San Francisco Chronicle

Social media’s impact on election

All the digital discourse may have generated more heat than light

- By Caleb Garling

When President Obama used Twitter as his first communicat­ion medium to claim re-election Tuesday, the tweet became the most amplified message the service had ever seen.

It was the culminatin­g moment in an election in which the role of the Internet was a constant, sometimes deafening, presence. CNN broadcast with on-screen hashtags — a way of linking tweets. Obama and his Republican challenger Mitt Romney boasted 33 million and 12 million Facebook fans, respective­ly. Countless election “memes” — digital snippets of pop culture — propagated on media sites

like YouTube and Tumblr. The president even took time to have a question-and-answer session on ultra-geek social network Reddit.

What isn’t clear days later — besides the final results in Florida — is the extent to which our growing digital connectivi­ty actually affected the election’s outcome. For all the incessant tweeting, posting, commenting and live-blogging across the Internet, did it ultimately change anything?

Polling outfit Pew Research published a study on the day of the election entitled “Social Media and Voting” that found that 22 percent of registered voters told friends and followers how they voted on sites such as Facebook or Twitter. But those conversati­ons might just as well have happened elsewhere, and, more importantl­y, might have had different outcomes.

Fighting words

While the Internet overflows with informatio­n — true and otherwise — both liberal and conservati­ve Americans are increasing­ly homogeneou­s in the way they access it, argues Michael Heaney, a political science professor at the University of Michigan. The December issue of American Behavioral Scientist will feature a paper in which he argues this trend is spreading further into offline activities as well.

“Social media helps to reinforce the already strong tendency toward polarizati­on — and worsens those trends in the electorate,” he says. “People search out these little groups where everyone agrees with them,” rather than engaging in authentic debate.

A March poll from Pew reinforces Heaney’s observatio­n. The study found that 9 percent of social-network users have gone so far as to block or “unfriend” someone who posted a disagreeab­le political opinion. Conversely, the study also reported that a fifth of users don’t talk about their politics for fear of offending others.

Internet companies reflect this trend back onto the electorate. Microsoft went so far as to tweak its Bing search engine to give users an option to “customize” their political news by selecting “left,” “right” or center. Search engines like Google look at past search histories and revise future results to better suit what it determines the user wants to find. Repeatedly searching for “Barack Obama” returns more sites relevant to Obama in the future.

This does not mean Google is politicall­y biased, says Danny Sullivan, editor of Search Engine Land, a website that tracks search technologi­es; it’s just a product of the way the software works.

“Every search engine has biases that are a natural consequenc­e of trying to rank web pages based on a wide range of factors,” he wrote recently. “But by no means are these results somehow doing something favorable for Obama over Romney.”

Clearly, digital connectivi­ty their interest is vested in users branching out from their own political views.

Campaign tool

The presidenti­al campaigns used similar strategies when pushing their candidates to the public, and used social media as a listening tool.

Software company Evidon shines a light on the Web’s monitoring tools with a piece of browser software called Ghostery. When a user goes to a new Web page, Ghostery examines all the third-party pieces of software that are registerin­g metrics like clicks or the amount of time someone stays on a particular page. Myriad companies, from Google to DoubleClic­k, build and sell tools for companies to get a better picture of who is visiting their sites — and serve up better advertisem­ents.

But campaigns can use them, too. From May to September, Ghostery users experience­d 87 different tracking technologi­es on BarackObam­a.com; MittRomney.com had 48. Most of the tracking devices were ad-related software, notes Andy Kahl, consumer products director at Evidon. The candidates weren’t serving up ads, however, but using the ad-targeting software to present relevant or new campaign messaging.

With each passing election, social media are becoming less novel — another tool alongside during the election did have important benefits. Issues arising at polling stations were immediatel­y made public. New Jersey voters could easily monitor their state’s altered procedures in the wake of Superstorm Sandy. There were countless resources to help voters register and find polling places. Being able to check e-mail, text, read and watch movies on a smartphone no doubt kept some voters in long lines when they might otherwise have gone home out of boredom or impatience.

And digital resources served to check the veracity of political claims. Organizati­ons like FactCheck.org at the University of Pennsylvan­ia or Pulitzer Prizewinni­ng Politifact parse the rhetoric and shed light on the truth. News aggregatin­g sites like AllSides try to capture the entirety of a political story by doing side-byside comparison­s of stories from across the political spectrum.

False informatio­n

But with online discourse so heavily siloed, these watchdogs’ efforts can go for naught. Heaney believes the 2012 presidenti­al campaigns generated the most false informatio­n ever — but the campaigns knew they could get away with it.

“This strategy only works if there is no meaningful cross-talk,” he says.

The Internet also magnified supposed campaign “turning points.” Much was made, for instance, of the presidenti­al debates, especially of Obama’s lackluster performanc­e in the first one.

However, the political projection site Electoral Vote — which aggregates data from pollsters like Rasmussen and Gallup — predicted Obama would win the election with 332 Electoral College votes on Oct. 3 — the day before the first debate. More than a month and three debates later, that appears to be precisely the number of votes the president will carry.

Business entities

There are factors at play on the other side of the Internet curtain. The public tends to view social networks like YouTube, Facebook and Twitter as free services and public forums. But these organizati­ons are, in fact, businesses — many of which have to answer to the demands of Wall Street. Jason Benlevi, a tech marketing veteran and the author of “Too Much Magic,” a book critical of cyberutopi­an visions of the Internet as a panacea, says he still believes in technology as a societal solution.

“I just don’t believe in the people pushing it,” he says.

It’s in the business interest of social networks and most websites to keep us talking and bantering back and forth. Revenue and data come from not only increasing the number of users, but users’ engagement with friends and followers. Online businesses tinker with mathematic­al representa­tions of the way people influence each other — sometimes called “social graphs” — and leverage them to achieve both goals.

In turn, they use the informatio­n to sell advertisem­ents or additional services. These proprietar­y calculatio­ns are well guarded so it’s difficult to say how much of TV ads and door-to-door canvassing. In 2008, the Obama campaign honed its Twitter and Facebook use for fundraisin­g and to drum up support, especially from younger voters, as the John McCain troops struggled to catch up. Though Romney used them far less than Obama in the 2012 campaign — perhaps a reflection of his campaign’s target electorate — tweets and Facebook updates were still commonplac­e.

Superficia­l dialogue

Micah Sifry, director of Personal Democracy Media and professor at Harvard’s Kennedy School of Government, doesn’t believe that any of the digital outlets — from Tumblr and Instagram to Facebook “townhalls” and Google “hangouts” — had any noticeable effect on the course of the election. He says they did little beyond foster the often superficia­l dialogue.

“Both sides benefited: The politician­s got a little ‘Internet buzz’ for their appearance­s, and the tech companies got some welcome and cheap marketing,” he wrote on election day. “And with a few exceptions, the political reporters who cover the election campaigns went merrily along for the ride.”

 ?? Shannon May / The Chronicle ??
Shannon May / The Chronicle
 ?? Lionel Bonaventur­e / AFP / Getty Images ?? A cell phone user in Paris checks out President Obama’s victory tweet and a computer image announcing the Democrat’s re-election a day after his hard-fought campaign against Republican Mitt Romney ended.
Lionel Bonaventur­e / AFP / Getty Images A cell phone user in Paris checks out President Obama’s victory tweet and a computer image announcing the Democrat’s re-election a day after his hard-fought campaign against Republican Mitt Romney ended.
 ?? Mladen Antonov / AFP / Getty Images ?? A delegate to the Republican National Convention poses in the Facebook photo booth in Tampa, Fla.
Mladen Antonov / AFP / Getty Images A delegate to the Republican National Convention poses in the Facebook photo booth in Tampa, Fla.

Newspapers in English

Newspapers from United States