Sen. Bob Casey leads GOP challenger Tom Smith by just 1.2 points, 45.2 percent to 44. Libertarian Rayburn Smith takes 2.2 percent.
As for approval numbers, Casey’s numbers are positive but lackluster. 46 percent of respondents gave him positive marks, 42 percent gave negative. Smith has favorable numbers, 43 percent to 32.
It’s not the first CU poll in Pa. this cycle. The group released a survey during the GOP primary, too. Smith, a longtime donor to many conservative candidates and causes, contributed to Citizens United in the past ($800). The CU PAC endorsed Smith in the primary and contributed $10,000 to his campaign.
The group is perhaps best known for its legal victory, in which the U.S. Supreme Court struck down limits on campaign spending by corporate donors.
The firm surveyed 1,000 likely Pa. voters from Oct. 22 to 23 via live telephone interviews. The margin of error is plus or minus 3.07 percent.
44 percent of respondents identified as Democrats while 40 percent identified as Republicans, making it among the more GOP-friendly polling samples of the Senate race so far.
The pollster also said that men and women are both evenly split in the race – a departure from the near-universal finding of a significant advantage for Casey among women voters.
It’s the latest of several internal polls, commissioned by Republicans or GOP-favoring groups, that shows the race in at or near a tie. As with any internal poll, take it with a grain of salt.
Independent polls tell a slightly different story. Casey leads by 6.5 points according to an average by Real Clear Politics.
Casey Campaign Manager Larry Smar made note of difference.
“Who knew there was such a growth industry in messaging polls from right-wing groups that have endorsed Tom Smith?” he said. “Independent public polls show the lead around ten points. It is clear that Tom Smith would rather talk process than meet with editorial boards or face scrutiny by the press on his extreme Tea Party agenda.”
Here’s the full memo, and you can click here to read the toplines: