A pair of new polls put President Barack Obama well in front of challenger Mitt Romney in their quest to win Pennsylvania and the White House.
President Obama has a 12-point lead -- 54 percent to 42 percent -- over Romney in a random telephone survey of likely voters in Pennsylvania, according to a poll by Quinnipiac University, CBS News and the New York Times.
The same poll also found the president with growing leads in two other battleground states, Florida and Ohio.
"The wide difference between the two candidates is not just a result of Romney's bad week," said Peter A. Brown, assistant director of the Quinnipiac University Polling Institute. "In Ohio and Florida, voters are basically split down the middle on whether the county and they and their families are worse or better off than they were four years ago. If voters don't think they are worse off, it is difficult to see them throwing out an incumbent whose personal ratings with voters remains quite high."
The Quinnipiac survey, conducted between Sept. 18 and 24, has a margin of error is 2.9 percentage points.
Another poll, conducted by Franklin & Marshall College in Lancaster, found President Obama with a 9-point lead -- 52 percent to 43 percent -- among likely voters and an 11-point lead -- 50 percent to 39 percent -- among registered voters over Mitt Romney.
The poll also found President Obama being seen as better able to handle foreign policy issues, better able to understand the concerns of ordinary Americans, better able to handle the job as military chief, closest to voters' views on abortion and gay marriage, and more prepared to fix the nation's economic problems.
In the U.S. Senate race, the Franklin & Marshall poll found incumbent Democrat Bob Casey Jr. leading the Republican challenger, Tom Smith, by a wide margin -- 46 percent to 34 percent -- although 14 percent of voters remain undecided. The race appears to be closer -- 48 percent to 38 percent -- among likely voters, eight percent of whom are still undecided.
The survey of 632 Pennsylvania voters was conducted from Sept. 18 through 23. It has a sample error of 3.9 percentage points. The survey included 392 interviews with likely voters, with a sample error of 4.9 percentage points.