My guess is that the importance of the so-called Bradley effect has
been vastly overestimated. The Bradley effect happened in the early
1980s, right? Things have changed a lot since then. I can't see the
reasoning behind it anyway. The idea is that people don't admit to
pollsters that they won't vote for Obama because he's black. But that
makes no sense. They could just say they plan to vote for McCain, and
give any old reason. Why should people be afraid or ashamed to tell an
anonymous pollster over a telephone who they intend to vote for? That
the Bradley effect could account for 4-5% of the vote seems impossible
to me. I think it's largely a myth. If I have this wrong, please
explain it to me. 

--- In FairfieldLife@yahoogroups.com, "boo_lives" <[EMAIL PROTECTED]> wrote:
>
> --- In FairfieldLife@yahoogroups.com, "shempmcgurk" <shempmcgurk@>
> wrote:
> >
> > 
> > Why can't Barky close the deal?
> > 
> > The latest Gallup poll puts him ahead by 2 measely points...well
within
> > the margin of error.
> > 
> > Of course, this is without the Bradley Effect factored in:
> > 
> > 
> > STRESS BY GALLUP: OBAMA +2
> >
> Why can't Sidney close the deal?  Anyone who takes the daily gallop
> poll seriously knows nothing about polls.  But that's the one shemp
> picks while ignoring all the reputable ones!  At least he didn't
> repeat his master drudge's attempt to make it a close election by
> citing some make believe vote done at a grade school.  The election
> will be close because that's the nature of the country these days, but
> an avg of reputable polls shows obama holding on to about a 6-7 pt
> lead.  Racist republicans will create a bradley effect of about 4-5% I
> think, so it's close.  Unless the superior obama run campaign GOTV
> operations cancels out some or all of the bradley - this is what
> insiders tell me, but I won't believe it till I see it.  I don't have
> it in front of me, but my model currently has obama winning with about
> 285 EVs (you need 270 to win), so I'm still very nervous.
>


Reply via email to