wrote in message
...
On Aug 29, 4:12 pm, "Ed Huntress" wrote:
But the city figures are still unreliable for reasons that aren't worth
going into here. The one category of crime that is, and always has been,
extremely accurate is the murder rate. You can't easily hide murders. So
if
you want to be fussy about it, start with that.
But the criticism of M-Q really should be directed at the FBI and at the
local reporting agencies for UCR. M-Q just reports what those agencies
report -- of course it's a "rehash of readily available statistics."
That's
what CQ Press does -- report summaries of official statistics.
Overall, their reports are generally accurate. When you have a murder rate
of 9.0/100k, like Nevada, and compare it with some state that has a rate
of
8.9, the comparisons are speculative. But when it's 9.0 (Nevada) versus
4.9
(New Jersey), or 1.2 or whatever (Hawaii), those UCR errors are not a
factor.
The criticism about relative city sizes and demographics is something for
the social scientists to worry about. This is just a state ranking. If you
live in Nevada, you're roughly twice as likely to be murdered than you are
if you live in NJ -- and the city issues and demographics actually work
against Nevada in the comparison. Those are just the facts, not
sociological
theorizing.
--
Ed Huntress
Very confusing to me. The city figures are unreliable. The state
figures are compiled from the city figures, but somehow are more
reliable?
No, they aren't just cities. They're compiled from every reporting
jurisdiction.
The bigger cities have been more resistant to using the uniform reporting
procedures, although that's more of an issue in regard to things like
domestic crimes, and to the definitions for clearing crimes. They're still
not 100% compliant.
However, that's not the big issue with the UCR. It isn't the accuracy of the
numbers that's questioned; it's the conclusions. The big issue is the
mistake of drawing simple conclusions from complex data.
Here's a simplified explanation by the FBI:
http://www.fbi.gov/ucr/word.htm
I agree the murder rates ought to be more reliable than say armed
robbery rates. But some murders might get plea bargained to
manslaughter. The number probably varies from state to state.
The data is based on crime incidences known to police, not on convictions.
I knew a guy that robbed a gas station with a shotgun, but was
persuaded to turn himself in. I forget what he was convicted of, but
it was something much less than armed robbery. Se he probably did not
raise the violent crime rate.
Yeah, he did, if the original crime was reported as an armed robbery.
At any rate there is probably no halo effect. In general criminal do
not think they are going to get caught. If they did they would not
commit crimes. In the same way, I do not think that they think about
the possibility that a intended victim might be armed.
The only place it seems to have an effect is in house robberies. The
percentage of homes in the US in which there is a gun is very high;
convicted criminals reported some fear of being shot when they were
interviewed in a DOJ study a decade or more ago.
But that doesn't seem to apply to concealed carry at all. The incidence of
concealed carry is so extremely low (typically around 1% of adults in
shall-issue states; even the total CCW licensing usually is around 2%) that
it's not likely to figure into their judgments. And the data seems to
support that idea.
I also think that there may be a correlation between the number of
violent crimes and the number of CCW permits. But it is more likely
that a high rate of violent crimes causes an increase in CCW permits.
Very possible. You'd have to spend a lot of time with the data to be sure.
If New Jersey had a high rate of violent crime, the government would
be under a lot of pressure to issue more CCW permits.
I don't know about that. In the late '80s and early '90s, when violent crime
rates were going through the roof, the response of states like NJ and NY was
to *tighten* restrictions on gun ownership.
--
Ed Huntress