Note on David Frum's piece on the Florida theater shooting Monday by a retired police captain

David Frum at the Daily Beast wrote:

The 71-year-old former Tampa police captain had founded the city’s SWAT team. Retired from the force, Reeves still carried a .38 caliber handgun. On Monday, he carried his gun with him to a movie theater in Wesley Chapel, Florida, an exurban community 26 miles north of Tampa. Reeves became annoyed by a man in the row directly ahead of him who texted before the show. Reeves complained first to the man, then to the theater manager. A confrontation erupted. Voices were raised. Popcorn was thrown. And suddenly: a man was dead. . . .

There was at least one adult who carried a gun in the theater in which Oulson was shot to death. Perhaps Reeves imagined that he might use his weapon to prevent some terrible crime. Instead, he committed one.

One statistic often tossed about in the gun debate is the claim that guns are used for self-defense some 2.5 million times a year, once every 13 seconds. That statistic is based on a set of surveys conducted before 1995 in which gun owners were asked whether they could remember using a gun to meet any kind of threat over periods that varied from one year to as many as five years. The phrasing of the questions could include anything from confronting an armed intruder to picking up a shotgun before investigating a squawk in the chicken coop.  This kind of hazy self-reporting, conducted almost a generation ago, is not likely to generate any kind of reliable information. . . .

Michael Dunn, a 45-year-old man, pulled into a Jacksonville gas station in November 2012. The next car over was occupied by four teenagers playing loud music. Dunn is white; the four young men, black. Dunn ordered the teens to turn down their music. They refused. An argument erupted. Dunn drew a gun and fired eight or nine shots at the teens, killing one of them. Dunn claimed that he had glimpsed a gun inside the teens’ car and therefore felt threatened. . . .

Part of an email that I sent to David:

1) Do you really want to use anecdotal evidence when there are currently about 11 million concealed handgun permit holders in the US?  I agree based on the evidence that has thus far been presented the Jacksonville gas station case looks extremely bad and so does this movie theater case.  Since you are raising two cases from Florida, there are about 1.2 million active permit holders in Florida.  168 have had their permit revoked for any fires related violations.  If you look at the revocations from January 2008 to July 31, 2013, there were four additional revocation (however, obviously the cases that you mention won't be included in those numbers).  That comes to an annual revocation rate of less than 1/10,000th of a percentage point.  When I have called up the licensing bureau they have told me that the vast majority of revocations have been violations where no one has been hurt, namely people accidentally carrying guns into gun-free zones or forgetting to have their carry permit with them when they are carrying.

Since the movie theater case involves a retire cop, is the conclusion that trained police shouldn't be allowed to carry when they are off-duty or when they retire?  Do on-duty police make mistakes?  Sure, but I assume that you wouldn't argue that police shouldn't be allowed to carry a gun, right?  The question isn't whether we can identify mistakes.  The question is what is the net effect of police being able to carry.

2) I know of at least four national surveys on defensive gun uses after 1995.  I would have to look it up, but the minimum estimated number of defensive gun uses in those later four surveys was at least 1.3 million.  Minor note: all the surveys in 1995 or earlier asked people about events over at least 5 years.  And, as you know, I agree that is problematic.  Only surveys after 1995 asked about events over previous year.  That said, surveys are only of the most limited use and must be used carefully.  That is the reason why I turn to regressions that look at changes in crime rates when people are allowed to own or carry guns for protection.