When The MRC's Right-Wing Narratives Fail, Part 2

The Media Research Center doesn't like it when others blow up its partisan narrative about purported social media bias against conservatives -- but it spends a lot of time raging about it and very little time actually rebutting the criticism.

By Terry Krepel
Posted 1/27/2022

A Feb. 1 item by Corinne Weaver complained that a study blew up one of the MRC's biggest narratives, that conservatives are uniquely being "censored" on social media:

A study released by the NYU Stern Center for Business and Human Rights decided that the argument that conservatives are being censored by Big Tech is “not legitimate.”

“False Accusation: The Unfounded Claim that Social Media Companies Censor Conservatives,” levied the accusation that conservative censorship is a myth. Deputy Director Paul M. Barrett and Law + Research Fellow J. Grant Sims wrote that “the claim of anti-conservative animus is itself a form of disinformation: a falsehood with no reliable evidence to support it.” The study referred to the overall bans of former President Donald Trump on Facebook and Twitter as “reasonable responses to Trump’s repeated violation of platform rules.”

The piece accused Rep. Jim Jordan (R-OH), conservative commentator Ben Shapiro and Fox News host Tucker Carlson of spreading “The false contention that conservatives are throttled online.” The problem with criticizing online censorship now, according to the study, is that it delegitimizes efforts made by platforms “when they’re actually experimenting with more aggressive forms of fact-checking and content moderation.”

The study relied on NewsGuard’s classification of “manipulators.” It noted, “All of the top five manipulators, in terms of their engagement levels on Facebook, were right-leaning: Fox News, The Daily Wire, Breitbart, The Blaze, and Western Journal.”

Furthermore, Barrett and Sims relied on the leftist Oxford Internet Institute’s report on “junk news.” That report had classified several conservative websites as “junk news,” including Drudge Report, NewsBusters, CNSNews.com, MRCTV, Breitbart, The Daily Caller, The Washington Free Beacon, LifeNews, National Review, the Red State, and The Federalist. These sites were smeared as “unprofessional,” “counterfeit,” “biased” and “emotionally driven.”

Note that all Weaver did here is name-calling -- dismissing the study as "liberal" and claiming it relied on a "leftist" group's previous report, then accused the Stern Center of having "a liberal advisory board." At no point did Weaver even attempt to rebut any claim actually made in the report.

There's a lot in the Stern Center report that's pretty damning of the MRC's victimization narrative, though Weaver will never admit it. It stated that "Even anecdotal evidence of supposed bias tends to crumble under close examination" -- and offered examples. It takes to task Robert Epstein -- a favorite of the MRC for his dubious research claiming that Google manipulates search results for the purpose of "switching" votes from Democrats to Republicans:

The basic question Epstein asks—how might internet searching affect voting—is potentially important. But his extrapolation to hard numbers of purposefully changed votes seems highly questionable. Francesca Tripodi, a social media scholar at the University of North Carolina who has reviewed Epstein’s work, says in an interview that he lacks evidence of either Google’s intent to manipulate elections or that the company has distorted search results toward that end. In a November 2020 article in Slate, she writes that “his hypothesis that Google influenced U.S. elections has never been rigorously tested or reviewed by political or information scientists.”

In fact, there is other research that clashes with Epstein’s findings. A study released in 2019 by researchers at Stanford University concluded that Google’s search algorithm is not biased along political lines and instead emphasizes authoritative sources. In a separate inquiry published the same year, The Economist came to a similar conclusion. The magazine compared news sites’ actual proportion of search results in Google’s News tab with a predictive model of that proportion based on factors Google says its search rankings rely on—namely, a site’s reach, output, and accuracy. “If Google favored liberals, left-wing sites would appear more often than our model predicted, and right-wing ones less,” The Economist said. “We saw no such trend.”

Epstein counters that his “work is meticulous. My standards are very high.” He points out that in 2015, he co-authored an article on SEME for the peer-reviewed Proceedings of the National Academy of Sciences. That piece, however, didn’t make any allegations against Google or point to any actual vote manipulation.

The report also blew up a key MRC narrative that Twitter exclusively bans conservatives for expressing mainstream conservative views:

Conservatives do get suspended or banned for violating Twitter’s rules against such things as harassment, hateful conduct, or, as in Trump’s case, glorifying violence. But liberals are excluded in this fashion, as well. Pinning down precise proportions is impossible because Twitter doesn’t release sufficient data.

[...]

More broadly, Twitter has suspended or banned individuals and groups of highly disparate political persuasions. In 2018, the platform excluded some 80 accounts belonging to activists affiliated with the left-leaning Occupy movement. According to some of these activists, Twitter revoked the accounts without giving a reason. In February 2020, Twitter banned 70 accounts affiliated with Mike Bloomberg’s short-lived campaign for the Democratic presidential nomination, reportedly for violating the platform’s policy against platform manipulation and spam.

On the right, Twitter doesn’t target conservatives or Republicans as such, but people who violate its rules by calling for violence, harassing others, or advocating hateful ideologies. Among the right-leaning users who have faced enforcement action are white nationalists like Richard Spencer, Jared Taylor, and David Duke, as well as white nationalist organizations such as the American Nazi Party, the neo-Nazi Traditionalist Worker Party, and American Renaissance magazine.

Weaver could not possibly offer a response to this because "media research" isn't what the MRC does -- its job is to push right-wing narratives.

Hating (then loving) the OII

Weaver has previously attacked the Oxford Internet Institute as "leftist" for blowing up right-wing narratives about media -- in part by pointing how just how biased and unreliable some right-wing websites are. In 2018, Weaver attacked an OII study concluding that "junk news" is disproportionately created by right-wing websites; she offered no rebuttal of that claim, instead whining that "Liberal media will go a long way to portray conservatives as liars -- all the way to England." In 2019, she bashed the OII for keeping an eye on right-wing "junk news" sites like the MRC's NewsBusters, where it so happens Weaver's work is posted.

Weaver huffed in an August 2020 post about another OII study: "Academics funded by Big Tech only argue for more censorship of conservative thought." The headline on her post dishonestly claimed that OII director Philip Howard said it would be a "public service" to "take down conservative sites"; in fact, Howard was talking about websites that publish "junk news and disinformation," saying, "They are hucksters, fraudsters, peddling misinformation. It would be a public service to take them down." While Weaver made a point of complaining that several right-wing websites fell under the study's "junk news and disinformation" category, Weaver made no effort to disprove the claim they contain junk news and disinformation; further, the category included some non-right-wing sites as well, such as Russian-run sites RT and Sputnik News. Weaver further played victim by complaining that the study was "using 'fact-checkers' with a history of political bias like Politifact and Newsguard" and whined of a study complaint about Google referring traffic to junk-news sites that "Google already suppresses conservative content."

In October 2020, Weaver complained that the OII once again called out NewsBusters for serving "junk news":

In the briefing published on Oct. 5, 2020, the institute slammed nine pieces written and published in American conservative outlets such as NewsBusters, The Daily Caller, The Heritage Foundation’s The Daily Signal, BizPac Review, The Daily Wire and The Blaze. According to the Oxford Internet Institute (OII), these sites are considered “junk news.”

OII defined “junk news” as “... sources deliberately publish misleading, deceptive or incorrect information purporting to be real news about politics, economics or culture. This content includes various forms of propaganda and ideologically extreme, hyperpartisan or conspiratorial news and information.”

[...]

The briefing named a piece by NewsBusters managing editor Curtis Houck as an example of the “sharp criticism ... levied against the reactions of mainstream media.” Even though the article in question was merely a wrap-up of quotes from NBC concerning the presidential debate that took place on Sept. 30, OII considered it “one of the best-fitting articles in the topic model of the previous section.” The issue with this characterization of the piece as “junk news” is that in the previous section, the only model given about pieces concerning the debate was that it “included words such as ‘Biden’, ‘debate’, ‘Trump’, ‘Wallace’, and ‘election’. This topic concerned the first US Presidential debate held on Tuesday night.”

Weaver is being disingenuous about the nature of OII's criticism of Houck's piece. It pointed out how right-wing websites' reactions to the debate fell into certain patterns: attacking moderator Chris Wallace, criticizing Biden's language without criticizing Trump's similar language, and bashing the media's reaction to the debate. That last point is what Houck's post got dinged on. Here's what OII wrote, since Weaver won't tell you:

Further, sharp criticism was levied against the reactions of mainstream media. A Daily Wire article with over 20,000 engagements detailed a CNN’s panel response to the debate, and although it mentioned Trump’s refusal to denounce white supremacy, it also claimed that Biden was the one that sunk to personal insults. Another article from NewsBusters that had comparatively few engagements at over 3,000 but was one of the best-fitting articles in the topic model of the previous section employed a similar strategy with NBC’s panel reaction.

Weaver didn't offer any rebuttal to OII's criticism -- just complained that it was made. Instead, she noted OII's definition of "junk news" and huffed, "By this definition, BuzzFeed would be 'junk news.'"

This is what happens when you put a partisan narrative first, as the MRC has done with its anti-social media war: you get swiftly and painfully owned by actual media researchers.

But in June, the MRC suddenly -- and hypocritically -- decided that the OII was credible, at least when it turned up a finding that aligned with the MRC's political agenda. Three posts that month referenced an OII study citing "inauthentic amplification" of Chinese-related content on Facebook, with the goal of attacking Facebook. The flip-flop on OII wasn't explained.

Narrative blown up again

In July, the MRC found its victimization narrative getting blown up again -- this time by Axios, which pointed out in an article on Donald Trump's lawsuit against social media for banning him that "To date, Trump and other conservative critics have not presented any substantial evidence that either platform is biased against conservatives in its policies or implementation of them," MRC chief Brent Bozell had a meltdown, as described in a July 7 item by Alexander Hall:

Bozell raked Axios over the coals by describing its coverage as “[b]latant lies from the leftist media.”

Free Speech America also scorched Axios for denying Big Tech censorship by summarizing some of the most infamous examples:

• Facebook and Twitter censored the Hunter Biden story and interfered in the 2020 election
• Trump’s banned on 10 Big Tech sites.
• Our http://CensorTrack.org database has already amassed more than 2,500 examples of censorship.

What Axios didn’t tell readers is that Facebook and Twitter don’t reveal how their algorithms work. So that restricts how conservatives can respond. But it is undeniable that both Facebook and Twitter did indeed censor the Hunter Biden story in the lead up to the 2020 election.

[...]

A post-election poll from the Media Research Center, conducted by McLaughlin & Associates, also sheds light on how censorship impacted our nation. The poll showed that 36 percent of Biden voters were NOT aware of the evidence linking then-presidential candidate Joe Biden to potentially corrupt financial dealings with China through his son Hunter. Thirteen percent of these voters (or 4.6 percent of Biden’s total vote) said that had they known these facts, they would not have voted for the former Vice President.

The MRC also won't tell you that the poll it paid for was done by Trump's own election pollster -- raising questions about its honesty and accuracy -- or that its CensorTrack database is highly selective, choosing only instances of "censorship" that advance the MRC's victimhood narrative and is in no way a comprehensive examination that proves social media "censors" conservative content solely and exclusively.

Hall also rehashed the MRC's bogus talking points: "Twitter alone censored Trump 625 times between May 31, 2018 and January 4, 2021 before he was kicked off of the platform. It has since refused to allow him back on the platform. Twitter did not censor Biden at all during the same period of time. Trump was banned from at least nine other platforms after the Jan. 6 riot at the U.S. Capitol, during which the former President called for 'peace.'" The more accurate way to say it is that Trump violated Twitter's terms of service 625 times while Biden did not violate them at all, and it's an absolute lie for Hall to claim that Trump was kicked off social for calling for "peace" at the riot he helped instigate.

The ultimate evidence that the MRC knows its talking point is bogus is that it sent out an email in April bragging about how well its content does on Facebook. Would it be making such extensive use of Facebook -- and bragging about its reach there -- if the "censorship" narrative was at all true? Doubtful.

As with the rest of its promotion of Trump's lawsuit, this is all just partisan blathering not meant to be taken seriously outside the MRC's ideological bubble.