Stop Blaming Social Media for Everything: The Minimal Effects of Facebook in the 2024 U.S. Presidential Election

Social media are ubiquitous, with billions across the world using these platforms to connect with others and learn about the world around them. Many accuse these platforms of spreading misinformation, polarizing polities, undermining democracy, and making their users miserable. We conducted a randomized field experiment during the 2024 United States presidential election (N=4,971 assigned, N=4,278 observed) that incentivized half of participants to deactivate their Facebook accounts for the two weeks before the election. The factorial experiment design also encouraged a randomly selected half of the participants to install an application that provided them with daily reminders to engage with the ideologically balanced news that it provided them. In line with previous Facebook deactivation studies, we find little evidence for concerns that it has dramatic and drastic damaging effects on society. Moreover, encouragements to read ideologically balanced news also had mostly null effects. We conclude with thoughts about why research on social media should move past an often implicit theoretical model that treats ordinary people as passive users and treat them as active agents.