In the 2008 general election, Florida voters cast some 35,635 provisional ballots on Election Day.  That’s but a fraction of the more than 8.3 million ballots cast in the election, but in close elections, local, state House and Senate, or even presidential, they could determine the outcome an election.

But unlike regular ballots cast by voters, provisional ballots–despite what we’re told–often don’t count.  In fact, in the 2008 general election, less than half of all provisional ballots cast were actually deemed to be valid.  Days after the polls closed on Tuesday, November 4, 2008, and long after the unofficial results were posted by the Secretary of State and broadcast by the media, local three-member canvassing boards in the state’s 67 counties opened thousands of envelopes containing provisional ballots and began to tabulate them.

Whether they count, is another question altogether.  Of the 35,635 provisional ballots cast in the 2008 general election, local canvassing boards validated only 17,312, or less than 50%.

The dirty little secret in the Sunshine State is that provisional ballots often don’t count. Or at least they don’t count as frequently in some counties as in others. There are innumerable reasons for the disparity, but the disparity exists. For whatever reason, provisional ballots cast by registered voters don’t have an equal shot of being accepted by local canvassing boards. The assault on voting rights by the Florida legislature in 2011, with the passage of HB1355, will likely increase the proportion of provisional ballots cast in the 2012 general election, and could very well lead to an even lower likelihood that provisional ballots will be validated.

In the 2008 general election there was a tremendous amount of variation across the state’s 67 counties regarding the number of provisional ballots cast and the percentage that were actually added to the final tabulation.  In six counties, all of them largely rural, all of the provisional ballots cast (a total of 54) were deemed to be valid by the county canvasing boards (Baker (0/0); Dixie (11/11); Hamilton (12/12); Holmes (13/13); Lafayette (3/3); and Suwannee (15/15)).

Other counties, as this Provisional Ballots Chart reveals, also had high percentages of validated provisional ballots.  For example, over 82 percent of the 731 provisional ballots cast in St. Johns County, 72 percent of the 411 provisional ballots cast in Pasco County, and nearly 60 percent of the 4,659 provisional ballots cast in Hillsborough (a Section 5 Voting Right Act county) were added to the total vote.

This 2008 Provisional Ballot Plot, crafted by my collaborator, Dartmouth University Professor Michael Herron, helps the visualization of where provisional ballots were cast in Florida in the 2008 general election. The proportion of the total votes cast in each county that were provisional ballot runs along the horizontal axis, and the percentage of provisional ballots cast in each county that were validated by the 67 county canvassing boards runs up the vertical axis. The size of the dot is proportional to the total number of provisional ballots cast, as distributed across the 67 counties.

There are several outliers, but two are pretty dramatic: Broward County, with its paltry acceptance rate of cast provisional ballots, and Osceola County, with its exceptionally high proportion of provisional ballots cast.

As I’ve written elsewhere with Dr. Herron, the rate of provisional ballots, the acceptance rate of provisional ballots, and the variation across counties should all be of grave concern as we head into the 2012 general election.

In the coming months, we’ll be investigating why there might be so much variation in the casting and counting of provisional ballots in Florida.  I suspect it’s quite likely that these clear disparities across Florida’s 67 counties are not out of the ordinary when it comes to voting provisional ballots in other states.