Provisional ballot use increased in 72 counties, remained the same in seven counties, and decreased in nine counties. While the number of overall provisional ballots grew in 2011, the proportion of provisional ballots counted decreased. The study found that most of the provisional ballots were rejected for one of three reasons:
- + 40.2 percent were rejected because the voter was not listed as registered to vote in Ohio;
- + 21.2 percent were rejected for voting at the wrong precinct and wrong polling place; and
- + 16.2 percent were rejected for voting at the wrong precinct within the correct polling place [often called the “right church, wrong pew” problem – ed.].
Provisional ballots are always interesting for a couple of reasons. First, they are a strong indicator of the health of the voter list; the more provisional ballots cast, the more people who went to the polls thinking they were registered but were not. Second, in the majority of states (like Ohio) where provisional ballots must be cast in the proper polling place, provisional ballots can be an indication of voter confusion – resulting in the rejection of ballots even for those individuals who are otherwise validly registered to vote. Often, the focus in the field (especially in the media) is on how many of the provisional ballots cast actually get counted – and suggest that that the higher this number is, the better the jurisdiction is performing. In this way, provisional ballots are seen as a kind of safety net for voters.
In truth, however, what we should really be focusing on is reducing the number of provisional ballots cast – and more importantly, ensuring that very few of the (hopefully) very few provisional ballots cast are counted – because those voters were never eligible in the first place. In this model, potential problems are prevented before they occur.