“Obama Won %108 of Registered Voters in Ohio County”*

That’s one headline that made its way through the conservative blogosphere this weekend. While many conservatives (or apostates, depending on who you ask) like David Frum and Steve Schmidt have taken stock in the outcome of the 2012 election and determined that the rabid vitriol, paranoia, and trutherism of the far right must be reigned in order for mainstream conservative thought to stand a chance, that message appears not to have made its way to Twitter or FreeRepublic or any number of other prominent far-right bastions. But what if it were true? What if there really were more registered voters than eligible voters in a heavily-populated Ohio County? Even if not proof of intentional fraud, it would surely be evidence of significantly flawed voter rolls in Ohio, which anyone, regardless of political affiliation might find troubling. I wanted to know more.

The link directed me to a brief piece at FrontPage Magazine, which, in turn, appeared to be sourced entirely from a blog posting at Pundit Press*:

Mr. Obama won Wood County in Ohio this year. That’s right, Mr. Obama won the majority of Wood County’s 108% of registered voters. That’s not a typo.

In 2012, 106,258 people in Wood County are registered to vote out of an eligible 98,213. But it certainly must all be a coincidence, right?

That does sound strange. (Although it’s a far cry from his winning 108% of the vote, which is what is suggested by the title of the item floating around Twitter.) Pundit Press referenced a Cleveland Plain Dealer story. So I followed the link to learn more.

Hm. The Plain Dealer story focused on Ohio’s attempts to “clean up” its voter rolls, resulting in nearly half a million fewer registered voters in the state, as compared to 2008. Of the 490,000 registrations purged from Ohio’s rolls, the Plain Dealer noted that 44% of those purged were from Cuyahoga County, a county that, based on previous elections and polling data, was thought to be strong for President Obama. (He ultimately won 69% of the vote there.) So what was the basis of the PunditPress blurb?

At the end of the Plain Dealer article, a chart offers a thumbnail sketch of the Ohio electorate on a county-by-county basis. And there, two rows from the end comes the entry for Wood County:

So. That does seem a bit odd. How can there be 98,213 voting age people living in Wood County and 106,258 registered voters? Is this finally proof of the voter fraud that so many Republicans have warned against?

Well, not so fast.

First of all, the figure for voting age persons comes from an extrapolation of 2010 Census data. The total number of persons counted in the 2010 Census in Wood County was 125,488. To get the 98,213 figure cited by the Plain Dealer, they apparently subtracted out 21.3% of the population (based on the 2010 Census’ figure that 21.3% of Wood County’s population was under 18). But there are a couple of flaws in this calculation. For one, it ignores the nearly 1% population growth forecast by the 2010 Census for Wood County. Additionally, it fails to address whether and to what extent Wood County’s population aged—clearly, some fraction of the young people counted in 2010 have reached age 18 by 2012. Additionally, there have historically been difficulties in obtaining accurate Census counts of college students, with confusion about whether they should be counted as residents of their parents’ homes or of their school-year residences.

This is particularly important in Wood County, the county seat of which is Bowling Green, the home of Bowling Green State University. As of 2011, the undergraduate population of BGSU was 15,059, with a total student population in excess of 20,000. Under Ohio law, students may vote where they are students so long as they have a “present intent to remain” there and intend to make their school residence their principle home. Thus, it’s entirely possible that there are a large number of students who registered to vote but were not counted by the 2010 Census. To rely exclusively on the Census snapshot is problematic at best in determining what the eligible voter population is.

Additionally, it is entirely possible that the voter rolls in Wood County have not been purged since 2008 that the purge was minimal. Under the Help America Vote Act and the National Voter Registration Act of 1993, administration of voter rolls must strike a balance between purging ineligible voters and preventing improper purging of eligible voters. States may purge voters only when 1) they have failed to vote in at least two federal general elections and fail to respond to a notice of removal; 2) they have lost the right to vote under state law due to status as a felon; 3) they have died; or 4) the state receives notice from another state that the voters have registered in that state. Because so many of Wood County’s registered voters are likely current or former students, it is unsurprising that the rolls are so large. In striking that balance between accuracy and fairness, I would argue that federal policy wisely errs on the side of retaining ineligible voters (who may or may not ever cast ballots) rather than purging eligible ones.

Finally, based on figures reported by the Ohio Secretary of State, I think it is worthy of note that Wood County’s voter turnout was 57.4%. In contrast, the state-wide voter turnout rate was 67.2%. If fraudsters were truly targeting Wood County as ground-zero for their feared voting shenanigans, they surely weren’t very good at it.

So what does all this mean? Well, while it doesn’t definitively disprove the persistent far-right narrative of rampant voter fraud, without more, this single data point proves nothing, and is itself substantially more nuanced than is being reported in the right-wing blogosphere. What seems clear is that voter rolls are and ever will be imprecise snapshots. The more mobile a population group is, the more likely they are to cause inaccuracies in voter rolls.

I’m a liberal living in a Southern, perpetually red state. So why does this matter to me? Well, I have friends and loved ones who are already writing their personal histories of this election. And the story they tell isn’t pretty—voter fraud and intimidation, lies and coverups all leading to a terrible travesty of an election. I started this afternoon looking neither to prove or disprove the claims of the brief article I read. It took me nearly four hours to conduct my research and draft my findings. I was willing to be persuaded that troubling irregularities had marred the election in Wood County. But I simply can’t draw that conclusion after reviewing all the facts (as opposed to hostile innuendo) I’ve been able to gather.

Is there hope for an open dialogue? For an agreement at least about the basic facts, if not the conclusions to be drawn from them? I don’t know. But every once in a while I think it’s healthy to challenge our own assumptions and to look at the arguments advanced by others. If I’m willing to do it, I hope my friends and relatives will be, too. What’s difficult to imagine is how people of good will are supposed to find the time to do this kind of digging and come to reasoned, informed conclusions. We used to rely on the news media for this, but it seems no one trusts reporters or analysts anymore. How should responsible people stay informed and separate the signal from the noise? Is it even possible anymore?

* Update: 11/12/2012—I’ve provided the links I followed to read these stories on 11/10/2012, but many of them have since been taken down. For now, at least, you can still read the content of the FrontPageMag story here.