United States, no gun is ever used defensively".). And what about robberies and burglaries and car theft or other crimes than "violent attacks"? There is likely a reason for the data selection.)

Perhaps the greatest unstated assumption is that a Constitutionally-guaranteed right can be "regulated" to the point of non-existence because of some data showing possible negative consequences (without really considering the full positive benefits). While that might be an assumption that some people share, not all do, especially when applied of rights other that Second Amendment rights (freedom of speech, freedom from unwarranted searches, right to legal representation, etc.).

quote from the article....
----
"The synthetic control approach, a research method now widely applied in economics and political science, uses an algorithm that combines crime patterns from several non-RTC states – or during the time before states adopted RTC – to create an artificial or synthetic state."
--------

They couldn't get any real stats so they used synthetic stats....like this

Donahue has spent his entire career attempting to prove that guns are bad. It is virtually impossible that he would ever produce any "research" that would support any other conclusion. That doesn't necessarily mean that his work is bogus or his conclusions erroneous. But probably (at least according to my rigorous data analysis and model projections).

We'll have to wait for Dr. Lott to fisk the methodological flaws of the study, as very few people could knowledgeably comment on that aspect.

We could comment on the "commonsense" aspects of what was and wasn't included as data in the study, but how many of us want to pay $5 for the privilege of reading the paper? I don't want to give them any money. I'll write the authors and ask for free copy... often they send them if personally requested.

I can almost guarantee that at some point in the paper they qualify that they have not demonstrated "causality", but only an "association", meaning that RTC laws do not cause an increase in whatever violent crimes they measured. I'm sure Lott will point out that by manipulating data and "models" that one can demonstrate virtually any association one wants (e.g. there's an association between larger shoe size and increased reading ability... i.e. 8-year-olds read better than 4-year-olds, but the shoe size isn't the cause of the increased ability.).

I'm extremely suspicious when anyone claims that every single RTC state had a greater crime increase than every single state that didn't have RTC. Not that it couldn't be true, just seems statistically improbable. I'd also like to see what constitutes RTC and not RTC. Every single state has RTC and they all issue to at least some degree, except for our beloved Hawaii of course. That's gonna be some " nuanced" "defining", aka "data manipulation". How would they explain one state having a 2% population rate of CCW having the same increase in crime as a state that had a 14% rate of population CCW? That seems "counterintuitive" if the actual CCW has something to do causally with the increase in crime, which is what they "imply". Otherwise you are left with the mere fact of "passing a RTC law" as being the causal factor for increased crime, not the consequences of the law in terms of what percentage of the population then legally carries.

I'd like to see their definition of "defensive gun use", as I suspect from the statistic they cite that they 1. mean the discharge of a firearm, and/or 2. they have not included attempted or suspected violent crimes that were not initiated (i.e. the would-be attacker was stopped by the brandishing of a weapon by the potential victim prior to the actual onset of the attack (read their language to see what I'm talking about: "in 99.2 percent of the violent attacks in the United States, no gun is ever used defensively".). And what about robberies and burglaries and car theft or other crimes than "violent attacks"? There is likely a reason for the data selection.)

Perhaps the greatest unstated assumption is that a Constitutionally-guaranteed right can be "regulated" to the point of non-existence because of some data showing possible negative consequences (without really considering the full positive benefits). While that might be an assumption that some people share, not all do, especially when applied of rights other that Second Amendment rights (freedom of speech, freedom from unwarranted searches, right to legal representation, etc.).

Well said....


Thank you

Seems like a flawed anti-gun study as they don't mention what is considered violent crime and is any committed by legal gun owners.

http://news.stanford.edu/2017/06/21/violent-crime-increases-right-carry-states/

The actual study probably defined violent crime, just gotta do more digging. I don't see a link to the actual study though.

The actual study probably defined violent crime, just gotta do more digging. I don't see a link to the actual study though.

You really do need to have that Googler of yours looked at.  It's been broken for far too long.

https://works.bepress.com/john_donohue/163/download/

Professor Donohue sent me that link to the free download of the paper: https://works.bepress.com/john_donohue/ (to see the 44 other articles in the "Gun Control" section).

The press person for Stanford who was also listed at the end of the article only sent me the link to the pay site.

It's 102 pages long so I probably won't have much to say about it for another week or so... my eyes glaze over after about three pages of that statistical modeling stuff... though I might skimthrough it and look for the relevant definitions...

Professor Donohue sent me that link to the free download of the paper: https://works.bepress.com/john_donohue/ (to see the 44 other articles in the "Gun Control" section).

The press person for Stanford who was also listed at the end of the article only sent me the link to the pay site.

It's 102 pages long so I probably won't have much to say about it for another week or so... my eyes glaze over after about three pages of that statistical modeling stuff... though I might skimthrough it and look for the relevant definitions...

Hidden in the middle of the study is a statement that carry stats can't be shown to cause increases or decreases in violent crime.  Other factors like hiring more or fewer police, changes in sentencing or laws, and so on are potentially having an impact on violence.

Correlation doesn't prove causation.

We've been presenting the anti2A with facts for years.
Now they responded with bogus data. 

Child gets killed by ND and it's all in the papers and in studies.  Woman prevents an assault, crickets.

You really do need to have that Googler of yours looked at.  It's been broken for far too long.

https://works.bepress.com/john_donohue/163/download/

I didn't try looking, I just pointed out that the article failed to report the definition while the study likely did have one.

We've been presenting the anti2A with facts for years.
Now they responded with bogus data. 

Not necessarily bogus data though news reports may certainly be guilty of presenting it a certain as saying more than it actually shows.

Not necessarily bogus data though news reports may certainly be guilty of presenting it a certain as saying more than it actually shows.

They used a state that doesn't exist to run the statistics on -- in other words, they couldn't prove their conclusions with real data, so they took a composite that provided data more friendly to their cause.

It's so obvious.  With every argument they come up with based on statistics, they use national, state, city or county data, depending on whether the data supports their conclusions.

Look at Chicago.  You can't use state totals because that skews the raw data having a city with extremely high gun crimes.  Due to local laws, local gangs, and other local factors, the only way to identify commonalities in any crime study is at the local level.  State laws might affect all the cities in it, but the crime stats are not the same from city to city. 

Yet, here this study is, creating a notional state using other state-level data.  It's invalid on its face.

But the question should be how many who legally conceal carry have committed violet crimes with handguns. That would be the only valid argument, if a study showed people who conceal carry are the ones committing all the violet crimes but I doubt that is the case. Who really cares unless it can be proved that conceal carry people are the ones committing all the violet crime, thus the increase in violet crime.

I tried to read it on the flight back from the mainland but only got part way. There were too many good shows to watch on their airline screen. They seem to equate crime rate with gun ownership, police hiring, and incarceration rates. The implication is that RTC states naturally also have higher police hiring and incarceration rates which sounds fishy to me. They also talk about their synthetic control approach giving them "comfort" data. WTF?

It said that the SCA was used in economic and political science. Using this method to compare crime rates with RTC states sounds spurious. I understand what they're trying to do. In a controlled environment, you take a sample and then experiment with them in a lab to eliminate confounding factors which would be problematic when analyzing gun violence. I guess in a traditional setting, you would put three similar groups in three different rooms then give them many guns, few guns, or no guns and see how they compare after several days. There might be ethical concerns brought up. To solve this, they create a synthetic control group to compare different cities like trying to compare apples and grapes to an imaginary cranberry. Perhaps it makes more sense if you squint.

Dr. Lott published a critique a couple of days ago. Somewhat lengthy, but required to point out the shortcomings and (deliberately misleading?) methodologies of this "study".

THE FLAWED AND MISLEADING DONOHUE, ANEJA, & WEBER STUDY CLAIMING RIGHT-TO-CARRY LAWS INCREASE VIOLENT CRIME

[Intro]

A new, unpublished study by John Donohue, Abhay Aneja, and Kyle Weber has received a lot of attention for supposedly finding some evidence that right-to-carry laws increase overall violent crime rates.  It has been covered in Newsweek, The Atlantic, Bloomberg, Vice, Snopes, and many newspapers such as Newsday and the Salt Lake City Tribune.  As is typical of Donohue’s work, there is no attempt to mention or respond to prior criticisms, and he just repeats the same, seriously biased methods and errors.

Publications such as Time and Newsweek would always interview critics when they ran stories on Lott’s original research.  But when studies have the right political biases, reporters no longer get both sides of the story.  But especially when the media explicitly describes a study as “debunking” John Lott’s previous research, you might think a reporter would call up Lott and get his take on it.  It has been two weeks after the Donohue-led research started getting attention, and not a single reporter has contacted him.

The bottom line is pretty clear: Since permit holders commit virtually no crimes, right-to-carry laws can’t increase violent crime rates.  You can’t get the 1.5 to 20 percent increases in violent crime rates that a few of their estimates claim with only thousandths of one percent of permit holders committing violent crimes.  To put it differently, states would have to be miss reporting 99%+ of crimes committed by permit holders for their results to be possible.

The synthetic control tests where they use anything from two to four states to predict the changes in another state’s violent crime rates are extremely arbitrary.  For example, would you look almost exclusively to Hawaii to predict violent crime rate changes in Idaho, Minnesota, Mississippi, Nebraska, and Utah?  Would you look almost exclusively at Illinois to predict changing violent crime rates in South Carolina?  Remember that half of Illinois’ violent crime occurs in Chicago and an even larger majority of the changes in Illinois’ changing violent crime rate is due to Chicago. Would you look at California and New York to predict changing violent crime rates in Georgia?

ETA: Links

Dr. Lott's fairly lengthy original critique of Donohue, et al. is here: https://crimeresearch.org/2017/07/badly-flawed-misleading-donohue-aneja-weber-study/

And a follow-up, wherein Lott critiques Donohue's responses to Lott's criticism is here: https://crimeresearch.org/2017/07/responding-john-donohues-responses-evaluation-new-study/