Readers with a negative view of media less likely to spot a fake headline, study finds

par Joshua Benton
21 nov 2018 dans Combating Mis- and Disinformation
Man with newspaper

Don’t like the media? Think it’s all “lies” or “fake”? Then you’re probably not as good at reading the news as your less perpetually annoyed peers.

That’s one finding from a new study from the News Co/Lab at Arizona State, in collaboration with the Center for Media Engagement at the University of Texas. Those who have negative opinions of the news media are less likely to spot a fake headline, less likely to differentiate between news and opinion — but more confident in their ability to find the information they need online.

The study surveyed 4,854 people in three cities — Kansas City, Fresno, and Macon, Georgia — and asked them what was the first word that came to mind when they saw the word “news.” About 62 percent responded with something negative — “fake,” “lies,” “untrustworthy,” and “BS” were the sample responses given. The remaining 38 percent responded with something positive or neutral (like “factual”).

That divide — a positive or negative reaction to “news” — mapped onto a number of other elements the researchers surveyed.

For instance, people were given three at least somewhat plausible headlines and ledes that might appear in their local newspaper. Two were real; one was fake. Those with positive attitudes fared better in figuring out which was which. In Kansas City, 82 percent of the half-glass-full types figured out which was fake, versus only 69 percent of the half-glass-empties. (The fake headline? “New study: Nearly half the nation’s scientists now reject evolution.”)

Another question asked people to categorize stories as news, opinion, analysis, or sponsored content. The negatives were less likely than the positives to correctly identify the news — though not by a super-wide margin, 74 percent to 80 percent.

So do the people who are bad at reading the news know they’re bad at it? Not so much. Another question asked which of these best described them:

  • “I do not need help finding the information I need online.”
  • “I could occasionally use some help in finding the information I need online.”
  • “I frequently need help finding the information I need online.”

Those with negative reactions to the word “news” were less likely to say they ever needed help (34 percent) than those with positive or neutral reactions (42 percent).

Beyond all that, the report — by Gina Masullo Chen, Caroline Murray, Eric Newton, Dan Gillmor, Kristy Roschke and Natalie Jomini Stroud — confirms a number of findings other studies have also found.

Does education make you more able to detect a fake headline? Yes. College grads sussed it out at a higher rate (68 percent) than those with less than a college degree (57 percent).

Age and income? Yes, people aged 65 and up were somewhat less likely to identify the fake headline than those 18 to 64 (60 percent vs. 66 percent). And those making over US$150 thousand were better at it than those making less than US$30 thousand (71 percent verses 54 percent).

How about partisan identity? In Kansas City and Macon, Democrats were more likely to pick out the fake than Republicans — by a +12 margin in K.C. and +18 in Macon. In Fresno, interestingly, there was no statistically significant difference between the parties.

Of course, all of these factors — partisan identity, age, income, education, dislike of the media — overlap and intersect in a number of ways, so it’s harder to pin the credit/blame on any one. Democrats were much less likely to use a negative word to describe news than Republicans — 26 percent verses 75 percent.

“We’re seeing a divide in news literacy among specific groups that may diminish their ability to fully understand what’s happening in the world,” said Chen, a past Nieman Lab contributor.

There’s a whole other section to the report — surveying local journalists about a variety of issues — that’s also interesting. You can check it and the rest of the report here.


This article was originally published by NiemanLab. It was republished on IJNet with permission.

Main image CC-licensed by Unsplash via rawpixel.