[On Sep 20 2002 I posted this to firearmsregprof and emailed it to Lott.]
James Lindgren writes:
After my post to this list saying that "a big national study doesn't just disappear without a trace" because a computer crashes, John Lott called me and told me a long story about how the study was done (which I don't choose to share just yet, if ever; Lott can speak for himself on his methods, if he wishes). He didn't ask me to do anything about it, and I wasn't planning on posting anytime soon, but given the recent posts I thought I would.Whether the study should be given credence is a different question from whether it was ever done. The latter question is the important one for me.
While it will probably be possible to get a lot of circumstantial evidence about Lott's data losses on various projects in 1997 (I got an email from David Mustard to this effect and will check with Bill Landes myself), the crucial question on which direct evidence is needed is whether study was done. Lott reports that he used UC undergrad volunteers to do the calling, using a sample drawn from a CD ROM with phone numbers.
Volunteers? Based on Kleck's surveys you would need 300 person days of calling. I can imagine a student doing a day or three as a volunteer, but more than that? How could you get enough volunteers?
When his computer crashed, he decided not to publish the study and says that he made only passing reference to it in his book.
In the 1st edition of MGLC Lott writes: "If national surveys are correct, 98 percent of the time that people use guns defensively, they merely have to brandish a weapon to break off an attack." He doesn't mention his survey *anywhere* in the 1st edition. In the 2nd edition of MGLC (published in 2000) Lott changed it to: "If a national survey that I conducted is correct, 98 percent of the time that people use guns defensively, they merely have to brandish a weapon to break off an attack." As far as I know, this is his first published mention of the survey.
This would not be a big deal, except that Lott has made his 98% claim on literally dozens of occasions. His claim promotes the "magic talisman" theory of gun self-defence -- suggesting that all you need to do is wave a gun around a little. This theory is dangerous -- it could potentionally lead to people being injured or killed.
Lott does not remember the names of the undergrads. I offered to contact the UC alumni office to see if I could email the undergrads who were juniors and seniors in the 1996-97 academic year to locate the callers. I intend to do just that in the next 3-4 weeks, if UC will go along.
Thank you so much for your assistance.
However, I find it disturbing that after discussing with you the means by which he could demonstrate that this survey was actually conducted, the best Lott could come up with was to attempt to contact students whose names he cannot remember. As has been discussed here before, there should be plenty more evidence -- phone bills, survey questions, people he discussed the survey with whose names he recalls and so on.