Let's Calm Down About Twitter Being Able To Predict Elections, Guys


Can Twitter help predict an election? Please, please, let the answer be "no." But Fabio Rojas, an associate professor of sociology at Indiana University, argues that it can in a recent Washington Post editorial. "Modern politics happens when somebody comments on Twitter or links to a campaign through Facebook," he writes, adding, "this new world will undermine the polling industry." Oh, well, it's been nice knowing you, polling industry!

The editorial reads more like, "Rah, Rah! [INSERT BUZZWORD HERE]" than anything resembling a piece of cogent political science. But Rojas and his coauthors lay out their case in a research paper, in which they describe how they painstakingly analyzed 542,969 tweets about Democratic or Republican candidates who ran in 2010. These were all sorted into specific races, and the percentage of tweets that mentioned each candidate was calculated. When this calculation, termed "tweet share," was matched up between opponents, the "tweet share" victor matched the winner in "404 out of 406 competitive races," Rojas writes. This was, he says, "a strong correlation."

Correlation does not imply ... what was it again?

In Rojas' mind, what he's stumbled upon is revolutionary because it's inexpensive, and polling is not. Furthermore, Rojas asserts that polling "favors the established candidates" and pays "disproportionate attention to 'big' races."

Some congressional races are never polled. Social media analysis can be used to systematically gather data on any race at any time. Thus, people in smaller states no longer need to rely on polling organizations for information. A single citizen can harvest social media data and learn about the election in his or her area.

Terrific, I guess? I mean, as near as I can tell, a single citizen can access lots of polling data, too. Besides, one big reason that some congressional races are never polled is that some congressional races aren't much of a race.

Normally, when political scientists or journalists write about “competitive” races they are talking about contests where at least two candidates have at least some chance of victory. Obviously, there weren’t 406 “competitive” House races in 2010 under that definition — at the Rothenberg Political Report, we rated just more than 100 House races as “not safe,” and a far fewer number in the truly competitive categories — so Rojas must be using the term to describe contested races.

Most races aren’t real competitions, of course. Relatively few House challengers run robust campaigns, and voters generally are unfamiliar with challengers.

Since House re-election rates have been over 90 percent in 19 of the past 23 elections, you don’t need polls or tweet counts to predict the overwhelming majority of race outcomes. In most cases, all you need to know is incumbency (or the district’s political bent) and the candidates’ parties to predict who will win.

Rothenberg reckons that what "tweet share" can measure is name recognition, which is something that we tend to assert as fact without actually quantifying it in any way. (That said, I think that simple horse sense still usually wins out when evaluating name recognition.)

"But other than that," Rothenberg writes, "the idea that the content of tweets is irrelevant, and that it doesn’t matter if the tweets originate from inside a district or from people who cannot even vote in the race, seems to fly in the face of logic and everything that political scientists believe."

Oh, yeah, that's an important reminder: lots of people who write tweets about candidates are writing negative things about those candidates. Surely that makes raw "tweet share" completely useless as a measurement, right?

But Rojas says that it doesn't matter if the message is positive or negative.

We believe that Twitter and other social media reflect the underlying trend in a political race that goes beyond a district’s fundamental geographic and demographic composition. If people must talk about you, even in negative ways, it is a signal that a candidate is on the verge of victory. The attention given to winners creates a situation in which all publicity is good publicity.

Well, then, congratulations to the next Mayor of New York City, Anthony Weiner!

[Would you like to follow me on Twitter? Because why not?]

This story appears in Issue 63 of our weekly iPad magazine, Huffington, in the iTunes App store, available Friday, August 23.

Before You Go

Popular in the Community