[lbo-talk] the white vote

[KJ]

Seems to me that few serious conclusions can be drawn from the exit poll
data as tabulated at the national level given the geographical divides which
overlay the racial responses. At a cursory glance through some states, it
seems like the serious red states mirror the blue states but with different
proportions. So, e.g., the white lower income tended to vote for Obama more
than the higher income, and that's true of Kansas as of Ohio, except that
the proportions for Kansas are suppressed relative to the proportions for
Ohio. It seems to me that what's needed are cross-tabs by region and
income/education/sex/age/religion by race -- or better, regressions throwing
in all these. The results might be pleasantly surprising.
Will the data set be made available at some point?

2008/11/6 SA <s11131978 at gmail.com>


> Doug Henwood wrote:
>
>>
>> On Nov 5, 2008, at 12:44 PM, SA wrote:
>>
>>  Doug Henwood wrote:
>>>
>>>  McCain's margin among the under-$50,000 households was 4 points;
>>>>
>>>
>>>
>>> So McCain comfortably won the poorer half of whites - and in a year when
>>> the economy is melting down and the incumbent might as well have molested a
>>> child on live television. Surely there must be a lot of
>>> what's-the-matter-with-kansas-ing going on. No?
>>>
>>
>> Bartels says they agree with Rep econ policy. And there's also the race
>> angle, about which Tom Frank has almost nothing to say.
>>
>
> I thought Bartels' analysis on that was pretty perfunctory. So the bottom
> half wants to privatize Social Security, shrink Medicare, cut food stamps,
> make it harder to join a union? I wasn't convinced of any of that just by
> some GSS numbers on a generic "more government services"/"less government
> services" question.
>
> You're right about the race angle. Actually, I think about 80% of
> everything under the heading "cultural issues" is actually about race if you
> look deep enough.
>
> SA
>
>