Guns do not cause crime - liberals do

[funkedup]

First, almost all of the factors listed by Miko (crime, illegitimacy and deviancy) linearly increase with population density. They also linearly decrease with increasing income. The chances are pretty good that you're not going to see a lot of petty crime among people making substantial sums of money, after all.

More cause and effect problems.
The same type of character flaws and antisocial tendencies that cause people to have difficulty earning an honest income can also cause people to commit crimes (and to vote for Democrats ).

[Dead Man Flying]

Originally posted by funkedup
Holy cause and effect issues Batman!
You are inferring a causal relationship from a simple correlation.  Tsk tsk tsk.

Regression tests a theoretically causal relationship.  So it's true that one must frame the regression based on theoretically-sound principles -- I could have just as easily argued that income increases as one becomes more conservative.  But that just doesn't make any sense.  

Unless you're arguing that other factors beyond merely income determine ideology, or that this requires structural equation modelling; and of course you'd be right.  Many factors determine one's ideological disposition beyond merely income (parents' ideology, peer groups, etc).  As income does not correlate very highly with these other factors, we can reasonably expect that the error term contains this unexplained variance rather than biasing income.

-- Todd/Leviathn

[Dead Man Flying]

Originally posted by funkedup
The same type of character flaws and antisocial tendencies that cause people to have difficulty earning an honest income can also cause people to commit crimes (and to vote for Democrats ).

I challenge you to operationalize "character flaws" and "antisocial tendencies" in a model, my friend.  

-- Todd/Leviathn

[funkedup]

I challenge you to visit a jail or juvenile detention facility and not come out of there thinking that those people would have a hard time earning a living even if they didn't have a record.

[lazs2]

I think a case could be made that the big cities have liberal ideas about how people should behave and what types of gun control and other laws should be in force... I believe that a case could be made that they screw up their own cities so .....

we shouldn't allow them to make decisons that affect rural people.
lazs

[Dead Man Flying]

Originally posted by funkedup
I challenge you to visit a jail or juvenile detention facility and remain not come out of there thinking that those people would have a hard time earning a living even if they didn't have a record.

I don't doubt it, but I'd be hard pressed to consider it, say, a genetic character flaw.

-- Todd/Leviathn

[funkedup]

Originally posted by Dead Man Flying
Regression tests a theoretically causal relationship.  So it's true that one must frame the regression based on theoretically-sound principles -- I could have just as easily argued that income increases as one becomes more conservative.  But that just doesn't make any sense.  

Unless you're arguing that other factors beyond merely income determine ideology, or that this requires structural equation modelling; and of course you'd be right.

-- Todd/Leviathn

I'm arguing that ideology and income are at least partially determined by intelligence and moral integrity (self control) and social skills.

[funkedup]

Originally posted by Dead Man Flying
I don't doubt it, but I'd be hard pressed to consider it, say, a genetic character flaw.

-- Todd/Leviathn

Maybe not genetic, but definitely passed on from one generation to the next.  Basically what I'm talking about here is addiction.  Things like addiction and child abuse are passed on from generation to generation, and most people in prison have sufferred one or both.

[lazs2]

so everyone agrees?   women and big city dwellers should not be allowed to vote?
lazs

[funkedup]

Originally posted by lazs2
I think a case could be made that the big cities have liberal ideas about how people should behave and what types of gun control and other laws should be in force... I believe that a case could be made that they screw up their own cities so .....

we shouldn't allow them to make decisons that affect rural people.
lazs

I definitely agree with that.  Too many "least common denominator" laws in this country.  Why should my liberties be restricted just because a few losers can't handle responsibility?

[Eagler]

Originally posted by lazs2
so everyone agrees?   women and big city dwellers should not be allowed to vote?
lazs

skinny city dwellers either

[Dead Man Flying]

Originally posted by funkedup
I'm arguing that ideology and income are at least partially determined by intelligence and moral integrity and social skills.

I'm not so sure about ideology (at least directly), but I could see where intelligence, integrity, and social skills would impact on income.  The problem, of course, is measuring such things in a nonobtrusive way.  You can't just ask someone if they're smart or dumb, and you can't really measure moral integrity or social skills in a poll.  The result is something that prima facie makes sense, but there's just no way to test it beyond looking at anecdotal examples.

Any significant impact that intelligence, moral integrity, and social skills would then play on ideology would occur indirectly through income.  Measuring that would require a structural model with multiple regression equations, and there's no way in hell I'm investing that kind of energy here.  

-- Todd/Leviathn

[funkedup]

Originally posted by Dead Man Flying
Measuring that would require a structural model with multiple regression equations, and there's no way in hell I'm investing that kind of energy here.  

That's Ph.D.-speak for "OK you win."  

[Dead Man Flying]

Originally posted by funkedup
Maybe not genetic, but definitely passed on from one generation to the next.  Basically what I'm talking about here is addiction.  Things like addiction and child abuse are passed on from generation to generation, and most people in prison have sufferred one or both.

I'd be curious to see how addiction corresponds to something like income.  Though obviously addicts appear at all income percentiles, the lowest income level appears to contain a disproportionate number of addicts relative to everyone else.  I don't think this addiction is a matter of low intelligence, but perhaps it is related to a sense of helplessness (either real or imagined).

OTOH, I have to wonder how many hopelessly addicted criminals actually vote in presidential elections.  My guess is pretty few.

-- Todd/Leviathn

[Dead Man Flying]

Originally posted by funkedup
That's Ph.D.-speak for "OK you win."  

hehe Like hell.    The important thing to remember about indirect effects is that they'll almost always be less than the direct effect, and it's also possible that those variables have no statistically signficant direct effect on ideology whatsoever.  In order to test this, I'd need variables I don't have, and I'd have to dig out books to remember how to test structural equations.  Bah!

-- Todd/Leviathn

PS, we're geeks