Is Twitter biased against conservatives? A Study

by kenjackson • 4 years ago

26 63 4 years ago

yablak• 4 years ago

"... the observation that Republicans were more likely to be suspended than Democrats provides no support for the claim that Twitter showed political bias in its suspension practices. Instead, the observed asymmetry could be explained entirely by the tendency of Republicans to share more misinformation. While support for action against misinformation is bipartisan, the sharing of misinformation –at least at this historical moment –is heavily asymmetric across parties. As a result, our study shows that it is inappropriate to make inferences about political bias from asymmetries in suspension rates."

DamnYuppie• 4 years ago

The key here is who is defining misinformation. As the COVID lab leak and Hunter Biden laptop story show there has been a tendency to loudly dismiss or condemn any dissenting opinions of those that don't align with the mainstream narrative as "misinformation". Only to subsequently find out later that these were not as far fetched or misinformed as previously believed.

busterarm2• 4 years ago

And the Russian DNC hack. Crowdstrike's President told the public one thing and then testified to the House Intelligence Committee that there's no evidence of Russian involvement.

This didn't seem to stop multiple MSNBC hosts from shouting about Russian election meddling, long after the Justice Department dropped such claims and related charges.

aNoob7000• 4 years ago

Has the Covid lab leak been proven true?

As for the Hunter Biden laptop story, what's the issue? I just want to know specifically.

amscanne5• 4 years ago

I’m not sure why you have been downvoted. It may be true that the Republicans share information from sketchier sources, but the Hunter Biden laptop story suppression was so bad that I don’t think Democrats have a leg to stand on here. The message that it was “Russian disinformation” was completely evidence-free and baseless, and the fact that Twitter blocked the URL was a clear signal that the standards for different political views is different.

jandrese4• 4 years ago

The study defined misinformation as links to articles that were deemed to be false by fact checkers.

lambada• 4 years ago

As the paper says in respect of politically motivated definitions of mis-information

Helping to address concerns about potential liberal bias among fact-checkers, we also examined untrustworthiness ratings from politically-balanced crowds of demographically representative (quota-sampled) American laypeople recruited via Lucid (15), rather than professional fact-checkers.

jonathan-adly2• 4 years ago

I am not sure why are you downvoted. But yes, who labels what's harmful misinformation and what is a legitimate dissenting voice is a big factor and the source of bias.

Zoom away from the polarizing US "current thing" BS politics and it becomes even clearer. Things like ESG, Nuclear energy, Middle Eastern conflicts, religious freedoms, criticizing China, etc.

The world is mostly gray, and whoever sets the boundaries of good/bad is king (that's why Media is valuable to billionaires and dictators).

skoskie• 4 years ago

Chronology matters here. Any conspiracy theory, no matter how seemingly unhinged could become factual in time. And when you spew misinformation and lies everyday you are bound to have some lies eventually emerge as truth. None of that should lend credibility to the people who found themselves eventually and accidentally speaking the truth.

The key here is who is defining misinformation.

We should respect the processes that are used to establish truthful information, even if it is occasionally wrong, rather than the noise makers who are occasionally right.

In other words “who” should not be a factor in determining misinformation, but “how”.

agomez314• 4 years ago

From reading the abstract, it seems like an exercise in circular logic.

olsonjeffery2• 4 years ago

Please expand upon this argument.

laverya2• 4 years ago

"Republican leaning users were banned ~5x more often than Democratic leaning users, but they also shared misinformation more as decided by we-promise-they're-neutral fact checkers and the average American, so there's no discrimination against Republican leaning users"

Which to me means "there's no discrimination unless the fact checkers and average American discriminate", also known as "there's no discrimination unless there's discrimination", which is circular logic.

Tehchops2• 4 years ago

I see disproportionate amount of violence/threats from right-leaning personas online. Perhaps that is a factor?

Yes, I know it's anecdotal, but the LGBTQ+ community has been the victim of an increasing number of hate crimes(confirmed by law enforcement reporting), and the rhetoric online seems to mirror this.

Free speech does not cover overt or implied violence based on sexual or gender identity.

_-david-_9• 4 years ago

Do you have any evidence that it is disproportionately coming from the right? I see a lot more coming from the left so it seems a bit anecdotal.

jandrese4• 4 years ago

Can you post any link to left leaning people calling for violence against LGBTQ aligned individuals?

danaris• 4 years ago

Well, let's see here.

We have one "side" publicly and loudly proclaiming the same things as racist lynch mobs from the 1800s and literal Nazis.

We have another "side" publicly and loudly calling for dignity and respect for every human being.

It would seem logical to me that the burden of proof here would be on the person claiming that the latter is sending more hate and death threats than the former. Extraordinary claims require extraordinary evidence.

Tehchops2• 4 years ago

Are you saying it's more likely hate crimes are being perpetrated by people that identify as progressive?

pphysch3• 4 years ago

"Instead, the observed asymmetry could be explained entirely by the tendency of Republicans to share more misinformation."

...misinformation defined by whom?

Factually speaking, the vast bulk of the "Russiagate" scandal was a hoax designed to achieve a political agenda, pure and simple. Trump may be as corrupt as the next guy, but there is no evidence that implicates him as a Russian asset. But who was removed from social media for boosting this blatant misinformation? Literally no one.

olsonjeffery2• 4 years ago

If you read the study you would know the answer, because they literally define the very thing you’re asking about.

As helpfully posted elsewhere in this thread:

Helping to address concerns about potential liberal bias among fact-checkers, we also examined untrustworthiness ratings from politically-balanced crowds of demographically representative (quota-sampled) American laypeople recruited via Lucid (15), rather than professional fact-checkers.

pphysch3• 4 years ago

The study pivots around popular perceptions of QAnon/Pizzagate as the archetype of "misinformation" and has zero mention of its political mirror-image, Russiagate. Both are false and ridiculous conspiracy theories aimed at defaming the opposing leadership. By omitting one of the two elephants (donkeys?) in the room, it has hard to take this study seriously. Very poor experimental design.

alphabettsy4• 4 years ago

Didn’t the Trump Camp lie about meeting with Russian operatives?

Didn’t people end up in jail for lying and refusing to cooperate?

It was absolutely political, but isn’t anything concerning an investigation of a sitting President?

pphysch3• 4 years ago

The bottom line is that the Steele Dossier was a QAnon-tier piece of disinformation, yet was endlessly paraded as "mostly true" by establishment media. That is an unexcusable double standard.

jonathan-adly2• 4 years ago

Basically whoever gets to labels what's harmful misinformation and what is a legitimate dissenting voice is the source of bias. Everything else is noise. The study is a great example of that noise.

Zoom away from the polarizing US "current thing" BS politics and it becomes even clearer. Things like ESG, Nuclear energy, Middle Eastern conflicts, religious freedoms, criticizing China, etc.

The world is mostly gray, and whoever sets the boundaries of good/bad is king (that's why Media is valuable to billionaires and dictators).

busterarm2• 4 years ago

This is why the Saudi leadership invests so heavily in companies like Twitter and other media companies.

silicon2401• 4 years ago

whoever gets to labels what's harmful misinformation and what is a legitimate dissenting voice is the source of bias

"misinformation" is the cleverest weapon against free speech in a while, despite being an old trick. It lets people support censorship while feeling like they're just opposing something evil. Catholic Europe said "heresy" or "blasphemy", McCarthyist USA said "communist", Islamic societies said "haram", secular dictatorships said "degenerate", etc etc.

padobson• 4 years ago

I'd like to see a study that correlates use of Twitter's "report" button with political persuasion.

My hypothesis is that lefties are more likely to appeal to authority and hit the "report" button when they see something they don't like, and right wingers are more likely to go vigilante and hit "reply" and attempt to persuade/shout the person into agreeing with them.

orwin• 4 years ago

I don't think you know a lot of lefties, or that your lefties are at worst socdems (barely leftwing)

ddingus• 4 years ago

I break it down on authoritarian lines. Left, right, indie are a separate thing.

Authoritarians do these things most often:

Brigade, mass report, preemptive blocks, use of lists.

Malicious Compliance and or using, gaming rules to get other people into problem positions.

Tribe vs tribe, guilty by association, and or forcing others into false choice scenarios.

Mixing authoritarianism with political alignment complicates things unnecessarily in my view. There can be poor quality people favoring any ideology, and or multiple ones, same as there can be better quality people.

I urge you to give it some thought. You may find it helps to understand others better.

zmibes• 4 years ago

Might as well ask the Catholic church to conduct a study into whether God exists. (They are the experts, right?)

Is Twitter biased against conservatives? A Study

Related Stories