In article ,
msnyder{at}redhat.com says...
>
> "Mark Borgerson"  wrote in message
> news:MPG.1cb1cb5095e09b3d98972f{at}news.comcast.giganews.com...
> > In article ,
> > msnyder{at}redhat.com says...
> > > Hello there, soc.men-ites,
> > >
> > >  "Lin which it was reported that
> > > 40% of spouse murders are committed by women (and that
> > > their average sentence if convicted is 6 years, compared
> > > to 16 years for men)?
> > >
> > > Well, I've found an independant source of data, and been
> > > able to confirm that result, plus I've eliminated the
> > > "self defense" argument with which feminists have attempted
> > > to discredit the finding.
> > >
> > > Figured you folks might find it interesting:
> > >
> > > The Chicago Homicide Database at Northwestern University
> > > An Online Treatise by Michael Snyder
> > >
> > > Chapter 1: Murder/Suicide
> > >
http://groups-beta.google.com/group/alt.true-crime/msg/73ccacd3e70e8f24
> > > Chapter 2: Self Defense
> > >
http://groups-beta.google.com/group/alt.true-crime/msg/977d6ba9b0b6a146
> > > Chapter 3: Spousal Self Defense
> > >
http://groups-beta.google.com/group/alt.true-crime/msg/b8119f617a000c33
> > > Chapter 4: Crime and Punishment
> > >
http://groups-beta.google.com/group/alt.true-crime/msg/3283cf27fcef38a0
> > >
> > > The Chicago Homicide Database at Northwestern University:
> > > http://homicide.northwestern.edu/database
> > >
> > > Tables for Chapter 2:
> > > Homicides, female-on-female
> > > http://www.sonic.net/~msnyder/femonfem.html
> > > Homicides, male-on-female
> > > http://www.sonic.net/~msnyder/monfem.html
> > > Homicides, female-on-male
> > > http://www.sonic.net/~msnyder/femonm.html
> > >
> > > Tables for Chapters 3 and 4:
> > > Homicides, wife-on-husband
> > > http://www.sonic.net/~msnyder/wife.html
> > > Homicides, husband-on-wife
> > > http://www.sonic.net/~msnyder/husband.html
> > >
> > >
> > In chapter 4 you say:
> >
> > "These findings, based on a very large and representative sample
> > of homicides from all over the US (but all from the same year,
> > 1988), have never been independantly verified, and have been
> > informally 'refuted' with the argument that the wives were
> > probably defending themselves (since "everybody knows" that
> > husbands attack wives far more often than wives attack husbands). "
> >
> >
> > But the study DOJ says that the data is from "Large Urban
Counties",
> > and your database seems to be the "Chicago Homicide Database".
> > While these two data sets may correlate well, are they really
> > a "very large and representative sample of homicides from all over the
> > US".
>
> Theirs is.  I urge you to read there preface, in which they discuss
> the representativity of their data.  Essentially, the vast majority
> of people in the US live in a large urban county, and the large urban
> counties included in their study constitute a very large and representative
> subset of such counties, so that the percentage of US citizens who are
> either directly or indirectly represented by the sample is truly astounding.
>
> >  Will your findings from these studies be representative
> > of the results in the rest of the US--where 2/3s of the population
> > lives in cities of less than 100,000 people?
>
> Mine won't -- but theirs will.  I can't do justice to the methodology
> section of the DOJ report, again I urge you to read it yourself.
>
> >  I don't know whether
> > the patterns of homicide differ between large urban counties and
> > smaller rural and suburban counties.
> >
> > I wonder why the DOJ picked large urban counties?  Perhaps it
> > was because it simplified the task of collecting a reasonable
> > sample---easier to pull the records from a few dozen large
> > jurisdictions that to get a similar number of case records
> > from a thousand smaller courts, I suppose.
>
> They picked them because they COULD obtain a very large
> and very representative sample from them.  It wasn't a few
> dozen -- it was a significant fraction of *all* of the large urban
> counties in the US, and "large" doesn't mean "bigger
than a million".
>

A large sample---yes.  Representative of the whole US?   I doubt it,
or they would have said so.  In this case large does mean "greater
than half a million",  (or greater that 689,000 if you use
2001 census data)

http://www.census.gov/popest/archives/2000s/vintage_2001/CO-EST2001-09.html
>
>   There is also
> > the possibility that the DOJ felt that the patterns in those
> > large jurisdictions were different from those in rurual
> > and suburban areas.
>
> Read it.  They address the issues that you raise.

Some of them, yes.  They do not conclude that the data represents
all spousal murders in the US.
>
>
> > I think you have an excellent analysis of the data you have
> > reviewed,  but I still wonder if the patterns of homicide
> > vary from urban to rural and suburban areas.
>
> Read it.  Here's the URL:
>
> http://www.ojp.usdoj.gov/bjs/abstract/spousmur.htm

I read the full article at :
http://www.ojp.usdoj.gov/bjs/pub/pdf/spousmur.pdf

There is nothing in the report to indicate that its results should
be applied to smaller jurisdictions.  The report data was taken
from the 75 largest counties in the US---the smallest of which now
has about 800,000 people.  Those 75 counties account for 105 million
people---about 1/3 the population of the US.  They also apparently
account for more than 1/2 the murders in the US.  That implies
a significantly lower murder rate in the other counties in the
US and it may also mean that patterns of spousal murder are different.

There is nothing in the report that covers the pattern of spousal
homicides in smaller counties.

> > I think that
> > the conviction and sentencing patterns  are probably less variable, as
> > the judges and lawyers involved probably have much the same
> > background as their counterparts in the large cities.  After
> > all,   most law schools are located in  or near the large
> > urban counties.
>
> Don't speculate -- read.
>
I did.  While the reports are broadly applicable, I think the confidence
intervals of many of the  statistics would be different if you applied
the conclusions to the other 2/3s of the US population.


Mark Borgerson



--- UseNet To RIME Gateway {at} 3/28/05 8:46:47 PM ---