It pretty much goes along party lines until you begin to factor in the independents and as of now it swings the election to Romney. http://www.gallup.com/poll/153902/R...cation&utm_content=morelink&utm_term=Politics But Like it says in the article, it's Spring and still early and history has shown anything can happen.
You have to watch out with these polls. Many use too large a democrat sample. I will question any poll which does not explain how many dems vs reps or Is they sample. And if this were a random call poll - gallup explained at the bottom of the page they use weighting. Monday, April 16, 2012 The Rasmussen Reports daily Presidential Tracking Poll for Monday shows that 24% of the nation's voters Strongly Approve of the way that Barack Obama is performing his role as president. Forty percent (40%) Strongly Disapprove, giving Obama a Presidential Approval Index rating of -16 (see trends). Most voters (56%) continue to favor repeal of the presidentâs health care plan. Even in Massachusetts, the reviews are mixed. Forty-six percent (46%) of Bay State voters would like the health care law repealed, while an equal number are opposed. In a hypothetical Election 2012 matchup, presumptive Republican nominee Mitt Romney earns 47% of the vote, while President Obama attracts support from 44%. Matchup results are updated daily at 9:30 a.m. Eastern (sign up for free daily e-mail update). See tracking history. http://www.rasmussenreports.com/pub...l_tracking_poll
There are 42 million registered democrats in The US,30 million registered republicans If a poll randomly calls 1000 registered voters odds are there will be more registered democrats in the poll then registered republicans because there are more registered democrats in the country.If a poll selectively takes out democrats that were randomly called and replace them with republicans then the poll would be rigged. http://www.usatoday.com/news/politics/story/2011-12-22/voters-political-parties/52171688/1 Registered Democrats still dominate the political playing field with more than 42 million voters, compared to 30 million Republicans and 24 million independents
Swing state and national polls that dont "weigh" or purposely take democrats out of their poll results
Eliminate all polls of "registered" voters and use only polls of "likely" voters. This will give you a much more accurate picture. And you're right... anything can happen between now and November.
How does USA Today calculate those numbers since only 28 states register voters by party affiliation?
Accurate polls do not "selectively take out democrats." They go through a very arduous process of determining "likely" voters. It costs almost three times as much to determine "likely" voters and most polling organizations won't do it this long before the election. The reason that polls of "likely" voters is important is that democrats are much more likley to register and then not vote. So most polls of registered voters do not accurately reflect what will happen in the election. How big the "gap" is varies from year to year.
+1 However, with the exception of Rasmussen, most polls currently are using registered voters, which is less accurate. Most switch to likely voters when the election date is near. They should use likely voters now.