In article , "Paul Barnett"
 wrote:


> With or without the presence of strict gun control in a city, Colin Loftin
> and Ellen MacKenzie have estimated that as few as only 3% of rapes are
> actually measured by the NCVS.  As has been noted by other people, since the
> NCVS doesn't ask about any defensive gun use if the interviewee refuses to
> talk about the crime, this results in a significant under-reporting.

What you are saying makes no sense, Paul. First of all, the NCVS is
precisely designed to elicit reports of crimes underreported to law
enforcement, like rape.  Second, suppose the views of some feminists, and
others, are correct, that rape is so horrible a crime that a vast majority
of victims will not talk about it, even to the Bureau of the Census (or
for that matter, if you think about it, to Kleck).  If so, then the true
extent of rape is likely unknowable.  But (third) the unknowable is not at
issue.  We have two figures, both derived from people who *are* willing to
talk about it. First, we have the people who are willing to talk about it
to the NCVS, and from that we can deduce that there are x number of
thousands of rape victims a year. There is no reason to suppose that 40%
of them (which I seem to remember, without looking it up, is implied by
Kleck) resisted with guns, or that *more than 100% of them*, as a study
says which uses Kleck's own methodology, resisted with guns.  

What these anomalies tell us is not that the NCVS significantly
underreports rapes, which is probably untrue, but in any case irrelevant
(since all of the surveys under consideration derive their results from
people who *are* willing to talk about the crime), but that studies using
Kleck's methods are susceptible to extreme inflation in DGUs produced by a
relatively small number of false positives. 

Best regards,

-Jim McCulloch