Social media, as a subaltern public sphere (Fraser, 1990), have a democratic function in providing an alternative platform for minorities and marginalized to defy mainstream discourses in the public sphere. However, social media have been found to have an echo chamber effect, which may be detrimental to democracy. They may help to accelerate the ascendancy of a “post-truth” era in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief. A study on political polarization, however, showed that selective exposure and avoidance in social media are weak indicators of polarization (Johnson et al., 2017). This study examines the role of social media in democracy and partisan politics. The authors considered that despite the echo chamber effect, social media have a limited part to play in the formation of polarized stances compared with other factors, such as demographics, political orientation, and mass media use. The study tested two main hypotheses: H1: Social media use is associated with political stance that is marginalized in the mainstream media; H2: Political orientation has a stronger relationship than social media use with the stance toward political values and social issues. The results supported both hypotheses. Social media are associated with political stance that is marginalized in the mainstream media. However, when compared with other factors, the relationship between social media and stance becomes less obvious. Although the echo chamber effect may reinforce the original stance, social media do not exhibit a strong relationship with the stance toward political values and social issues. Partisan orientation and use of partisan mass media are found to have stronger links with variations in stance. Social media, however, provide a subaltern public sphere for those excluded from the dominant public sphere, thus extending the public sphere to accommodate multiple opinions and perspectives.
Echo chamber effects of social media
Many previous studies found that the segregation and polarization of online media users in consuming one-sided content matched their positions or beliefs (Adamic and Glance, 2005; Bimber and Davis, 2003; Kushin and Kitchener, 2009; Mutz and Martin, 2001; Stroud, 2010). People tended to choose media content according to their interests and likes. The theories of “selective exposure” (Sear and Freedman, 1971) and “cognitive dissonance” (Festinger, 1957) have long considered this phenomenon. However, the emergence of social media has exacerbated the anxiety about the polarization of opinions and the balkanization of politics, both of which are considered harmful to democratic development. A recent study that investigated Facebook and YouTube, for example, found that 94% of Facebook users and 88% of YouTube users were polarized, that is, they concentrated their online reading and interactions (at least 95%) on a specific narrative related to a given controversial topic, such as climate change. The study further found that although the majority of the group was initially open to both sides of content, the participants eventually moved to consuming only one type of information, thus becoming polarized toward one narrative (Uzzi, 2017). Another study on Twitter use by Demos, a cross-party think tank in the United Kingdom, found that the “echo chamber effect” was the strongest among those furthest from the political mainstream. Supporters of the Scottish National Party and the UK Independence Party were much less likely to engage with people who held different beliefs or to retweet material from outlets with opposing editorial stances (Cheshire, 2017). A panel study in South Korea found that participants who actively used social media were more likely to engage in political processes, which led them to develop political attitudes that were more extreme than the attitudes held by those who did not use social media (Lee, Shin, and Hong, 2018). Social media may help to accelerate the ascendancy of a “post-truth” era in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief. The echo chamber effect can reinforce existing views and inclinations embraced by subaltern publics. They may become overly critical and extreme in their discourses and actions as they refuse to expose themselves to opposing or alternative views. The polarization and extremity of public discourses were demonstrated in the Trump–Clinton Presidential Election in 2016. A study showed that unfounded stories about Clinton, such as “Clinton set up Satanic Network” and “Clinton had Parkinson’s disease,” received much attention in social media by Trump supporters (Sillito, 2016). Supporters on both sides in the election chose to see and believe what they wanted to see and believe. However, some studies have challenged the idea that social media has negative effects on public life and democracy. In a metaanalysis of 36 current studies on social media use and participation, Boulianne (2015) found that the metadata suggested a positive relationship between social media use and participation in civic and political life. More than 80% of the coefficients were positive although only half of them were statistically significant. The study also found that social media did not affect the likelihood of voting or participation in an election campaign. The metadata provided little evidence to support the claim that social media were successful in changing the levels of participation in election campaigns. In other words, these findings indicate that social media have little influence on voters’ decisions although its use can increase participation in civic and political life.