Understanding “fake news,” & why defeating it isn’t a fix-all

After the dust from our toxic post-election discourse settled, the talk of traditional and social media turned to “fake news” – a term that has taken on new meaning in recent years, and new prominence in the 2016 presidential race.

In this iteration, fake news doesn’t refer to satire like The Daily Show or The Onion. Nor does it refer to news that is biased in its selection and interpretation of facts. No, for now we’re fighting a much simpler to identify foe – the peddling of information that is blatantly, demonstrably false and intentionally deceptive.

Stuff like these sensational – and completely fictional – headlines that circulated in the months leading up to the election:

Pope Francis shocks world, endorses Donald Trump for president, releases statement [Ending The Fed]

FBI agent suspected in Hillary email leaks found dead in apparent murder-suicide [Denver Guardian]

WikiLeaks CONFIRMS Hillary sold weapons to ISIS… Then drops another BOMBSHELL! Breaking news [The Political Insider]

Thousands of fake ballot slips found marked for Hillary Clinton! TRUMP WAS RIGHT!! [Donald Trump News]

President Obama confirms he will refuse to leave office if Trump is elected [Burrard Street Journal]

BREAKING: Hillary Clinton to be indicted… Your prayers have been answered [World Politic US]

Rupaul claims Trump touched him inappropriately in the 1990s [World News Daily Report]

This sort of nonsense has been around for a long time, previously circulating via your crazy relatives’ email inboxes. But it found new prominence this election cycle, on Facebook. Craig Silverman and his team at Buzzfeed compared Facebook engagement metrics on the top 20 fake news stories and the top 20 stories from a sampling of traditional media outlets across the final three quarters of the 2016 election. They found that after lagging well behind for most of the year, the most popular fake news out-engaged the most popular real news in the final three months of the race. (All of the headlines above were among the top 20 in that time period.)

*There are caveats to this method, and if you care, I discuss them at the end of this post. The point is that engagement with fake news has risen dramatically.

That has invited three questions – where is fake news coming from, does it have an effect, and what can be done to stop it? Continue reading “Understanding “fake news,” & why defeating it isn’t a fix-all”

Advertisements

Sunday morning talk shows and portrayals of public opinion during the 2012 presidential campaign

The 2016 presidential campaign has been unique thus far, to say the least. It makes the 2012 cycle look downright boring. Yet, one aspect of the 2012 campaign that stood out to me was the use of public opinion polling by media to frame the race.

This paper was presented August 7, 2016 at the Association for Education in Journalism and Mass Communication Annual Conference in Minneapolis, Minn. An early version was presented February 27, 2016 at the Association for Education in Journalism and Mass Communication Midwinter Conference in Norman, Okla.

The 2016 presidential campaign has been unique thus far, to say the least. It makes the 2012 cycle look downright boring. Yet, one aspect of the 2012 campaign that stood out to me was the use of public opinion polling by media to frame the race. Leading up to Election Day, it seemed pretty clear that President Obama would have the electoral votes to win a second term. Election forecasting guru Nate Silver thought so, and most polling data agreed.

However, a completely different picture was painted in conservative media – at least in a few anecdotal instances. Fox News contributor Dick Morris infamously predicted a “landslide” victory for Mitt Romney, while Karl Rove’s refusal to accept Obama’s victory-sealing win in Ohio made for awkward Election Night coverage for the cable news ratings leader. Both had evidence on their side – poll numbers that made it look like Romney was indeed going to win Ohio and the White House. But those polls were in the minority, and they were wrong.

This matters. Previous research suggests that publishing of public opinion polls can actually influence public opinion, and eventually, voting. To be fair, these findings have always been tough to untangle. Does a poll showing a candidate with a big lead create a bandwagon effect where everyone wants to vote for the inevitable winner, or does it spur an underdog effect in which the losing candidate’s supporters mobilize to close the gap? Does depicting a close race boost turnout, while voters skip out on a projected blowout? There’s evidence of all of these. Continue reading “Sunday morning talk shows and portrayals of public opinion during the 2012 presidential campaign”

Newspaper portrayals of celebrity suicide: Examining coverage of Robin Williams

The suicide of Robin Williams shocked his fans around the world. His depression and history of addiction, though not uncommon to celebrity biopics or news of suicide, did not pair with the Williams we thought we knew. We have a relationship with Williams through the characters he portrayed in film, from the Genie in Aladdin to a man who literally healed through humor in Patch Adams. Though roles for an actor, they become a public persona – one that we don’t associate with depression and suicide.

icaPresented May 23, 2015 at the annual conference of the International Communication Association, San Juan, P.R., Journalism Studies division. Information about accessing this paper and associated materials available here, or by visiting the Academia page on dylanmclemore.com.

The suicide of Robin Williams shocked his fans around the world. His depression and history of addiction, though not uncommon to celebrity biopics or news of suicide, did not pair with the Williams we thought we knew. We have a relationship with Williams through the characters he portrayed in film, from the Genie in Aladdin to a man who literally healed through humor in Patch Adams. Though roles for an actor, they become a public persona – one that we don’t associate with depression and suicide.

Research on media coverage of suicide mostly deals with the risk of imitation, what has been referred to as the Werther effect. Celebrity suicide, in particular, increases the amount and prominence of coverage, and the parasocial relationships fans have with those celebrities might exacerbate imitation.

Numerous advocacy organizations issue guidelines for journalists covering suicides, suggesting that the content of the story, and not simply the subject, may be at fault. However, this assumption turns out to be woefully under-investigated. What we do know is that media –both in news and entertainment – have a less than stellar track record of accurately representing people with mental illnesses. They are depicted as being disproportionately dangerous, and their identity is rarely examined beyond their illness.

The Williams suicide presented a unique case study to look at how the story of a prominent celebrity suicide was told through the media. This particular study considered coverage from the 20 highest circulation newspapers in the U.S., including tabloids, over the week immediately following Williams’ death. It looked at the types of sources and content used in articles, as well as how prominent those articles were in the newspaper. Findings were then compared to World Health Organization (WHO) guidelines for reporting on suicide and existing literature on media stigmatization of mental illness.

The degree of detail needed for this type of study required some intensive work on the part of our coders. The team analyzed almost 6,000 individual sentences within 174 stories that cited over 550 sources. If you’re one for reliability statistics and methodology, drop me a line and I’d be glad the details of achieving reliability with so many moving parts.

Among the Findings:

– The suicide was largely told through the lens of Williams’ celebrity – his career and reaction from fellow movie stars – rather than focusing on issues of mental health.

– Coverage became less celebrity-oriented as time passed, but the frequency and prominence of articles also declined considerably, meaning that while coverage became more substantive, it also became harder to find.

– While medical experts weren’t widely cited, they were relied upon for most health-related information. Only twice was sourcing problematic. Medical sources were lacking in articles regarding addiction and speculating as to possible reasons for suicide.

– Tabloids were behind most WHO guidelines violations. They were far more likely than traditional newspapers to speculate about external “triggers” for the suicide. These “reasons” were almost always attributed to celebrities or friends of Williams, oftentimes anonymously. Tabloids also described the suicide in the most graphic detail – something that, aside from taste, is thought to increase imitation. This New York Daily News cover pretty well sums up the problems with tabloid coverage:

hanged

Overall, print coverage of the suicide departed from stigmatizing presentations. As a result of Williams’ stature, he was not a faceless victim; rather, his identity was explored rather deeply. Aside from the tabloids, attempts to link the suicide to some “trigger” event were rare, and suicide as a form of escape was not promoted, despite that message being circulated by some very prominent sources. Most impressively, the stigma that people with mental illness were dangerous did not appear a single time in almost 6,000 sentences analyzed. This was especially surprising when drugs and alcohol entered the discussion.

However, there was still room for improvement. It took time for a medical angle to emerge and for medical experts to appear as sources. That leaves room for unwarranted speculation in the immediate aftermath, when audience attention is at its highest. Print coverage also appeared to miss an opportunity to talk about depression, addiction, and suicide in a more general context. Such extrapolation was isolated.

The real question is whether coverage was unique to the uniqueness of its subject. Is the press getting better at explaining mental illness, or was coverage more delicate because of the beloved actor who made us all laugh? Future research will tell us more, though we hope it is a long time before a context for investigation presents itself again.

This research was conducted by the University of Alabama Health Communication Lab, a part of the College of Communication and Information Sciences. McLemore served as the lead researcher for this study.

Institutional Isomorphism and the Community Structure Approach in Visual Framing of the Trayvon Martin Shooting

jpcover#ICantBreathe. #BlackLivesMatter. #Ferguson. One does not have to look far lately to find prominent discussions of race and social justice. You could argue that the present movement began in 2012, with the shooting of Trayvon Martin by George Zimmerman. I looked at how elite and local newspapers covered the shooting and its aftermath, specifically in terms of the images used to represent the story.

I wrote more about the findings back in May, when an earlier version of the paper was presented at the International Communication Association annual conference in Seattle.

The article has now been published in Journalism Practice. You can access it here.

 

The published version provided a chance to think a great deal more about sociological explanations for news framing. In particular, it argues that institutional isomorphism offers a well-developed, multidisciplinary theoretical framework in which intermedia agenda setting can be positioned and strengthened. From a theory-building standpoint, the Martin shooting left something to be desired, but nevertheless demonstrated the functionality of the approach. I’m hoping to replicate the design in the future.

As a heavily quantitative media effects guy, this project began outside of my comfort zone. But I’m glad I was encouraged to follow it through to publication. It made the process of mass communication become that much bigger – thinking not just about the sender and receiver, but about the entire ecosystem surrounding them. And for that, I owe Wilson Lowrey a great deal of gratitude. He (and the tremendous number of books and articles he recommended) stretched me as a scholar, and made my approach more well-rounded as a result. Truly the sort of experience doctoral programs are meant to foster.

Access the full article.

A Cross-Cultural Comparison of Campaign Tweets in the 2012 U.S. and South Korean Presidential Elections

aejmc14Presented August 8, 2014 at the annual conference of the Association for Education in Journalism and Mass Communication, Montreal, Qc., Canada, Political Communication Interest Group.

For more information about this paper available here, or by visiting the Academia page on dylanmclemore.com.

 

As America embarked on its second social media election, changes to South Korea’s election laws permitted the highly digital nation to have its first. Amidst concerns that reform might lead to Americanized campaigning in South Korea, we sought to compare the Twitter activity between candidates in the two countries.

We conducted an extensive content analysis of over 4,500 Tweets from the accounts of presidential candidates, from frontrunners like Barack Obama and Park Geun-hye to third-partiers or independents like Jill Stein and Kang Ji-won. Specifically, we were curious as to what topics candidates Tweeted about, whether they used collectivist or individualistic language, and how often they used Twitter to engage in opposition attacks.

Among the findings:

– U.S. candidate Twitter feeds focused on issues and candidate image, while South Korean feeds spent more time promoting campaign events.

– Surprisingly, South Korean feeds featured almost entirely third-person language. They were far less likely to use collective “we/you” language than American candidates. U.S. feeds were also more likely to engage in individualistic “I” language. This might be less of a result of cultural differences in communication styles and more a function of South Korean usage of Twitter as a campaign calendar more than a platform for ideas.

– American candidates used Twitter to attack opponents more often than South Korean candidates. However, given cultural and political norms in the two nations, the number of attacks in the U.S. was less than what one might expect in other campaign communications, while the noticeable presence of negative campaigning in South Korea was new and somewhat surprising.

Together, the findings denote some differences between the two countries in Twitter campaign communication. However, similarities also emerged, and tended to point toward an Americanization of political discourse.

Institutional Isomorphism and the Community Structure Approach in Visual Framing of the Trayvon Martin Shooting

icaPresented May 25, 2014 at the annual conference of the International Communication Association, Seattle, Wash., Journalism Studies division.

To request the accompanying visual aids for this paper, email Dylan.

To read the abstract, go to the Academia page on dylanmclemore.com.

UPDATE: This conference paper has since been published.
DOI: 10.1080/17512786.2014.988993

 

The Trayvon Martin shooting and the legal (in)actions that followed, became perhaps the first major American news story of 2012. Narratives surrounding Martin and George Zimmerman widely varied, as did the images available to depict them. There’s a big difference between Zimmerman in an orange prison jumpsuit (for an unrelated charge for which he was never tried) and Zimmerman smiling in a suit and tie. Images of Martin depicted a boy much younger than the 17-year-old involved in the incident. Given the impact of imagery on the framing of a news story, this study considered competing explanations for why editors from newspapers serving racially distinct communities may have selected particular photographs to represent Martin and Zimmerman in their coverage. The method specifically sought to measure institutional isomorphism – a field-level homogeneity fed by stabilization and risk-reduction – and the community structure approach – variances at the local level based on the demographics of the market.

Among the findings:

– The story was far more likely to be depicted visually with images of or relating to Martin in the sample period (Feb. 27-Apr 27, 2012, or, from the day after the shooting to four days after Zimmerman’s not guilty plea in court).

– These depictions of or relating to Martin were overwhelmingly positive, while portrayals of Zimmerman were neutral-to-negative. Image valence held true across publications.

– Images of Martin himself quickly gave way to images of his family, and supportive demonstrators around the country. This initial spike of intense visual framing toward Martin diminished over time, and an increase in images of Zimmerman became apparent as he made more public appearances. Once again, these trends were consistent across publications.

What does it mean?

The findings observe a fairly homogenous media depiction of the Martin shooting. This is indicative of institutional isomorphism, though clear evidence of mimetic inter-media agenda setting was not identified. The results may be better explained by normative isomorphism, as media outlets quickly moved away from images captured outside of the context of the story. The apparent strength of journalistic norms in the face of a story that presented so many salacious angles offers some comfort to those concerned with the profit motive of the press affecting editorial decisions.

No support was found for the community structure approach. Despite the availability of images that portrayed Martin and Zimmerman in starkly different ways, newspapers serving predominately Black, Hispanic, and White communities employed similar presentations. Across the board, Martin was portrayed more frequently and more positively than Zimmerman, though Zimmerman’s legal battles seemed to be developing more frequent and nuanced coverage, a trend that should be followed beyond the sample in this particular study for a fuller understanding.

Covering the Conventions: Bias in Pre and Post-speech Media Commentary during the 2012 Presidential Nominating Conventions

aejmc

Presented August 10, 2013 at the annual conference of the Association for Education in Journalism and Mass Communication (AEJMC), Washington, D.C., Political Communication Interest Group.

This paper was previously presented at the 2013 AEJMC Southeast Colloquium in Tampa, Fla., where it received the Top Paper Award for the Electronic News Division. AEJMC permits re-submission of regional papers to the national conference.

To request the accompanying poster for this paper, email Dylan.

To read the abstract and request the full paper, go to the Academia page on dylanmclemore.com.

SYNOPSIS:

Party nominating conventions used to be messy and meaningful. Today, they’re more pomp and circumstance. It’s no wonder the broadcast television networks pay far less attention to them now than in decades past. For cable news, on the other hand, the conventions provide fuel – or at least a moving backdrop – to the 24-hour news cycle. And in the pauses between speakers, the talking heads weigh-in with their analysis.

There is a healthy amount of research suggesting that the party nominating conventions can influence voters, as well as a stack of studies that indicate media analysis of political events can influence voters. However, the specific cross-section between conventions and commentary has not been evaluated.

Does instant media commentary affect perceptions of convention speeches? This study lays the foundation for that investigation by looking at how favorably (or unfavorably) different news networks covered the 2012 Republican and Democratic national conventions.

We looked at a large sample of live convention coverage – all six nights of primetime (10 p.m. E.T.) on the three major broadcast (ABC, CBS, NBC) and cable (CNN, Fox News, MSNBC) networks. Here’s what we learned:

  • Convention coverage was generally favorable to the host party.
  • Bias was indicated by an exaggeration of this positive commentary, and near absence of negative commentary (for instance, coverage of the DNC on MSNBC was 3% negative; the RNC on Fox News was only 2% negative). In other words, unbalanced coverage was not the result of tearing one side down, but by disproportionately praising the other.
  • The largest differences in valence were observed on Fox News & MSNBC, though some statistical tests revealed evidence of bias in traditional network broadcasts.

We look forward to enriching this study with further data from the content analysis, including potential explanatory mechanisms. Next, we desire to test the effects of such instant media commentary on the audience (presently in the data collection phase). We extend our thanks to the reviewers, moderator, and discussant for taking the time to read our paper and provide valuable feedback.