AZ Capitol Times: Provisional ballot voter profile apparent with medical marijuana measure tip?

PHOENIX - As I was grocery shopping the evening after the election, I bumped into Andrew Myers, the head of the Arizona Medical Marijuana Policy Project and the main proponent for Prop. 203. At that point, the measure to legalize medical marijuana was trailing several thousand votes, though the outcome was still too close to call.

We chatted for a bit and when I asked him how he felt about the situation, he sounded confident. Now, of course, if you're working on a campaign, your job includes sounding confident, but Myers made a very interesting case that evening for why he thought the vote count would turn around, delivering an eventual win for Prop. 203.

Myers said he expected the outstanding ballots to favor the medical marijuana measure at a much higher rate than the ballots that had been counted up to that point. One of the reasons he thought that was because the votes yet to be counted included all the provisional ballots.

For the next week and a half, I watched as the gap narrowed in ebbs and flows, then finally tipped toward Prop. 203 Friday night. Now, it appears Prop. 203 has a large enough lead that it’s almost impossible for the remaining votes to be counted could change the outcome. It appears Arizonans have voted, once again, to legalize medical marijuana.

But between the time Myers made his prediction about provisional ballots and the time when the vote flipped, I’ve been thinking about what can be said about voters who cast provisional ballots.

Provisional ballots are used when it’s unclear if the voter is registered. Most often, it’s because a voter shows up to the wrong polling location. In that example, the poll workers won’t have a record of the voter on their voter list, so he or she will be given a provisional ballot.

After that, county elections officials check the ballot (and the more than 100,000 other provisional ballots) against the voter database. If the person is registered and the signature on the ballot matches the one already on file, the vote is tallied.

I’ve been searching for some sound research on whether a correlation has been observed between provisional ballot voters and certain causes, campaigns or partisan stances, but I haven’t found any yet. As I’ve spoken with my colleagues in the Capitol Times newsroom and other journalists, we’ve snickered about the absentmindedness that accompanies some of the reasons for needing a provisional ballot, and the way that those votes turned the measure for legalizing marijuana.

Is it possible that provisional ballots should be expected to favor those who would support the use of marijuana - medical or otherwise? Hopefully, the data that comes from the official canvas of this year’s election will help provide some of that information and helps us draw some conclusions about why this particular measure was able to pull off a come-from-behind victory.