MIAMI (AP) _ Voters in Florida were focused on the economy, with more than 6 in 10 calling it the nation’s top problem, according to results from exit polling in the state done for The Associated Press. Among the findings:


The economy was foremost on the minds of Florida voters. Health care, the deficit and foreign policy lagged far behind.


When it came to economic issues facing voters, rising prices and unemployment were the top concern of Floridians casting their ballots. The state has an 8.7 percent unemployment rate.


President Barack Obama was favored among female, black, Hispanic and young voters in Florida, while Republican challenger Mitt Romney was favored by white, male and older voters. Obama was supported by more than half of female voters in the Sunshine State. He was favored by more than 9 of 10 black voters and 3 of 5 Hispanic voters in Florida. The president also was the choice of two-thirds of voters under age 30. Romney was favored by more than 3 of 5 white voters and almost the same percentage of voters over age 65.


More than half of voters who described themselves as moderate chose Obama. Independent voters split for Obama and Romney, compared to four years ago when they decisively favored Obama.


Voters who picked having a vision as a top candidate quality slightly favored Romney, but voters who chose having a candidate who cares about them as their top trait overwhelmingly picked Obama.


Florida voters made up their minds early. More than two-thirds of voters said they had made up their minds before September.


Half of Florida voters believe that Obama’s health care overhaul should be repealed. Slightly more than half of voters also support an amendment to the state constitution that encourages state leaders to resist the implementation of Obama’s health care overhaul.


Slightly more than half of Florida voters believe President George W. Bush is to blame for the nation’s economic problems. Four in 10 blamed Obama.

The survey of 4,255 voters was conducted for the AP by Edison Research. This includes results from interviews with voters at 50 polling places statewide; 811 who voted early or absentee were interviewed by landline or cellular telephone from Oct. 25 through Nov. 2. Results for the full sample were subject to sampling error of plus or minus 3 percentage points; it is higher for subgroups.