It’s back
Tim’s post on Kellermann, which I mentioned here, is back. Tim, regarding my objection to Kellermann, notes:
Kellermann clearly notes this problem in his 1986 study, which also discounts gun misuse that does not result in death. This was one of the reasons why he did the 1993 study which does measure self-protection that does not result in death.
I have not read the 1993 study and am unable to look for it right now. I will do it later. But, to be clear, the 1986 study was creation science. Kellermann knew the conclusion he wanted and worked the data to get those results. He can, of course, fall back on the fact his claims are technically correct (they are) while his supporters and detractors misrepresent the results. Gun control types make the incorrect claim and pro gun people debunk the bogus claim (because it’s easy). Or has XRLQ says in comments:
In any event, I do think there is something wrong with studying an issue almost no one cares about, but which is likely to be mistaken for a different issue people really do care about. Any potential gun owner wishing to weigh the pros and cons of gun ownership for self defense will be ill served by the study. Come to think of it, even the rare sicko who dreams of “bagging a burglar” will be, as the study does not control for the fact that most gun owners don’t think (or act) that way.
Assuming that the data was sound and the methodology was impeccable, the most Dr. Kellermann may have accomplished is to prove once and for all that if you ask a stupid question, you really will get a stupid answer.
September 3rd, 2004 at 2:44 pm
There are a lot of problems with both of Kellerman’s studies. Here’s a quick sample:
1 The study does not deal with “family members”, but rather “acquiantances”. “Acquaintance” as defined in the study is a broad category that includes, for example, members of a rival gang.
2 It is possible to legitimately use a firearm for self-defense against a family member; consider domestic violence, for example. The Kellerman study does not account for any legitimate uses in self-defense.
3 Kellerman’s sample was invalid, because he acquired it in a manner which clearly skewed the data.
4 Kellerman neglected to account for which gun actually killed the victims in his study; he counted those who were murdered by an attacker who brought their own firearm.
I used to have a whole bunch of links on this topic. I’ll relocate the betters ones and post something, since people seem to be citing Kellerman again (still?).
September 6th, 2004 at 3:47 am
Going through your comments.
1. There were two studies. Neither one just dealt with acquaintances.
2. The first study did include defensive killings of family members. The second study measured protection by seeing if gun ownership reduced your chances of being murdered.
3. The sample was found by looking at all homicides in the home. I don’t see why this would skew the data.
4. The extra risk of gun ownership was associated with murders by residents, not outsiders.