There are some statistically significant shifts from June to July in ARG's survey results among likely Democratic primary voters.
All three of these polls were taken July 26-30 and have a margin of error of plus or minus 4 percent.
In order of primary calendar:
Iowa
Clinton 30 (-2 from June)
Edwards 21 (-8 from June)
Obama 15 (+2 from June)
Richardson 13 (+8 from June)
Undecided 15 (+1 from June)
New Hampshire
Obama 31 (+6 from June)
Clinton 31 (-3 from June)
Edwards 14 (+3 from June)
Richardson 7 (+1 from June)
Undecided 13 (+1 from June)
All others have 2 percent or less
South Carolina
Obama 33 (+12 from June)
Clinton 29 (-8 from June)
Edwards 18 (-4 from June)
Undecided 12 (+1 from June)
All others have 3 percent or less
Fine print, links, and some commentary after the jump...
The only thing that polls tell us this early out is which candidates are moving up or down (and it is advisable to only take movement of four points or more as statistically significant, given the four point "margin of error.") We can also speculate as to why the numbers are moving as they are.
All three of these states were polled July 26-30, after the July 23 debate.
Statistically significant movements:
Iowa: Edwards down 8 and Richardson up 8. It appears that the Clinton-Obama flap didn't help Edwards (as many pundits speculated) but did, interestingly, help Richardson (who is like the kitten outside the restaurant door, lapping up the spilled milk.) A major shot in the arm for Richardson, a wake-up call for Edwards, and if the next polls confirm this trend it would place Clinton in the unenviable position of being "front-runner" in Iowa, making the expectations game tougher for her: if this trend continues, even second place in Iowa could signal a campaign death knell. Here's the link.
Sample Size: 600 completed telephone interviews among a random sample of likely Democratic caucus goers living in Iowa (475 Democrats and 125 no party (independent) voters).
Sample Dates: July 26-30, 2007
Margin of Error: ± 4 percentage points, 95% of the time, on questions where opinion is evenly split.
New Hampshire: Obama up 6 and tied for the lead. In anti-war New Hampshire, his message is getting through. The endorsement by US Rep. Paul Hode (D-NH) may be helping. The post-debate flap clearly helped him here and did nothing good for Clinton. No big movement for Richardson yet, but clearly a surprise second place or better showing by Richardson in Iowa could radically change the dynamic, first in the Nevada caucus and then the New Hampshire primary. Here's the link.
Sample Size: 600 completed telephone interviews among a random sample of likely Democratic primary voters living in New Hampshire (400 Democrats and 200 undeclared (independent) voters).
Sample Dates: July 26-30, 2007
Margin of Error: ± 4 percentage points, 95% of the time, on questions where opinion is evenly split.
South Carolina: Obama up 12, Clinton down 8, Edwards down 4: Obama's been in the radio targetting black women that had been for Clinton and the strategy appears to be working. But look: He's also taking votes away from Edwards. Could it be that this year's "southern candidate" will be a Senator from Illinois? Here's the link.
Sample Size: 600 completed telephone interviews among a random sample of likely Democratic primary voters living in South Carolina (483 Democrats and 117 independent voters).
Sample Dates: July 26-30, 2007
Margin of Error: ± 4 percentage points, 95% of the time, on questions where opinion is evenly split.