Thursday, August 09, 2007

New Poll Shows Obama in Third,..Among Iowa Republicans

There was a new poll released yesterday by the University of Iowa. The poll is interesting to take a look at because it breaks down caucus goers based on how likely they will attend the caucuses.

However, the most shocking finding in the poll comes from the question about presidential preference that included candidates from both parties...

Obama actually finished third as the preferred general election candidate of registered Republicans, at 6.7 percent, behind Romney (21.8 percent), and Giuliani (10 percent), but ahead of Thompson (5.2 percent) and McCain (1.8 percent).
Another interesting thing to note is that the results among the least likely caucus goers in Iowa are pretty similar to the national poll numbers.

From Open Left...
I am struck by how similar the “least likely” caucus goers in Iowa seem to break along roughly the same lines as national polls. Would anyone bat an eyelash at a national poll showing Clinton 33%, Obama 25%, Edwards 10%, and Richardson 6%? I don’t think so. Even though Clinton’s number has been higher lately, the similarity is interesting, and might imply that once the campaign heats up post-Iowa, there could be as much flexibility in national numbers as there was in 2004. Of course, there are only 106 in the least likely caucus goers sample, so it might be best not to put too much stock in those numbers.
And finally here are the results among most likely caucus goers...
Mitt Romney leads with 28% support, followed by Rudy Giuliani at 12%, Fred Thompson trails at 8%, Rep. Tom Tancredo at 5%, Sen. Sam Brownback at 4%, Sen. John McCain at 3% and Mike Huckabee at 2%.

John Edwards leads with 26%, followed by Sen. Hillary Clinton at 25%, Sen. Barack Obama at 19%, and Gov. Bill Richardson at 9%.

1 comment:

Anonymous said...

Here's a straw poll that actually means something, because plurality (vote for one) voting performs horrendously when you have more than two candidates. That's because candidates Y and Z can be preferred by a huge majority to candidate X, but can split the vote such that X still "wins". Here's a severe hypothetical example to illustrate this phenomenon.

The solution is Range Voting.