As a Matter of Fact . . .

Posted Tuesday, July 23, 2013 - 19:49
How we interpret data can affect public perception and policy
Joseph Janes

As always, it was a pleasure to welcome friends and colleagues to Seattle for the Midwinter Meeting this past January. Two topics of conversation often arose, unexpectedly: our recent ballot measures on marriage equality and marijuana legalization. Of course, we’d been living with those issues all through the election, so the fact that anybody else wanted to talk about them took me by surprise.

We passed both of those relatively comfortably, though not without a lot of discussion and more than a little soul searching on the part of many. A particularly intriguing component of that discussion was the fine Living Voters Guide, an ongoing project that brings together a number of partner organizations interested in civic engagement and that for the 2012 election had its fact-checking provided by reference librarians at Seattle Public Library.

It appears now that the Washington Post is working on implementing an automatic real-time fact-checking system for political speeches. Stories about this have a sort of breathless, world-of-tomorrow kind of feel to them (“a Shazam for truth,” as an executive producer from the Post  calls it), though tinged with serious questions about the technology, the approach, the relevance, and the challenges.

What is only hinted at is the information angle, which we know is critical to any attempt to “validate” or “verify” facts. It immediately reminded me of a class exercise in which I ask students to independently check a variety of things found in the World Almanac. It starts benignly, asking for alternative lists of birthstones—which are numerous, mainly from a variety of jewelry associations, all understandably trying to sell gems.

They then work on other examples, exploring the “official,” if almost certainly imprecise, recent election vote totals, trying to determine who actually measures the world’s longest railway tunnels and whether anybody other than MLB.com can be definitive on all-time home run leaders (no, though the Hall of Fame plaque for Negro Leagues star Josh Gibson mentions his “almost 800”).

More potentially fraught are questions of how many homeschooled students there are, what kinds of organizations and institutions are interested in that question, and what their various and likely competing motivations might be. By the time my students get to “some notable new books for children,” courtesy of ALA’s Association for Library Service to Children (ALSC), and the delicately titled table “Examples of Genocides Since 1900” with its very large and very round numbers and no sources whatsoever, a number of furrowed brows and scrunched looks appear around the room.

The point of the exercise is mainly about how we typically substitute authority for accuracy because it’s easier to figure out. The second question, not very far beneath the surface, is, “What’s a fact?” Trivially, if my birthstone is opal rather than aquamarine, who really cares? But shouldn’t Josh Gibson get credit if he really hit more home runs than Barry Bonds, asterisks notwithstanding? Couldn’t very different numbers, perhaps based on different definitions of homeschoolers, potentially affect public perception, professional practice, and policy questions? And if my genocide is your war of liberation, how does one properly record and memorialize that?

Is there an app for that? Not yet.

It’s fascinating to contemplate how one might work and how that would affect political coverage, not to mention daily life. If Guinness World Records was meant to prevent pub fights, what would a made-up factoid lead to? Who knows what the state of the art on factual interpretation will be by the time Midwinter rolls around here again in 2019, not to mention views on marijuana and marriage . . . but that’s another story.

JOSEPH JANES is associate professor and chair of the MLIS program at the Information School of the University of Washington in Seattle.

Issue: 
Column: