Do nearly half of Americans live in precincts that suffer from long lines?

When I saw this headline from a story on a site operated by a subsidiary of AOL Inc., I thought it couldn't be right. And here was the claim: "Nearly half of Americans live in precincts where long lines at the voting booth were a problem in the 2012 election cycle, according to a survey conducted by President Barack Obama’s Presidential Commission on Election Administration."

Except, that it isn't true.

Yesterday, the commission held an event. You can watch the video of the actual event, which is, inconveniently, not linked in the piece that makes the claim above. And the video contains the actual survey results. The relevant question asked, around the 1 hour, 9 minute mark of the video, is as follows:

Q: Did your jurisdiction experience long lines (approximately one hour or more) at any precincts or early voting sites in the 2012 general election?

A (weighted by eligible voters):

Yes, long lines were common and widespread: 1.9%
Yes, but only at some locations: 26.2%
Yes, but only at one or two locations: 21.3%
There were no appreciable lines in my jurisdiction: 48.0%
Don't know: 2.6%

Strictly speaking, if you live in a jurisdiction (a large city or a small town) that had one precinct with long lines (which would be about 21% of eligible voters), you would, under the headline of the piece, "live in places where election officials admit long lines are a problem." But, that is fairly deceptive: you live in such a place, but it, in all likelihood, does not affect you. There is, of course, the risk it affects you. And it obviously affects some voters there, and some quite substantially. But it certainly does not affect "nearly half" of voters in a way that the sensationalist headline of the story implies.

The text of the story is worse, because it contains an outright inaccuracy. It says that "nearly half of Americans live in precincts where long lines at the voting booth were a problem," (emphasis added) which isn't what the survey said. "Jurisdictions" are far larger in scope than "precincts," which is why a "jurisdiction" may have "some locations" or "only one or two locations" with long lines.

This is not to say that long lines are not a problem. They clearly are in some places, and Dr. Charles Stewart has conducted excellent research about drilling down the locations and length of lines. (There's very good discussion over the 10 minutes in the video: he notes that it indicates that long lines are "rare," but that it might affect a high volume of voters in that "rare" precinct, and that the risk is higher in big jurisdictions.)

But I don't think this sensationalist "journalism" is productive when it contains an important inaccuracy and fails to link to the relevant source. As the nonpartisan commission is attempting to identify problems with actual facts, distorting them (perhaps in the name of partisanship, or in an effort for sensationalism) doesn't move the ball much.