Newsgroups: talk.politics.guns
Path: tribune.usask.ca!decwrl!ames!agate!howland.reston.ans.net!sol.ctr.columbia.edu!news.mtu.edu!bogus!me
From: cescript@mtu.edu (Charles Scripter)
Subject: Re: Crime Stats
Message-ID: <7456546718359@mtu>
Sender: news@mtu.edu
Nntp-Posting-Host: physerver.phy.mtu.edu
Organization: Michigan Tech
X-Newsreader: TIN [version 1.1 PL9]
References: <249iol$6bu@hw_chevy.agcs.com>
Distribution: na
Date: Wed, 18 Aug 1993 07:03:11 GMT
Lines: 92
On 10 Aug 1993 18:41:41 -0700, Andrew Ford (forda@agcs.com) wrote:
> In article <1993Aug6.170751.10138@mdd.comm.mot.com>,
> booth@mdd.comm.mot.com (Greg Booth) writes:
> > A change in gun laws in parts of Australia during/after WWI has been
> > claimed to have reduced homicide (or violent crime) rates, but
> > there enough other variables (i.e shipping young men off to war) to
> > render that conclusion suspect. Not to start that argument again. :-)
Actually, if you examine Tim Lambert's model (which I did), you need
only examine the data from 1900-1937 to see that the data is
correlated beyond the 2 parameter model that Lambert proposed.
Lambert's specious argument was based on data from 1910-1930, whereas
the from 1900-1937 gives exactly the same averages as Lambert's choice
of years. The expanded data makes it apparent that a single parameter
is insufficient to describe his data before OR after. The 2 data
sets, 1900-1920 and 1921-1937 each show a sequence of 6 consecutive
values which lie on ONE SIDE of the average:
The average before: a=2.3 [for both (1910-20) and (1900-20)]
Data for 1907-1912: 3.0, 3.0, 3.1, 2.6, 2.7, 2.4
The average after: b=1.5 [for both (1921-30) and (1921-37)]
Data for 1926-1931: 1.6, 1.6, 1.9, 1.7, 1.7, 1.6
A proper distribution will be equally divided about the average, and
will not show this correlation. The probability that this occurrance
is is due to random fluctuations is (1/2)**6 = 1/64, thus the
probability that random fluctuation caused this trend in BOTH halves
of the data is (1/2)**12 = 1/4096
[if your fit results in _many_ consecutive data points lying on one
side of your of your average, then your fit is probably wrong]
> Such claim is BS.
Agreed.
> In fact, it is more likely that these laws (passes in
> 1919, I think) actually precipitated an increase in crime.
1921.
It seems that some folks found it worthwhile to examine Lambert's
model closer. I concluded that his model was FUBAR based on
examination of the correlation above, and the trends apparent in the
graph below.
The constants "a" and "b", indicated between the "==>", "<==" arrows.
Each "*" corresponds to 0.2 units.
before (1900-20) ave=2.3; after (1921-37) ave=1.5
WW I ends, 1918
|
* | War vets come home?
* *** /
* *** * **
* ***** * **
===> * * ****** ** ** <== a=2.3
* * *********** **
* * ************ ** * *
* ************** *** *** * *
==> * * ************** **** ****** * * * <== b=1.5
*** ******************* ************
*** ******************* *************
**************************************
**************************************
**************************************
**************************************
**************************************
1 1 1| 1 1
9 9 9 \ 9 9
0 1 2 \ 3 3
0 0 0 \ 0 7
|
Year the law passed, 1921
--
Charles Scripter * cescript@phy.mtu.edu
Dept of Physics, Michigan Tech, Houghton, MI 49931
---------------------------------------------------------------------
"... And what country can preserve its liberties, if it's rulers are
not warned from time to time, that this people preserve the spirit of
resistance? Let them take arms. The remedy is to set them right as
to the facts, pardon and pacify them. What signify a few lives lost
in a century or two? The tree of liberty must be refreshed from time
to time, with the blood of patriots and tyrants. It is its natural
manure." Thomas Jefferson, Nov. 13, 1787