A new study from Arizona State’s News Co/Lab, working with the University of Texas’ Center for Media Engagement (PDF download), claims to show that people with a negative opinion of the media are less likely to detect fake news headlines.
This is, of course, just a polite way of saying those who dislike the media are ignorant rubes.
So, how did they come their finding? We’ll let News Co/Lab tell you in their own words:
“When given two real headlines and one fake headline, 68 percent of survey participants with a college degree or more were able to successfully identify the fake news headline, compared to 57 percent of participants with less than a college degree. Note that this means that more than 30 percent of college-educated participants were unable to identify the fake headline.
We also asked survey participants to share the first word they think of when they hear the word “news”. Seventy-four percent of those identifying as Republicans used a negative word, such as “fake” or “biased,” whereas only 26 percent of those identifying as Democrats used negative words. The negative association was also found to be a differentiating factor in spotting fake headlines. Participants who associated negative words with the word “news” were less able to spot fake headlines and to distinguish news from opinion, analysis or advertising.”
- Three headlines.
- Two are real. One is fake.
- Spot the fake.
- Finally, a game of word association with “news.”
- People who associate the word “news” with negativity had a lower rate of identifying the fake headline.
Simple, straightforward stuff, right?
Ah, But There’s A Catch (You Knew There Would Be) . . .
In two of the three tests, the fake news headline addressed highly-contentious, highly-partisan issues: gang violence (with race and gender tossed in for good measure . . . oh boy!), and evolution.
This is important for two reasons.
First, we know there’s a strong ideological divide between how the media is viewed. Those on the left view the media in a far more positive light than those on the right. This research confirms this fact: of those who associated negative words with “news,” 74% identified as Republican (versus only 26% Democrat).
Secondly, we know people are far more likely to believe things that support their beliefs, and less likely to believe things that don’t. This is a well-known psychological concept known as confirmation bias:
“Confirmation bias, also called confirmatory bias or myside bias, is the tendency to search for, interpret, favor, and recall information in a way that confirms one’s preexisting beliefs or hypotheses. It is a type of cognitive bias and a systematic error of inductive reasoning. People display this bias when they gather or remember information selectively, or when they interpret it in a biased way. The effect is stronger for emotionally charged issues and for deeply entrenched beliefs.”
Now can you see the issue with this research using contentious, partisan issues in their fake news headlines?
If not, continue reading.
Whoever Finds The Fake News Headline More Favorable, Loses
If you want somebody to believe something that’s false, make them want to believe it, and they (probably) will.
It’s really that simple.
This begs one important question: who would find the researchers’ fake news headlines more favorable, Republicans or Democrats?
Let’s take a close look at each.
Fake Headline #1: “Macon gang initiation: Shoot white women at mall.”
This fake headline used in Macon, Georgia.
How could a headline based around gang violence induce confirmation bias? After all, isn’t gang violence a nonpartisan issue?
Well . . . mostly.
There was that issue several months back where Democrats bizarrely defended MS-13, arguably the most-violent gang in the world, over Donald Trump calling them animals.
Other than that, I can’t find any research or polls which show an ideological divide among the issue of gang violence.
The real issue with this fake headline is the strange addition of “white women.” Gang initiations aren’t just targeting anyone, but white women.
What’s the significance of this addition?
Anti-white violence is a common theme among Republicans. You may deny the reality of the threat, but there’s no denying the belief of the threat (often spurred on by disgusting racist comments from those on the left).
This is confirmed by a poll from Reuters/Ipsos, via The Week:
“The poll, conducted in conjunction with the University of Virginia Center for Politics, found that 39 percent of respondents strongly or somewhat agreed that “white people are currently under attack in this country,” while 38 percent disagreed.”
Knowing that anti-white violence is a concern for many Republicans, but much lesser so for Democrats, confirmation bias (and common sense) says that, yes, they’re more susceptible to believing fake news which delves into the issue.
Which would explain why they were (slightly) less likely to identify the fake headline.
Fake Headline #2: “New study: Nearly half the nation’s scientists now reject evolution.”
The fake headline was used in Kansas City, Missouri.
What makes this headline interesting is how politically-divided the century-old issue of evolution is.
According to a 2014 article from the Pew Research Center, only 43% of Republicans accept the claim that humans have evolved over time, compared to 67% of Democrats.
The notion that nearly half the nation’s scientists would reject evolution is much more appealing to Republicans, the majority of whom are evolution skeptics, than to democrats, two-thirds of whom accept the evolution hypothesis.
Could it be the reason those who view the media in a negative light (read: mostly Republicans) found this claim more believable is because they found it more favorable? Isn’t that a better explanation than the claim that they’re less-capable of spotting fake news?
Confirmation bias says that’s probably the case.
Do you know what would strongly support my hypothesis?
If the researchers asked a non-political, non-ideological question, and the accuracy rate between those with a negative and non-negative association of the media disappeared . . .
What Happened When The Fake Headline Went Non-Political?
The third and final headlines, used in beautiful Fresno, California, went with an apolitical, geographic-based fake headline:
“‘By population, Fresno is now the 10th largest city in the U.S.'”
And what was the result of this politically-neutral test?
“In Fresno, however, 59.6 percent correctly identified the false headline, and there were no statistically significant differences by partisanship.”
Now that’s interesting. A politically-neutral headline gave neutral results.
So, what happened? Did those who think negatively of the media get better at detecting fake news, or did those who think positively of the media get worse? Maybe it was a fluke?
Or maybe the politically-neutral nature of the question made confirmation bias a non-factor, giving us the fairest, and therefor most-accurate result of the entire study.
I think I nailed it.
What’s Up With Those Fake Headlines, Anyway?
Did the researchers intentionally use fake headlines designed to exploit confirmation bias, giving them their desired results?
I highly doubt it.
They seem to be honest researchers, working at respectable organizations. Accusations of pushing personal agenda would be unfair and unwarranted.
I think the more likely (and more charitable) interpretation is that they simply goofed.
Their fake headlines ended up one-sided, and it produced one-sided results.
One thing worth noting is that the researchers mention the clear distinction between the nature of the Macon and Kansas City fake headlines, and that of the one used in Fresno. From footnote 2 of the PDF:
We note the fake headline in Fresno differed from the fake headlines in Macon and Kansas City. In Fresno, the fake headline was geographical in nature, written in an effort to appeal to a person who thought the city did not get the attention it deserved. In Macon and Kansas City, the headlines were rooted in racial or ideological beliefs. In addition, the language used in the fake headline in Fresno was not incendiary.
They gave us their reasoning for why they used the Fresno fake headline. What about the other two?
Do they disagree that confirmation bias may have played a role in their research?
What do they think the results would have been had they used a fake headline such as, “climate change leads to extinction of rare horse on the island of Tonga”
Would the researchers consider doing another study using a more ideologically-balanced, or ideologically-neutral fake headlines? Repeatability is a requirement of good research, after all. The results of this followup research would either strengthen their original findings (and blow my rebuttal to smithereens), or support my position, and give us a whole new set of data with which to glean valuable insight.
Food for thought.
Our Oh-So-Humble Media . . .
There was at least one other interesting finding from the research: the spectacular discrepancy between how the local media views itself, and how the public views the local media.
Look at this glorious chart from @EngagingNews. It's a survey of 4,854 people about fake news.
Go through it slowly.
Journalists think they're beloved heroes, unbiased, credible, accurate.
Readers hate journalists and don't trust them.
— Ezra Levant 🍁 (@ezralevant) November 29, 2018
How could anyone ever think negatively about such wonderful, unbiased, fair and accurate people? Heh.
One Last Thing . . .
If you enjoyed this article, or just want to help out a young independent writer, please share it. It’s getting harder and harder for right-wing media to have their voices heard online, and it would be much appreciated. Thanks!