Tim Lambert (deltoidblog AT gmail.com) is a computer scientist at the University of New South Wales.
J. Neil Schulman writes:
When a dozen surveys which are specifically attempting to quantify DGU's finds DGU's an order of magnitude larger than the NCVS, then you have your answer.
None of those surveys other than Kleck's were designed to quantify DGU's and they all have problems when used for that purpose. See Kleck's paper.
And even if those surveys were designed like Kleck's, what you have then is just a larger sample size, still much smaller than the NCVS.
since (a) the NCVS wasn't properly designed to get an accurate count of DGU's and
The professional criminologists in the BJS don't seem to agree with you.
If the criminologists at the BJS think so little of Kleck's abilities to do accurate surveys, then why do they contract with him to do surveys for them?
Non sequiter. Please consult report NCJ-147003
Here's an extract:
Average annual number of victimizations
in which victims used firearms to defend
themselves or their property
________________________________________
Attacked Threatened
Total offender offender
________________________________________
All crimes 82,500 30,600 51,900
If BJS criminologists don't think that the NCVS provides any meaningful estimate of DGUs, why did they publish this report?
It may be that Kleck is right and 97% of gun-defenders lied to the NCVS. Or it may be that the NCVS is right and 4% of those who did not use a gun defensively lied to Kleck. If only a small percentage lie on surveys then it follows that Kleck is wrong.
Provide me a plausible explanation for people inflating their DGU claims --- when every incentive is to minimize it to avoid legal problems --- and you might have an argument.
It is well known that people in surveys often try to tell the surveyors what they believe the surveyors want to hear. Conduct a survey about DGU and some people will make up a DGU to make you happy. Conduct a survey about alien abductions and some people will tell about being abducted (2% in a 91 survey of adult Americans.)
In his essay "A Nation Of Cowards" Jeffrey Snyder argues that it is a citizen's civic duty to fight back against crime with a gun, and it is cowardly not to. In answering surveys people tend to give answers that make themselves look good. Someone who did not want to be thought a coward might invent a DGU, or appropriate someone else's.
Someone who was politically opposed to gun control has an incentive to make DGUs look more common.
Some people might consider a DGU to be an exciting macho experience and brag about it, even though they did not have one.
Some people will lie just for the hell of it.
Bear in mind that only a very small percentage have to make up or change the details of a DGU to account for Kleck's result.
Yes? Tell me precisely by what method the NCVS has been "validated"?
I'm not going to explain elementary stuff to you. Try reading Biderman's book on the NCVS.
You impudent jerk. I find myself responsing to your ad hominem with more ad hominem because it seems to be the only thing you're capable of understanding through the barriers of your superstitions.
You asked a technical question about the NCVS. I've told you where you can find the answer. You indulge in yet another of your bouts of name-calling.
My father defended himself with his gun five times. Not once was he ever surveyed by the NCVS.
Oh good grief. You don't even understand the most basic principle of random sampling --- that you don't have to survey the entire population.
And "good grief" back at you. You are congenitally incapable of reading a comment within the intended context.
Listen: you are arguing that a survey database is "real."
No I am not. I joined this thread when you claimed that a survey database was "overwhelmingly convincing" scientific evidence. I pointed out that survey research has limitations. You then claimed that I wanted to throw out all of social science...
My father's experience may be an anecdote to you. But when one man can use a gun in a DGU five times and NEVER have a policeman show up to take a report ... when such a man wouldn't have admitted he owned a gun to ANYONE taking a survey ... and when I have run into hundreds of people just like my father as a consequence of writing a book on gun rights (which means that gun defenders will talk to me who wouldn't talk either to the NCVS or Kleck in a million years), THEN, yes, I have a base of anecdotal evidence that gives me expectations of what I would expect science to find.
Yes, surveys aren't going to find gun defenders that are unwillingly to talk to strangers about it. This is hardly an argument for the accuracy of one survey over another.
When it does confirm it, not once but dozens of times
Huh? You claim that these people wouldn't talk to Kleck. How can his survey confirm their experiences?
--- and when there is a rational explanation of why NCVS would fail to reify the dozen other surveys
There is no such rational explanation. To quote Philip Cook: "I don't understand why people would be so much more forthcoming with Kleck's survey callers than the government's. I find that absurd."
--- then it's no longer an interesting question to me, except to throw it in the faces of cranks and crackpots who keep pushing gun control laws WHEN THERE IS NO SCIENTIFIC EVIDENCE FROM ANY PLACE ON EARTH THAT IT DOES ANYTHING BUT TEND TO INCREASE VIOLENT CRIME.
Yeah right. Let me speculate: If a correlation is found between gun control and less violent crime you'll claim that is not scientific evidence because "other factor" caused it. Or you'll claim the whole thing was a fraud. On the other hand, if a correlation is found between gun control and more violent crime, then that's proof that gun control causes more violent crime. How am I doing?
And that is my last word on this subject. I don't enjoy debating with people who are only interested in scoring points by any means possible, and have no actual interest in the truth.
Nor I.
