Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Gun control

6 views
Skip to first unread message

Jilllian

unread,
Nov 7, 2001, 8:24:50 PM11/7/01
to
I'm have to debate the pros of gun control for one of my classes. As a WW
fan, I know that the show is an advocate for gun control but most of the
shows they discuss this I don't have on video. So I'm asking for help for
little facts and arguments. Thanks.
Jillian


Jerri

unread,
Nov 7, 2001, 8:26:37 PM11/7/01
to
"Jilllian" <jillk...@accesstoledo.com> wrote

I wouldn't use WW as a source, but I would use some web sites dedicated to
the topic:

www.bradycampaign.org
www.wagc.com
www.aclu.org/library/aaguns.html
www.opensecrets.org/news/guns/
www.jurist.law.pitt.edu/gunlaw.htm
www.guncite.com/
www.nra.org/
www.vpc.org/

... and for those who would suspect a bias, I just picked the first 8 that
came up on a search of google for "gun control".
Jerri


Dan Kimmel

unread,
Nov 7, 2001, 9:56:06 PM11/7/01
to

"Jilllian" <jillk...@accesstoledo.com> wrote in message
news:tujn7bc...@corp.supernews.com...

Handgun Control, Inc. drives the NRA batty, so you know they must be doing
something right. The link to the part of their website you'll find helpful
is

http://www.bradycampaign.org/facts/index.asp

DaveMoore

unread,
Nov 7, 2001, 11:45:07 PM11/7/01
to

I second the comment that WW is not the best place to
do your research.
On the Web, try:
http://www.saf.org/
http://www.saf.org/LawReview
The Second Amendment foundation. The Law Review
link gets you to a database of articles concerning
the right to keep and bear arms (RKBA).

http://www.ciar.org/~ttk/politics/guns/
An excellent collection of links to important articles.
I strongly recommend "The Embarrasing Second Amendment".
Also check out the first article, "If you only read one
article about gun control..."

http://www.ca5.uscourts.gov/opinions/pub/99/99-10331-cr0.htm
The opinion in a recent court case, Emerson, which has
an excellent analysis of whether or not the RKBA is an
individual right or an state's right. (Hint: the first
amendment does not protect the right to free speech
merely for government officials.)

Try Google searchs on Daniel Polsby, a noted gun control
scholar; Gary Kleck, who has researched the effects of
gun control on crime, and of course, John Lott, another
important researcher. Some papers by all these three
appear at the SAF and CIAR links.

In the bookstore or library, check out John Lott's _More
Guns, Less Crime_; this is the key study on guns and
crime, which looked at data from every county in the
U.S.A.

--
Dave Moore == djm...@uh.edu == I speak for me.
In the wrong hands, sanity is a dangerous weapon.

Robert Long

unread,
Nov 7, 2001, 11:57:23 PM11/7/01
to

"Dan Kimmel" <dan.k...@worldnet.att.net> wrote in message
news:avmG7.162044$3d2.6...@bgtnsc06-news.ops.worldnet.att.net...

And this is a good thing?

Read Yale Economics Professor John Lott's book More Guns, Less Crime. After
reading that, if you think Gun Control is the answer to crime, you are an
idiot.


Sam Worf

unread,
Nov 8, 2001, 2:21:19 AM11/8/01
to
"Robert Long" <rob...@mediaone.net> wrote:

<snip>


>
>And this is a good thing?
>
>Read Yale Economics Professor John Lott's book More Guns, Less Crime. After
>reading that, if you think Gun Control is the answer to crime, you are an
>idiot.

True, except that no one believes gun control is the answer to crime.


--
Remove spamblock to reply. // / Sam Worf
GCV 3.12 \X/ / samworf-AT-newsguy-DOT-com
GFA d- s-: a28 C+(+++)>$ UI(--) W+(-) N+ w+(--) M+(--) PS++
PE- Y+ t+(---) 5+++(-) X+ R tv+ b++ DI++ D++ G e++ !y+
Pure electric EV-1 / Accept no substitute!
---------------------------/-------------------------------
I thought the truth would matter,
I thought the truth would win;
While I was busy thinkin' They let the liars in...

Jaquandor

unread,
Nov 8, 2001, 8:37:13 AM11/8/01
to
>> > I'm have to debate the pros of gun control for one of my classes. As a
>WW
>> > fan, I know that the show is an advocate for gun control but most of the
>> > shows they discuss this I don't have on video. So I'm asking for help
>for
>> > little facts and arguments. Thanks.
>>
>> Handgun Control, Inc. drives the NRA batty, so you know they must be doing
>> something right. The link to the part of their website you'll find
>helpful
>> is
>>
>> http://www.bradycampaign.org/facts/index.asp
>>
>>
>>
>
>And this is a good thing?
>
>Read Yale Economics Professor John Lott's book More Guns, Less Crime. After
>reading that, if you think Gun Control is the answer to crime, you are an
>idiot.

If only it were that clear-cut. Anyway, here is a response to Lott:

www.cse.unsw.edu.au/~lambert/guns/lott/


--
-Jaquandor

"The proof of a poem is not that we have never forgotten it, but we knew at
sight we never could forget it." --Robert Frost

Mary Rosh

unread,
Nov 8, 2001, 2:50:44 PM11/8/01
to
jaqu...@aol.comsinatra (Jaquandor) wrote in message news:<20011108083713...@mb-fo.aol.com>...

> >> > I'm have to debate the pros of gun control for one of my classes. As a
> WW
> >> > fan, I know that the show is an advocate for gun control but most of the
> >> > shows they discuss this I don't have on video. So I'm asking for help
> for
> >> > little facts and arguments. Thanks.
> >>
> >> Handgun Control, Inc. drives the NRA batty, so you know they must be doing
> >> something right. The link to the part of their website you'll find
> helpful
> >> is
> >>
> >> http://www.bradycampaign.org/facts/index.asp
> >>
> >>
> >>
> >
> >And this is a good thing?
> >
> >Read Yale Economics Professor John Lott's book More Guns, Less Crime. After
> >reading that, if you think Gun Control is the answer to crime, you are an
> >idiot.
>
> If only it were that clear-cut. Anyway, here is a response to Lott:
>
> www.cse.unsw.edu.au/~lambert/guns/lott/

Lott's book (Chp. 9) points some of the inaccuracies at that web site.
By the way, if is points are at all accurate, why doesn't he try to
publish them in an academic journal?

Bruce Mills

unread,
Nov 8, 2001, 3:10:16 PM11/8/01
to

>
> >> > I'm have to debate the pros of gun control for one of my classes. As a
> >WW
> >> > fan, I know that the show is an advocate for gun control but most of the
> >> > shows they discuss this I don't have on video. So I'm asking for help
> >for
> >> > little facts and arguments. Thanks.
> >>
> >> Handgun Control, Inc. drives the NRA batty, so you know they must be doing
> >> something right. The link to the part of their website you'll find
> >helpful
> >> is
> >>
> >> http://www.bradycampaign.org/facts/index.asp

HCI drives the NRA batty because they deliberately lie and
misrepresent the facts.

userb3

unread,
Nov 8, 2001, 3:22:59 PM11/8/01
to

They were pretty weak on the pro-gun control arguments. But they were
pretty weak on the other side, too, so I'd skip the tape as a good
research tool

--
Userb3
What is the connection between guns and crime?
See http://www.lib.uchicago.edu/~llou/guns.html#study
See http://www.amazon.com/exec/obidos/ISBN=0226493636/t/3782-9387712-550397


userb3

unread,
Nov 8, 2001, 4:16:55 PM11/8/01
to
On 08 Nov 2001 13:37:13 GMT, Jaquandor wrote:

>If only it were that clear-cut. Anyway, here is a response to Lott:
>
>www.cse.unsw.edu.au/~lambert/guns/lott/

Unfortunately, Lambert's arguments start with faulty assumptions.
--
userb3

She was a Jung girl who was easily Freudened.


Chris Crandall

unread,
Nov 8, 2001, 5:07:01 PM11/8/01
to
: On 08 Nov 2001 13:37:13 GMT, Jaquandor wrote:
: >If only it were that clear-cut. Anyway, here is a response to Lott:
: >www.cse.unsw.edu.au/~lambert/guns/lott/

userb3 (use...@yahoo.com) wrote:
: Unfortunately, Lambert's arguments start with faulty assumptions.


: --
: userb3
: She was a Jung girl who was easily Freudened.

Well, I've checked out your post pretty carefully, but I couldn't find out
which faulty assumptions you were talking about.

Andrew Ryan Chang

unread,
Nov 8, 2001, 5:33:16 PM11/8/01
to
Chris Crandall <cran...@lark.cc.ku.edu> wrote:
>userb3 (use...@yahoo.com) wrote:
>: Unfortunately, Lambert's arguments start with faulty assumptions.
>: --
>: userb3
>: She was a Jung girl who was easily Freudened.
>
>Well, I've checked out your post pretty carefully, but I couldn't find out
>which faulty assumptions you were talking about.

Is this gonna be like when userb3 didn't present any evidence for
his/her statements of fact about the estate tax 'n' farms?

--
Computer Scientists love ed, not just because it comes first
alphabetically, but because it's the standard. Everyone else loves ed
because it's ED!
-Patrick J. Lopresti

userb3

unread,
Nov 8, 2001, 6:11:06 PM11/8/01
to
On 8 Nov 2001 22:07:01 GMT, Chris Crandall wrote:

>: On 08 Nov 2001 13:37:13 GMT, Jaquandor wrote:
>: >If only it were that clear-cut. Anyway, here is a response to Lott:
>: >www.cse.unsw.edu.au/~lambert/guns/lott/
>
>userb3 (use...@yahoo.com) wrote:
>: Unfortunately, Lambert's arguments start with faulty assumptions.
>

>Well, I've checked out your post pretty carefully, but I couldn't find out
>which faulty assumptions you were talking about.

When Lambert writes "3. even if there was more guns and less crime,
more guns did not cause less crime" he loses my interest. And a review
of his postings (handily availible from a link on his site) don't
impress me with his scholarship.

userb3

userb3

unread,
Nov 8, 2001, 6:12:34 PM11/8/01
to
On 8 Nov 2001 22:33:16 GMT, Andrew Ryan Chang wrote:

> Is this gonna be like when userb3 didn't present any evidence for
>his/her statements of fact about the estate tax 'n' farms?

Do I smell an ad hominem attack? Did I claim to have specific figures
or evidence that I didn't provide? Please be specific.

Andrew Ryan Chang

unread,
Nov 8, 2001, 7:10:18 PM11/8/01
to
userb3 <use...@yahoo.com> wrote:
>On 8 Nov 2001 22:33:16 GMT, Andrew Ryan Chang wrote:
>> Is this gonna be like when userb3 didn't present any evidence for
>>his/her statements of fact about the estate tax 'n' farms?
>
>Do I smell an ad hominem attack? Did I claim to have specific figures
>or evidence that I didn't provide? Please be specific.

You said that anyone could provide multiple examples of farms lost
because of the estate tax, but didn't. When Chris Crandall said that no
farms had been lost to the estate tax, you replied that that was false,
without any further elaboration. I followed up by quoting an essay on the
estate tax that mentions that the American Farm Bureau Federation also has
not provided (cannot provide?) any examples of farms lost because of the
estate tax; you didn't reply to that either.

--
Bart: Actually, we were just planning the father-son river rafting
trip.
Homer: Heh heh, you don't have a son.
-- People unclear on the concept, "Boy Scoutz 'N the Hood"

Andrew Ryan Chang

unread,
Nov 8, 2001, 7:15:55 PM11/8/01
to
userb3 <use...@yahoo.com> wrote:
>>Well, I've checked out your post pretty carefully, but I couldn't find out
>>which faulty assumptions you were talking about.
>
>When Lambert writes "3. even if there was more guns and less crime,
>more guns did not cause less crime" he loses my interest. And a review

I believe that's a section heading that summarizes his conclusion,
not a stated assumption. It hyperlinks to the chapter discussing it in
more depth.

userb3

unread,
Nov 8, 2001, 7:41:59 PM11/8/01
to
On 9 Nov 2001 00:10:18 GMT, Andrew Ryan Chang wrote:

>userb3 <use...@yahoo.com> wrote:
>>On 8 Nov 2001 22:33:16 GMT, Andrew Ryan Chang wrote:
>>> Is this gonna be like when userb3 didn't present any evidence for
>>>his/her statements of fact about the estate tax 'n' farms?
>>
>>Do I smell an ad hominem attack? Did I claim to have specific figures
>>or evidence that I didn't provide? Please be specific.
>
> You said that anyone could provide multiple examples of farms lost
>because of the estate tax, but didn't.

What, you wanted names and phone numbers? If you won't accept my word
that I have a number of customers and neighbors who lost their farms to
the estate tax, then so be it. This is usenet, and not a court of law.

> When Chris Crandall said that no
>farms had been lost to the estate tax, you replied that that was false,
>without any further elaboration.

We had to sell a portion of my own grandfather's farm. Happy?

>I followed up by quoting an essay on the
>estate tax that mentions that the American Farm Bureau Federation also has
>not provided (cannot provide?) any examples of farms lost because of the
>estate tax; you didn't reply to that either.

I didn't see that.

Kenneth Crudup

unread,
Nov 8, 2001, 9:00:23 PM11/8/01
to
In article <9sevm5$q9j$1...@news.cc.ukans.edu>,
cran...@lark.cc.ku.edu (Chris Crandall) says:

>Well, I've checked out your post pretty carefully, but I couldn't find out
>which faulty assumptions you were talking about.

<mock surprise>

Why, you're in favor of g*n c*ntr*l too?!

-Kenny

--
Kenneth R. Crudup Sr. SW Engineer, Scott County Consulting, Washington, D.C.
Home1: PO Box 914 Silver Spring, MD 20910-0914 ke...@panix.com
Home2: 38010 Village Cmn. #217 Fremont, CA 94536-7525 (510) 745-0101
Work: 5141 California Suite 200, Irvine, CA 92612 (949) 737-6200

Tim Lambert

unread,
Nov 8, 2001, 8:05:48 PM11/8/01
to
"userb3" <use...@yahoo.com> writes:

> On 8 Nov 2001 22:07:01 GMT, Chris Crandall wrote:
>
> >: On 08 Nov 2001 13:37:13 GMT, Jaquandor wrote:
> >: >If only it were that clear-cut. Anyway, here is a response to Lott:
> >: >www.cse.unsw.edu.au/~lambert/guns/lott/
> >
> >userb3 (use...@yahoo.com) wrote:
> >: Unfortunately, Lambert's arguments start with faulty assumptions.
> >
> >Well, I've checked out your post pretty carefully, but I couldn't find out
> >which faulty assumptions you were talking about.
>
> When Lambert writes "3. even if there was more guns and less crime,
> more guns did not cause less crime" he loses my interest.

I'm always keen to improve the quality of my presentation, but I don't
understand what you are objecting to. You di understand what the
"post hoc" fallacy is, don't you.

> And a review
> of his postings (handily availible from a link on his site) don't
> impress me with his scholarship.

It would move the discussion on just a tad if you gave us an example
of what you considered poor scholarship on my part.

Tim

Tim Lambert

unread,
Nov 8, 2001, 8:08:39 PM11/8/01
to
mary...@aol.com (Mary Rosh) writes:

> jaqu...@aol.comsinatra (Jaquandor) wrote in message news:<20011108083713...@mb-fo.aol.com>...

> > If only it were that clear-cut. Anyway, here is a response to Lott:


> >
> > www.cse.unsw.edu.au/~lambert/guns/lott/
>
> Lott's book (Chp. 9) points some of the inaccuracies at that web site.

Lott's response to my critique and my reply are available from the
link above. Read them both and decide for yourself who is inaccurate.

Tim

Jaquandor

unread,
Nov 8, 2001, 9:53:31 PM11/8/01
to
>>: >If only it were that clear-cut. Anyway, here is a response to Lott:
>>: >www.cse.unsw.edu.au/~lambert/guns/lott/
>>
>>userb3 (use...@yahoo.com) wrote:
>>: Unfortunately, Lambert's arguments start with faulty assumptions.
>>
>>Well, I've checked out your post pretty carefully, but I couldn't find out
>>which faulty assumptions you were talking about.
>
>When Lambert writes "3. even if there was more guns and less crime,
>more guns did not cause less crime" he loses my interest. And a review
>of his postings (handily availible from a link on his site) don't
>impress me with his scholarship.

So you read not more than the intro to the very first page? Then I don't think
you're qualified to comment on his work. If you want to refute his apparent
claim that Lott is employing "post hoc, ergo procter hoc" reasoning, then by
all means refute it. But scoffing at it without reading it is *not* refuting
it.

david

unread,
Nov 9, 2001, 2:23:23 AM11/9/01
to
On Wed, 7 Nov 2001 20:24:50 -0500, Jilllian <jillk...@accesstoledo.com> wrote:
> I'm have to debate the pros of gun control for one of my classes.

That's kinda rough, isn't it? As a student, shouldn't you be able to
start by debating the amateurs?

Mary Rosh

unread,
Nov 9, 2001, 8:25:17 AM11/9/01
to
Tim Lambert <lam...@cse.unsw.EDU.AU> wrote in message news:<tnpu6sh...@oolong.orchestra.cse.unsw.EDU.AU>...

Anyone who has the book can see that Lott's entire discussion is not
on Lambert's web site.

userb3

unread,
Nov 9, 2001, 11:27:31 AM11/9/01
to
On 09 Nov 2001 02:53:31 GMT, Jaquandor wrote:

>So you read not more than the intro to the very first page?

I read much further, and didn't find enough substance to warrant
further commentary.

userb3

unread,
Nov 9, 2001, 11:38:56 AM11/9/01
to
On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:

>It would move the discussion on just a tad if you gave us an example
>of what you considered poor scholarship on my part.

I'm not especially interested in a comprehensive critique of your
presentation. I can, however direct you to

http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm

Andrew Ryan Chang

unread,
Nov 9, 2001, 2:05:02 PM11/9/01
to
userb3 <use...@yahoo.com> wrote:
>On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:
>>It would move the discussion on just a tad if you gave us an example
>>of what you considered poor scholarship on my part.
>
>I'm not especially interested in a comprehensive critique of your
>presentation. I can, however direct you to
>
>http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm

As far as I can tell, it doesn't address Lambert's 3rd point
("correlation != causation") at all. His closing sentence is in fact
directly answered in that third section, in fact. It may be that Lambert
added that entire section after Lott's response (I dunno) but all the
same, it means that Lott's response currently does not address all of
Lambert's points.

--
"Ed is the standard text editor."

Ed, the greatest WYGIWYG editor of all.
-Patrick J. Lopresti

Mary Rosh

unread,
Nov 9, 2001, 8:43:46 PM11/9/01
to
arc...@sfu.ca (Andrew Ryan Chang) wrote in message news:<9sh9cu$s7b$1...@morgoth.sfu.ca>...

> userb3 <use...@yahoo.com> wrote:
> >On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:
> >>It would move the discussion on just a tad if you gave us an example
> >>of what you considered poor scholarship on my part.
> >
> >I'm not especially interested in a comprehensive critique of your
> >presentation. I can, however direct you to
> >
> >http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm
>
> As far as I can tell, it doesn't address Lambert's 3rd point
> ("correlation != causation") at all. His closing sentence is in fact
> directly answered in that third section, in fact. It may be that Lambert
> added that entire section after Lott's response (I dunno) but all the
> same, it means that Lott's response currently does not address all of
> Lambert's points.

I don't trust Lamber at all given how he mischaracterizes Lott's
positions, but, in any case, Lott's book has a long discussion about
this issue and he addresses it in his book in response to others who
have raised it. His evidence isn't simply that crime rates fall in
different states after the adoption of right to carry laws or that it
rises after other types of gun control laws are passed. In the case
of right-to-carry laws, the change in crime rates correspond very
closely with the number of permits issued. Those states that issue
the most permits have the biggest drops in crime and as more permits
are issued over time the size of the drop increases. Lott also points
out important differences across different types of crime. For
example, multiple victim public shootings are shown to decline by much
more than murders and that is exactly what the theory predicts. When
you have a lot of potential victims in a public place even if the
probability that any individual has a permit is small, the probability
that at least someone will be able to defend themselves is very high
and thus there is a greater level of deterrence. Violent crime rates
also decline relative to property crimes and again Lott argues that
this is what the theory predicts because violent crimes involve
contact between victims and criminals where the presence of a
concealed handgun can make a difference. Lott also makes a lot about
the work where he compares counties that border each other on opposite
sides of state lines where he sees what happens to the crime rates in
both the county that is in a state with the law and the adjacent
county in a state without the law. He finds that not only does the
violent crime rate fall for the county in a state with the law, but
just at that same time there is an increase in the crime rate in the
boarder county. He claims that the size of the increase is greatest
when you have two urban counties next to each other. There is also a
spillover for other types of counties but it is generally largest he
claims when the adjacent counties are most similar to each other. He
argues that some fraction of criminals stop committing crime after the
concealed handgun laws are passed, some switch to other types of
crime, and some move to adjacent areas where the cost of committing
crime is lowest. What Lott claims is that the more different types of
"qualitatively" different evidence that you give the more difficult it
is to think of an alternative explanation for why the effects are
occurring. Lott also points out how he and others have replicated the
results as more years have gone by. If it were simply by accident
over the 1987 to 1992 period, there should be little reason to see it
when you look at additional years. But he extended it to 1994 in the
first addition and then to 1996 in the second and he has a paper up on
the SSRN web site where he talks about the data through 1998. He has
also extended his multiple victim public shooting paper with data
through 1999. I think that Lott is very clear about how the number of
different test strengthens the confidence one can have in the
research. Why should Lott bother responding to a nothing like Lamber
who isn't in the area and who isn't particularly honest? I don't even
know why he responded to him once. In any case, if Lambert really
cared about the truth he would acknowledge that Lott has dealt
extensively with this discussion in his book. All I have done here is
parrrot what Lott wrote.

Tim Lambert

unread,
Nov 10, 2001, 10:06:52 AM11/10/01
to
mary...@aol.com (Mary Rosh) writes:

> Tim Lambert <lam...@cse.unsw.EDU.AU> wrote in message news:<tnpu6sh...@oolong.orchestra.cse.unsw.EDU.AU>...
> > mary...@aol.com (Mary Rosh) writes:
> >
> > > jaqu...@aol.comsinatra (Jaquandor) wrote in message news:<20011108083713...@mb-fo.aol.com>...
> >
> > > > If only it were that clear-cut. Anyway, here is a response to Lott:
> > > >
> > > > www.cse.unsw.edu.au/~lambert/guns/lott/
> > >
> > > Lott's book (Chp. 9) points some of the inaccuracies at that web site.
> >
> > Lott's response to my critique and my reply are available from the
> > link above. Read them both and decide for yourself who is inaccurate.

> Anyone who has the book can see that Lott's entire discussion is not


> on Lambert's web site.

Lott's response is not on my site at all. It's on David Friedman's
site. I have links to it so people can read it and make up their own
minds. Most of Lott's response to my critique also appears in chapter
9 of the second edition of his book. Lott's book is not available
on-line. His original paper is, and I have a link to it in the links
section and in the bibliography of my critique.

Tim

Tim Lambert

unread,
Nov 10, 2001, 11:07:12 AM11/10/01
to
mary...@aol.com (Mary Rosh) writes:

> arc...@sfu.ca (Andrew Ryan Chang) wrote in message news:<9sh9cu$s7b$1...@morgoth.sfu.ca>...
> > userb3 <use...@yahoo.com> wrote:
> > >On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:
> > >>It would move the discussion on just a tad if you gave us an example
> > >>of what you considered poor scholarship on my part.
> > >
> > >I'm not especially interested in a comprehensive critique of your
> > >presentation. I can, however direct you to
> > >
> > >http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm
> >
> > As far as I can tell, it doesn't address Lambert's 3rd point
> > ("correlation != causation") at all. His closing sentence is in fact
> > directly answered in that third section, in fact. It may be that Lambert
> > added that entire section after Lott's response (I dunno) but all the
> > same, it means that Lott's response currently does not address all of
> > Lambert's points.
>
> I don't trust Lamber at all given how he mischaracterizes Lott's
> positions,

I do no such thing.

> but, in any case, Lott's book has a long discussion about
> this issue and he addresses it in his book in response to others who
> have raised it.

See
http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html
for my response to that discussion.


> His evidence isn't simply that crime rates fall in
> different states after the adoption of right to carry laws or that it
> rises after other types of gun control laws are passed. In the case
> of right-to-carry laws, the change in crime rates correspond very
> closely with the number of permits issued.

It also "corresponds very closely" with the CPI. That's because both
numbers increase with time.

> Those states that issue
> the most permits have the biggest drops in crime

Not true. Lott has data on handgun permits for three states (table
4.7). This shows that the percentage of the population with permits
in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
However, this is exactly the opposite ordering of the changes in the
violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
Pennsylvania.

> Lott also points
> out important differences across different types of crime. For
> example, multiple victim public shootings are shown to decline by much
> more than murders and that is exactly what the theory predicts.

Well, no, actually. The theory can predict either a greater or lesser
decline.

> When
> you have a lot of potential victims in a public place even if the
> probability that any individual has a permit is small, the probability
> that at least someone will be able to defend themselves is very high
> and thus there is a greater level of deterrence.

Of course, the folks doing these shootings usually kill themselves, so
are unlikely to be deterred. So, the theory can predict either a
greater or lesser decline.

> Violent crime rates
> also decline relative to property crimes and again Lott argues that
> this is what the theory predicts because violent crimes involve
> contact between victims and criminals where the presence of a
> concealed handgun can make a difference.

Table 9.1 of Lott's book shows similar results for violent crime
(-2.3%) and property crime (-1.6%).

> Lott also makes a lot about
> the work where he compares counties that border each other on opposite
> sides of state lines where he sees what happens to the crime rates in
> both the county that is in a state with the law and the adjacent
> county in a state without the law. He finds that not only does the
> violent crime rate fall for the county in a state with the law, but
> just at that same time there is an increase in the crime rate in the
> boarder county.

Table 4.14 shows that there was no change in the violent
crime rate of these border counties. And I don't mean that the change
was not statistically significant. The number (middle of top row of
the table) is exactly zero. It is not possible to have a better
example of a "no effect" result".

> Lott also points out how he and others have replicated the
> results as more years have gone by. If it were simply by accident
> over the 1987 to 1992 period, there should be little reason to see it
> when you look at additional years. But he extended it to 1994 in the
> first addition and then to 1996 in the second and he has a paper up on
> the SSRN web site where he talks about the data through 1998.

Since the original data is still included, if I am right what you
would see when extra data is added is a lessening of the effect. I
reviewed Lott's results to see how they changed as the time period got
longer. Unfortunately, since the specification of the model in
table 9.1 is different from that in tables 4.8 and 4.13 the only
comparison that can be made is that between the results in table 4.8
and 4.13. This shows the effect on violent crime lessening from -0.9%
to -0.5% when the period of time studied increased by two years.

> Why should Lott bother responding to a nothing like Lamber
> who isn't in the area and who isn't particularly honest?

Your unfounded slurs reflect on your character, not mine.

> In any case, if Lambert really
> cared about the truth he would acknowledge that Lott has dealt
> extensively with this discussion in his book.

Here's the URL again:

http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html

I respond, at length, to Lott's discussion. I provide links to Lott's
response online. How can you possibly believe that I'm not
acknowledging its existence?

Tim

Mary Rosh

unread,
Nov 10, 2001, 10:37:41 PM11/10/01
to
Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3g07mv...@cycloid.localdomain>...

> mary...@aol.com (Mary Rosh) writes:
>
> > arc...@sfu.ca (Andrew Ryan Chang) wrote in message news:<9sh9cu$s7b$1...@morgoth.sfu.ca>...
> > > userb3 <use...@yahoo.com> wrote:
> > > >On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:
> > > >>It would move the discussion on just a tad if you gave us an example
> > > >>of what you considered poor scholarship on my part.
> > > >
> > > >I'm not especially interested in a comprehensive critique of your
> > > >presentation. I can, however direct you to
> > > >
> > > >http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm
> > >
> > > As far as I can tell, it doesn't address Lambert's 3rd point
> > > ("correlation != causation") at all. His closing sentence is in fact
> > > directly answered in that third section, in fact. It may be that Lambert
> > > added that entire section after Lott's response (I dunno) but all the
> > > same, it means that Lott's response currently does not address all of
> > > Lambert's points.
> >
> > I don't trust Lamber at all given how he mischaracterizes Lott's
> > positions,
>
ÿ I do no such thing.

Well, you make several misleading and inaccurate statements in this
posting.

>
> > but, in any case, Lott's book has a long discussion about
> > this issue and he addresses it in his book in response to others who
> > have raised it.
>
> See
> http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html
> for my response to that discussion.
>

Well, I would be nice if you could start by accurately discussing
Lott's work.

>
> > His evidence isn't simply that crime rates fall in
> > different states after the adoption of right to carry laws or that it
> > rises after other types of gun control laws are passed. In the case
> > of right-to-carry laws, the change in crime rates correspond very
> > closely with the number of permits issued.
>
> It also "corresponds very closely" with the CPI. That's because both

ÿ numbers increase with time.

Are you stating that the CPI only started increasing in the states
that passed the law and did so only after the laws were passed?
Because as you know, even if you are unwilling to publicly state it,
Lott finds that the trends in crime rates change before and after the
concealed handgun laws are passed. So were prices also falling in
states that passed the concealed handgun laws prior to their passage
and then prices were rising after the laws were passed? These are
also changes relative to states that didn't pass the law.

>
> > Those states that issue
> > the most permits have the biggest drops in crime
>
> Not true. Lott has data on handgun permits for three states (table
> 4.7). This shows that the percentage of the population with permits
> in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> However, this is exactly the opposite ordering of the changes in the
> violent crime rate: -4% in Florida, -3% in Oregon, and -1% in

ÿ Pennsylvania.

A typical incomplete or inaccurate misrepresentation. Lott's second
edition examines measures of permit issuance across many states.

>
> > Lott also points
> > out important differences across different types of crime. For
> > example, multiple victim public shootings are shown to decline by much
> > more than murders and that is exactly what the theory predicts.
>
> Well, no, actually. The theory can predict either a greater or lesser

ÿ decline.

If the theory is that the greater the level of deterrence, the greater
the relative drop in violent crime, the results conform very well to
that.

>
> > When
> > you have a lot of potential victims in a public place even if the
> > probability that any individual has a permit is small, the probability
> > that at least someone will be able to defend themselves is very high
> > and thus there is a greater level of deterrence.
>
> Of course, the folks doing these shootings usually kill themselves, so
> are unlikely to be deterred. So, the theory can predict either a

ÿ greater or lesser decline.

As usual you don't quite get things right. Lott and Landes say that
these killer frequently die either by being killed by others or
through their own hand. If you have other evidence you should state
what it is rather than making assertions. This little
misrepresentation is crucial as usual to your claims. But in any
case, the data is very consistent with greater probabilities of
stopping these killers from accomplishing their goals does have a
greater deterrence.

>
> > Violent crime rates
> > also decline relative to property crimes and again Lott argues that
> > this is what the theory predicts because violent crimes involve
> > contact between victims and criminals where the presence of a
> > concealed handgun can make a difference.
>
> Table 9.1 of Lott's book shows similar results for violent crime

ÿ (-2.3%) and property crime (-1.6%).

Lott provides a lot of results, but in all cases the violent crime
drop is greater than for property crimes. As is the case even for the
one result that you pick.

>
> > Lott also makes a lot about
> > the work where he compares counties that border each other on opposite
> > sides of state lines where he sees what happens to the crime rates in
> > both the county that is in a state with the law and the adjacent
> > county in a state without the law. He finds that not only does the
> > violent crime rate fall for the county in a state with the law, but
> > just at that same time there is an increase in the crime rate in the
> > boarder county.
>
> Table 4.14 shows that there was no change in the violent
> crime rate of these border counties. And I don't mean that the change
> was not statistically significant. The number (middle of top row of
> the table) is exactly zero. It is not possible to have a better

ÿ example of a "no effect" result".

Typical misrepresentation. Lott has written over-and-over again about
the differences between differences in averages and trends. If anyone
looks at the year by year changes in the figures (see especially the
second edition), you see the increase in violent crime rates starting
right at the time their neighbors adopt their law.

>
> > Lott also points out how he and others have replicated the
> > results as more years have gone by. If it were simply by accident
> > over the 1987 to 1992 period, there should be little reason to see it
> > when you look at additional years. But he extended it to 1994 in the
> > first addition and then to 1996 in the second and he has a paper up on
> > the SSRN web site where he talks about the data through 1998.
>
> Since the original data is still included, if I am right what you
> would see when extra data is added is a lessening of the effect. I
> reviewed Lott's results to see how they changed as the time period got
> longer. Unfortunately, since the specification of the model in
> table 9.1 is different from that in tables 4.8 and 4.13 the only
> comparison that can be made is that between the results in table 4.8
> and 4.13. This shows the effect on violent crime lessening from -0.9%
> to -0.5% when the period of time studied increased by two years.
>

When Lott added the data for the 1993 to 1994 period the strength of
the results went up. When the additional two years were added they
tended to back down, but again if you were even remotely honest you
would mention that Lott discussed how the states that added their laws
in later years had relatively more restrictive laws and tended to
issue fewer permits, there was thus a smaller expected drop in crime
rates for those later states.

>
>
> > Why should Lott bother responding to a nothing like Lamber
> > who isn't in the area and who isn't particularly honest?
>

ÿ Your unfounded slurs reflect on your character, not mine.

I believe that your continued misrepresentations and selective
interpretations of Lott's work shows that it is you has the character
problem.

>
> > In any case, if Lambert really
> > cared about the truth he would acknowledge that Lott has dealt
> > extensively with this discussion in his book.
>
> Here's the URL again:
>
> http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html
>
> I respond, at length, to Lott's discussion. I provide links to Lott's
> response online. How can you possibly believe that I'm not
> acknowledging its existence?
>

As I have shown here, you may write a lot, but you consistently
misrepresent what Lott wrote.

Tim Lambert

unread,
Nov 12, 2001, 12:16:09 PM11/12/01
to
Mary's standard tactic is to throw out many accusations of
dishonesty. When these accusations are proven false, instead of doing
the decent thing and apologizing, she justs offends again, throwing
out more false accusations.

For example, Mary insinuated that I had deliberately ignored Lott's
discussion on causation when I had in fact already directly responded
to it. When this was pointed out to her, she failed to apologize, and
instead threw out more false accusations.

Another example (from another thread) was her response to Kleck's
finding that surveys indicated that 5-11% of US adults admit to
carrying guns for self-protection, dwarfing the 1% or so of the
population that obtained concealed-weapon permits. Mary insisted that
this number was including mere transportation of guns and not just
carrying for protection, and claimed that the statistic was made up.
When I posted the exact page number of Kleck's book that the statistic
came from, Mary did not apologize or admit to any error.

Now, on to a posting chock full of Mary Rosh's false accusations.

mary...@aol.com (Mary Rosh) writes:

> > > His evidence isn't simply that crime rates fall in
> > > different states after the adoption of right to carry laws or that it
> > > rises after other types of gun control laws are passed. In the case
> > > of right-to-carry laws, the change in crime rates correspond very
> > > closely with the number of permits issued.
> >
> > It also "corresponds very closely" with the CPI. That's because both
> ÿ numbers increase with time.
>
> Are you stating that the CPI only started increasing in the states
> that passed the law and did so only after the laws were passed?

No.



> Because as you know, even if you are unwilling to publicly state it,
> Lott finds that the trends in crime rates change before and after the
> concealed handgun laws are passed.

That's a different issue. I was addressing your claim that "the change


in crime rates correspond very closely with the number of permits issued."

> So were prices also falling in


> states that passed the concealed handgun laws prior to their passage
> and then prices were rising after the laws were passed?

Oh, so are you claiming that the number of permits was falling in states
that passed the concealed handgun laws prior to their passage?

> > > Those states that issue
> > > the most permits have the biggest drops in crime
> >
> > Not true. Lott has data on handgun permits for three states (table
> > 4.7). This shows that the percentage of the population with permits
> > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > However, this is exactly the opposite ordering of the changes in the
> > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
> ÿ Pennsylvania.
>
> A typical incomplete or inaccurate misrepresentation. Lott's second
> edition examines measures of permit issuance across many states.

Oh, then it should be very easy for you to post the data on the permit
rates from Lott's book for those other states. Put up or shut up.


> > > Lott also points
> > > out important differences across different types of crime. For
> > > example, multiple victim public shootings are shown to decline by much
> > > more than murders and that is exactly what the theory predicts.
> >
> > Well, no, actually. The theory can predict either a greater or lesser
> ÿ decline.
>
> If the theory is that the greater the level of deterrence, the greater
> the relative drop in violent crime, the results conform very well to
> that.

Sigh. You don't get it. If the theory can predict an increase or a
decrease then you can't use an increase (or a decrease) as evidence
that the theory is true.


> > > When
> > > you have a lot of potential victims in a public place even if the
> > > probability that any individual has a permit is small, the probability
> > > that at least someone will be able to defend themselves is very high
> > > and thus there is a greater level of deterrence.
> >
> > Of course, the folks doing these shootings usually kill themselves, so
> > are unlikely to be deterred. So, the theory can predict either a
> ÿ greater or lesser decline.
>
> As usual you don't quite get things right. Lott and Landes say that
> these killer frequently die either by being killed by others or
> through their own hand.

That hardly contradicts what I wrote.

> If you have other evidence you should state
> what it is rather than making assertions.

It comes from my review of a list of cases of mass public shootings.

> This little misrepresentation is crucial as usual to your claims.

It is not a misrepresentation and is not crucial in any way shape or
form. If the shooter ends up dead, whether from his own hand, or from
others, then concealed handguns cannot provide any additional penalty.


> > > Lott also makes a lot about
> > > the work where he compares counties that border each other on opposite
> > > sides of state lines where he sees what happens to the crime rates in
> > > both the county that is in a state with the law and the adjacent
> > > county in a state without the law. He finds that not only does the
> > > violent crime rate fall for the county in a state with the law, but
> > > just at that same time there is an increase in the crime rate in the
> > > boarder county.
> >
> > Table 4.14 shows that there was no change in the violent
> > crime rate of these border counties. And I don't mean that the change
> > was not statistically significant. The number (middle of top row of
> > the table) is exactly zero. It is not possible to have a better
> ÿ example of a "no effect" result".
>
> Typical misrepresentation. Lott has written over-and-over again about
> the differences between differences in averages and trends.

Lott does not provide any numbers for the trends here, just the
averages. You have a lot of gall to accuse me of misrepresention for
looking at the figures he gives and not on some nonexistant ones that
he doesn't present.

> If anyone
> looks at the year by year changes in the figures (see especially the
> second edition), you see the increase in violent crime rates starting
> right at the time their neighbors adopt their law.

Wrong. This is an example of the reification fallacy. I went into
detail about what is wrong with Lott's graphs in my critique.
Look at
http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html#SECTION00032800000000000000
They do not show that there was an increase in the violent crime rate
at the time that the law was passed. The graphs do NOT show the year
to year changes, rather they show a curve fitted to the data. On the
web page above I have a Java applet that fits similar curves to random
data. You get similar looking graphs even though the data is
completely random and did not change when the law was passed.

> > > Lott also points out how he and others have replicated the
> > > results as more years have gone by. If it were simply by accident
> > > over the 1987 to 1992 period, there should be little reason to see it
> > > when you look at additional years. But he extended it to 1994 in the
> > > first addition and then to 1996 in the second and he has a paper up on
> > > the SSRN web site where he talks about the data through 1998.
> >
> > Since the original data is still included, if I am right what you
> > would see when extra data is added is a lessening of the effect. I
> > reviewed Lott's results to see how they changed as the time period got
> > longer. Unfortunately, since the specification of the model in
> > table 9.1 is different from that in tables 4.8 and 4.13 the only
> > comparison that can be made is that between the results in table 4.8
> > and 4.13. This shows the effect on violent crime lessening from -0.9%
> > to -0.5% when the period of time studied increased by two years.

> When Lott added the data for the 1993 to 1994 period the strength of
> the results went up.

Wrong. The difference between table 4.8 and table 4.13 was just the
adding of data for 1993 and 1994.

> When the additional two years were added they
> tended to back down, but again if you were even remotely honest you
> would mention that Lott discussed how the states that added their laws
> in later years had relatively more restrictive laws and tended to
> issue fewer permits, there was thus a smaller expected drop in crime
> rates for those later states.

Oh, please. You argue that if the results were accidental then X
would happen. I point out that in fact, X did happen. Suddenly you
claim that your theory also predicts X. Even if your claim is true,
it doesn't matter. Your argument that the results weren't accidental
has been refuted.

Tim

Mary Rosh

unread,
Nov 12, 2001, 3:25:20 PM11/12/01
to
Where are you Mr. Lambert?

Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3g07mv...@cycloid.localdomain>...

> mary...@aol.com (Mary Rosh) writes:
>
> > arc...@sfu.ca (Andrew Ryan Chang) wrote in message news:<9sh9cu$s7b$1...@morgoth.sfu.ca>...
> > > userb3 <use...@yahoo.com> wrote:
> > > >On 09 Nov 2001 12:05:48 +1100, Tim Lambert wrote:
> > > >>It would move the discussion on just a tad if you gave us an example
> > > >>of what you considered poor scholarship on my part.
> > > >
> > > >I'm not especially interested in a comprehensive critique of your
> > > >presentation. I can, however direct you to
> > > >
> > > >http://www.best.com/~ddfr/Lott_v_Teret/Response_to_Lambert.htm
> > >
> > > As far as I can tell, it doesn't address Lambert's 3rd point
> > > ("correlation != causation") at all. His closing sentence is in fact
> > > directly answered in that third section, in fact. It may be that Lambert
> > > added that entire section after Lott's response (I dunno) but all the
> > > same, it means that Lott's response currently does not address all of
> > > Lambert's points.
> >
> > I don't trust Lamber at all given how he mischaracterizes Lott's
> > positions,
>

ÿ I do no such thing.

Well, you make several misleading and inaccurate statements in this
posting.

>

> > but, in any case, Lott's book has a long discussion about
> > this issue and he addresses it in his book in response to others who
> > have raised it.
>
> See
> http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html
> for my response to that discussion.
>

Well, I would be nice if you could start by accurately discussing
Lott's work.

>

> > His evidence isn't simply that crime rates fall in
> > different states after the adoption of right to carry laws or that it
> > rises after other types of gun control laws are passed. In the case
> > of right-to-carry laws, the change in crime rates correspond very
> > closely with the number of permits issued.
>
> It also "corresponds very closely" with the CPI. That's because both

ÿ numbers increase with time.

Are you stating that the CPI only started increasing in the states
that passed the law and did so only after the laws were passed?

Because as you know, even if you are unwilling to publicly state it,

Lott finds that the trends in crime rates change before and after the
concealed handgun laws are passed. So were prices also falling in


states that passed the concealed handgun laws prior to their passage

and then prices were rising after the laws were passed? These are
also changes relative to states that didn't pass the law.

>

> > Those states that issue
> > the most permits have the biggest drops in crime
>
> Not true. Lott has data on handgun permits for three states (table
> 4.7). This shows that the percentage of the population with permits
> in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> However, this is exactly the opposite ordering of the changes in the
> violent crime rate: -4% in Florida, -3% in Oregon, and -1% in

ÿ Pennsylvania.

A typical incomplete or inaccurate misrepresentation. Lott's second
edition examines measures of permit issuance across many states.

>

> > Lott also points
> > out important differences across different types of crime. For
> > example, multiple victim public shootings are shown to decline by much
> > more than murders and that is exactly what the theory predicts.
>
> Well, no, actually. The theory can predict either a greater or lesser

ÿ decline.

If the theory is that the greater the level of deterrence, the greater
the relative drop in violent crime, the results conform very well to
that.

>

> > When
> > you have a lot of potential victims in a public place even if the
> > probability that any individual has a permit is small, the probability
> > that at least someone will be able to defend themselves is very high
> > and thus there is a greater level of deterrence.
>
> Of course, the folks doing these shootings usually kill themselves, so
> are unlikely to be deterred. So, the theory can predict either a

ÿ greater or lesser decline.

As usual you don't quite get things right. Lott and Landes say that
these killer frequently die either by being killed by others or

through their own hand. If you have other evidence you should state
what it is rather than making assertions. This little
misrepresentation is crucial as usual to your claims. But in any
case, the data is very consistent with greater probabilities of
stopping these killers from accomplishing their goals does have a
greater deterrence.

>

> > Violent crime rates
> > also decline relative to property crimes and again Lott argues that
> > this is what the theory predicts because violent crimes involve
> > contact between victims and criminals where the presence of a
> > concealed handgun can make a difference.
>
> Table 9.1 of Lott's book shows similar results for violent crime

ÿ (-2.3%) and property crime (-1.6%).

Lott provides a lot of results, but in all cases the violent crime
drop is greater than for property crimes. As is the case even for the
one result that you pick.

>

> > Lott also makes a lot about
> > the work where he compares counties that border each other on opposite
> > sides of state lines where he sees what happens to the crime rates in
> > both the county that is in a state with the law and the adjacent
> > county in a state without the law. He finds that not only does the
> > violent crime rate fall for the county in a state with the law, but
> > just at that same time there is an increase in the crime rate in the
> > boarder county.
>
> Table 4.14 shows that there was no change in the violent
> crime rate of these border counties. And I don't mean that the change
> was not statistically significant. The number (middle of top row of
> the table) is exactly zero. It is not possible to have a better

ÿ example of a "no effect" result".

Typical misrepresentation. Lott has written over-and-over again about

the differences between differences in averages and trends. If anyone


looks at the year by year changes in the figures (see especially the
second edition), you see the increase in violent crime rates starting
right at the time their neighbors adopt their law.

>

> > Lott also points out how he and others have replicated the
> > results as more years have gone by. If it were simply by accident
> > over the 1987 to 1992 period, there should be little reason to see it
> > when you look at additional years. But he extended it to 1994 in the
> > first addition and then to 1996 in the second and he has a paper up on
> > the SSRN web site where he talks about the data through 1998.
>
> Since the original data is still included, if I am right what you
> would see when extra data is added is a lessening of the effect. I
> reviewed Lott's results to see how they changed as the time period got
> longer. Unfortunately, since the specification of the model in
> table 9.1 is different from that in tables 4.8 and 4.13 the only
> comparison that can be made is that between the results in table 4.8
> and 4.13. This shows the effect on violent crime lessening from -0.9%
> to -0.5% when the period of time studied increased by two years.
>

When Lott added the data for the 1993 to 1994 period the strength of
the results went up. When the additional two years were added they


tended to back down, but again if you were even remotely honest you
would mention that Lott discussed how the states that added their laws
in later years had relatively more restrictive laws and tended to
issue fewer permits, there was thus a smaller expected drop in crime
rates for those later states.

>
>

> > Why should Lott bother responding to a nothing like Lamber
> > who isn't in the area and who isn't particularly honest?
>

ÿ Your unfounded slurs reflect on your character, not mine.

I believe that your continued misrepresentations and selective
interpretations of Lott's work shows that it is you has the character
problem.

>

> > In any case, if Lambert really
> > cared about the truth he would acknowledge that Lott has dealt
> > extensively with this discussion in his book.
>
> Here's the URL again:
>
> http://www.cse.unsw.edu.au/~lambert/guns/lott/node18.html
>
> I respond, at length, to Lott's discussion. I provide links to Lott's
> response online. How can you possibly believe that I'm not
> acknowledging its existence?
>

As I have shown here, you may write a lot, but you consistently
misrepresent what Lott wrote.

david

unread,
Nov 12, 2001, 8:39:07 PM11/12/01
to
On 13 Nov 2001 04:16:09 +1100, Tim Lambert <lam...@cse.unsw.edu.au> wrote:
>
> Another example (from another thread) was her response to Kleck's
> finding that surveys indicated that 5-11% of US adults admit to
> carrying guns for self-protection, dwarfing the 1% or so of the
> population that obtained concealed-weapon permits. Mary insisted that
> this number was including mere transportation of guns and not just
> carrying for protection, and claimed that the statistic was made up.
> When I posted the exact page number of Kleck's book that the statistic
> came from, Mary did not apologize or admit to any error.

Since you've brought it up, I've always wondered what this statistic
refers to. Since you've mentioned the book, could you clear it up?

Is it a national survey, or regional? Does it distinguish between
carrying a gun on one's person as opposed to carrying it in a vehicle?
And does it break down the numbers by regularity, e.g., does the
original study distinguish between the guy who carries a gun daily as
opposed to a shop owner who carries a gun when he takes his monthly
income to the bank? Or specifically, what was the sample size, was it
randomly chosen, and what exactly was the question asked?

I'm not arguing either side of this issue. I've just seen this stat a
few times and I've always wondered what exactly the study tried to
measure, and you seem to know a bit about it.


Mary Rosh

unread,
Nov 12, 2001, 10:00:35 PM11/12/01
to
Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3n11r2...@cycloid.localdomain>...

> Mary's standard tactic is to throw out many accusations of
> dishonesty. When these accusations are proven false, instead of doing
> the decent thing and apologizing, she justs offends again, throwing
> out more false accusations.
>
> For example, Mary insinuated that I had deliberately ignored Lott's
> discussion on causation when I had in fact already directly responded
> to it. When this was pointed out to her, she failed to apologize, and
> instead threw out more false accusations.
>
> Another example (from another thread) was her response to Kleck's
> finding that surveys indicated that 5-11% of US adults admit to
> carrying guns for self-protection, dwarfing the 1% or so of the
> population that obtained concealed-weapon permits. Mary insisted that
> this number was including mere transportation of guns and not just
> carrying for protection, and claimed that the statistic was made up.
> When I posted the exact page number of Kleck's book that the statistic
> came from, Mary did not apologize or admit to any error.
>

These surveys are not based upon concealed carry but based upon any
carrying of a gun. Lott addresses this in his book p. 230. These are
not surveys for self defense as you claimed, but surveys involving any
type of carrying guns for any reason (e.g., hunting, moving
residences, etc.).


> Now, on to a posting chock full of Mary Rosh's false accusations.
>
> mary...@aol.com (Mary Rosh) writes:
>
> > > > His evidence isn't simply that crime rates fall in
> > > > different states after the adoption of right to carry laws or that it
> > > > rises after other types of gun control laws are passed. In the case
> > > > of right-to-carry laws, the change in crime rates correspond very
> > > > closely with the number of permits issued.
> > >
> > > It also "corresponds very closely" with the CPI. That's because both
> > ÿ numbers increase with time.
> >
> > Are you stating that the CPI only started increasing in the states
> > that passed the law and did so only after the laws were passed?
>
> No.

But prices are going up the entire time and concealed handgun permits
are not. So your comparison makes absolutely no sense.

>
> > Because as you know, even if you are unwilling to publicly state it,
> > Lott finds that the trends in crime rates change before and after the
> > concealed handgun laws are passed.
>
> That's a different issue. I was addressing your claim that "the change
> in crime rates correspond very closely with the number of permits
> issued."

that is not a different issue. You claim that Lott might as well be
picking the CPI, yet the CPI could not have given the type of pattern
that Lott found. In addition, the CPI is rising for all states, but
the number of permits are only increasing (starting in different
years) for a set of states.

>
> > So were prices also falling in
> > states that passed the concealed handgun laws prior to their passage
> > and then prices were rising after the laws were passed?
>
> Oh, so are you claiming that the number of permits was falling in states
> that passed the concealed handgun laws prior to their passage?

Are you on drugs? The only way you could get a constant negative
relationship between prices and crime rates is if prices were falling
prior to the adoption of concealed handgun laws and rising afterward.
In any case, your reference to the CPI is nonsensical because price
changes are going to be very highly correlated across states over
time. In you world of price changes, prices were falling in Texas
prior to 1996 and simultanously rising in florida and other states
starting in 1987.

>
> > > > Those states that issue
> > > > the most permits have the biggest drops in crime
> > >
> > > Not true. Lott has data on handgun permits for three states (table
> > > 4.7). This shows that the percentage of the population with permits
> > > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > > However, this is exactly the opposite ordering of the changes in the
> > > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
> > ÿ Pennsylvania.
> >
> > A typical incomplete or inaccurate misrepresentation. Lott's second
> > edition examines measures of permit issuance across many states.
>
> Oh, then it should be very easy for you to post the data on the permit
> rates from Lott's book for those other states. Put up or shut up.

The discussion is on pages 174 to 177 in Lott's book. So is the list
of states.

>
>
> > > > Lott also points
> > > > out important differences across different types of crime. For
> > > > example, multiple victim public shootings are shown to decline by much
> > > > more than murders and that is exactly what the theory predicts.
> > >
> > > Well, no, actually. The theory can predict either a greater or lesser
> > ÿ decline.
> >
> > If the theory is that the greater the level of deterrence, the greater
> > the relative drop in violent crime, the results conform very well to
> > that.
>
> Sigh. You don't get it. If the theory can predict an increase or a
> decrease then you can't use an increase (or a decrease) as evidence
> that the theory is true.

A theory predicts a relative drop in violent crime, I then look to see
if the relative drop occurred as confirmation of the theory.

>
>
> > > > When
> > > > you have a lot of potential victims in a public place even if the
> > > > probability that any individual has a permit is small, the probability
> > > > that at least someone will be able to defend themselves is very high
> > > > and thus there is a greater level of deterrence.
> > >
> > > Of course, the folks doing these shootings usually kill themselves, so
> > > are unlikely to be deterred. So, the theory can predict either a
> > ÿ greater or lesser decline.
> >
> > As usual you don't quite get things right. Lott and Landes say that
> > these killer frequently die either by being killed by others or
> > through their own hand.
>
> That hardly contradicts what I wrote.


You want to claim that there is no deterrence effect because everyone
was going to commit suicide. You therefore claim that most commit
suicide. The evidence in fact says that a lot are killed on the
scene, which you conveniently left out. Getting killed seems like
deterrence to me, though you might not believe that the death penalty
is deterrence.

>
> > If you have other evidence you should state
> > what it is rather than making assertions.
>

ÿ It comes from my review of a list of cases of mass public shootings.


Given how inaccurate you are about most things, I see no reason for
trusting your informal "review" of the data.

>
> > This little misrepresentation is crucial as usual to your claims.
>
> It is not a misrepresentation and is not crucial in any way shape or
> form. If the shooter ends up dead, whether from his own hand, or from
> others, then concealed handguns cannot provide any additional penalty.
>
>

It is a misrepresentation if you believe that the loss of life is a
deterrence.


> > > > Lott also makes a lot about
> > > > the work where he compares counties that border each other on opposite
> > > > sides of state lines where he sees what happens to the crime rates in
> > > > both the county that is in a state with the law and the adjacent
> > > > county in a state without the law. He finds that not only does the
> > > > violent crime rate fall for the county in a state with the law, but
> > > > just at that same time there is an increase in the crime rate in the
> > > > boarder county.
> > >
> > > Table 4.14 shows that there was no change in the violent
> > > crime rate of these border counties. And I don't mean that the change
> > > was not statistically significant. The number (middle of top row of
> > > the table) is exactly zero. It is not possible to have a better
> > ÿ example of a "no effect" result".
> >
> > Typical misrepresentation. Lott has written over-and-over again about
> > the differences between differences in averages and trends.
>
> Lott does not provide any numbers for the trends here, just the
> averages. You have a lot of gall to accuse me of misrepresention for
> looking at the figures he gives and not on some nonexistant ones that
> he doesn't present.
>

Are you serious? The graphs provide you with obvious information on
the change in trends. Why would one believe the simple averages and
not the trends? Somehow Lott will get the one accurate and not the
other? In any case, when you are dealing with more complicated trends
like the case of assault it is difficult to summarize it with a simple
number.

You however are dishonest because Lott provides the before and after
averages but he continually explains why one should rely on more
complicated information regarding trends.

> > If anyone
> > looks at the year by year changes in the figures (see especially the
> > second edition), you see the increase in violent crime rates starting
> > right at the time their neighbors adopt their law.
>
> Wrong. This is an example of the reification fallacy. I went into
> detail about what is wrong with Lott's graphs in my critique.
> Look at
> http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html#SECTION00032800000000000000
> They do not show that there was an increase in the violent crime rate
> at the time that the law was passed. The graphs do NOT show the year
> to year changes, rather they show a curve fitted to the data. On the
> web page above I have a Java applet that fits similar curves to random
> data. You get similar looking graphs even though the data is
> completely random and did not change when the law was passed.


Lott has done this both ways with most of his data. In any case these
are nonlinear lines. Random data is a joke. If it is random you are
very unlikely to get crime rates falling before and rising after.
That is why Lott provides things like f-tests to see whether these
types of changes are random or not. Lott mentions that this paper
with Bronars was published in the American Economic Review and you can
look up the detailed statistical tests there if you are interested.


>
> > > > Lott also points out how he and others have replicated the
> > > > results as more years have gone by. If it were simply by accident
> > > > over the 1987 to 1992 period, there should be little reason to see it
> > > > when you look at additional years. But he extended it to 1994 in the
> > > > first addition and then to 1996 in the second and he has a paper up on
> > > > the SSRN web site where he talks about the data through 1998.
> > >
> > > Since the original data is still included, if I am right what you
> > > would see when extra data is added is a lessening of the effect. I
> > > reviewed Lott's results to see how they changed as the time period got
> > > longer. Unfortunately, since the specification of the model in
> > > table 9.1 is different from that in tables 4.8 and 4.13 the only
> > > comparison that can be made is that between the results in table 4.8
> > > and 4.13. This shows the effect on violent crime lessening from -0.9%
> > > to -0.5% when the period of time studied increased by two years.
>
> > When Lott added the data for the 1993 to 1994 period the strength of
> > the results went up.
>
> Wrong. The difference between table 4.8 and table 4.13 was just the
> adding of data for 1993 and 1994.

Boy, you are a dishonest person. You know as well as I do the
addition for 1995 and 1996 were done in Chapter 9.

>
> > When the additional two years were added they
> > tended to back down, but again if you were even remotely honest you
> > would mention that Lott discussed how the states that added their laws
> > in later years had relatively more restrictive laws and tended to
> > issue fewer permits, there was thus a smaller expected drop in crime
> > rates for those later states.
>
> Oh, please. You argue that if the results were accidental then X
> would happen. I point out that in fact, X did happen. Suddenly you
> claim that your theory also predicts X. Even if your claim is true,
> it doesn't matter. Your argument that the results weren't accidental
> has been refuted.

Lott has a theory that the rate of permit issuance is important in
explaining the degree of deterrence. This is not a new theory by Lott
but has been what he has argued all along. The evidence compiled in
chapter 9 confirms that by showing that those states that have the
strickest issuing rules issue the smallest number of permits and have
smaller drops in violent crime. You claim to have read Lott's work.
I believethat you have, but unfortunately if I believe that I can't
simultaneously believe that you are an honest person because Lott
makes these arguments.

Tim Lambert

unread,
Nov 13, 2001, 7:19:54 AM11/13/01
to
david <da...@example.com> writes:

> On 13 Nov 2001 04:16:09 +1100, Tim Lambert <lam...@cse.unsw.edu.au> wrote:
> >
> > Another example (from another thread) was her response to Kleck's
> > finding that surveys indicated that 5-11% of US adults admit to
> > carrying guns for self-protection, dwarfing the 1% or so of the
> > population that obtained concealed-weapon permits. Mary insisted that
> > this number was including mere transportation of guns and not just
> > carrying for protection, and claimed that the statistic was made up.
> > When I posted the exact page number of Kleck's book that the statistic
> > came from, Mary did not apologize or admit to any error.
>
> Since you've brought it up, I've always wondered what this statistic
> refers to. Since you've mentioned the book, could you clear it up?
>
> Is it a national survey, or regional?

It is a summary of the results of seven national surveys.


> Does it distinguish between
> carrying a gun on one's person as opposed to carrying it in a vehicle?

Two of the surveys made that distinction

> And does it break down the numbers by regularity, e.g., does the
> original study distinguish between the guy who carries a gun daily as
> opposed to a shop owner who carries a gun when he takes his monthly
> income to the bank?

One survey asked how often the gun was carried for protection.

> Or specifically, what was the sample size, was it
> randomly chosen, and what exactly was the question asked?

The survey that gave the most details was Kleck's own survey. This
had a random sample of 5000 adults. The specific question asked was:
"In the last twelve months, have you ever carried a gun away from
home, either on your person, or in a vehicle, for protection against
crime? Do not count carrying for recreation or in connection with
duites in law enforcemnt, work as a security guard, or in the armed
forces" If the answer was "yes", there were follow up questions on
whether it was in a vehicle, and on how frequently the gun was carried
for protection.



> I'm not arguing either side of this issue. I've just seen this stat a
> few times and I've always wondered what exactly the study tried to
> measure, and you seem to know a bit about it.

That's cool. Now you know more about it than Lott does.

Tim

Tim Lambert

unread,
Nov 13, 2001, 10:15:22 AM11/13/01
to
mary...@aol.com (Mary Rosh) writes:

> Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3n11r2...@cycloid.localdomain>...
> > Mary's standard tactic is to throw out many accusations of
> > dishonesty. When these accusations are proven false, instead of doing
> > the decent thing and apologizing, she justs offends again, throwing
> > out more false accusations.

> > Another example (from another thread) was her response to Kleck's


> > finding that surveys indicated that 5-11% of US adults admit to
> > carrying guns for self-protection, dwarfing the 1% or so of the
> > population that obtained concealed-weapon permits. Mary insisted that
> > this number was including mere transportation of guns and not just
> > carrying for protection, and claimed that the statistic was made up.
> > When I posted the exact page number of Kleck's book that the statistic
> > came from, Mary did not apologize or admit to any error.
> >
>
> These surveys are not based upon concealed carry but based upon any
> carrying of a gun. Lott addresses this in his book p. 230. These are
> not surveys for self defense as you claimed, but surveys involving any
> type of carrying guns for any reason (e.g., hunting, moving
> residences, etc.).

Yes, Lott makes that claim, but anyone who reads chapter 6 of
"Targetting Guns" can see that Lott does not know what he is talking
about. I am surprised to find that Lott is unaware of current
research on the frequency with which guns are carried for
self-protection.


> > mary...@aol.com (Mary Rosh) writes:

> > > > > In the case of right-to-carry laws, the change in crime
> > > > > rates correspond very closely with the number of permits
> > > > > issued.

> > > So were prices also falling in


> > > states that passed the concealed handgun laws prior to their passage
> > > and then prices were rising after the laws were passed?
> >
> > Oh, so are you claiming that the number of permits was falling in states
> > that passed the concealed handgun laws prior to their passage?
>
> Are you on drugs? The only way you could get a constant negative
> relationship between prices and crime rates is if prices were falling
> prior to the adoption of concealed handgun laws and rising afterward.

Sigh. You have claimed that "the change in crime rates correspond
very closely with the number of permits issued." You have also
asserted that the crime rate was rising before the laws were passed.
If both of your statements are true, then it follows that the number
of permits was decreasing before the laws were passed.

> > > > > Those states that issue
> > > > > the most permits have the biggest drops in crime
> > > >
> > > > Not true. Lott has data on handgun permits for three states (table
> > > > 4.7). This shows that the percentage of the population with permits
> > > > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > > > However, this is exactly the opposite ordering of the changes in the
> > > > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
> > > ÿ Pennsylvania.
> > >
> > > A typical incomplete or inaccurate misrepresentation. Lott's second
> > > edition examines measures of permit issuance across many states.
> >
> > Oh, then it should be very easy for you to post the data on the permit
> > rates from Lott's book for those other states. Put up or shut up.
>
> The discussion is on pages 174 to 177 in Lott's book. So is the list
> of states.

Those pages do not contain data on permit rates for any state. The
only data Lott provides on actual permit rates is in table 4.7 as I
have posted. Table 4.7 contains data for three sites and I included
all of them in my analysis. Your assertion that my analysis is
incomplete is false. If you are privy to some data that does not
appear in Lott's book, please share it with us.


> > > > > When
> > > > > you have a lot of potential victims in a public place even if the
> > > > > probability that any individual has a permit is small, the probability
> > > > > that at least someone will be able to defend themselves is very high
> > > > > and thus there is a greater level of deterrence.
> > > >
> > > > Of course, the folks doing these shootings usually kill themselves, so
> > > > are unlikely to be deterred. So, the theory can predict either a
> > > ÿ greater or lesser decline.
> > >
> > > As usual you don't quite get things right. Lott and Landes say that
> > > these killer frequently die either by being killed by others or
> > > through their own hand.
> >
> > That hardly contradicts what I wrote.
>
>
> You want to claim that there is no deterrence effect because everyone
> was going to commit suicide.

"everyone was going to commit suicide"??? Where did I say that?



> > > > > Lott also makes a lot about
> > > > > the work where he compares counties that border each other on opposite
> > > > > sides of state lines where he sees what happens to the crime rates in
> > > > > both the county that is in a state with the law and the adjacent
> > > > > county in a state without the law. He finds that not only does the
> > > > > violent crime rate fall for the county in a state with the law, but
> > > > > just at that same time there is an increase in the crime rate in the
> > > > > boarder county.
> > > >
> > > > Table 4.14 shows that there was no change in the violent
> > > > crime rate of these border counties. And I don't mean that the change
> > > > was not statistically significant. The number (middle of top row of
> > > > the table) is exactly zero. It is not possible to have a better
> > > ÿ example of a "no effect" result".
> > >
> > > Typical misrepresentation. Lott has written over-and-over again about
> > > the differences between differences in averages and trends.
> >
> > Lott does not provide any numbers for the trends here, just the
> > averages. You have a lot of gall to accuse me of misrepresention for
> > looking at the figures he gives and not on some nonexistant ones that
> > he doesn't present.

> Are you serious? The graphs provide you with obvious information on
> the change in trends. Why would one believe the simple averages and
> not the trends? Somehow Lott will get the one accurate and not the
> other?

I already pointed out that Lott only presents numbers for averages
here and not for trends. It is extremely silly for you to acuse me of
misrepresentation for looking at the numbers that Lott presents
instead of ones he doesn't present.

> You however are dishonest because Lott provides the before and after
> averages but he continually explains why one should rely on more
> complicated information regarding trends.

Gee, then why doesn't he present that trend data?


> > > If anyone
> > > looks at the year by year changes in the figures (see especially the
> > > second edition), you see the increase in violent crime rates starting
> > > right at the time their neighbors adopt their law.
> >
> > Wrong. This is an example of the reification fallacy. I went into
> > detail about what is wrong with Lott's graphs in my critique.
> > Look at
> > http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html#SECTION00032800000000000000
> > They do not show that there was an increase in the violent crime rate
> > at the time that the law was passed. The graphs do NOT show the year
> > to year changes, rather they show a curve fitted to the data. On the
> > web page above I have a Java applet that fits similar curves to random
> > data. You get similar looking graphs even though the data is
> > completely random and did not change when the law was passed.
>
>
> Lott has done this both ways with most of his data. In any case these
> are nonlinear lines.

I never heard of a nonlinear line before. Is it something like a
square circle?

> Random data is a joke. If it is random you are
> very unlikely to get crime rates falling before and rising after.

Random data will show crime rates falling before and rising after 25%
of the time. You really need to study some basic probability.

> That is why Lott provides things like f-tests to see whether these
> types of changes are random or not. Lott mentions that this paper
> with Bronars was published in the American Economic Review and you can
> look up the detailed statistical tests there if you are interested.

Well, some of them. They seem to have left out the results of the
test on the violent crime rate. I wonder why?

> > > > > Lott also points out how he and others have replicated the
> > > > > results as more years have gone by. If it were simply by accident
> > > > > over the 1987 to 1992 period, there should be little reason to see it
> > > > > when you look at additional years. But he extended it to 1994 in the
> > > > > first addition and then to 1996 in the second and he has a paper up on
> > > > > the SSRN web site where he talks about the data through 1998.
> > > >
> > > > Since the original data is still included, if I am right what you
> > > > would see when extra data is added is a lessening of the effect. I
> > > > reviewed Lott's results to see how they changed as the time period got
> > > > longer. Unfortunately, since the specification of the model in
> > > > table 9.1 is different from that in tables 4.8 and 4.13 the only
> > > > comparison that can be made is that between the results in table 4.8
> > > > and 4.13. This shows the effect on violent crime lessening from -0.9%
> > > > to -0.5% when the period of time studied increased by two years.
> >
> > > When Lott added the data for the 1993 to 1994 period the strength of
> > > the results went up.
> >
> > Wrong. The difference between table 4.8 and table 4.13 was just the
> > adding of data for 1993 and 1994.
>
> Boy, you are a dishonest person. You know as well as I do the
> addition for 1995 and 1996 were done in Chapter 9.

Huh? You claimed, quite falsely, that adding 1993 and 1994 cause the
strength of the results to go up. You just have to look at tables
4.8 and 4.13 to see this. 1995 and 1996 are different years from 1993
and 1994. You are making even less sense than usual. Or is is this
one of your nonlinear lines?

> > > When the additional two years were added they
> > > tended to back down, but again if you were even remotely honest you
> > > would mention that Lott discussed how the states that added their laws
> > > in later years had relatively more restrictive laws and tended to
> > > issue fewer permits, there was thus a smaller expected drop in crime
> > > rates for those later states.
> >
> > Oh, please. You argue that if the results were accidental then X
> > would happen. I point out that in fact, X did happen. Suddenly you
> > claim that your theory also predicts X. Even if your claim is true,
> > it doesn't matter. Your argument that the results weren't accidental
> > has been refuted.
>
> Lott has a theory that the rate of permit issuance is important in
> explaining the degree of deterrence. This is not a new theory by Lott
> but has been what he has argued all along. The evidence compiled in
> chapter 9 confirms that by showing that those states that have the
> strickest issuing rules issue the smallest number of permits and have
> smaller drops in violent crime.

It doesn't show that at all. You're committing the reification
fallacy again.

Tim

Mary Rosh

unread,
Nov 13, 2001, 5:00:21 PM11/13/01
to
Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3adxqn...@cycloid.localdomain>...

> mary...@aol.com (Mary Rosh) writes:
>
> > Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3n11r2...@cycloid.localdomain>...
> > > Mary's standard tactic is to throw out many accusations of
> > > dishonesty. When these accusations are proven false, instead of doing
> > > the decent thing and apologizing, she justs offends again, throwing
> > > out more false accusations.
>
> > > Another example (from another thread) was her response to Kleck's
> > > finding that surveys indicated that 5-11% of US adults admit to
> > > carrying guns for self-protection, dwarfing the 1% or so of the
> > > population that obtained concealed-weapon permits. Mary insisted that
> > > this number was including mere transportation of guns and not just
> > > carrying for protection, and claimed that the statistic was made up.
> > > When I posted the exact page number of Kleck's book that the statistic
> > > came from, Mary did not apologize or admit to any error.
> > >
> >
> > These surveys are not based upon concealed carry but based upon any
> > carrying of a gun. Lott addresses this in his book p. 230. These are
> > not surveys for self defense as you claimed, but surveys involving any
> > type of carrying guns for any reason (e.g., hunting, moving
> > residences, etc.).
>
> Yes, Lott makes that claim, but anyone who reads chapter 6 of
> "Targetting Guns" can see that Lott does not know what he is talking
> about. I am surprised to find that Lott is unaware of current
> research on the frequency with which guns are carried for
> self-protection.

Boy, that is a powerful and convincing response. Appeal to your
authority as an expert. Can you list for me the academic publications
that you have?

>
>
> > > mary...@aol.com (Mary Rosh) writes:
>
> > > > > > In the case of right-to-carry laws, the change in crime
> > > > > > rates correspond very closely with the number of permits
> > > > > > issued.
>
> > > > So were prices also falling in
> > > > states that passed the concealed handgun laws prior to their passage
> > > > and then prices were rising after the laws were passed?
> > >
> > > Oh, so are you claiming that the number of permits was falling in states
> > > that passed the concealed handgun laws prior to their passage?
> >
> > Are you on drugs? The only way you could get a constant negative
> > relationship between prices and crime rates is if prices were falling
> > prior to the adoption of concealed handgun laws and rising afterward.
>
> Sigh. You have claimed that "the change in crime rates correspond
> very closely with the number of permits issued." You have also
> asserted that the crime rate was rising before the laws were passed.
> If both of your statements are true, then it follows that the number
> of permits was decreasing before the laws were passed.

Let me explain this to you: permits were only increasing after the
right-to-carry law passed. There was a general upward trend in crime
rates before the law and then when permits started to be issued, the
crime began to fall.

By contrast, the CPI variable would be effecting all states equally
and prices are rising both before and after the right-to-carry laws
were enacted.


>
> > > > > > Those states that issue
> > > > > > the most permits have the biggest drops in crime
> > > > >
> > > > > Not true. Lott has data on handgun permits for three states (table
> > > > > 4.7). This shows that the percentage of the population with permits
> > > > > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > > > > However, this is exactly the opposite ordering of the changes in the
> > > > > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
> > > > ÿ Pennsylvania.
> > > >
> > > > A typical incomplete or inaccurate misrepresentation. Lott's second
> > > > edition examines measures of permit issuance across many states.
> > >
> > > Oh, then it should be very easy for you to post the data on the permit
> > > rates from Lott's book for those other states. Put up or shut up.
> >
> > The discussion is on pages 174 to 177 in Lott's book. So is the list
> > of states.
>
> Those pages do not contain data on permit rates for any state. The
> only data Lott provides on actual permit rates is in table 4.7 as I
> have posted. Table 4.7 contains data for three sites and I included
> all of them in my analysis. Your assertion that my analysis is
> incomplete is false. If you are privy to some data that does not
> appear in Lott's book, please share it with us.

This just goes to prove that you don't address this issue as a
scientist but as an advocate. Now you admit that Lott did address this
issue, but you claim now is that he didn't print out his data. Is
Lott supposed to put all of his data in the book?

You claimed that Lott hadn't found the result that states that issue
the most permits have the biggest drops in crime. Now you lamely
claim that well Lott didn't just include a table show permit data for
all the states that he got the data for.

If you don't believe the results, put the data together yourself or
ask Lott for it.

>
>
> > > > > > When
> > > > > > you have a lot of potential victims in a public place even if the
> > > > > > probability that any individual has a permit is small, the probability
> > > > > > that at least someone will be able to defend themselves is very high
> > > > > > and thus there is a greater level of deterrence.
> > > > >
> > > > > Of course, the folks doing these shootings usually kill themselves, so
> > > > > are unlikely to be deterred. So, the theory can predict either a
> > > > ÿ greater or lesser decline.
> > > >
> > > > As usual you don't quite get things right. Lott and Landes say that
> > > > these killer frequently die either by being killed by others or
> > > > through their own hand.
> > >
> > > That hardly contradicts what I wrote.
> >
> >
> > You want to claim that there is no deterrence effect because everyone
> > was going to commit suicide.
>
> "everyone was going to commit suicide"??? Where did I say that?

"the folks doing these shootings usually kill themselves" => since
"usually" means "most," and kill themselves means suicide, I take that
to mean that there is no deterrence effect because most commit
suicide. I was pointing out that 1) you have no evidence to back up
that claim and 2) getting killed when one wasn't going to commit
suicide is a deterrence.

I does present it in the graphs. Sometimes a picture is worth a
thousand words and it seems like a pretty straightforward way of
presenting such complicated information.


>
>
> > > > If anyone
> > > > looks at the year by year changes in the figures (see especially the
> > > > second edition), you see the increase in violent crime rates starting
> > > > right at the time their neighbors adopt their law.
> > >
> > > Wrong. This is an example of the reification fallacy. I went into
> > > detail about what is wrong with Lott's graphs in my critique.
> > > Look at
> > > http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html#SECTION00032800000000000000
> > > They do not show that there was an increase in the violent crime rate
> > > at the time that the law was passed. The graphs do NOT show the year
> > > to year changes, rather they show a curve fitted to the data. On the
> > > web page above I have a Java applet that fits similar curves to random
> > > data. You get similar looking graphs even though the data is
> > > completely random and did not change when the law was passed.
> >
> >
> > Lott has done this both ways with most of his data. In any case these
> > are nonlinear lines.
>
> I never heard of a nonlinear line before. Is it something like a
> square circle?
>
> > Random data is a joke. If it is random you are
> > very unlikely to get crime rates falling before and rising after.
>
> Random data will show crime rates falling before and rising after 25%
> of the time. You really need to study some basic probability.

25 percent if you drop out the option that things could remain
unchanged and you are only dealing with one change before and after
the change in law. Obviously we are dealing with multiple years
before and after so the odds are much, much lower than 25 percent.
When you add in the fact that there may be no change you are talking
about a tiny probability. The f-tests that Lott conducts show that
these things happening the way that do are statistically at
thousandths of one percent.

>
> > That is why Lott provides things like f-tests to see whether these
> > types of changes are random or not. Lott mentions that this paper
> > with Bronars was published in the American Economic Review and you can
> > look up the detailed statistical tests there if you are interested.
>
> Well, some of them. They seem to have left out the results of the
> test on the violent crime rate. I wonder why?
>

YOU ARE AMAZINGLY DISHONEST. HAVE YOU ABSOLUTELY NO SHAME? Lott and
Bronars say that "results that are not shown are available from the
authors upon request." My guess is that the editors put some space
constraints on the piece.

Gee are 1995 and 1996 really different from 1993 and 1994? The point
is that Lott has expanded his data set a couple of times. There is
even a new paper of his that expands it to 1998 (with similar
results). If the original findings were simply some accident and that
the drops in crime in different places just happened to be related to
the number of permits issued, the results would not hold up when more
years were used. In fact, the results have continued to be confirmed.

>
> > > > When the additional two years were added they
> > > > tended to back down, but again if you were even remotely honest you
> > > > would mention that Lott discussed how the states that added their laws
> > > > in later years had relatively more restrictive laws and tended to
> > > > issue fewer permits, there was thus a smaller expected drop in crime
> > > > rates for those later states.
> > >
> > > Oh, please. You argue that if the results were accidental then X
> > > would happen. I point out that in fact, X did happen. Suddenly you
> > > claim that your theory also predicts X. Even if your claim is true,
> > > it doesn't matter. Your argument that the results weren't accidental
> > > has been refuted.
> >
> > Lott has a theory that the rate of permit issuance is important in
> > explaining the degree of deterrence. This is not a new theory by Lott
> > but has been what he has argued all along. The evidence compiled in
> > chapter 9 confirms that by showing that those states that have the
> > strickest issuing rules issue the smallest number of permits and have
> > smaller drops in violent crime.
>
> It doesn't show that at all. You're committing the reification
> fallacy again.
>

Smaller percent of adults with permits, the smaller the drop in crime.
Some of the more recent states that have adopted concealed handgun
laws are relatively restrictive and have issued relatively few
permits. The drop in crime is smaller for those states. Gee this is
deep thinking. Sorry for stretching you intellectual capacity so
much.

Mary Rosh

unread,
Nov 13, 2001, 10:57:44 PM11/13/01
to
david <da...@example.com> wrote in message news:<slrn9v0udr...@panix2.panix.com>...

Good question. The survey simply asked whether one had carried a gun
outside of one's residence over the course of the previous year.

Tim Lambert

unread,
Nov 13, 2001, 11:39:57 PM11/13/01
to
mary...@aol.com (Mary Rosh) writes:

Untrue. The exact wording, from chapter 6 of Kleck's "Targetting
Guns", and cited by me several times so that you have absolutely no
excuse for your behaviour, is:

"In the last twelve months, have you ever carried a gun away from
home, either on your person, or in a vehicle, for protection against
crime? Do not count carrying for recreation or in connection with
duites in law enforcemnt, work as a security guard, or in the armed

forces."

Tim

Tim Lambert

unread,
Nov 14, 2001, 8:52:27 AM11/14/01
to
mary...@aol.com (Mary Rosh) writes:

Did you even read what I wrote? I cited C h a p t e r 6 o f
" T a r g e t t i n g G u n s". Did you think I wrote "Targetting
Guns"? Here's the complete reference (from the bibliography of my
critique):

Gary Kleck.
Targeting Guns: Firearms and their Control.
Aldine de Gruyter, New York, 1997.

On page 205, Kleck gives the actual question asked:

"In the last twelve months, have you ever carried a gun away from
home, either on your person, or in a vehicle, for protection against
crime? Do not count carrying for recreation or in connection with
duites in law enforcemnt, work as a security guard, or in the armed
forces"

> > > > > > > Those states that issue


> > > > > > > the most permits have the biggest drops in crime
> > > > > >
> > > > > > Not true. Lott has data on handgun permits for three states (table
> > > > > > 4.7). This shows that the percentage of the population with permits
> > > > > > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > > > > > However, this is exactly the opposite ordering of the changes in the
> > > > > > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in
> > > > > ÿ Pennsylvania.
> > > > >
> > > > > A typical incomplete or inaccurate misrepresentation. Lott's second
> > > > > edition examines measures of permit issuance across many states.
> > > >
> > > > Oh, then it should be very easy for you to post the data on the permit
> > > > rates from Lott's book for those other states. Put up or shut up.
> > >
> > > The discussion is on pages 174 to 177 in Lott's book. So is the list
> > > of states.
> >
> > Those pages do not contain data on permit rates for any state. The
> > only data Lott provides on actual permit rates is in table 4.7 as I
> > have posted. Table 4.7 contains data for three sites and I included
> > all of them in my analysis. Your assertion that my analysis is
> > incomplete is false. If you are privy to some data that does not
> > appear in Lott's book, please share it with us.
>
> This just goes to prove that you don't address this issue as a
> scientist but as an advocate.

You are projecting.

> Now you admit that Lott did address this issue,

I made no such admission. Lott has not shown that states that issue
the most permits have the biggest drops in crime. (If you think he
has, you're back to the reification fallacy again.)

> but you claim now is that he didn't print out his data. Is
> Lott supposed to put all of his data in the book?

He doesn't have to if he doesn't want to. I performed an analysis
using all the data that he provided. It shows that states with the
most permits have the *smallest* drops in crime. If have any data on
permits for other states, please contribute it, and I can include it
in my analysis.

> > > > > > > Lott also makes a lot about
> > > > > > > the work where he compares counties that border each other on opposite
> > > > > > > sides of state lines where he sees what happens to the crime rates in
> > > > > > > both the county that is in a state with the law and the adjacent
> > > > > > > county in a state without the law. He finds that not only does the
> > > > > > > violent crime rate fall for the county in a state with the law, but
> > > > > > > just at that same time there is an increase in the crime rate in the
> > > > > > > boarder county.
> > > > > >
> > > > > > Table 4.14 shows that there was no change in the violent
> > > > > > crime rate of these border counties. And I don't mean that the change
> > > > > > was not statistically significant. The number (middle of top row of
> > > > > > the table) is exactly zero. It is not possible to have a better
> > > > > ÿ example of a "no effect" result".

> > > > > If anyone


> > > > > looks at the year by year changes in the figures (see especially the
> > > > > second edition), you see the increase in violent crime rates starting
> > > > > right at the time their neighbors adopt their law.
> > > >
> > > > Wrong. This is an example of the reification fallacy. I went into
> > > > detail about what is wrong with Lott's graphs in my critique.
> > > > Look at
> > > > http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html#SECTION00032800000000000000
> > > > They do not show that there was an increase in the violent crime rate
> > > > at the time that the law was passed. The graphs do NOT show the year
> > > > to year changes, rather they show a curve fitted to the data. On the
> > > > web page above I have a Java applet that fits similar curves to random
> > > > data. You get similar looking graphs even though the data is
> > > > completely random and did not change when the law was passed.
> > >

> > > Random data is a joke. If it is random you are
> > > very unlikely to get crime rates falling before and rising after.
> >
> > Random data will show crime rates falling before and rising after 25%
> > of the time. You really need to study some basic probability.
>
> 25 percent if you drop out the option that things could remain
> unchanged

That has probability zero with a continuous distribution.

> and you are only dealing with one change before and after
> the change in law. Obviously we are dealing with multiple years
> before and after so the odds are much, much lower than 25 percent.

Untrue. The slope of the fitted curve is either positive or negative
at t=0, with a 50% chance of either outcome, hence a 25% cahnce of a
falling before, rising after outcome. Please try my applet at the
link above and see how often you get a curve that looks like one of
Lott's.

> > > That is why Lott provides things like f-tests to see whether these
> > > types of changes are random or not. Lott mentions that this paper
> > > with Bronars was published in the American Economic Review and you can
> > > look up the detailed statistical tests there if you are interested.
> >
> > Well, some of them. They seem to have left out the results of the
> > test on the violent crime rate. I wonder why?
> >
> YOU ARE AMAZINGLY DISHONEST. HAVE YOU ABSOLUTELY NO SHAME? Lott and
> Bronars say that "results that are not shown are available from the
> authors upon request."

That does not explain why they left it out. I also note that you
attacked me for not including it, even though it's not in Lott's book
or in Lott and Bronars.

> My guess is that the editors put some space constraints on the piece.

That's just your speculation. And, even if true, it doesn't explain
why that result was omitted and not some other one.

And, what really amuses me about your response is that in an earlier
discussion you produced this gem:

] Lambert referencing Duggan isn't serious. Why look at only the fourth
] largest gun magazine? You talk about cherry picking. If you want
] handguns, use one of the magazines that are purely on handguns. Guns
] and Ammo has a similar ratio of handgun to long gun reviews as the
] American Rifleman, yet Duggan doesn't include that one. Why? This
] isn't serious work.

And then you admitted that you had known that Duggan had explained the
reason for his choice: "Unlike the three gun magazines with greater
circulation,2 sales data for this magazine are available annually at
both the state and the county level." (A quote from Duggan's paper.)

> > > > > > Since the original data is still included, if I am right what you
> > > > > > would see when extra data is added is a lessening of the effect. I
> > > > > > reviewed Lott's results to see how they changed as the time period got
> > > > > > longer. Unfortunately, since the specification of the model in
> > > > > > table 9.1 is different from that in tables 4.8 and 4.13 the only
> > > > > > comparison that can be made is that between the results in table 4.8
> > > > > > and 4.13. This shows the effect on violent crime lessening from -0.9%
> > > > > > to -0.5% when the period of time studied increased by two years.
> >
> > > > > When Lott added the data for the 1993 to 1994 period the strength of
> > > > > the results went up.
> > > >
> > > > Wrong. The difference between table 4.8 and table 4.13 was just the
> > > > adding of data for 1993 and 1994.
> > >
> > > Boy, you are a dishonest person. You know as well as I do the
> > > addition for 1995 and 1996 were done in Chapter 9.
> >
> > Huh? You claimed, quite falsely, that adding 1993 and 1994 cause the
> > strength of the results to go up. You just have to look at tables
> > 4.8 and 4.13 to see this. 1995 and 1996 are different years from 1993
> > and 1994. You are making even less sense than usual. Or is is this
> > one of your nonlinear lines?
>
> Gee are 1995 and 1996 really different from 1993 and 1994?

Yes they are. Please explain why you feel it is somehow dishonest to
point out that the difference between table 4.8 and table 4.13 was just the


adding of data for 1993 and 1994.

> > > Lott has a theory that the rate of permit issuance is important in


> > > explaining the degree of deterrence. This is not a new theory by Lott
> > > but has been what he has argued all along. The evidence compiled in
> > > chapter 9 confirms that by showing that those states that have the
> > > strickest issuing rules issue the smallest number of permits and have
> > > smaller drops in violent crime.
> >
> > It doesn't show that at all. You're committing the reification
> > fallacy again.

> Some of the more recent states that have adopted concealed handgun


> laws are relatively restrictive and have issued relatively few
> permits. The drop in crime is smaller for those states.

Nope. Guess again. Do you know what the reification fallacy is?

Tim

Mary Rosh

unread,
Nov 15, 2001, 8:14:31 PM11/15/01
to
You are a very funny person tim. You deny that Lott has done tests
even when he reports the results because he doesn't reprint all the
data in the book. I spent a few minutes and found a couple of web
site that provide additional permit issuance data so you can't claim
that Lott is making up data that doesn't exist. Incidentally, you
raise the issue of Duggan and defend him without explaining why he
selectively picks one gun magazine to measure gun ownership when other
gun magazines meet or exceed the criteria that he picks to limit the
data. I have raised this issue before but you continually ignore it.
By the way, did Duggan reprint all his magazine sales data in his
paper? I guess that we should say that he preformed the tests that he
said he did until he does so. Unlike Lott, there is the claim that
Duggan has refused to give out his data to others. May you could ask
him if that is true?


Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3u1vxt...@cycloid.localdomain>...

I read what you wrote and I was referring to your last argument. I
cited Lott and you cited Kleck. Your last claim was simply about how
we should trust your judgment, which for obvious reasons I am
unwilling to do.

> > > > > > > > Those states that issue
> > > > > > > > the most permits have the biggest drops in crime
> > > > > > >
> > > > > > > Not true. Lott has data on handgun permits for three states (table
> > > > > > > 4.7). This shows that the percentage of the population with permits
> > > > > > > in 1994 was 1% in Florida, 1.4% in Oregon, and 4% in Pennsylvania.
> > > > > > > However, this is exactly the opposite ordering of the changes in the
> > > > > > > violent crime rate: -4% in Florida, -3% in Oregon, and -1% in

> > > > > > Ë™ Pennsylvania.


> > > > > >
> > > > > > A typical incomplete or inaccurate misrepresentation. Lott's second
> > > > > > edition examines measures of permit issuance across many states.
> > > > >
> > > > > Oh, then it should be very easy for you to post the data on the permit
> > > > > rates from Lott's book for those other states. Put up or shut up.
> > > >
> > > > The discussion is on pages 174 to 177 in Lott's book. So is the list
> > > > of states.
> > >
> > > Those pages do not contain data on permit rates for any state. The
> > > only data Lott provides on actual permit rates is in table 4.7 as I
> > > have posted. Table 4.7 contains data for three sites and I included
> > > all of them in my analysis. Your assertion that my analysis is
> > > incomplete is false. If you are privy to some data that does not
> > > appear in Lott's book, please share it with us.
> >
> > This just goes to prove that you don't address this issue as a
> > scientist but as an advocate.
>
> You are projecting.
>
> > Now you admit that Lott did address this issue,
>
> I made no such admission. Lott has not shown that states that issue
> the most permits have the biggest drops in crime. (If you think he
> has, you're back to the reification fallacy again.)

Let me explain this to you. The hypothesis is that those states that
issue the most permits have the biggest relative drops in violent
crime. Lott has presented the hypothesis consistently throughout his
work. I cite you the pages where he shows statistical results that
confirm this. Now you may not want to accept Lott's results, but it
is bizarre to argue that Lott hasn't provided this test and shown that
the data confirms that hypothesis.

REIFICATION fallacy -- Description: To reify something is to convert
an abstract concept into a concrete thing. Reification is a Fallacy of
Ambiguity. Reification is also sometimes known as a fallacy of
"hypostatization."

This is a nonsensical argument in this case. You use words without
even understanding whether they apply. Deterrence is an abstract
concept, sure. And the notion that greater deterrence can result in
fewer crimes is also an abstract concept, but that does not prevent it
from being test. If you continue to argue this, I will view it as
grounds that it is not worth arguing with you.


>
> > but you claim now is that he didn't print out his data. Is
> > Lott supposed to put all of his data in the book?
>
> He doesn't have to if he doesn't want to. I performed an analysis
> using all the data that he provided. It shows that states with the
> most permits have the *smallest* drops in crime. If have any data on
> permits for other states, please contribute it, and I can include it
> in my analysis.

The only sense that I can make of this claim is that you don't believe
that data exists on permit level in the states that Lott says he uses.
Since you are apparently too lazy and demand that Lott provide you
with this data without you even making the effort to ask, I decided to
take a few minutes and check for issuance rates. Here are a couple of
places I found:
Texas
http://txdps.state.tx.us/administration/crime_records/chl/demographics.htm
North Carolina
http://sbi2.jus.state.nc.us/crp/public/other/chpsum.htm

Finally, let me note that Lott has provided his data to the academics
who have asked. No one has said otherwise. Can you provide any
reference that contradicts this? Of course, you cannot. Instead you
waste an incredible amount of space first denying that Lott even did a
certain test and then claiming while it may appear that he did the
test, he really didn't because he didn't reprint all the data in a
table for you in the book. You might not understand this but the fact
that Lott provides his data to academics and doesn't reprint all his
data in the book is not the same thing as saying he didn't do a test.

>
> > > > > > > > Lott also makes a lot about
> > > > > > > > the work where he compares counties that border each other on opposite
> > > > > > > > sides of state lines where he sees what happens to the crime rates in
> > > > > > > > both the county that is in a state with the law and the adjacent
> > > > > > > > county in a state without the law. He finds that not only does the
> > > > > > > > violent crime rate fall for the county in a state with the law, but
> > > > > > > > just at that same time there is an increase in the crime rate in the
> > > > > > > > boarder county.
> > > > > > >
> > > > > > > Table 4.14 shows that there was no change in the violent
> > > > > > > crime rate of these border counties. And I don't mean that the change
> > > > > > > was not statistically significant. The number (middle of top row of
> > > > > > > the table) is exactly zero. It is not possible to have a better

> > > > > > Ë™ example of a "no effect" result".

Let's see. During the period after the law is passed, there is a
chance of a statistically significant increase, a chance of no
statistically significant change, a chance of a statistically
significant decline. During the period before the law is passed, there
is a chance of a statistically significant increase, a chance of no
statistically significant change, a chance of a statistically
significant decline. If you just had one state and were just
examining three years (one year before and after the law and the year
of change),

Before After
Down Up
Down Constant
Down Down
Constant Up
Constant Constant
Constant Down
Up Up
Up Constant
Up Down

I count nine possibilities for even this very limited set of
possibilities. If you have two states these nine possibilities become
81 possibilities (e.g., Down, up; Down, up or Down constant or . . .).
Lott gets his result only if it is Up Down for both states. By my
calculations that is one possibility out of 81 cases. Now include
lots of states and lots of years before and after a law is adopted.
If you understood statistics, you would realize that there is a reason
why the F-tests show statistical significance levels of better than 1
in 10,000.

>
> > > > That is why Lott provides things like f-tests to see whether these
> > > > types of changes are random or not. Lott mentions that this paper
> > > > with Bronars was published in the American Economic Review and you can
> > > > look up the detailed statistical tests there if you are interested.
> > >
> > > Well, some of them. They seem to have left out the results of the
> > > test on the violent crime rate. I wonder why?
> > >
> > YOU ARE AMAZINGLY DISHONEST. HAVE YOU ABSOLUTELY NO SHAME? Lott and
> > Bronars say that "results that are not shown are available from the
> > authors upon request."
>
> That does not explain why they left it out. I also note that you
> attacked me for not including it, even though it's not in Lott's book
> or in Lott and Bronars.

Let me explain something to you. Lott provided the relevant graphs in
the book so that you could see whether there was a pattern. Any one
who looks at the graph can see that there was a big change that
occurred at the time of the law. He and Bronars said in their paper
that they are happy to provide more the rest of the information to
those who ask. Have you asked? Why don't you contact Bronars? He is
at the University of Texas.

>
> > My guess is that the editors put some space constraints on the piece.
>
> That's just your speculation. And, even if true, it doesn't explain
> why that result was omitted and not some other one.

If you published in journals you would know that very frequently,
authors are not allowed to put in the text everything that they have
done. As a result they frequently say results for certain things are
available from the authors. Journals have space constraints, but they
have offered to make the results available. If this were really a
problem, don't you think that Lott's academic critics would have made
an issue of this. Can you point to one academic who sees this issue
the same way that you do?

>
> And, what really amuses me about your response is that in an earlier
> discussion you produced this gem:
>
> ] Lambert referencing Duggan isn't serious. Why look at only the fourth
> ] largest gun magazine? You talk about cherry picking. If you want
> ] handguns, use one of the magazines that are purely on handguns. Guns
> ] and Ammo has a similar ratio of handgun to long gun reviews as the
> ] American Rifleman, yet Duggan doesn't include that one. Why? This
> ] isn't serious work.
>
> And then you admitted that you had known that Duggan had explained the
> reason for his choice: "Unlike the three gun magazines with greater
> circulation,2 sales data for this magazine are available annually at
> both the state and the county level." (A quote from Duggan's paper.)

Typical distortion from you. What I said was that this explanation
isn't true because magazine sales are available at the county level
for other magazines and I gave you the web site to check it out. If
he wanted to look at primarily handguns, why pick a magazine that has
only 50 percent of its reviews dealing with handguns when county level
magazine sales are available for Handguns magazine which is 100
percent handguns? I know that you didn't want to answer this before.
You still refuse to address the issue that I raised.


>
> > > > > > > Since the original data is still included, if I am right what you
> > > > > > > would see when extra data is added is a lessening of the effect. I
> > > > > > > reviewed Lott's results to see how they changed as the time period got
> > > > > > > longer. Unfortunately, since the specification of the model in
> > > > > > > table 9.1 is different from that in tables 4.8 and 4.13 the only
> > > > > > > comparison that can be made is that between the results in table 4.8
> > > > > > > and 4.13. This shows the effect on violent crime lessening from -0.9%
> > > > > > > to -0.5% when the period of time studied increased by two years.
>
> > > > > > When Lott added the data for the 1993 to 1994 period the strength of
> > > > > > the results went up.
> > > > >
> > > > > Wrong. The difference between table 4.8 and table 4.13 was just the
> > > > > adding of data for 1993 and 1994.
> > > >
> > > > Boy, you are a dishonest person. You know as well as I do the
> > > > addition for 1995 and 1996 were done in Chapter 9.
> > >
> > > Huh? You claimed, quite falsely, that adding 1993 and 1994 cause the
> > > strength of the results to go up. You just have to look at tables
> > > 4.8 and 4.13 to see this. 1995 and 1996 are different years from 1993
> > > and 1994. You are making even less sense than usual. Or is is this
> > > one of your nonlinear lines?
> >
> > Gee are 1995 and 1996 really different from 1993 and 1994?
>
> Yes they are. Please explain why you feel it is somehow dishonest to
> point out that the difference between table 4.8 and table 4.13 was just the
> adding of data for 1993 and 1994.

Do you know what a rhetorical question is? Of course, they are
different years and that is why Lott added the additional years on to
see what difference the results made. Lott has continued to add to
the data and continued to get strong results.

>
> > > > Lott has a theory that the rate of permit issuance is important in
> > > > explaining the degree of deterrence. This is not a new theory by Lott
> > > > but has been what he has argued all along. The evidence compiled in
> > > > chapter 9 confirms that by showing that those states that have the
> > > > strickest issuing rules issue the smallest number of permits and have
> > > > smaller drops in violent crime.
> > >
> > > It doesn't show that at all. You're committing the reification
> > > fallacy again.
>
> > Some of the more recent states that have adopted concealed handgun
> > laws are relatively restrictive and have issued relatively few
> > permits. The drop in crime is smaller for those states.
>
> Nope. Guess again. Do you know what the reification fallacy is?
>

Abstract concepts can be tested. There is nothing wrong with taking
an abstract concept like deterrence and see whether crime actually
falls. There is no problem in seeing whether greater deterrence
produces an even greater drop in crime. This is Lott's hypothesis.
It has been his hypothesis all along.

Tim Lambert

unread,
Nov 16, 2001, 1:30:35 AM11/16/01
to
I've put this into a separate posting because some of the arguments
are a bit technical, but I think anyone should be able to read the
actual survey question and decide whether Lott's opinion about it is
correct.

The survey asks:

"In the last twelve months, have you ever carried a gun away from
home, either on your person, or in a vehicle, for protection against
crime? Do not count carrying for recreation or in connection with

duties in law enforcemnt, work as a security guard, or in the armed
forces"

Lott claims "The survey results mentioned by Lambert refer to all
transportation or carrying of guns by Americans. It includes not only
carrying concealed handguns (whether legally or illegally) but also
people who have a gun with them to go hunting or who may simply be
transporting a gun between residences"

The actual words of the question show that Lott's claim is false.
Mary refuses to admit this.

mary...@aol.com (Mary Rosh) writes:

> > duties in law enforcemnt, work as a security guard, or in the armed


> > forces"
> >
>
> I read what you wrote and I was referring to your last argument. I
> cited Lott and you cited Kleck. Your last claim was simply about how
> we should trust your judgment, which for obvious reasons I am
> unwilling to do.

Yet again you fail to follow an extremely simple argument. If I had,
out of the blue, without offering any supporting cites or arguments,
asserted that Lott was unfamilar with current research on the
frequency of gun carrying, then that would be asking people to trust
my judgement. Instead, I gave a cite to a source where the questions
and methodology of the survey were described. Since it is crystal
clear from the question that recreational carrying was specifically
excluded, I concluded that Lott had not actually looked at the survey
question.

I hope that you get it now.

Tim

Chris Crandall

unread,
Nov 16, 2001, 11:41:26 AM11/16/01
to
Mary Rosh (mary...@aol.com) wrote:
: Incidentally, you raise the issue of Duggan and defend him without

: explaining why he selectively picks one gun magazine to measure gun
: ownership when other gun magazines meet or exceed the criteria that he
: picks to limit the data.

Tim L. did explain why, and quite well. He quoted the original author as
stating that the gun magazine was the only one reporting county-level
circulation numbers. Or rather, the largest one to do that.

Mary--I recommend actually reading his reponses before you attack him.
Might save you a little embarassment.

Mary Rosh

unread,
Nov 16, 2001, 1:54:25 PM11/16/01
to
Does this new tread mean that you are conceeding all the other points
that we were debating?

Why are you running away from all the other issues? You accuse Lott
of doing tests even when he reports the results because he doesn't


reprint all the data in the book. I spent a few minutes and found a
couple of web
site that provide additional permit issuance data so you can't claim

that Lott is making up data that doesn't exist. While it is very
typical of the types of distortions that you make, rather than finally
admitting how inaccurate your claims are you run away and start a new
thread.

You raise the issue of Duggan and defend him without explaining why he


selectively picks one gun magazine to measure gun ownership when other
gun magazines meet or exceed the criteria that he picks to limit the
data. I have raised this issue before but you continually ignore it.
By the way, did Duggan reprint all his magazine sales data in his
paper? I guess that we should say that he preformed the tests that he
said he did until he does so. Unlike Lott, there is the claim that

Duggan has refused to give out his data to others. May be you could


ask
him if that is true?

Putting aside all the five points that we have been debating which you
want to ignore, I have cited an expert who claims that the survey is
much broader than you what you cite. Since both references have the 5
to 11 percent claim and given the debate that Lott references, it
appears that both are dealing with the same evidence. You disregard
the expert that I reference, fine. Your inability to correct your
many false claims about Lott or Duggan and your twisting of statements
that I have made imply to me that there is no evidence that I could
cite you in any case.

Tim Lambert <lam...@cse.unsw.EDU.AU> wrote in message news:<tny9l7b5...@oolong.orchestra.cse.unsw.EDU.AU>...

Chris Crandall

unread,
Nov 16, 2001, 2:02:31 PM11/16/01
to
Distribution:
X-Newsreader: TIN [version 1.2 PL2]

Mary Rosh (mary...@aol.com) wrote:
: Does this new tread mean that you are conceeding all the other points
: that we were debating?

Mary is a lot of fun, isn't she?

I'll answer for Tim, in one small, slow word: No.

david

unread,
Nov 16, 2001, 2:35:06 PM11/16/01
to
On 16 Nov 2001 17:30:35 +1100, Tim Lambert <lam...@cse.unsw.EDU.AU> wrote:
> I've put this into a separate posting because some of the arguments
> are a bit technical, but I think anyone should be able to read the
> actual survey question and decide whether Lott's opinion about it is
> correct.
>
> The survey asks:
> "In the last twelve months, have you ever carried a gun away from
> home, either on your person, or in a vehicle, for protection against
> crime? Do not count carrying for recreation or in connection with
> duties in law enforcemnt, work as a security guard, or in the armed
> forces"
>
> Lott claims "The survey results mentioned by Lambert refer to all
> transportation or carrying of guns by Americans. It includes not only
> carrying concealed handguns (whether legally or illegally) but also
> people who have a gun with them to go hunting or who may simply be
> transporting a gun between residences"

I probably shouldn't get into this, but IMHO the survey question
supports neither Lott's interpretation nor your own.

For Lott's basic thesis, the question you really want answered is "for
a random selection of people at a given time, how many are carrying
guns?". The survey question above is mostly tangential to this.

I'm not surprised that the number of carriers revealed by the survey
dwarfs the number of permits. After all, you're counting things like
the guy who takes a gun on vacation once a year for protection on the
long drive. That type of "one-off" carry has a negligible effect on
the amount of carriers at any given time, but counts in the survey
just as highly as the person who carries a gun every day.

Mary Rosh

unread,
Nov 17, 2001, 2:04:01 AM11/17/01
to
cran...@lark.cc.ku.edu (Chris Crandall) wrote in message news:<9t3ns7$opk$4...@news.cc.ukans.edu>...

OK, fine. Then have either Mr. Lambert or yourself explain why his
previous claims are true. Lott does provide evidence that those
places that issue the most permits have the biggest drops in violent
crime. Lambert can claim that Lott doesn't as many times as he wants,
just like he makes up and distorts everything else Lott writes, but
anyone who has read chapter 9 in Lott's book can see that Lambert is
full of it. As to Lambert's cites of Duggan's justification for only
using sales of one gun magazine (the fourth largest) to measure gun
ownership when other magazines better meet the criteria that he claims
he is following (e.g., if you want to focus on handguns, use a purely
handgun magazine not one that is 50 percent long guns), I would like
Lambert to explain his defense of Duggan. Lambert continually
misquotes and distort's Lott's work and it is high time that he be
forced to fess up. I admit that I get angry when I come across
someone is as continually dishonest as Lambert.

Tim Lambert

unread,
Nov 17, 2001, 8:53:47 AM11/17/01
to
david <da...@example.com> writes:

> On 16 Nov 2001 17:30:35 +1100, Tim Lambert <lam...@cse.unsw.EDU.AU> wrote:
> > I've put this into a separate posting because some of the arguments
> > are a bit technical, but I think anyone should be able to read the
> > actual survey question and decide whether Lott's opinion about it is
> > correct.
> >
> > The survey asks:
> > "In the last twelve months, have you ever carried a gun away from
> > home, either on your person, or in a vehicle, for protection against
> > crime? Do not count carrying for recreation or in connection with
> > duties in law enforcemnt, work as a security guard, or in the armed
> > forces"
> >
> > Lott claims "The survey results mentioned by Lambert refer to all
> > transportation or carrying of guns by Americans. It includes not only
> > carrying concealed handguns (whether legally or illegally) but also
> > people who have a gun with them to go hunting or who may simply be
> > transporting a gun between residences"
>
> I probably shouldn't get into this,

Please join in. I welcome discussion with someone who is actually
prepared to consider the question that was actually asked.


> but IMHO the survey question
> supports neither Lott's interpretation nor your own.
>
> For Lott's basic thesis, the question you really want answered is "for
> a random selection of people at a given time, how many are carrying
> guns?". The survey question above is mostly tangential to this.

Lott claimed that the survey included carrying for hunting when the
wording clearly excluded such uses.

>
> I'm not surprised that the number of carriers revealed by the survey
> dwarfs the number of permits. After all, you're counting things like
> the guy who takes a gun on vacation once a year for protection on the
> long drive. That type of "one-off" carry has a negligible effect on
> the amount of carriers at any given time, but counts in the survey
> just as highly as the person who carries a gun every day.

That's why the survey went on to ask the people that reported carrying
guns for self-protection how often they carried. The average for
people who carried guns on the person (as opposed to in a vehicle) was
138 days per year.

If we are comparing this with the number of permit holders, we should
take into acount the fact that permit holders will not carry their gun
every day either.

Tim

Tim Lambert

unread,
Nov 17, 2001, 10:14:18 AM11/17/01
to
mary...@aol.com (Mary Rosh) writes:

> Does this new tread mean that you are conceeding all the other points
> that we were debating?

No. I'm sorry if that was not obvious to you. You continue to insist
that the survey included carrying for recreation when the actual
question asked was:


"In the last twelve months, have you ever carried a gun away from
home, either on your person, or in a vehicle, for protection against
crime? Do not count carrying for recreation or in connection with
duties in law enforcemnt, work as a security guard, or in the armed
forces"

I think that anyone can look at the question and what you are saying
about it and form their own judgement about your credibility.


> Why are you running away from all the other issues? You accuse Lott
> of doing tests even when he reports the results because he doesn't
> reprint all the data in the book.

I made no such accusation. Just once, just once, I wish you would
accurately describe my arguments.

> Putting aside all the five points that we have been debating which you
> want to ignore, I have cited an expert who claims that the survey is
> much broader than you what you cite. Since both references have the 5
> to 11 percent claim and given the debate that Lott references, it
> appears that both are dealing with the same evidence. You disregard
> the expert that I reference, fine.

God forbid you should read the actual question and figure out which of
Kleck or Lott is correct.

Tim

Tim Lambert

unread,
Nov 17, 2001, 10:37:01 AM11/17/01
to
mary...@aol.com (Mary Rosh) writes:
>
> Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3u1vxt...@cycloid.localdomain>...

> > Lott has not shown that states that issue
> > the most permits have the biggest drops in crime. (If you think he
> > has, you're back to the reification fallacy again.)

> REIFICATION fallacy -- Description: To reify something is to convert


> an abstract concept into a concrete thing. Reification is a Fallacy of
> Ambiguity. Reification is also sometimes known as a fallacy of
> "hypostatization."
>
> This is a nonsensical argument in this case. You use words without
> even understanding whether they apply. Deterrence is an abstract
> concept, sure.

That's not where the fallacy lies. This one is a little complicated
to explain, so it is better to look at another example where you fall
for the fallacy:

> If anyone
> looks at the year by year changes in the figures (see especially the
> second edition), you see the increase in violent crime rates starting
> right at the time their neighbors adopt their law.

Lott's graphs show a curve fitted to the data (the year to year
changes). Lott's graphs do not show the actual year to year changes.
You are committing the reification fallacy when you act as if the
fitted curve (an abstract model) is the same as the year to year
changes (concrete data).

Now back to the claim that states that issue the most permits have the
biggest drops in crime. You first need to answer this question and
then we can proceed:
If a random variable X is correlated to a r.v. Y and X is also
correlated with Z, does it follow that Y is corrrelated with Z?

Tim

david

unread,
Nov 17, 2001, 3:55:20 PM11/17/01
to
On 18 Nov 2001 00:53:47 +1100, Tim Lambert <lam...@cse.unsw.edu.au> wrote:

> david <da...@example.com> writes:
>
>
>> but IMHO the survey question
>> supports neither Lott's interpretation nor your own.
>>
>> For Lott's basic thesis, the question you really want answered is "for
>> a random selection of people at a given time, how many are carrying
>> guns?". The survey question above is mostly tangential to this.
>
> Lott claimed that the survey included carrying for hunting when the
> wording clearly excluded such uses.

I stipulated that his response wasn't responsive and inaccurately
described the study (I'm being careful about the way I say that not to
cast doubt on your characterization, but to avoid the crossfire.).

>> I'm not surprised that the number of carriers revealed by the survey
>> dwarfs the number of permits. After all, you're counting things like
>> the guy who takes a gun on vacation once a year for protection on the
>> long drive. That type of "one-off" carry has a negligible effect on
>> the amount of carriers at any given time, but counts in the survey
>> just as highly as the person who carries a gun every day.
>
> That's why the survey went on to ask the people that reported carrying
> guns for self-protection how often they carried. The average for
> people who carried guns on the person (as opposed to in a vehicle) was
> 138 days per year.
>
> If we are comparing this with the number of permit holders, we should
> take into acount the fact that permit holders will not carry their gun
> every day either.

Right. And it's those questions that get you into the realm where
they are directly relevant to Lott's original thesis. My only point
was that the original question was tangential at best, and that more
detailed questions are necessary for the point being made.

Also, it appears on first reading that the 5%-11% figure you quote
refers to this first question, and not the later follow-ups. Is that
right?

Mary Rosh

unread,
Nov 17, 2001, 10:43:02 PM11/17/01
to
Stop editing my responses! It is amazing enough that you distort and
lie about what Lott says. But do not edit my responses again. You
cut out the first part of my response to your claim that "Lott has not

shown that states that issue the most permits have the biggest drops
in crime."

The part of my response which you edited out was: "Let me explain this
to you. The hypothesis is that those states thatissue the most
permits have the biggest relative drops in violentcrime. Lott has


presented the hypothesis consistently throughout his work. I cite you
the pages where he shows statistical results that confirm this. Now
you may not want to accept Lott's results, but it is bizarre to argue
that Lott hasn't provided this test and shown that the data confirms
that hypothesis."

You also refuse to defend all your other claims about such as the
supposed reasons for why Duggan looked at only the fourth largest gun
magazine when there are purely handgun magazines.

Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3bsi1p...@cycloid.localdomain>...


> mary...@aol.com (Mary Rosh) writes:
> >
> > Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3u1vxt...@cycloid.localdomain>...
> > > Lott has not shown that states that issue
> > > the most permits have the biggest drops in crime. (If you think he
> > > has, you're back to the reification fallacy again.)
>

> > FIRST PART OF MY RESPONSE HERE IS EDITED OUT BY LAMBERT.


> >
> > REIFICATION fallacy -- Description: To reify something is to convert
> > an abstract concept into a concrete thing. Reification is a Fallacy of
> > Ambiguity. Reification is also sometimes known as a fallacy of
> > "hypostatization."
> >
> > This is a nonsensical argument in this case. You use words without
> > even understanding whether they apply. Deterrence is an abstract
> > concept, sure.
>
> That's not where the fallacy lies. This one is a little complicated
> to explain, so it is better to look at another example where you fall
> for the fallacy:
>
> > If anyone
> > looks at the year by year changes in the figures (see especially the
> > second edition), you see the increase in violent crime rates starting
> > right at the time their neighbors adopt their law.
>
> Lott's graphs show a curve fitted to the data (the year to year
> changes). Lott's graphs do not show the actual year to year changes.
> You are committing the reification fallacy when you act as if the
> fitted curve (an abstract model) is the same as the year to year
> changes (concrete data).
>

What Bronars and Lott were examining in this case was whether crime
trends changed before and after the adoption. They found strong
evidence that was indeed the case. They go beyond just fitting a
linear relationship, and see what the nonlinear relationship looks
like. A simple linear relationship is going to pick up general
changes in trends. It is hard to see why the nonlinear examination
isn't going to pick up any important changes.

Lott has given out his data to lots and lots of academics. If some
how this examination of trends was misleading, I am sure that one of
the many left-wingers in academia would have pointed out some problem
with this. Is there even one academic who has challenged Bronars and
Lott's work or Lott's work or approach on this topic? Indeed why has
no one challenged this set of findings (a set of findings Lott has
referred to several times as his most powerful evidence)?


> Now back to the claim that states that issue the most permits have the
> biggest drops in crime. You first need to answer this question and
> then we can proceed:
> If a random variable X is correlated to a r.v. Y and X is also
> correlated with Z, does it follow that Y is corrrelated with Z?

If X is positively correlated with Y and X is negatively correlated
with Z (or the reverse set of correlations is true), there is not
necessarily any systematic correlation between Y and Z. While your
statement is not very clearly stated, I assume that you mean that X is
correlated with Y in the same way that X is correlated with Z.

Mary Rosh

unread,
Nov 17, 2001, 11:02:24 PM11/17/01
to
Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3g07dp...@cycloid.localdomain>...

> mary...@aol.com (Mary Rosh) writes:
>
> > Does this new tread mean that you are conceeding all the other points
> > that we were debating?
>
> No. I'm sorry if that was not obvious to you.

OK, fine then address these claims. Nothing what you write below
deals with any of the false claims of yours that I asked you to
defend. What about your claim that Duggan had consistent reasons for
only examining the fourth largest gun magazine when other gun
magazines focus solely on handguns. Duggan claimed that he wanted
only magazines that focused on handguns when the magazine that he
picked only had half of its product reviews dealing with handguns.
Other magazines are 100 percent handguns but they were not included.
He claimed that he wanted to look at magazine sales data that was
broken down at the county level, yet many gun magazines including some
purely handgun magazines (e.g., Handguns Magazine) which have county
level sales data.

What about your bizarre claim that Lott never tested whether the
greater issuance of permits was associated with a greater drop in
violent crime? You don't conceed, fine. Just try to justify your
earlier claims.


You continue to insist
> that the survey included carrying for recreation when the actual
> question asked was:
> "In the last twelve months, have you ever carried a gun away from
> home, either on your person, or in a vehicle, for protection against
> crime? Do not count carrying for recreation or in connection with
> duties in law enforcemnt, work as a security guard, or in the armed
> forces"
>
> I think that anyone can look at the question and what you are saying
> about it and form their own judgement about your credibility

Lott claims that the question reads differently. I believe Lott, you
can believe Kleck.


.
>
>
> > Why are you running away from all the other issues? You accuse Lott
> > of doing tests even when he reports the results because he doesn't
> > reprint all the data in the book.
>
> I made no such accusation. Just once, just once, I wish you would
> accurately describe my arguments.
>

What a joke. You even edit my past responses to you.

You first claimed that Lott hadn't tested this, then you claimed that
Lott didn't even have the data to test this. You even keep on asking
me if I had the data to test whether the states with the most permits
had the biggest drops in violent crime.

> > Putting aside all the five points that we have been debating which you
> > want to ignore, I have cited an expert who claims that the survey is
> > much broader than you what you cite. Since both references have the 5
> > to 11 percent claim and given the debate that Lott references, it
> > appears that both are dealing with the same evidence. You disregard
> > the expert that I reference, fine.
>
> God forbid you should read the actual question and figure out which of
> Kleck or Lott is correct.
>

Again, I accept Lott's view here.

Tim Lambert

unread,
Nov 18, 2001, 9:30:05 AM11/18/01
to
david <da...@example.com> writes:

> Also, it appears on first reading that the 5%-11% figure you quote
> refers to this first question, and not the later follow-ups. Is that
> right?

Yes. Since permit holders aren't going to carry all the time, I
believe that is the most reasonalbe comparison. As far as I can tell,
Kleck, who designd the survey, agrees with me on this point.

Tim

Tim Lambert

unread,
Nov 18, 2001, 11:05:32 AM11/18/01
to
mary...@aol.com (Mary Rosh) writes:

> Stop editing my responses!

Sorry, but I will continue to only quote the parts of your posting
that I am responding to. It makes it much easier on readers.

> It is amazing enough that you distort and
> lie about what Lott says.

I do no such thing.

> But do not edit my responses again.

I don't wish to hurt your feelings, but I will continue to follow what
I believe to be correct netiquette.

> You cut out the first part of my response to your claim that
> "Lott has not shown that states that issue the most permits have
> the biggest drops in crime."

Because you were just repeating his claim -- there was nothing to
respond to.



> You also refuse to defend all your other claims about such as the
> supposed reasons for why Duggan looked at only the fourth largest gun
> magazine when there are purely handgun magazines.

I don't need to. I'm not worried if you get the last word on something.


> Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3bsi1p...@cycloid.localdomain>...
> > mary...@aol.com (Mary Rosh) writes:
> > > REIFICATION fallacy -- Description: To reify something is to convert
> > > an abstract concept into a concrete thing. Reification is a Fallacy of
> > > Ambiguity. Reification is also sometimes known as a fallacy of
> > > "hypostatization."

> > > If anyone


> > > looks at the year by year changes in the figures (see especially the
> > > second edition), you see the increase in violent crime rates starting
> > > right at the time their neighbors adopt their law.
> >
> > Lott's graphs show a curve fitted to the data (the year to year
> > changes). Lott's graphs do not show the actual year to year changes.
> > You are committing the reification fallacy when you act as if the
> > fitted curve (an abstract model) is the same as the year to year
> > changes (concrete data).
> >
>
> What Bronars and Lott were examining in this case was whether crime
> trends changed before and after the adoption. They found strong
> evidence that was indeed the case. They go beyond just fitting a
> linear relationship, and see what the nonlinear relationship looks
> like. A simple linear relationship is going to pick up general
> changes in trends. It is hard to see why the nonlinear examination
> isn't going to pick up any important changes.

None of this relevant to what I wrote. I don't think you understood
at all what I wrote.


> Lott has given out his data to lots and lots of academics. If some
> how this examination of trends was misleading, I am sure that one of
> the many left-wingers in academia would have pointed out some problem
> with this.

Ooh, that was a give away, wasn't it? You think only a left-winger
would dare to criticize Lott. I don't know whether he is a
left-winger or a right-winger, but here is Robert Ehrlich, from
http://reason.com/hod/debate1.1.shtml

"John Lott's 1998 book, More Guns, Less Crime contains many points
with which I agree. For example, I believe that many criminals are
leery of approaching potential victims who may be armed -- an idea
at the core of his deterrence theory that guns help to prevent
crime. I also believe that violent criminals are not typical
citizens, and that the possession of a gun by a law abiding
citizen is unlikely to turn him into a crazed killer. Lott also
has a point when he speaks of the over-reporting by the media of
gun violence by and against kids and the corresponding
under-reporting of the defensive use of guns to prevent crime. As
a gun owner myself, I was quite prepared to accept Lott's thesis
that the positive deterrent effect of guns exceeds their harmful
effects on society, but as a scientist I have to be guided by what
the data actually show, and Lott simply hasn't made his
case. Here's why:

"Lott misrepresents the data. His main argument
that guns reduce crime is based on the impact on various violent
crime rates of "concealed carry laws," which allow any legal gun
owner to carry concealed weapons. Since these laws were passed at
different dates in different states, he looks at how the crime
rates change at t=0, the date of the law's passage in each
state. Lott's book displays a series of very impressive looking
graphs that show dramatic and in some cases immediate drops in
every category of violent crime at time t=0. The impact on
robberies is particularly impressive, where a steeply rising
robbery rate suddenly turns into a steeply falling rate right at
t=0 -- almost like the two sides of a church steeple. As they say,
when something looks too good to be true, it probably is. Lott
neglects to tell the reader that all his plots are not the actual
FBI data (downloadable here), but merely his fits to the data.

"The actual data are much more irregular with lots of ups and
downs, and they show nothing special happening at time t=0. Lott
has used the data from 10 states in his book. When we look at
changes in the robbery rate state by state, only two of the states
(West Virginia and Georgia) show decreases at t=0, while the other
eight show increases. Overall, averaging the 10 states, there is
a small but not statistically significant increase in the robbery
rate at t=0, certainly not the dramatic decrease Lott's fits
show. In fact, Lott's method of doing his fits is virtually
guaranteed to produce an "interesting" result at time t=0. What he
does is fit a smooth curve (actually a parabola) to the data
earlier than t=0, and fit a separate curve to the data after t=0.

"Given a completely random set of data, Lott's fitting procedure is
virtually guaranteed to yield either a drop or a rise near time
t=0. Only if the data just happened to lie on a single parabola on
both sides of t=0 would the fits show nothing special at that
time. Since random data would show a drop or a rise equally often
at t=0, we have a 50 percent chance of finding a drop -- not a
very good argument for the drop being real. The fact that all
categories of violent crime (murder, rape, assault, robbery) show
drops is also not particularly surprising, since the causes of
violent crime (whatever they are) probably affect the rates in all
the separate categories. Similarly, it is no more mysterious that
when the overall stock market rises or falls dramatically the
individual sectors (industrials, utilities, etc) are more likely
than not to move in the same direction."

And I should mention again that I have a Java applet at
http://www.cse.unsw.edu.au/~lambert/guns/lott/node12.html
that demonstrates Ehrlich's point -- fitting curves like Lott's to
random data shows something happening at t=0, even though nothing did.

> > Now back to the claim that states that issue the most permits have the
> > biggest drops in crime. You first need to answer this question and
> > then we can proceed:
> > If a random variable X is correlated to a r.v. Y and X is also
> > correlated with Z, does it follow that Y is corrrelated with Z?
>
> If X is positively correlated with Y and X is negatively correlated
> with Z (or the reverse set of correlations is true), there is not
> necessarily any systematic correlation between Y and Z. While your
> statement is not very clearly stated, I assume that you mean that X is
> correlated with Y in the same way that X is correlated with Z.

Alright, let's make it more concrete: X is positively correlated with
Y, r=0.7. X is positively correlated with Z, r=0.7. Does it follow
that Y is positively corrrelated with Z?

Tim

Tim Lambert

unread,
Nov 18, 2001, 11:17:01 AM11/18/01
to
mary...@aol.com (Mary Rosh) writes:

> Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3g07dp...@cycloid.localdomain>...

> > You continue to insist
> > that the survey included carrying for recreation when the actual
> > question asked was:
> > "In the last twelve months, have you ever carried a gun away from
> > home, either on your person, or in a vehicle, for protection against
> > crime? Do not count carrying for recreation or in connection with
> > duties in law enforcemnt, work as a security guard, or in the armed
> > forces"
> >
> > I think that anyone can look at the question and what you are saying
> > about it and form their own judgement about your credibility
>
> Lott claims that the question reads differently. I believe Lott, you
> can believe Kleck.

No, I believe what the question says. Have you ever considered
reading it yourself and forming your own opinion on it? The question
is right there, a few lines above.

Tim

Mary Rosh

unread,
Nov 19, 2001, 12:47:21 AM11/19/01
to
I am perfectly happy to have you cut out a particular question and a
response, but if you do so, I expect that you will retain my complete
answer and not edit the answer to that question.

Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3elmwf...@cycloid.localdomain>...


> mary...@aol.com (Mary Rosh) writes:
>
> > > Now back to the claim that states that issue the most permits have the
> > > biggest drops in crime. You first need to answer this question and
> > > then we can proceed:
> > > If a random variable X is correlated to a r.v. Y and X is also
> > > correlated with Z, does it follow that Y is corrrelated with Z?
> >
> > If X is positively correlated with Y and X is negatively correlated
> > with Z (or the reverse set of correlations is true), there is not
> > necessarily any systematic correlation between Y and Z. While your
> > statement is not very clearly stated, I assume that you mean that X is
> > correlated with Y in the same way that X is correlated with Z.
>
> Alright, let's make it more concrete: X is positively correlated with
> Y, r=0.7. X is positively correlated with Z, r=0.7. Does it follow
> that Y is positively corrrelated with Z?
>

Yes, of course.

Tim Lambert

unread,
Nov 19, 2001, 7:48:19 AM11/19/01
to
mary...@aol.com (Mary Rosh) writes:

> Tim Lambert <lam...@cse.unsw.edu.au> wrote in message news:<m3elmwf...@cycloid.localdomain>...
> > mary...@aol.com (Mary Rosh) writes:
> >
> > > > Now back to the claim that states that issue the most permits have the
> > > > biggest drops in crime. You first need to answer this question and
> > > > then we can proceed:
> > > > If a random variable X is correlated to a r.v. Y and X is also
> > > > correlated with Z, does it follow that Y is corrrelated with Z?
> > >
> > > If X is positively correlated with Y and X is negatively correlated
> > > with Z (or the reverse set of correlations is true), there is not
> > > necessarily any systematic correlation between Y and Z. While your
> > > statement is not very clearly stated, I assume that you mean that X is
> > > correlated with Y in the same way that X is correlated with Z.
> >
> > Alright, let's make it more concrete: X is positively correlated with
> > Y, r=0.7. X is positively correlated with Z, r=0.7. Does it follow
> > that Y is positively corrrelated with Z?
> >
>
> Yes, of course.

Nope. Here's a simple counterexample: Let Y and Z be independant
r.v.s and let X = Y+Z. Then X is positively correlated with
Y, X is positively correlated with Z and Y and Z are not correlated.

Tim

0 new messages