"This race is a statistical tie," said Jim Lee, president of Voter Survey Service, which conducted the poll. "The key is what Scott can do to grow his support with Republicans and independents."
While Scott wins the Republican vote by a 69-15 margin and Sink holds Democrats 75-14, Sink leads among independents 47-36, with 17 percent undecided.
Lee said Scott's future hangs with those undecideds.
"Sixteen percent of Republicans are undecided, and this represents 51 percent of all undecided voters in the poll, so if Scott can inch upward with GOP voters, this could be enough to put him over the top," Lee said.
Scott's costly and caustic primary battle with Attorney General Bill McCollum left the former health-care executive with high negatives. Even his running-mate choice of state Rep. Jennifer Carroll, a McCollum supporter, has done little to heal the intraparty wounds.
Scott's rating among Republicans is just 49 percent favorable vs. 28 percent unfavorable.
Sink, by contrast, enjoys a solid 63-12 favorable/unfavorable ranking among Democrats.
Similarly, Sink, the state's elected chief financial officer, has a much better image with independents (33-29 positive) while Scott is inverted -- 51/22 negative to positive.
"This could really be a factor if Scott can’t appeal to these critical swing voters down the stretch," Lee speculated.
The news is even grimmer for Scott, considering that more respondents in the Sunshine State News Poll identified themselves as Republicans -- 45 percent GOP, 41 percent Democratic and 14 percent independent.
Nonetheless, there are indications that Scott can recover.
Among “excellent” voters -- those most likely to vote -- the race flips narrowly in Scott’s favor (44-43), while among “good” and “fair” voters Sink leads 53-27.
"Good" and "fair" voters are disproportionately Democratic and much less likely to vote on Nov. 2.
Geographically, Scott leads in the North (49-36) while Sink leads in the South (52-36). The Central region -- which accounts for nearly half the total vote -- is a dead heat (Scott 44-42).
"The 'intensity' to vote favors the Republicans (91 percent are 'excellent') over the Democrats (only 79 percent say their chances of voting are 'excellent'), so this means if Scott can solidify the GOP vote a little more and Democrats remain less enthusiastic, the advantage would flip back to Scott from a voter-turnout perspective," Lee said.
Commenting in advance of the release of the poll results, Sink spokeswoman Kyra Jennings said, “Alex Sink is receiving support from Republicans, Democrats and independents, including the support of Florida’s law enforcement officers, because she will be an honest, commonsense governor who will hold Tallahassee politicians accountable.
"The more Floridians learn about Alex Sink’s decades as a respected Florida business leader and her detailed plans to create jobs, support small businesses and improve our schools, the more her support will continue to grow," Jennings said.
Trey Stapleton, a spokesman for the Scott campaign, said, "Alex Sink has had a free pass so far during this race. As we approach Election Day, voters will know that they have a clear choice between a political outsider with a proven track record of creating jobs and a Tallahassee insider who wants to bring Obama’s liberal agenda to Florida."
The Sunshine State News/VSS Poll surveyed 1,016 likely Florida voters Sept. 1-5 and Sept. 7, and had a margin of error of 3 percent.
--
Contact Kenric Ward at kward@sunshinestatenews.com or at (772) 801-5341.
METHODOLOGY
This survey was commissioned by Sunshine State News and conducted Sept. 1-5 and Sept. 7 by Voter Survey Service, a division of Pennsylvania-based Susquehanna Polling and Research in Harrisburg. Calls were automated, meaning that voters were contacted using a pre-recorded set of questions instructing voters to complete the survey by pressing their telephone key pads to score their answers; no live interviewers were used. The universe includes a random sample of likely voters from a statewide voter registration list of registered voters who have prior vote history in either the 2008 or 2006 general elections, with most interviews conducted among voters who voted in both elections. A vote-intensity screen was also applied to gauge interest in the upcoming general election, with interviews being exclusively conducted among those who indicated they have an “excellent,” “good” or “fair” chance of voting, while those who indicated their chances of voting were “poor” were disqualified. Interviews are also monitored to ensure a representative sample of the state’s voters are interviewed proportionally based on demographics like geographic region, age and gender consistent with what we believe will be the likely voter turnout model on Nov. 2. Results are sometimes statistically weighted to reflect under or over samples within various demographic groups. The margin of error for a sample size of 1,016 interviews is +/- 3.07% at the 95% confidence level, but higher for sub-groups of respondents.