> You may be able to make a case comparing
> states and
> > counties, but you are comparing apples & oranges.
> They have
> > different histories. For example, I'm grew up in
> Mass. just
> > north of Boston, and I believe that your
> comparison between
> > N.H. & Mass. if fundamentally flawed. N.H. is
> > has been a rural state except in a few "mill
> towns". Mass.
> > is a fairly industrial state in the Eastern half
> (the part
> > bordering on N.H. I suspect Mass. has a much
> higher population
> > of "urban" poor, as compared to N.H. which may
> have "rural" poor.
> > The religious percentages in the two states are
> different as well.
> > I'm not sure if this is a factor, but it points
> out that *you* can't
> > go making the claims you have been making.
> Southeastern NH is as urbanized today as the
> adjacent territory in
> Massachusetts is.
That's not the whole state...
> > If you want to do this, you have to put *all* the
> > into a statistical regression model and tease out
> > exactly what the correlations are -- *and* then
> > you have the correlations, you still have work to
> > because correlations are *not* causation.
> Well, though I stated general perceptions, others
> have come up with
> solid stats, showing that the crime rate per 100,000
> population is 3
> times higher in Mass than NH. rates per 100,000
> should be the same no
> matter what your population density if all other
> factors are equal, so
> obviously something is not equal.
This *assumes* that density has no effect.
These figures are statewide *means*, so you cannot account for regional variations within the state, which might show density effects.