Does Facebook drive polarisation?

Would logging off of social media reduce the political polarisation that has increasingly become part of modern political life? New evidence suggests the answer is no. A study co-authored by Drew Dimmery, Professor of Data Science for the Common Good at the Hertie School’s Data Science Lab, shows that deactivating Facebook and Instagram accounts before the 2020 US presidential election had little effect on political attitudes or polarisation. Logging off did, however, reduce the attention participants paid to political news (both on- and offline). The study, a collaborative project between academics and researchers at Meta, randomised whether 35,000 Meta users deactivated their accounts for the six weeks prior to the 2020 election.

The article, titled “The effects of Facebook and Instagram on the 2020 election: A deactivation experiment”, was published on 13 May in the journal Proceedings of the National Academy of Sciences. 

Study reveals no significant change in polarisation

The Meta study was conducted to address widespread concerns that social media platforms had driven political polarisation leading up to the 2016 US presidential election, and that this would increase in 2020. The researchers evaluated two types of polarisation: affective polarisation, which measured participants’ favourability of party supporters, candidates, and the perceived ‘smartness’ of a party and its supporters, and issue polarisation, which measured political opinions on eight different topics, including immigration, gender equality and policing. Their results showed that deactivating Meta accounts before the 2020 election had no statistically significant effect on affective polarisation, and only minimal effects on issue polarisation. Similarly, they found that deactivation had no significant impact on the perceived legitimacy of the election and candidate favourability. 

Effects on political knowledge and participation 

When it comes to news knowledge, however, the picture becomes blurrier. On the one hand, deactivating Facebook slightly reduced overall knowledge of the news; on the other, it also decreased belief in misinformation. “We found that participants not using Facebook consumed less news and were worse at answering questions about the news,” comments Dimmery. “But they were also less likely to believe some widely circulated misinformation.”

Besides political knowledge, the researchers found that deactivating Facebook and Instagram reduced some forms of political participation, especially online activities. “Social media provides the most important online forum for political discussions, so it could directly increase the likelihood of online participation,” says the study. Although deactivation decreased online political participation, it did not have a significant effect on voter turnout. 

Relevance for future elections

Looking to future elections, Dimmery says: “To bring large online platforms under democratic control, it’s critical to conduct this kind of rigorous impact assessment. What would these results have looked like if users had deactivated social media ahead of the 2016 campaign? Or if users deactivated TikTok for the 2024 election campaign? Without running these kinds of tests, we may never know.”

About the study

The research was part of the US 2020 Facebook and Instagram Election Study, a collaboration between academics and researchers at Meta that allowed unprecedented access to Meta platform data and algorithms while also including extensive safeguards to guarantee the integrity of the research. The study involved 19,857 Facebook and 15,585 Instagram users who agreed to stop using the platforms ahead of the 2020 election. About a quarter of these users deactivated their accounts for six weeks before the November vote; the remaining users were a control group that logged off for just one week.

The study provides the largest-scale evidence available to date on the effect of Facebook and Instagram access on political knowledge, attitudes and behaviour in a presidential election season.

The article “The effects of Facebook and Instagram on the 2020 election: A deactivation experiment” was published in Proceedings of the National Academy of Sciences on 13 May and was authored by Hunt Allcott, Matthew Gentzkow, Winter Mason, Arjun Wilkins, Pablo Barberá, Taylor Brown, Juan Carlos Cisneros, Adriana Crespo-Tenorio, Drew Dimmery, Deen Freelon, Sandra González Bailón, Andrew M. Guess, Young Mie Kim, David Lazer, Neil Malhotra, Devra Mohler, Sameer Nair-Desai, Houda Nait El Barj, Brendan Nyhan, Ana Carolina Paixao de Queiroz, Jennifer Pank, Jaime Settle, Emily Thorson, Rebekah Tromble, Carlos Velasco Rivera, Benjamin Wittenbrink, Magdalena Wojcieszak, Saam Zahedian, Annie Franco, Chad Kiewiet de Jonge, Natalie Jomini Stroud, and Joshua A. Tucker.

The Hertie School is not responsible for any content linked or referred to from these pages. Views expressed by the author/interviewee may not necessarily reflect the views and values of the Hertie School.