New data show that the polarization of online political discourse has largely remained the same since the end of 2020. This is probably not surprising if you have seen it all over the internet in the past year. But the data also show an underlying pattern in which individual subjects – such as abortion and immigration – continue to divide. While people were constantly mad online about political issues, the issues that provoked the conversation changed dramatically throughout the year.
Data from a joint project between Signal Labs, a social media intelligence platform, the University of Southern California Annenberg School and PR firm Golin, helps explain why political discourse in 2021 seemed like a never-ending carousel. Outrage.
Zignal, USC Annenberg and Golin teamed up to create a polarization index, which measures the connection to polarizing content on Twitter and calculates polarization scores. Since the index began tracking conversations last year, there have been major political events, such as the January 6 uprising, the transition from Trump to the Biden administration, and most of the Covid-19 vaccine rollouts. All in all, the PI score has barely progressed.
While Twitter is far from a complete proxy for broad divisions, online platforms play a vital role in shaping political discourse. Social media platforms such as Meta (formerly Facebook) have come under the microscope again this year, leading to new doubts about the ethics of these platforms and how they can address misinformation, extremism and hate speech online.
There has been a long-running academic debate on how to measure polarization and no clear standard has yet emerged. The index generates average polarization scores for 10 political subjects – immigration, policing, ethnic equity, abortion, voting integrity, gun law, climate change, minimum wage, Covid-19 vaccines and health-care reform (from 100 to 100). ) Is 100 absolute polarization on a scale of). Polarization scores are calculated by associating the share of news links on Twitter with the bias and credibility ratings of media sources publishing distributed content, assuming that “no” is a less-credible source on both ends. Political bias spectrum Does more polarization.
The group of media sources according to bias and credibility comes from Ad Fontes Media Bias Chart, an independent news content rating company that determines political trends and evaluates credibility based on factual reports.
Why did it feel so bad to be online this year?
The polarization index began at the end of 2020 with a score of 85.5, which researchers called a “critical” level. The score dropped just 3 points in early 2021 and has been steady ever since.
Currently, immigration is the most polarized subject measured by the index, followed by policing policy, ethnic equity and gun laws. At the subject level, changes in polarization were more common, and the degree of polarization seemed to vary from subject to subject, keeping overall scores high.
Voting integrity, for example, was the second most controversial issue in Q4 2020, and then rose from fifth to sixth out of 10 in the second half of 2021.
Research published with the Polarization Index also found that news articles shared on most polarizing topics are more likely to come from unreliable, right-wing sources. The report says that “engagement with right-leaning sources was more likely to push the conversation in an increasingly polarized direction.”
For example, this was the case for immigration, the most polarized topic: from the end of 2020 to the third quarter of 2021, right-wing sources of middle and low credibility dominated the conversation, and the polarization score increased to 84.8. Up to 100.3 during the year. The pattern is consistent with other highly polarized subjects.
What’s to come
Consistent with the results of Zignal’s research, it is well documented that even more extreme content is more misleading.
“A lot of disinformation is top-down,” says Anya Schiffrin, director of technology, media and communications at Columbia University. It comes from the heads of state, it comes from the politicians. Shiffrin also pinned the problem on the lack of a “gatekeeper” to police the flow of content. Instead, algorithmic recommendation systems on social media platforms expand the extreme content, which Shiffrin says leads to more “extreme internet”.
Due to the extreme digital environment this year saw dramatic demonstrations of real-world violence. Examples of this relationship include Facebook’s role in post-insurgency violence in Myanmar and the January 6 uprising in the United States, which was the result of a flurry of misinformation about the election results.
At the request of the MIT Technology Review, Zignal conducted an analysis focusing specifically on how people were periodically connected to various media sources on the issue of electoral confidence and voter integrity. Data show that engagements with less credible sources on both the left and right were closest to the election and around the events of January 6.
In late 2020, engagement with sources, especially those with less credible right-wing tendencies, dominated online conversations about voter integrity. This was also the time when the polarization score of voter integrity was at its highest level, reaching 95. According to the report, the “events of January 6 in the Capitol” were caused by a high level of dissent over the split over voter integrity.
Notably, the most reliable right-leaning sources accounted for only .017% of the total engagement on the subject of voter integrity, while the most reliable left-leaning sources accounted for about 36%.
According to a Pew research study in late November 2020, 79% of Trump voters said the 2020 presidential election did not go well, compared to 6% of Biden voters.
Another election year is approaching, and conversations about the health of American democracy are coming to the fore again, putting renewed pressure on social media.
Some reasons for optimization, however, can be found in the Atlantic. The European Union is considering two major bills in the first half of 2022, known as the Digital Services Act and the Digital Markets Act, led by the French government. The bill seeks to curb hate speech and the underlying advertising model, which is generally considered to be one of the most fundamental challenges in preventing the spread of misinformation.