USA Today/Gallup Survey: Obama Up By 13
From USA Today's On Politics blog:
Sen. Barack Obama has opened up a 13 percentage point lead over Sen. Hillary Rodham Clinton in the battle for votes in Tuesday's New Hampshire primary, according to a USA TODAY/Gallup Poll conducted in the state from Friday through this afternoon. The results were just released.
On the Republican side, Sen. John McCain is 4 percentage points ahead of Mitt Romney.
The surveys of 776 New Hampshire residents who are "likely" to vote in the Republican primary and 778 New Hampshire residents who are "likely" to vote in the Democratic primary were all completed after the news from Thursday's Iowa caucuses had been reported.
The rundowns:
Democrats.
• Obama: 41%; up from 32% in the last USA TODAY/Gallup New Hampshire poll, taken in mid-December.
• Clinton: 28%; down from 32%.
• John Edwards: 19%; up from 18%.
• Gov. Bill Richardson: 6%; down from 8%.
• No one else above 3%.
Republicans.
• McCain: 34%; up from 27% in mid-December.
• Romney: 30%; down from 34%.
• Mike Huckabee: 13%; up from 9%.
• Rep. Ron Paul: 8%; down from 9%.
• Rudy Giuliani: 8%; down from 11%.
• No one else above 3%.
Each figure has a margin of error of +/- 4 percentage points. So Obama's lead is "outside" that range, while McCain's is not.
A new CNN/WMUR poll gives Obama a 10-point advantage over Clinton, and has McCain leading Romney by 6 points.
Soruces say Hillary has requested an important one on one meeting with Obama. She has intructed him to meet her " Alone" in Ft Marcy Park in Washington DC.
You know, I've been very impressed with the quality of the polling during this cycle. They called Iowa really well, and I also note that almost every poll released today (over a half dozen) all basically agree with each other, once one takes into account the margin of error. That tells me that most of the pollsters are probably looking at the same universe the same way, thus serving to reinforce the likelihood that they are all accurate.
About the only outlier I've spotted is the latest from Suffolk U, taken on the 4th and 5th. Suffolk U. claims a margin of error of +/-4.38 points, and reports a Clinton figure of 35, meaning that it's saying Clinton is somewhere between 30.62 and 39.38. This directly contradicts both Rasmussen (1/4-5) and ARG (also 1/45), which are in agreement that she is no higher than 30 (ARG's error margin is +/-4 with a Clinton figure of 26, while Rasmussen's error margin is with a Clinton figure of 27).
However, other than the Suffolk U. survey, all other current extant surveys which included January 5th for their field work, all six of them, CNN (1/4-5), ARG, the Concord Monitor, Rasmussen, Gallup, and CNN (1/4-6), have overlapping agreement with each other for all the candidates. The current picture appears to be as follows:
NH 1/4-6 Comp.
1. Obama 37-38 +2.6-5
2. Clinton 28-30
3. Edwards 18-21
4. Kucinich 0-5
4. Richardson 5-7
4. Undecided 5
7. Gravel 0-3
The previous universe of polls that agreed with each other covered the survey dates of 1/2-5. Comparing the two collections of surveys reveals that only Obama has moved a statistically significant amount between the previous and current universe of polls. So I added that movement figure to the Obama listing above. Note that he continues to move up, though it appears Hillary has stopped bleeding for the time being (she was trending down sharply in the previous 1/2-5 universe compared to the poll releases just previous, dating from 12/31-1/4).