Facebook friends get out the vote in large numbers

WASHINGTON (AP) — Here’s some­thing most politi­cians can “like”: Face­book friends played a big role in get­ting hun­dreds of thou­sands of peo­ple to vote in 2010, a new sci­en­tific study claims.

Face­book researchers and sci­en­tists at the Uni­ver­sity of Cal­i­for­nia, San Diego con­ducted a mas­sive online exper­i­ment in the mid-term con­gres­sional elec­tion to test and mea­sure the polit­i­cal power of online peer pressure.

They found that peo­ple who got Face­book mes­sages that their friends had voted were a bit more likely to go to the polls than those who didn’t get the same reminder. And from there the effect mul­ti­plied in the social net­work, they reported in Thursday’s jour­nal Nature.

The friend-prodding likely increased voter turnout by as much as 340,000 in the non-presidential elec­tion that voted in a new Repub­li­can con­gress, the sci­en­tists cal­cu­lated. They said that it could poten­tially change the out­come of close elections.

“Our study is the first large-scale sci­en­tific test of the idea that online social net­works affect real world polit­i­cal behav­ior,” said study lead author James Fowler, a pro­fes­sor of med­ical genet­ics and polit­i­cal sci­ence at the Uni­ver­sity of Cal­i­for­nia, San Diego.

He has stud­ied friend and social media influ­ences on pub­lic health and pol­i­tics over the past decade. While pun­dits have pointed to social media-inspired rev­o­lu­tions in the Arab world, this is more ver­i­fi­able sci­en­tif­i­cally because it is a con­trolled study com­par­ing groups that had dif­fer­ent inputs. It’s the vot­ing equiv­a­lent of test­ing real drugs ver­sus sugar pills.

Out­side experts say the new study makes sense and fits with other research about how effec­tive get-out-the-vote dri­ves are, but say Fowler’s num­bers may be a bit high. That’s because they fac­tor in a large indi­rect effect, cal­cu­la­tions which some didn’t find as convincing.

Nearly every Amer­i­can of vot­ing age who logged into Face­book on Elec­tion Day 2010 was part of the exper­i­ment, even though they didn’t know it.

Most of them — more than 60 mil­lion — saw an announce­ment on top of their Face­book news feed: Today is Elec­tion Day. It showed how many Face­book users as well as their friends had clicked an “I voted” but­ton and showed up to six pic­tures of those friends. It also linked to a list of polling places.

Researchers com­pared voter turnout with two groups that didn’t receive that same mes­sage. One group of 611,000 peo­ple sim­ply got a generic announce­ment encour­ag­ing vot­ing, but no pic­tures or count of friends. Another 613,000 users didn’t receive any message.

Those who got the peer pres­sure mes­sage were less than half a per­cent (0.39 per­cent) more likely to vote than those who got no mes­sage or the generic one. While that seems like a very small increase, it is sta­tis­ti­cally sig­nif­i­cant and it adds up, Fowler said.

There was no dif­fer­ence in vot­ing found between the generic and no-message groups. Nor was there any dif­fer­ence seen in friend-prompted turnout between self-identified con­ser­v­a­tives and lib­er­als, Democ­rats and Republicans.

Fowler and col­leagues didn’t just take the word of peo­ple who clicked the “I voted” but­ton. They checked pub­lic vot­ing records in 13 states for that elec­tion, and found about 4 per­cent of those who said they voted hadn’t really cast ballots.

Of those who saw the peer pres­sure post­ing, Fowler cal­cu­lated that 60,000 voted who wouldn’t have. On top of that, he said, another 280,000 peo­ple voted who wouldn’t have because their friends saw the online mes­sage and spread the get-out-the-vote word.

It’s a form of social con­ta­gion with peo­ple notic­ing that the orig­i­nal mes­sage recip­i­ent voted, so the mes­sage spread in per­son, by word of mouth and online, he said.

“The net­work is key,” he said.

Colum­bia Uni­ver­sity polit­i­cal sci­en­tist Don­ald Greene said the 60,000 direct voter num­ber makes sense and fits with other research done, which shows that the more per­son­al­ized the appeal the bet­ter the result. But he said he had dif­fi­culty buy­ing the cal­cu­la­tions used to come up with the 280,000 indi­rect votes.

George Mason Uni­ver­sity polit­i­cal sci­ence pro­fes­sor Michael McDon­ald, an expert in voter turnout, said the study seemed rea­son­able to him, adding “any­thing we can do to increase turnout is a good thing.”

In the 2010 elec­tion, about 38 per­cent of the vot­ing age pop­u­la­tion cast bal­lots, up from about 37 per­cent in 2006 and 36 per­cent in 2002. Vot­ing is usu­ally much higher in pres­i­den­tial elec­tions, so the over­all effect of a social media get-out-the-vote push might be lower in 2012 because peo­ple were already more likely to vote, Fowler and oth­ers said.

Fowler and Face­book sci­en­tist Cameron Mar­low said no deci­sion has been made about doing a sim­i­lar study or vot­ing drive on Face­book this November.

The study was ini­ti­ated by Fowler, who got Face­book involved, and was funded by the Uni­ver­sity of Notre Dame and two pri­vate foundations.

Posted by Matt Echelberry on Sep 12 2012. You can follow any responses to this entry through the RSS Feed. Both comments and pings are currently closed.