View this email in your browser
Monday, March 12, 2018
The virality of falsehoods

On Thursday, Science magazine published findings from the largest ever study of online misinformation. Three researchers at the Massachusetts Institute of Technology analyzed more than 126,000 fact-checked pieces of content that had been shared on Twitter more than 4.5 million times between 2006 and late 2016, and found that the false information in their dataset — as identified through reports from six fact-checking organizations — spread more quickly and extensively than true information. The study also showed that accounts with automated “bot” patterns shared roughly equal amounts of true and false information, and so were not responsible for driving this trend.

The study, which was funded by Twitter but kept independent of its influence, affirmed a lot of the existing research on how and why people spread false information on social media. For example, it reinforced the finding that false information is often more novel and shocking — presumably by design, since outright fictions are not bound by reality — and, in turn, get a disproportionate amount of attention. It also reinforced a related trend: False information often elicits more volatile emotions, such as disgust and outrage, than true information. In addition, it found that content about politics, whether true or false, gets significantly more engagement than any other topic.

But the study also provided new insights into Twitter’s misinformation ecosystem — for example, users who shared falsehoods had far fewer followers, followed a far smaller number of accounts, were much less likely to be “verified” Twitter users and had been on Twitter “for significantly less time,” while those who shared accurate information generally had larger followings and sent more tweets.

Though the study has been well-received, commentators have raised a few concerns about interpretations of the findings. First, because the study’s dataset is limited to contested information — pieces of content with claims that have been either verified or debunked by major fact-checking organizations — it doesn’t provide insight into the engagement metrics of the overwhelming majority of credible news reports. As the researchers acknowledge, their study doesn’t account for the fact that the supply of credible information about a given topic is considerably larger than the supply of false information, which could work to dilute the engagement metrics of credible content.

Still, the study is an important, evidence-based reminder of the ways that types of misinformation exploit our innate biases toward information that is novel and shocking. It also highlights the fact that social media platforms — which are optimized for focusing attention on “engaging” content — can tend to intensify our worst tendencies and amplify misinformation as they seek to capture (and sell) our attention, a fact that also makes them well-suited vectors for organized disinformation campaigns.

  • Idea: Produce a variation of this study with students by having them collect and analyze a random sample of tweets from a hashtag of their choice. They will need to isolate tweets that share links to content, then sort the true or verified content from false content before analyzing the engagement trends of each type.
    • Note: You can set the sample size to fit your integration of this activity, but should remind students that smaller datasets produce less reliable results than larger datasets. Also, some hashtags may contain significantly more misinformation than others.
    • Also note: This activity can serve as a great “side door” for student inquiry into the event or issue that the hashtag is focused on.
  • Related: “YouTube, the Great Radicalizer” (Zeynep Tufekci, The New York Times)
    • Discuss: How is the tendency of YouTube’s algorithm to introduce increasingly radical content related to the findings of the MIT study about false information on Twitter?
What kind of news/information/media literacy?

In her keynote address last Wednesday at the SXSW EDU conference, social media and technology scholar danah boyd (she does not capitalize her name) raised important questions about the goals of media literacy education. (While there are differences among “media literacy,” “news literacy” and “information literacy” as education content areas, I am using “media literacy” here to refer to the work in a general sense because it is the term used by boyd, a principal researcher at Microsoft Research.) Rather than summarize her talk, I wanted to share a few of the questions it raises, both directly and indirectly, as important reflection points for educators.

  • Discuss:
    • How can we teach students to apply critical thinking to all the information they encounter while still emphasizing the dramatic differences in credibility among types of information sources? (Put another way: How can we teach students to question all the information they encounter without teaching them that all information is equally questionable?)
    • How can we also teach students to question their own perceptions of information, or to recognize the ways that their assumptions (or, as boyd refers to them, their “priors”) about news (or news organizations) or controversial issues can distort their evaluation of information?
    • How do we teach students the importance and value of fact-based evidence without denigrating or insulting some of the other “ways of knowing” (epistemologies) that may be active in their lives, such as experience, traditions, feelings and faith-based beliefs?
    • What does it really mean to teach students to think critically? What voids can thinking critically about previously unquestioned assumptions open up, and how are students filling them?
    • How can media literacy help students better understand people they disagree with?
    • “What kind of media literacy should we be working towards?” (This question is taken directly from boyd’s address.)
Looking for The Sift archives? You can find them at:

Viral rumor rundown

NO: — a project of the University of Pennsylvania’s Annenberg Public Policy Center — did not “expose”, another fact-checking website, as “an extremely liberal propaganda site” in 2009. This baseless claimposted to Reddit on Feb. 15 and included in The Sift’s viral rumor rundown on Feb. 19 — continues to circulate.  

  • Misinformation pattern: Rumors attacking the credibility and political neutrality of have been circulating since at least late 2008.
NO: Michele Bachmann, a former member of Congress who sought the Republican nomination for president in 2012, did not say that Jesus created assault rifles. This fake quote meme was created — and watermarked — by the satirical group Christians for Michele Bachmann.

  • Fact-check bell-ringer: Display this meme and elicit ideas for the approaches that could be used to fact-check it: Search for the quote or claim; observe details (the watermark), then search for the source (Christians for Michele Bachmann); search for other instances of the meme using a reverse image search; or verify the attribution (Fox & Friends on March 2).
    • Note: An important lesson here is that there are multiple ways to fact-check most viral rumors. A search of the claim should quickly turn up the entry for this meme; a search of the source will quickly flag the meme as satire; a reverse image search will show other outrageous quotes using the same meme layout; and searching for archived Fox & Friends episodes would show that Bachmann did not appear on or call in to the show on March 2, 2018.
    • Also note: The Internet Archive’s TV News Archive is a powerful fact-checking tool and instructional resource.
  • Alternate bell-ringer: Assign students individually or in groups one of the four approaches listed above, then have them share their fact-checking paths.
  • Bell-ringer bonus: Where and when was the picture of Bachmann that is used in the meme taken?
    • Note: Use observation skills and reverse image search. Where else does this image appear? Can you find attribution? What is in the background? When and where did she wear that particular combination of clothing and jewelry?
NO: The flu shot is not “laced with cancer-causing ingredients” and is not “designed to spread cancer.” This claim was made, citing no evidence, by alternative medicine website Meddaily and contradicts overwhelming scientific consensus on the safety of the flu shot.

NO: President Donald Trump did not remove a Muslim federal judge in Dearborn, Mich., for trying to implement Sharia law. This is a fiction published by the website As American As Apple Pie (whose “about” page starts with “Everything on this website is fiction”). That site is one of several run by Christopher Blair that have been described as “a collection of ‘satirical’ sites that post false and extremely inflammatory articles.”

NO: President Trump did not sign an executive order that allows veterans to receive free treatment at any hospital they want. YES: In April 2017, Trump signed legislation removing the August 2017 expiration date from the Veterans Choice Program, which began in 2014 and allows eligible veterans to receive medical services from providers outside the VA system.
Sinclair promo script leaked

Last Tuesday, Sinclair Broadcast Group, the largest owner of local television stations in the country, instructed its stations to air what it called a “journalistic responsibility message” to be read by local news anchors during their newscasts. The script, which was leaked to CNN the following day, states that “biased and false news” is a problem on social media, and that “national media outlets are publishing these same fake stories without fact checking first.” It goes on to say that “some members of the national media” are “using their platforms to push their own personal bias and agenda to control ‘exactly what people think’” and calls this “extremely dangerous for democracy.”

This isn’t the first time that the company has required its local television stations to air a message like this. A year ago, the stations were sent a recorded version of an almost-identical statement read by Sinclair’s now-senior vice president of news, Scott Livingston, and were told that it must run. Last May, the company announced its intention to buy Tribune Media — a deal that would add 42 local television stations to Sinclair’s existing 173. The deal is being reviewed closely by the Federal Communications Commission and the U.S. Department of Justice over concerns that it violates FCC ownership rules and antitrust laws.

Local journalists at some Sinclair-owned stations are uncomfortable with the message and unhappy with what they see as a violation of their editorial independence. On Thursday, Livingston responded to CNN’s publication of the memo, saying that the promos are aimed at unreliable content on social media and “not network or cable news.”

  • Note: Sinclair has been providing “must-run” pieces of content to its stations’ news divisions since November 2015. Journalists at some of those stations have spoken out against the practice; some have said that they have tried to lessen the impact of the prepared pieces by, for example, airing them at less-popular viewing times or adjacent to commercial breaks, where viewers might see them as just another ad.
  • Discuss:
    • Spots promoting the news division of local stations have been a staple of the industry for decades. Is this one really any different?
    • Since this is a promotional spot and not a news report, is this an example of a parent company breaching the “firewall” that is supposed to keep newsrooms independent from management or from the parent company’s other interests?
    • Why is it significant that some journalists at Sinclair-owned stations have pushed back against its content mandates? Why do you think this one was leaked and became news?
    • The Sinclair promo script ends with this statement: “We understand Truth is neither politically ‘left or right.’ Our commitment to factual reporting is the foundation of our credibility, now more than ever.” Is “the truth” always nonpartisan? Can neutral, factual reporting be perceived as politically slanted? What is the political center, and who decides? Is the political center a fixed concept, or does it shift over time?
Please share this newsletter with others who may find this information useful (subscribe here). For more examples and ideas like these, you can follow me on Twitter (@PeterD_Adams). Also follow @NewsLitProject and @MrSilva.

If you have suggestions for future issues of The Sift, please share them here.

If you're looking for engaging and effective news literacy resources, check out NLP's checkology® virtual classroom. We’re giving away student licenses for 1:1 functionality for the rest of the 2017-18 school year. Yes, it’s free.

Copyright © 2018 The News Literacy Project. All rights reserved.
You are receiving this email because you signed up for resources and updates from the News Literacy Project.

5335 Wisconsin Ave. NW
Suite 440
Washington, DC 20015

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.