Students Have 'Dismaying' Inability To Tell Fake News From Real, Study Finds
If the children are the future, the future might be very ill-informed.
That's one implication of a new study from Stanford researchers that evaluated students' ability to assess information sources and described the results as "dismaying," "bleak" and "[a] threat to democracy."
As content creators and social media platforms grapple with the fake news crisis, the study highlights the other side of the equation: What it looks like when readers are duped.
The researchers at Stanford's Graduate School of Education have spent more than a year evaluating how well students across the country can evaluate online sources of information.
Middle school, high school and college students in 12 states were asked to evaluate the information presented in tweets, comments and articles. More than 7,800 student responses were collected.
In exercise after exercise, the researchers were "shocked" — their word, not ours — by how many students failed to effectively evaluate the credibility of that information.
The students displayed a "stunning and dismaying consistency" in their responses, the researchers wrote, getting duped again and again. They weren't looking for high-level analysis of data but just a "reasonable bar" of, for instance, telling fake accounts from real ones, activist groups from neutral sources and ads from articles.
"Many assume that because young people are fluent in social media they are equally savvy about what they find there," the researchers wrote. "Our work shows the opposite."
A professional appearance and polished "About" section could easily persuade students that a site was neutral and authoritative, the study found, and young people tended to credulously accept information as presented even without supporting evidence or citations.
The research was divided by age group and used 15 different assessments. Here's a sample of some of the results:
Most middle school students can't tell native ads from articles.
The researchers showed hundreds of middle schoolers a Slate home page that included a traditional ad and a "native ad" — a paid story branded as "sponsored content" — as well as Slate articles.
Most students could identify the traditional ad, but more than 80 percent of them believed that the "sponsored content" article was a real news story.
"Some students even mentioned that it was sponsored content but still believed that it was a news article," the researchers wrote, suggesting the students don't know what "sponsored content" means.
Most high school students accept photographs as presented, without verifying them.
The researchers showed high school students a photograph of strange-looking flowers, posted on the image hosting site Imgur by a user named "pleasegoogleShakerAamerpleasegoogleDavidKelly. The caption read "Fukushima Nuclear Flowers: Not much more to say, this is what happens when flowers get nuclear birth defects."
Sam Wineburg, a professor of education and history at Stanford University and the lead author of the study, spoke to NPR on Tuesday.
"The photograph had no attribution. There was nothing that indicated that it was from anywhere," he said. "We asked students, 'Does this photograph provide proof that the kind of nuclear disaster caused these aberrations in nature?' And we found that over 80 percent of the high school students that we gave this to had an extremely difficult time making that determination.
"They didn't ask where it came from. They didn't verify it. They simply accepted the picture as fact."
Many high school students couldn't tell a real and fake news source apart on Facebook.
One assessment presented two posts announcing Donald Trump's candidacy for president — one from the actual Fox News account, with a blue checkmark indicating it was verified, and one from an account that looked like Fox News.
"Only a quarter of the students recognized and explained the significance of the blue checkmark, a Stanford press release noted. "And over 30 percent of students argued that the fake account was more trustworthy."
Most college students didn't suspect potential bias in a tweet from an activist group.
The researchers sent undergraduate students a link to a tweet by MoveOn about gun owners' feelings on background checks, citing a survey by Public Policy Polling.
New polling shows the @NRA is out of touch with gun owners and their own members https://t.co/wm6weuaCbd #NRAfail pic.twitter.com/y4K4r5EcYX— MoveOn (@MoveOn) November 18, 2015
They asked students to evaluate the tweet and say why it might or might not be a good data source.
"Only a few students noted that the tweet was based on a poll conducted by a professional polling firm," which might make it a good source, the researchers wrote.
At the same time, less than a third of students cited the political agenda of MoveOn.org as a reason it might be a flawed source.
And more than half of the students didn't even click on the link within the tweet before evaluating the usefulness of the data.
Most Stanford students couldn't identify the difference between a mainstream and fringe source.
The American Academy of Pediatrics, which publishes the journal Pediatrics, has more than 65,000 members and has been around since 1930.
The American College of Pediatricians (ACPeds) split from AAP in 2002, over objections to parenting by same-sex couples. ACPeds claims homosexuality is linked to pedophilia. It's classified as a hate group by the Southern Poverty Law Center, which estimates that ACPeds has about 200 members.
In an article in Education Week, Wineburg and his colleague Sarah McGrew explain that they directed Stanford undergrads to articles on both organizations' sites. The students spent up to 10 minutes evaluating them, and were free to click links or Google anything they liked.
"More than half concluded that the article from the American College of Pediatricians ... was 'more reliable,' " the researchers wrote. "Even students who preferred the entry from the American Academy of Pediatrics never uncovered the differences between the two groups."
You can see in-depth examples of some of the exercises — including sample responses — at the study's executive summary.
The project began before the recent uproar over the prevalence of fake news online. But its relevance is immediately clear.
Wineburg told NPR on Tuesday that the study demonstrates that U.S. classrooms haven't caught up to the way information is influencing kids daily.
"What we see is a rash of fake news going on that people pass on without thinking," he said. "And we really can't blame young people because we've never taught them to do otherwise."
In fact, as Wineburg and McGrew wrote in Education Week, some schools have filters directing students to valid sources, which doesn't give them practice learning to evaluate sources for themselves.
The solution, they write, is to teach students — or, really, all Internet users — to read like fact checkers.
That means not just reading "vertically," on a single page or source, but looking for other sources — as well as not taking "About" pages as evidence of neutrality, and not assuming Google ranks results by reliability.
"The kinds of duties that used to be the responsibility of editors, of librarians now fall on the shoulders of anyone who uses a screen to become informed about the world," Wineburg told NPR. "And so the response is not to take away these rights from ordinary citizens but to teach them how to thoughtfully engage in information seeking and evaluating in a cacophonous democracy."