Facebook, Mainstream Media, Confirmation Bias, and the Trump Trauma

Lots of Americans, like those out here in liberal-leaning, Hillary Clinton-supporting California, are suffering what I call the “Trump Trauma”. They were sure she would win, easily, and are shocked at the unexpected outcome. It’s all disbelief, like someone suddenly died without warning. They were unprepared and now mourn the death of the Clinton candidacy. How could this come to be?

During our pre-election Frak That! podcast, on Nov. 7, 2016, cohost Randall Kennedy and I discussed the social media election. He expressed surprise at the “speed with which information travels”. I interrupted: “The speed with which disinformation travels now”, later describing social media interaction as something like “Borg sentience”, in context of phenomenon “confirmation bias“. The group mind—perpetuated by Facebook, news media reports, and political polls over-weighted to fit the narrative booming from the Echo Chamber—led many Americans, regardless of party affiliation, to misguided expectations about whom would be President-elect.

Lies, Damn Lies
I call Election 2016 the “perfect storm of disinformation”. Within the social media sphere, Facebook is largest culprit. Throughout the Presidential campaign, most subscribers’ News Feeds favored positive news about Clinton and negative news about Trump—the presumption being to influence the outcome. But I contend the tactic backfired, by obscuring information about the growing groundswell of support for the Republican candidate and understanding about who these voters are.

I shouldn’t be surprised someone else reaches similar conclusions. While Googling something else for this analysis, I found like-minded, but more smartly articulated, opinion from Christopher Hooten, writing for The Independent:

Instead of trying to reach out and understand why people felt moved to vote for Trump, too many admonished them. We retreated to our echo chambers, where we bellow our opinions at people who are already in agreement and enjoy having them repeated back to us. Our prismatic view of public opinion through tailored social media feeds not only hid from us the confused, angry people we needed to try and reason with, but it gave us a warped view of their motives. And this is where it gets particularly scary: Trump voters are fully aware he is sexist and xenophobic, they just don’t care.

That last sentence goes to something more fundamental that I will expand on later in the essay: People who were convinced no one could vote for Trump wrongly presumed the entire electorate, or even a large part of it, shared their values. Eh, no.

For now, I want to focus on confirmation bias and the resounding Echo Chamber as perpetuated by news and social medias and the influence on political polling, starting with this clarification: Whether or not Facebook, using its algorithm, overtly biased News Feeds for Clinton is immaterial to the outcome. Structurally, the service is engineered to drive revenue-generating traffic in a way that causes confirmation bias to flourish, often supported by fake news or other misinformation. Olivia Solon, writing for The Guardian better states the problem than I can:

Currently, the truth of a piece of content is less important than whether it is shared, liked, and monetized. These ‘engagement’ metrics distort the media landscape, allowing clickbait, hyperbole, and misinformation to proliferate. And on Facebook’s voracious News Feed, the emphasis is on the quantity of posts, not spending time on powerful, authoritative, well-researched journalism.

The more we click, like, and share stuff that resonates with our own world views the more Facebook feeds us with similar posts. This has progressively divided the political narrative into two distinct filter bubbles—one for conservatives and one for liberals (a blue feed and a red feed), pulling further and further apart in the run-up to election day.

I have been on the World Wide Web since early 1994, back in the Mosaic era, before Netscape entered public beta. Then, the Internet promised an epoch of openness and communication. But 22 years later, reality is something else: More and more people go online to confirm what they believe rather than to expand their viewpoints. One of the best explanations of this phenomenon—and there are many—is an analysis professor Tomas Chamorro-Premuzic wrote two-and-a-half years ago (also) for The Guardian: “How the web distorts reality and impairs our judgement skills“.

BTW, maybe Tomas foresaw something many point-of-view confirmers or political pollsters missed, tweeting on November 6: “It would not require a major polling error for Trump to be elected, and uncertainty is high”. But you wouldn’t know that from the disinformation ravaging the InterWebs even on election day: Clinton’s victory was all but assured. But as we know now, it wasn’t.

Social Media Demographics

I agree with Olivia: “Rather than connecting people—as Facebook’s euphoric mission statement claims—the bitter polarization of the social network over the last eighteen months suggests Facebook is actually doing more to divide the world”.

The phenomenon is larger than FB, of course, but with 1.79 billion active users (monthly), the network’s influence is larger than any other social media service—and I contend greater than all the others combined. According to Pew Research, 67 percent of U.S. adults use Facebook and 44 percent get news there. Additionally: “Of those who get news on at least one of the [social media] sites, a majority (64 percent) get news on just one—most commonly Facebook”.

According to Pew, the majority of Facebook users are female (57 percent); white (65 percent); Democrat or Independent (31 percent and 32 percent); under age 50 (18-29, 31 percent; 30-49, 38 percent); and have attended college (67 percent). Does anything about these numbers suggest likelier Clinton supporters to you? They sure do to me.

Truth Untold
In June 2009 analysis “Iran and the Internet Democracy“, I argued that then new online tools, like Facebook, Twitter, and YouTube (all about three years old) would take editorial control from newsrooms and give it to the masses. That prediction came true to a fault, as Election 2016 so clearly demonstrates.

But bias, flamed by online point-of-view confirmers in tandem with the traditional news media, turned Election 2016 into a screenplay, where the hero (Hillary Clinton) fought the villain (Donald Trump). Who doesn’t love a a good vs evil story, and Hollywood tells us that good triumphs. Perhaps it did for those Americans voting for Trump—not that we learned much about them; social mores and computer algorithms kept the group largely silent.

To say social media shares and daily news stories vilified the Republican candidate cannot be overstated. For example, consider endorsements and what they must mean for bias in the newsroom. Mother Jones kept a tally of them during Election 2016. Among 45 listed newspapers, only one—Las Vegas Review-Journal—endorsed Trump. Forty formally backed Clinton. They include the New York Times and Washington Post; I subscribe to both and found their lopsided anti-Trump reporting to be offensive. I also read Wall Street Journal, which endorsed no one—and, coincidentally or not, the editorial curation and tone was far more balanced than any other news source I regularly consult.

Think there isn’t inherent bias built into editorial endorsements? As a journalist, I can truthfully state that there most certainly is. Professor Pablo Boczkowski, writing for NeimanLab, rightly explains:

A clear indicator of the breadth of negative coverage of Trump is the distribution of newspaper and magazine endorsements of both candidates. Hillary Clinton was endorsed by 229 dailies and 131 weeklies, including news organizations that historically have not been identified with either party and others clearly representing a conservative ideology normally linked to Republican candidates. By contrast, Trump received the endorsement of nine dailies and four weeklies. That’s a 27-to-1 difference.

Eh, yeah.

The mainstream news media’s rally behind Clinton, while also demonizing Trump, occurred while Facebook courted media organizations to disseminate content over the network. For anyone seeking confirmation Clinton was good and Trump was evil, they needed look no further than their News Feed. No matter the source: Personal or professional. And remember Pew’s FB demographics: Majority are younger women who attended or graduated from college; voters who conceivably would favor Clinton.

But these aren’t the people who made Trump President-elect. Another America, silenced by social mores (that would get them ostracized as evil for supporting an alleged Hitler-like demagogue) and lack of news reporting about the group, brought different values to the polls. Accusations of Trump’s bigotry, corruption, misogyny, and sexism didn’t deter Another America’s support—viewpoints perhaps bolstered among this electorate’s confirmation bias circles.

Traditional news media lost touch with a large part of its audience, which in part explains so grossly misreading the whole electorate, while over-investing creating content that burned effigy Trump. Joshua Benton, (also) writing for NeimanLab laments; “That Donald Trump’s victory came as such a surprise—a systemic shock, really—to both journalists and so many who read or watch them is a marker of just how bad a job we did”. You think?

Perhaps most shocking is where Another America gathered in plain sight but was ignored by the media elite. Nick Bolton rightly explains for Vanity Fair:

There was, however, one big-data predictor that did seem to augur Trump’s landslide electoral victory: Twitter. We looked at the people replying to Trump on the social network and thought they were just a bunch of ignorant, angry men, and that they were outliers. We all looked at what the Trump supporters were saying on Twitter and thought it was just hate speech, and it just happened to be the smallest group who was screaming the loudest. Not the other way around.

Trump’s decision to regularly rant on Twitter may in the final forensic analysis be the linchpin to his victory. There he spoke plainly, and often offensively, but also approachably. He and his supporters used social media to bypass the mainstream.

The Echo Chamber
I agree with Mr. Benton: “Facebook has become a sewer of misinformation”. But, sadly, the cesspool is a partnership that includes the news media. Old and new media reverberated a resounding Echo Chamber of self-satisfying viewpoints: Donald Trump cannot be elected because he is unfit to lead; Hillary Clinton can make history as the first woman elected President of the United States. That’s the confirmation bias accepted by many Millennials—six out of 10 of which get political news from Facebook, according to Pew—Baby Boomers, and, most disastrously, the mainstream news media itself.

The economic, intellectual, political, and social elite agreed with the media. It’s the “We’re right, because everyone agrees we’re right” syndrome. But everyone saying the Earth is flat doesn’t make it so. The elites assumed that their values were shared across America. Instead, their peer pressure silenced Trump supporters. There’s no excuse for local newspapers’ failure to identify the character of their audiences. Confirmation bias and the Echo Chamber misplaced their priorities. Editors, reporters, and social sharers believed what they wanted to.

Bias blindness led the news media—my shameful profession—to too often ignore the Echo Chamber of self-confirming misinformation about their candidate Clinton from Trump supporters.  Through my FrakThat! podcast cohost, who avidly supports Trump, I got my first exposure to Breitbart and Alt-Right news sites and gossipers that spread damaging, disparaging information about Hillary Clinton among their audience and across the social media landscape, primarily Facebook, but also Reddit and Twitter.

The point: Confirmation bias isn’t an ailment belonging only to the liberal elite. As Olivia Solon observes, there are red and blue bias bubbles. But a large chunk of Millennials, other news consumers consulting Facebook, and the mainstream media self-confirmed “Hillary is good, Trump is bad” within an Echo Chamber that silenced their recognizing another viewpoint existed, and among a large number of Americans.

In a stunning commentary I discovered hours after first posting this analysis, Kyle Pope, writing for Columbia Journalism Review, sees what I do:

Journalism’s moment of reckoning has arrived. Its inability to understand Donald Trump’s rise over the last year, ending in his victory Tuesday night, clearly stand among journalism’s great failures, certainly in a generation and probably in modern times.

Reporters’ eagerness first to ridicule Trump and his supporters, then dismiss them, and finally to actively lobby and argue for their defeat have led us to a moment when the entire journalistic enterprise needs to be rethought and rebuilt. In terms of bellwether moments, this is our anti-Watergate.

Blind Leading the Blind
Something else: When there is final, forensic analysis of the disastrous Presidential political polling, which overwhelmingly favored Clinton, even on election day, confirmation bias and the Echo Chamber will emerge as fundamental mistakes.

Pollsters’ problems were a little larger, though, exacerbating bias:

  • Many first-time voters came out for Trump, but they weren’t polled.
  • Many others were reluctant to admit Trump (for fear of social stigma or personal embarrassment).
  • Then there was the large number of undecided voters, who represented more than 10 percent of people polled.

The Washington Post gives a final polling tally, matching expectations to reality (see chart). The disconnect disturbs.

presidential-poll-counts

Pollster John Zogby says there is an industry-wide “tendency to oversample Democrats and under-sample Republicans”, and looking at the final result “we can easily see that Clinton’s leads were based on faulty data”.

It’s my contention that, buoyed by news and social medias’ “Clinton cannot lose” hysteria, pollsters misread their data—and in many instances they mathematically over-weighted tallies for Democrats while under-weighting for Independents and Republicans. You do that when the data suggests something so different from what you see and hear across the echoing InterWebs. You disbelieve your data as having oversampled another group and compensate. Meanwhile, silenced Trump voters, by their absence, made the data interpretation all sensible.

But that most pollsters, like the mainstream media, missed Another America and the groundswell of dissatisfaction behind it, shows how flawed was their methodologies and, I contend, how overly they were influenced by popular sentiment as expressed across the news and social media Echo Chamber.

Looked at differently, Election 2016 was never about Donald Trump, Hillary Clinton, or any other candidate. What we see in this week’s voting is a populist revolt against the economic, media, political, and social elite—who, not coincidentally, largely bought into the “Trump is evil and unelectable” narrative. But for millions of disaffected Americans tired of the status quo, there is a different story unfolding. My profession failed to see it, or appropriately report about it.

Kyle Pope is right:

Journalism’s fundamental failure in this election, its original sin, is much more basic to who we are and what we are supposed to be. Simply put, it is rooted in a failure of reporting. As much of the country—indeed, apparently, a majority of the country—was roiling in anger and resentment and racism, too few reporters took the time to seek out people outside major cities on the coasts and listen to them. In this election, as in every election, the voters ultimately were the story that mattered, much more so than the campaign managers or the think-tank talkers or the reporter-cum-commentators with massive Twitter followings.

If you believe Facebook posters, news media stories, and pollsters, Donald Trump cannot deliver a happy ending. For them, for the present, this is true, because they are traumatized for expecting one thing and getting something so dramatically different. But the Trump Trauma may not be a permanent ailment. Much depends on the elected’s Presidency.

Election 2016 was a war of words. Now the measure is, or should be, action—what Trump does, and how objectively and accurately news media professionals report about it and how responsibly and actively they expose fake news that continues to spread across social media, particularly Facebook. To my peers: Leave behind partisan politics, along with the failed Clinton candidacy. Like it or not, you must accept two words a large number of Americans never expected to put together: President Trump.

Editor’s Note: If you find some portions of this essay to be repetitious, you’re right—and it’s deliberate to emphasize the key point, which when lost during election coverage could be easily ignored by some readers clinging to preconceived biases.