[lbo-talk] Southern vs. Northern violence

[Miles Jackson]

Michael McIntyre wrote:
> I did some quick checking of homicide rates to see if violence is  
> higher in northern or southern cities.  The strange answer, looking  
> at raw homicide rates, seems to be that border cities are  
> disproportionately bad.
> 2004 Unadjusted Ranking
> 1. New Orleans
> 2. Baltimore
> 3. Detroit
> 4. Washington, D.C.
> 5. St. Louis
> 
> 
> What happens, though, if you take out the "five highly  
> intercorrelated variables (factor loadings in parentheses): the  
> poverty rate (.934), male unemployment rate (.888), % black (.839), %  
> female-headed families w/own children under 18 (.928), and median  
> family income (-.862)".  (http://www.cjgsu.net/initiatives/ 
> HomRates-2005-12-06.htm) The adjusted homicide rates give a very  
> surprising ranking. (http://www.cjgsu.net/initiatives/ 
> HomRates-2005-12-06-Score.pdf)
> 1. San Francisco
> 2. Washington, D.C.
> 3. New Orleans
> 4. Baltimore
> 5. Kansas City

I don't get the point of this multiple regression model.  What the guy 
did was partial out various sociodemographic factors (not just those 
above, but also divorce rate, population, and residence stability) to 
create the "adjusted" ratings above.  So if the cities were equivalent 
on all these sociodemographic factors that are in fact integral to urban 
life in that city, the cities would show the "adjusted" rankings.  So 
what?  You're still more likely to be a homicide victim in New Orleans 
than in San Francisco.

Miles