Posts, Likes, and Ballots: Social Media’s Effects on Political Engagement, Polarization, and Institutional Trust
By Samantha Buonato
As of October 2025, 5.66 billion people, or 68.7 percent of the global population, were active social media users. The platforms they use have become far more than spaces for personal connection, but now shape how citizens consume news, form political opinions, and participate in civic life. As social media’s influence has grown, so too has scrutiny of its effects on democracy. Does it empower voters or manipulate them? Does it build community or breed division? This article will examine three interrelated dimensions of social media’s influence: polarization, trust in institutions, and political engagement. Understanding polarization first is essential because it sets the conditions through which trust is eroded and engagement is shaped.
Polarization
A core concern about social media is not simply what people see, but what kinds of ideas the platforms are designed to amplify. According to research from Brookings, divisive content tends to spread more widely and quickly on social media because posts expressing moral outrage or attacking opposing groups tend to be particularly effective at going viral. This phenomenon is not accidental but part of the platform design. Algorithms reward engagement, and outrage drives engagement. The result is a feedback loop in which politicians, media outlets, and influencers are economically incentivized to post inflammatory content.
For politicians specifically, this not only shapes how they communicate but what they choose to stand for. A candidate who posts a measured policy proposal and one who posts an attack on a political enemy are not competing on equal footing, as the attack travels further and faster throughout the public. Over time, this creates an underlying incentive to select a certain kind of politician: one who is skilled at performing conflict rather than resolving it. Understanding this pattern matters because it means political polarization may be driven not only by genuine disagreement among voters but by a media environment that systematically rewards the most combative voices and pushes them to the top.
Brookings has also documented that Facebook is fully aware of its role in promoting divisiveness and conducts extensive internal research on the problem. Yet, the company typically adjusts its algorithms only for limited time windows, such as after the November 2020 election or before the April 2021 Derek Chauvin verdict, because permanent changes would cut into user engagement. This pattern was made clear by the 2021 Facebook Papers, leaked by whistleblower Frances Haugen, which showed that company researchers had repeatedly flagged how the platform's recommendation systems were pushing users toward increasingly extreme content. However, Facebook’s leadership had largely ignored those findings.
Other research, however, complicates a simple story of social media as the cause of polarization: a 2017 study found that from 1996 to 2016, polarization rose most sharply among Americans aged 65 and older, the demographic least likely to use social media. One explanation is that this generation came of age during the rise of partisan cable news: Fox News launched in 1996, the same starting point of the study. These findings suggest that television may be a more powerful force than social media for older Americans. This assertion is also reflected in voting patterns: older Americans have shifted Republican over the past two decades, a trend that predates the dominance of social media and aligns more closely with the expansion of right-leaning cable news.
Although this connection may be challenged by the idea that, as people get older, they generally get more conservative, the media environment remains a more compelling answer than just age. For example, Pew Data shows that nearly half of Americans over 65 regularly get news from Fox, compared to just 28 percent of adults under 30, and 76 percent of older Republicans say they trust the network. Additionally, in a Yale Study, when committed Fox viewers were paid to watch CNN for four weeks, their views on multiple issues shifted meaningfully, suggesting that the rightward shift among older Americans is less a product of aging than of a media system that moves their audience right.
The gendered and partisan dimensions of youth media consumption are also striking. A CIRCLE analysis found that 37 percent of young Republican men encountered political information on YouTube compared to 20 percent of young Democratic women, while 38 percent of young Democratic women relied on TikTok versus 19 percent of young Republican men. These divergent phenomena may partly explain the notable 15-point rightward shift among young men in the 2024 election, as their specific digital ecosystem has been dominated by right-leaning content.
The polarization produced and amplified by these platforms does not stay confined to feelings and opinions. It creates the ground in which trust erodes and misinformation flourishes.
Trust in Institutions
Social media has become a vehicle for data exploitation and disinformation that erodes democratic trust. The 2018 Cambridge Analytica scandal revealed that data from up to 87 million Facebook profiles had been harvested without consent and used to target voters in elections across more than 200 countries. The scandal exposed to the public what had been building for years: platforms enable political actors to construct detailed profiles of voters by drawing on their browsing habits, purchase histories, and location data and then tailoring highly specific messages accordingly, often without voters’ knowledge.
The stakes are especially high for younger generations, who both rely more heavily on social media for news and are more susceptible to the misinformation circulating on it. Research shows that members of Gen Z report higher rates of victimization from identity theft, romance scams, and cyberbullying than Baby Boomers, and teens are more likely than adults to believe online conspiracy claims. At the same time, young people hold social media companies in remarkably low esteem, with only 19 percent of young people in a CIRCLE poll reporting trusting social media companies, ranking them last among all institutions surveyed.
This distrust of institutions, sparked by a polarized environment and deepened by scandal and manipulation, creates a backdrop against which political engagement must be understood.
Political Engagement
Given the polarized, low-trust media environment described above, it may seem surprising that social media’s direct effects on democratic participation are relatively low. A 2020 Facebook and Instagram Election Study, conducted in a collaboration between researchers at Meta and independent academics, recruited over 35,000 American Facebook and Instagram users and randomly assigned and paid participants to deactivate their accounts for six weeks leading up to the 2020 U.S. election. The study found that deactivation had almost no effect on polarization, voter turnout, or trust in elections, suggesting the forces driving these phenomena stem deeper than social media usage. Two outliers, however, were vote choice and informativeness. Facebook access seemed to shift preferences towards Trump as well as make people better informed about the news. At the same time, they also increased belief in misinformation as both reliable and false claims circulate on the platform. Overall, the findings suggest that Facebook and Instagram shape the surface of political life more than its foundation, influencing how informed or misinformed people are, but falling short of being the primary driver of democratic dysfunction.
Still, social media clearly shapes where people get their information. According to Pew Research, about one in five U.S. adults now gets their news from social media influencers, and rates are even higher among young people. In a CIRCLE poll, 77% of young people listed at least one social media platform among their top three news sources. Yet, despite this heavy reliance, young people did not rank influencers among their top motivations to vote in 2024. In fact, being persuaded by influencers or celebrities ranked last on their list. Instead, “[to] support a candidate who will do something about the issues I care about” ranked the highest according to the CIRCLE poll. Thus, the platforms drive exposure but not necessarily action.
Conclusion
Taken together, these three dimensions reveal a troubling pattern. Social media platforms are not passive conduits for political information but are systems that shape what ideas spread and what kind of civic culture emerges. Algorithms that reward outrage produce a polarized environment. That polarized environment creates the conditions for data exploitation and the erosion of institutional trust. Yet, despite the noise and manipulation, the platforms fail to translate exposure into meaningful democratic engagement. People are angrier and more informed yet less trusting and not more likely to vote.
Thus, social media is not destroying democracy single-handedly, but it is not neutral either. The evidence points to a system that rewards division, normalizes manipulation, and leaves users, especially young ones, navigating a difficult landscape where lies travel faster than facts. Reforming this system will require pressure on multiple fronts: regulatory action to increase algorithmic transparency and accountability, investment in digital and media literacy for young people, and a cultural effort to rebuild the in-person civic ties that platforms cannot replicate. The billions of people who rely on social media for their understanding of the world deserve platforms that are accountable to democracy, not indifferent to it.
Bibliography
Barrett, Paul, Justin Hendrix, and Grant Sims. "How Tech Platforms Fuel U.S. Political Polarization and What Government Can Do About It." Brookings Institution, September 27, 2021. https://www.brookings.edu/articles/how-tech-platforms-fuel-u-s-political-polarization-and-what-government-can-do-about-it/.
CIRCLE (Center for Information and Research on Civic Learning and Engagement). "25 Things We Learned About Young Voters: 2025." Tufts University, 2025. https://circle.tufts.edu/latest-research/25-things-we-learned-about-young-voters-2025.
CIRCLE (Center for Information and Research on Civic Learning and Engagement). "Youth Need Trust and Connections to Navigate the Digital Media Landscape." Tufts University, 2024. https://circle.tufts.edu/latest-research/youth-need-trust-and-connections-navigate-digital-media-landscape.
Allcott, Hunt, Matthew Gentzkow, Winter Mason, Arjun Wilkins, Pablo Barberá, Taylor Brown, Juan Carlos Cisneros, et al. "The Effects of Facebook and Instagram on the 2020 Election: A Deactivation Experiment." Proceedings of the National Academy of Sciences 121, no. 21 (2024): e2321584121. https://doi.org/10.1073/pnas.2321584121.
Lai, Samantha. "Data Misuse and Disinformation: Technology and the 2022 Elections." Brookings Institution, June 21, 2022. https://www.brookings.edu/articles/data-misuse-and-disinformation-technology-and-the-2022-elections/.
Overgaard, Christian Staal Bruun, and Samuel Woolley. "How Social Media Platforms Can Reduce Polarization." Brookings Institution, June 24, 2023. https://www.brookings.edu/articles/how-social-media-platforms-can-reduce-polarization/.
Statista. "Number of Social Media Users Worldwide from 2017 to 2027." Accessed March 2026. https://www.statista.com/statistics/617136/digital-population-worldwide/.