Questioning the sanity of the electorate

[Alcon]

Pouring over some old exit polls, I came across this, from 2000's CNN poll:

13% of the population characterized Bush as being "too liberal."

Of those people, 81% (or about 11% of the overall population) voted for Gore instead.  In other words, 11% of the population thought Bush was too liberal and voted for Gore - 1 in about 9 voters.

About 1 in 256 people thought Bush was too liberal, but voted for Nader.

Thus, about 14% of Nader voters (around 1 in 7)  thought Bush was "too liberal."

Sad, sad, sad.

[Alcon]

May be some of these people had an "unAmerican" notion of what "liberal" means. In Europe or in Latin America, or almost everywhere else in the world, for that matter. disliking "liberals" means disliking free markets.  US is pretty much unique in the (mis)use of the term.  If a large proportion of these people were Hispanic, the answer would be straigtforward: in the discourse south of the border Bush is "neo-liberal", and Kerry is not (or not to the same extent).  That's simple.

That's the first thought that went through my mind.  Perhaps it is primarily international immigrants that dislike Bush, and thus didn't want to say "just fine."  This and a mix of total idiots.

[Alcon]

Exit polls have become so unreliable it's hard to trust them anymore.  Remeber how they predicted Kerry victories in Ohio?  Then again, maybe we are a nation of stupid people.  A recent study showed that only a relatively small percentage of Americans could find Iraq on a map.

Exit polls are usually pretty accurate. The fact that the exit polls were off in the Ukraine "proved" that the election was stolen.

For something with probability near 0% or 100%, the MOE is smaller.

Normally exit polls can be inaccurate on the state level, but the national one in this case has 13,000 respondents, which is sufficient to achieve statistical integrity, I believe.