Meta has more questions than it answers

The First Results of Meta’s Social Media Dispatch: Demagogical Partisanship and False Information in the Age of Elections

Researchers theorize that Meta’s content- delivery algorithm puts information from people with like minded friends in front of the political right. The fear is that this system reinforces online echo chambers that encourage the spread of partisan and false information.

The researchers weren’t paid by Meta, but the company seemed pleased with the results, which were released today in four papers in Nature and Science. The president of global affairs at Meta said in a statement that there is little evidence that the key features of Meta’s platforms alone cause harmfulaffective polarization or have meaningful effects on political views.

Despite setbacks, Meta and researchers have continued to work together. On Thursday, the first research to come out of this work was published.

Guess said that they didn’t know what would happen had they been able to conduct the studies over a period of time. More importantly, he said, there is no accounting for the fact that many users have had Facebook and Instagram accounts for upwards of a decade now. “This finding cannot tell us what the world would have been like if we hadn’t had social media around for the last 10 to 15 years or 15 or 20 years.”

The time frame that the researchers were able to study was the run up to the election in an atmosphere of intense political enmity.

Michael Wagner is a professor of journalism and communication at the University of Wisconsin- Madison, and was a co-chairman of Meta.

The research challenges the assertion that the ability to reshare content on social media drives polarization, according to Meta.

Researchers weren’t so sure. Resharing increases content from unreliable sources according to one of the studies. The study shows that the misinformation caught by the third-party fact checkers is primarily consumed by conservative users, which has no equivalent on the other side of the political aisle.

The company long ago agreed, in principle, to do just that. It has been a rough path. The Cambridge Analytica data privacy scandal of 2018, which originated from an academic research partnership, has made Meta understandably anxious about sharing data with social scientists. The Social Science One project went nowhere, as it took so long to produce data that its biggest backers quit before producing anything of note. (Later it turned out that Meta had accidentally provided researchers with bad data, effectively ruining the research in progress.)

In each of the experiments, the tweaks did change the kind of content users saw: Removing reshared posts made people see far less political news and less news from untrustworthy sources, for instance, but more uncivil content. Replacing the algorithm with a chronological feed reduced the amount of hate and discrimination, even though it resulted in more unreliable content. Users in the experiments spent less time on the platforms than other users, suggesting they have become less compelling.

The findings don’t confirm the argument that Meta’s products play a leading role in the United States being at risk for democracy, because they are alone. But nor do they suggest that altering the feed in ways some lawmakers have called for — making it chronological rather than ranking posts according to other signals — would have a positive effect.

The title of Science’s package on the studies “Wired to Split” was changed to reflect uncertainty but not the presentation of the research, according to Horwitz.

There will be more research on the way. The four studies released today will be followed by 12 more covering the same time period. Perhaps, in their totality, we will be able to draw stronger conclusions than we can right now.

Data architecture and programming code are complex at corporations such as Meta, and academy collaboration research models don’t properly engage with this in their studies. “Simply put, researchers don’t know what they don’t know, and the incentives are not clear for industry partners to reveal everything they know about their platforms.”

Meta said it is committed to transparency, but that privacy obligations prevent it from making data available to external researchers. Meta said that it hopes that the research results will help policymakers as they set the rules of the road for the internet to benefit democracy and society as a whole.

independence is not independent at all according to the author. It’s a sign of things to come in the academy, as data and research opportunities are offered to a few researchers at the expense of true independence. Scholarship is not wholly independent when the data are held by for-profit corporations, nor is it independent when those same corporations can limit the nature of what it studied.”

The project will hopefully inspire more research, but it’s still up to Meta and the other social-media platforms, warns Tucker. “We very much hope that society, through its policymakers, will take action to make sure that this kind of research continues in the future,” Tucker says.

In a study, a group of over 22,000 social network users were divided into two groups: one that received good content and the other that received bad. Giving users the latest news and information would make it more apparent to them that they are seeing something.

Two other interventions, published in Science3 and Nature4, also showed little effect: one limited “reshared” content — which comes from outside a user’s network but is reposted by a connection or a group to which the user belongs — and the other limited content from “like-minded” users and groups.

For Joshua Tucker of New York University and a lead investigator on the project, the lesson is clear: proposed solutions for reducing online echo chambers probably wouldn’t have had much of an impact in the 2020 election. Tucker acknowledges that there are limits to the research’s findings.

The authors say that all of the data that were collected will be available for researchers. The model depends on Meta’s willingness to participate, which is why some academics question it.