Abstract
Teen users outpace adults in social media use across several platforms. Though much scholarship has considered the negative effects of social media use on teen well-being, this study considers how participation on Twitter, Instagram, Facebook, and TikTok is influencing teens’ political interests and behaviors. Compared to traditional resources, we find that social media use across these platforms positively correlated with political interest and civic online and offline engagement, while Twitter and Facebook use had positive relationships with affective polarization. TikTok and Instagram each correlated with higher levels of interest and civic engagement, and the platforms had no relationship with polarization. We discuss these implications and what they mean for political participation among teens online.
In 2023, an Ohio train derailment endangered the local community of East Palestine with hazardous materials, but it went nearly unnoticed by mainstream news media. However, it did not take long before the incident became one of the leading items on the U.S. public agenda. This is because TikTok users initially provided information about the crash when mass media organizations did not (Thalen, 2023). Social media has become one of the primary ways that young adults and teens learn about current events at local, national, and international levels. Between firsthand accounts of Ukrainian residents sharing the magnitude of Russia’s invasion in their communities to Philadelphia residents providing helpful guidelines about the city’s water shortage (Griffith, 2023; Sato, 2022), social media provides a quick way to gain information and learn about political events before journalists and political leaders can address them fully. Today, young users are increasingly using social media as a search engine to learn about the latest political events (Huang, 2022). It also allows citizens to bring issues to light when they receive inadequate attention from those in power.
Teens gravitate toward online spaces where older adults are not present (boyd, 2014). Consequently, teens’ social media usage shifted from an early preference for Myspace to Facebook to early adopters of newer image and video-based visual platforms like Instagram and TikTok (Pew Research Center, 2011, 2022). Parents and scholars alike rightfully concerned themselves with the effects of teens spending hours each day on these platforms. Using these sites often leaves teens vulnerable to potential threats. Previous research considered the role social media has on physical and mental well-being with varying alarming results (Boer et al., 2021; Coyne et al., 2020; Fardouly & Vartanian, 2016; Pantic, 2014) and concerns over the misinformation that spreads throughout these sites are equally important (Allcott et al., 2019; Wang et al., 2019).
But past scholarship also considers the potential benefits of social media for teens online (Bode et al., 2014; boyd, 2014; Kaskazi & Kitzie, 2021; Seongyi & Woo-Young, 2011). The relationship between social media and public opinion is especially important to consider for teens who are often barred or restricted from participating in politics in more traditional ways (Kaskazi & Kitzie, 2021). Much of this research considers the role sites like Facebook have in promoting political participation and cultivating political interest in teen users (Bode et al., 2014; Gil de Zúñiga et al., 2012; Heiss et al., 2020; Seongyi & Woo-Young, 2011). This study expands the research of social media and teen political participation by considering how increased social media use, rather than traditional resources, along several different platforms, Facebook, Twitter, Instagram, and TikTok, influences political interest, and online and offline civic engagement. We also consider the influence of social media on affective polarization, or the tendency of Democratic and Republican partisans to “increasingly dislike or distrust” members of their opposing political party (Iyengar et al., 2019, p. 130). Polarization is often regarded as a major threat to US democracy (Draughton, 2022; Kingzette et al., 2021), but as politics become more intertwined with our social identities (Mason, 2018; McCoy & Somer, 2019), a process that is likely compounded through the personalization of the internet (Liu, 2012), polarization among teen citizens could result from the early formation of political identity.
Through a survey of U.S. teens, we found significant relationships between social media use and civic engagement. Our results show that teen social media use has a positive relationship with civic engagement, especially when compared to traditional resources. We also find that Twitter and Facebook use had positive relationships with affective polarization among teens. We discuss the findings below and how these results show somewhat promising evidence of the role social media can play in fostering political interest and behaviors among the next generation of political participants.
Literature Review
Teen Political Participation on Social Media
Democratic systems rely on an informed, participating electorate (Delli Carpini & Keeter, 1996), but for teens, opportunities for participation are limited. In the United States, the legal voting age begins at 18 years, but political participation can also involve civic activities that have no age limits such as volunteering for social organizations, discussing politics with friends and family, or participating in political demonstrations. Much of this participation online involves reading about politics, joining digital political groups, or engaging with political campaigns and candidates. Recent studies show that young adults have become increasingly interested in politics and social issues in the last decade (Harvard Kennedy School Institute of Politics, 2021). This interest in 18- to 29-year-olds may be cultivated in their formative teen years. Increases in political interest may be in part credited to the internet’s increased accessibility to political information as well as social media’s influence on individual identity and expression.
Nowadays, teens have new means of engaging in politics, and much of that participation is taking place online. Many scholars considered the potential democratization of the internet and its influence on political engagement and discussion (Bentivegna, 2002; Dahlberg, 2001; Dutton, 2009; Hayles, 1999; Turkle, 1995). People are now more connected than ever, increasing the likelihood of exposure to others with different opinions or backgrounds (James et al., 2016). Stolle (1998) found that engagement in more diverse groups with weaker ties leads to more cooperation between individuals. However, offline inequalities persist online. The participatory theory of mass communication examines people’s varying levels of engagement, interaction, and access online and how these inequalities can limit the participation of disadvantaged individuals (Carpentier, 2016). By looking at the participation of users rather than the reactions of audiences, researchers are recognizing the shift in power from organizations to users. For example, researchers have found that the internet plays a key role in allowing marginalized groups, including teens, to produce their own content and share it online for others to see (Carpentier, 2011; Lee et al., 2013). Rather than simply consuming information, users now have some power to create and share content that would not typically be reported on by the mass media (Chaffee & Metzger, 2001).
Social Media as a Resource
Social media provides inexpensive ways to mobilize, network, or discuss politics (Carpentier, 2011). Who can participate in politics has been a prominent question in media studies. For example, Brady et al. (1995) foundational study looked at whether people with more traditional resources, such as money, time, and civic skills, would have more interest and greater access and ability to participate in politics. However, access to the internet may narrow the gap in participation between online users (Krueger, 2002). Past work finds that among internet users, traditional resources rarely predict political participation, while internet usage is associated with higher levels of political participation both on and offline (Anduiza et al., 2010; Best & Krueger, 2005; Quintelier & Vissers, 2008; Wang et al., 2018).
Sites like Facebook and Twitter are important platforms for promoting democratic discourse and participation (Bode et al., 2014). Social media provides information that is relevant to teens online (Kaskazi & Kitzie, 2021), and as such, teens and young adults are more likely than older generations to use social media as their primary source of news (Lee et al., 2013). Teens often prefer social media to traditional news outlets for news, because oftentimes, the information that is presented to teens on their social newsfeeds is more relevant to their interests and identities (Kaskazi & Kitzie, 2021). Similarly, social media has provided more access for teens, especially to participate in democratic discourse and voice their opinions on current affairs (Holt et al., 2013). Holt and colleagues (2013) found that younger people are more likely than older adults to use blogs to communicate about political issues, and teens often use sites like Instagram, Twitter, and WhatsApp to talk about politics with friends (Kaskazi & Kitzie, 2021).
Therefore, we predict teens’ social media use will influence their political interest and civic engagement more than traditional resources.
Participatory culture online can lead to more civic engagement by users (Clark & Marchi, 2017; Middaugh et al., 2017). Previous research supports the idea that online participation on social media sites will increase both political interest and participation offline among young adults and teens (Holt et al., 2013; Kahne & Bowyer, 2018; Lee et al., 2013; Seo et al., 2014), but most studies have yet to differentiate the relationships between each platform and its respective correlation with civic engagement both on and offline. Exceptions to this include Halpern et al. (2017), Valenzuela and colleagues’ (2018), and Theocharis and colleagues’ (2022) studies of the distinctions between Twitter and Facebook use and their influence on political participation, as well as Effing et al. (2011), look at Facebook, Twitter, and Youtube’s influence on voting, specifically.
Still, little is known about how the image-sharing platform Instagram or the popular video-sharing platform TikTok influence political participation offline. Teens tend to use these platforms at higher levels than they use other platforms like Facebook and Twitter (Pew Research Center, 2022). Social media platforms each offer unique affordances and, as such, have each had different effects on political participation. To expand the scope of the current understanding of the relationships between teen social media use and politics, our study looks at whether unique platforms are related to higher levels of political interest as well as online and offline civic engagement.
Political Identities and Affective Polarization
For teens, these early adolescent years mark a time in their life when they are highly impressionable to their environments and the opinions of others. Before social media, scholars analyzed the influence of parents’ partisan identities on teens’ political expression, finding that nearly always, children tend to adopt the same positions as their parents before achieving more independence in their early adult years (Hyman, 1959; Jennings & Niemi, 1968; Jennings et al., 2009). But polarization can also spark more debate and participation by citizens (Arora et al., 2022). Simon (1985) argued there are benefits to belonging to a political party. Oftentimes, people’s political expression is mediated by their fear of social isolation (Laird & White, 2020; Van Duyn, 2021). Belonging to a political party can increase levels of political expression because people feel safer talking to like-minded individuals. For teens who are in the early stages of developing their political identities, belonging to a group may result in higher levels of civic engagement because of the confidence and social support it lends them in these spaces online. As Törnberg and colleagues (2021) argue, the internet can influence polarization through the psychological formation of one’s social identity.
As people speak about politics, their confidence in their own ability increases (Van Duyn, 2021), leading to the formation of a stronger political ideology and stronger stances on issue positions. Polarization is a natural result of forming a partisan identity (Iyengar & Westwood, 2015). The internet has often been accused of increasing political polarization among highly active users online (Demszky et al., 2019; Iandoli et al., 2021; Kubin & von Sikorski, 2021; Yarchi et al., 2021). As some scholars argue, the population is becoming more politically polarized, and our social identities are becoming more closely tied to our partisan beliefs (Bruns, 2019; Mason, 2018). For many Americans, political identities overlap with their racial, gender, and socioeconomic identities (Mason, 2018). The terms “echo chambers” and “filter bubbles” were coined to describe the process in which online users separate themselves into partisan camps as a result of these social characteristics and selective exposure (Pariser, 2011), but other research challenges this perspective with evidence that social media instead forces people to see a wide range of opinions and views as a result of the open access to participation that these platforms provide to the public (Bruns, 2019; Dubois & Blank, 2018).
Political interest leads to higher levels of polarization (Strömbäck et al., 2013). As Mummolo (2016) found, people are highly motivated to find and read about topics that interest them, leading them to select content that aligns with their interests and social identities as well. People who are more interested in politics may engage in more political discussion (Prior, 2007), resulting in more exposure to disagreement and potential hostility (Sydnor, 2019). But as Dubois and Blank (2018) find, political interest can also lead people to be exposed to more diverse points of view and a wider range of content as users seek out varying information.
Affective polarization, the tendency of partisans to disapprove and even loathe one another (Druckman & Levendusky, 2019), is often represented by animosity and social distancing between citizens (Iyengar et al., 2012). While some studies find social media may reduce affective polarization by increasing exposure to a more diverse array of political views online (James et al., 2016; Jones-Jang & Chung, 2022), online users also have more power to make like-minded connections and selectively consume content that supports their own views. At its worst, selective exposure reinforces tendencies to consume partisan news, limits cross-cutting democratic discourse, and amplifies extreme ideological positions online (Bail et al., 2018; Lorenzano et al., 2018; Qureshi et al., 2020; Wang & Song, 2020).
For teens, who are just forming an interest in political expression, social media may accelerate ill feelings toward out-party members. Some platforms, like Facebook or Instagram, rely on a network of strong ties where people “friend” their friends, family, and coworkers. Settle’s (2018) study of Facebook looks at how the newsfeed intersects people’s personal and partisan identities, resulting in a more psychologically polarized public. Still, people are more likely to have Facebook friends that belong to their own community and offline networks, resulting in more homogeneity among users (Barberá, 2014; Yarchi et al., 2021). Users can also join groups and pages that appeal to their interests and offer the organization of niche communities online (Conroy et al., 2012). As such, Facebook’s platform leads to more extreme expressions of ideological beliefs but less interaction between partisans, reducing the hostility on the platform (Yarchi et al., 2021).
Other sites, like Twitter and TikTok, are shaped by the formation of weak-tie networks, where people may follow strangers, content creators, political leaders, pundits, and more. As a result, people are exposed to a much wider range of perspectives and sources of information on these sites. But this exposure to different perspectives could also lead to more animosity and conflict, resulting in negative feelings toward out-party members (Yarchi et al., 2021). Therefore, we examine the different platforms and their relationship with affective polarization with the following hypotheses:
Data Collection and Key Variables
To test our hypotheses, we fielded an online survey for one week from May 23 to May 30, 2022, through Lucid, a survey sampling service, to a sample of 1,315 U.S. teen residents between the ages of 13 and 17 years old (76.22% girls, 23.78% boys; M = 1.76, SD = .426). 1 The teen participants and their closest guardians each gave consent for the teens’ participation in the study. Of the sample, 44.64% of the teens identified as White or Caucasian, 16.96% as Black or African American, 7.83% as Asian American or Pacific Islander, 19.70% as Hispanic, Latinx, or of Spanish origin, and 10.88% indicated other ethnicities.
We asked the participants to identify which political party they and their guardians consider themselves as belonging to. Approximately 62% of the teens (N = 808) had some political preference; 23.89% of these teens identified as belonging to the Republican Party, 43.44% belonged to the Democratic party, and 32.67% identified as Independent. Participants received compensation for completing the study. The survey measured the teens’ social media use and asked the participants to provide self-reported measures regarding their political interests and civic engagement. In the OLS models, we controlled for gender, age, race/ethnicity, and partisan identification, and we weighted gender due to the large proportion of teen girls represented in our sample.
Measures
Social Media Use
First, we asked our teen participants if they use the following online platforms: Facebook, Instagram, BeReal, Tumblr, Twitter, Snapchat, TikTok, Twitch, and Truth Social, Parler, Gab, or Gettr. The five platforms most used by teens included: Instagram (85.29%), TikTok (83.03%), Snapchat (78.43%), Twitter (55.05%), and Facebook (45.40%). This is unsurprising, considering these platforms tout the highest membership numbers among teens, according to the Pew Research Center (2022).
To measure how active teens are on social media, we asked teens to report the frequency of different types of engagement on Instagram, TikTok, Twitter, and Facebook. For each of the four platforms, we asked participants how frequently on a scale of “Constantly” to “Never” they perform nine different types of social actions. Some actions include “Post[ing] updates related to [their] personal life, activities, and interests,” “Shar[ing] photos,” and “Reshar[ing] materials that someone else posted.” For the full list of each of the platform-specific scales, see Online Appendix B.
Traditional Resources
As we were surveying teens, we chose not to use traditional forms of measuring resources, such as education, income level, or occupation (Brady et al., 1995). Instead, we used the six-level family affluence scale designed specifically to ask teens about the material items they have at home: number of cars (2 or more), number of bathrooms (0, 1, 2, 3, or more), number of computers (0, 1, 2, > 2), unshared bedroom (no/yes), dishwasher (no/yes), and the number of days abroad during the last 12 months (0, 1, 2, 3, or more). To create a new variable indicating the teens’ level of resources, these responses are added together on a scale of 0–13. Correll and colleagues (1995) found that the scale correlated with surveyed adolescents’ household incomes.
We used the STATA package developed by Newson (2012) to calculate the ridit scores for affluence among our teen sample (N = 1,304). We used the Family Affluence Scale as a continuous variable (0–13) and as a categorical variable divided into low, medium, and high affluence. In our sample, the following cutoffs were used for the teens: low affluence (0–5); medium affluence (6–10); and high affluence (11–13). We used the ridit scores to break the teens into three groups: 15.18% of teens in the lowest (low affluence), 57.29% of teens in the middle (medium affluence), and 27.53% of teens in the highest (high affluence).
Political Interest
To measure how interested the teen participants were in politics, we used a four-point Likert-type ranked from “Very interested” to “Not at all interested” in political affairs.
Civic Engagement
To measure civic engagement, the teen participants reported how many times in the past 12 months of taking the survey they had done 11 unique forms of civic engagement, on a scale of “10 or more times” and “None.” Six of the 11 items asked teens about their online civic engagement (Cronbach’s α = .86). These items include digital forms of engagement like “Join[ing] a political or cause-related group on a social media site” and “Email[ing] or direct messag[ing] a politician online.” The other five items measured how often they engaged in offline forms of civic engagement (Cronbach’s α = .82), including “Volunteer[ing] to help with a political cause” or “Attend[ing] a protest or march in person.” For a full list of our civic engagement items, see Online Appendix B.
Affective Polarization
Affective polarization is just one dimension of political polarization. Rather than looking at issue position or the heterogeneity of groups online, affective polarization considers the attitudes of partisan individuals toward out-party members. In this study, we are specifically interested in the attitudes toward fellow citizens rather than political elites. Some research uses feeling thermometers to test an individual’s “warm or cold” feelings toward the Republican party, Democratic party, or political leaders (Iyengar et al., 2012). Instead, we are interested in teens’ social distancing attitudes. Affective polarization may inhibit democratic discussion and bipartisan compromise by pitting members of the Republican and Democratic parties against each other, especially when people equate a person’s partisanship with their social identity (Iyengar et al., 2019). This hostile view toward out-party members can lead to less legislative action and increasing turmoil at the citizen level. We seek to determine whether these hostile feelings toward out-party members are cultivated as early as the teen years.
We measured affective polarization by asking teens how much they “Strongly agree” to “Strongly disagree” with the statement, “I think it’s okay to be friends with people with different political beliefs.” This single-item approach suited the nature of our teen sample. As Druckman and Levendusky (2019) found, when survey respondents are asked to evaluate their feelings toward political parties, most individuals think of political elites rather than voters. Therefore, we approach affective polarization through the mechanism of social distancing, similar to measures asking how okay a respondent may be with their son/daughter marrying someone of the opposite party (Iyengar et al., 2012; Levendusky & Malhotra, 2016). Iyengar and colleagues (2019) call these social distancing measures “less obtrusive” and argue, “If partisanship is an important social identity in its own right, partisans should be averse to entering into close interpersonal relations with opponents” (p. 132). Furthermore, we did not want to compound this measure with other facets of polarization, like issue positions or feelings toward elites, since what we are truly testing is the willingness to interact and form relationships with out-party partisans.
Results
To begin our analysis, we examined the demographic differences in social media use. Using a series of t-tests, we find that gender does not significantly predict social media use on any of the four platforms, Twitter, Instagram, Facebook, and TikTok. This is somewhat surprising since our descriptive statistics in Table 1 depict a higher share of girls using platforms Instagram and TikTok and a higher share of boys using Twitter and Facebook (by a slight amount). The partisan identification of the teen users did have a significant influence on social media use, with Republican teens using Instagram, t(460) = 2.049, p = .041, Facebook, t(241) = 2.036, p = .043, and TikTok, t(441) = 2.096, p = .037, more than Democratic teens.
The Percentage of Teen Participants on Each Platform.
N = 1,315.
Next, we looked at race/ethnicity and its relationship to social media use. Consistent with prior research (Ahn, 2011; Madden et al., 2013; Pew Research Center, 2015; Tynes & Mitchell, 2014), we find that teens of color were more likely to use all four platforms compared to white teens. The biggest difference was on Twitter. Teens of color were significantly more likely to use Twitter, t(711) = 4.809, p < .000. Past research finds that Twitter provides a way for users of color, especially Black Twitter users, to form counterpublics, speak to shared experiences, and raise awareness around political and social issues that predominantly affect the Black community (Hill, 2018; Peterson-Salahuddin, 2022). Looking at Black teens specifically, this relationship persists, with Black teens more likely than other teens to post content and engage with others on Twitter, t(711) =-3.345, p = .001. We also tested this relationship with Hispanic teens but found that there was no significant difference in which Hispanic teens used any of the platforms. For the full t-test results, see Table C1 in Online Appendix C.
Traditional Resources
Our first hypothesis,
Teen Scores on the Family Affluence Scale.
N = 1,304.
We controlled for gender, age, partisanship (if noted), and race/ethnicity. We found that there was no significant relationship between traditional resources and political interest (β = .002, p = .234) or offline civic engagement (β = −.002, p = .183). Traditional resources had a negative relationship with online civic engagement (β = −.003, p = .020). See Table D1 in Online Appendix D for the full models. Though we would expect traditional resources to influence teens’ computer and smartphone use, technological access is becoming increasingly widespread and offers a level playing field for teens with low and high levels of resources. This supports prior research that social media levels access to information for teens who may be socioeconomically disadvantaged (Micheli, 2016).
Next, we found that Instagram (β = .109, p < .000), TikTok (β = .109, p < .000), Facebook (β = .069, p = .042), and Twitter (β = .060, p = .044) each have a positive significant relationship with political interest, with Instagram predicting the highest rates, TikTok the second highest, and so forth (see Table E1 in Online Appendix E). Next, we looked at whether social media use significantly predicted civic engagement. Twitter (β = .275, p < .000), Instagram (β = .273, p < .000), Facebook (β = .257, p < .000), and TikTok (β = .254, p < .000) predicted significant levels of online civic engagement among teens (see Table E2 in Online Appendix E; Figure 1).
Twitter, Instagram, TikTok, and Facebook use and online civic engagement.
In addition, all four platforms, Twitter (β = .308, p < .000), Facebook (β = .282, p < .000), Instagram (β = .282, p < .000), and TikTok (β = .236, p < .000) significantly predicted offline civic engagement (see Table E3 in Online Appendix E; Figure 2). According to these tests, we found that social media may predict higher levels of political interest and civic engagement overall compared to traditional resources, supporting H1.
Twitter, Instagram, Facebook, and TikTok use and offline civic engagement.
To further investigate the relationship between online and offline forms of civic engagement, we tested whether online civic engagement was a significant predictor of offline civic engagement. The findings suggest that as teens become more comfortable discussing and engaging in politics online, there is a positive relationship between that online action and the physical forms of civic engagement offline (β = .793, p < .000) they partake in, such as volunteering, discussing politics with friends or family, and participating in protests.
Affective Polarization
Before testing the relationship between social media use and affective polarization, we tested the differences in affective polarization by gender, race/ethnicity, and partisan identification. These tests mimic our analyses of the demographic differences in social media use. There were no differences in affective polarization between boys and girls or between Republicans and Democrats. The only significant difference was teens of color reported higher levels of polarization compared to white teens, t(1,306) = 3.170, p = .002, but the differences between Black and non-Black teens and Hispanic and non-Hispanic teens were not significant. Therefore, these findings should be treated with caution. See Table F1 in Online Appendix F for the full t-test results.
H2 predicts Facebook and Instagram, two strong-tie networks, will have no relationship with affective polarization among teens, while
First, we found that traditional resources did have a significant but minor, negative relationship with polarization (β = −.002, p < .000). Among the social platforms, Twitter use (β = .066, p < .000) and Facebook (β = .035, p = .042) were the only sites to significantly predict higher levels of polarization. Instagram (β = .021, p = .148), and TikTok (β = .014, p = .395) did not have a significant relationship with polarization (see Table F2 in Online Appendix F). These findings suggest that strong or weak-ties between online users may not be the main predictor of affective polarization in these environments.
Discussion
Social media has been lauded for the access it gives to marginalized groups to politics and information (Holt et al., 2013; Kahne & Bowyer, 2018), but critics have also recognized the way in which the algorithmic attention economy of the internet promotes extreme ideologies and fringe attitudes (Kubin & von Sikorski, 2021; Yarchi et al., 2021). Across the board, our data show that social media use among teens is related to increases in political interest and both online and offline civic engagement. At the same time, we also found that using TikTok and Instagram has no relationship with affective polarization. With teen social media use on the rise (Pew Research Center, 2022), our findings have important implications for the future of politics in the United States. Our results indicate that teens may be more interested and engaged in politics the more they use social media, regardless of their access to traditional resources.
Most teens are using social media. We find this use has a positive relationship with political interest. Interestingly, Instagram ranked the highest among teens as influencing political interest, and TikTok was a close second. These are the platforms that teens are using at the highest levels, but they have also received some of the most criticism in regard to body image concerns, data privacy laws, and online harassment (Fardouly & Vartanian, 2016; Harriger et al., 2023; Meral, 2021; Thomas et al., 2022; Vitis & Gilmour, 2017). However, both platforms are filled with content creators who introduce teens to new political information. For example, Gen-Z for Change, a nonprofit organization with 1.7 million followers on TikTok and 65,000 followers on Instagram, provides the latest updates on state and federal-level legislation as well as action items like signing online petitions or following other activist organizations.
Twitter, a weak-tie social network, provides new political information to users, but the uncivil environment on the application could be creating more negative and tense situations online for teens. Over 40% of Twitter users report receiving some level of online harassment (Matias et al., 2015), with threats being even prevalent when a user discusses politics (Nadim & Fladmoe, 2021). More research should also be done to measure whether teens are exposed to more political elites on sites like Twitter, leading to more dislike toward political parties as well a general sense of cynicism in politics (Kingzette et al., 2021). Facebook, as a strong-tie network, also had a positive relationship with affective polarization. As Settle (2018) argues, the collapse of personal and partisan information on the platform’s newsfeed could politicize traditionally nonpolitical characteristics, like what car a person drives, where they eat, or what music they listen to. Despite the platform having more strong ties between users, there appears to be enough information and exposure to varying interests that could also lead to higher levels of affective polarization.
Despite their significant relationship to polarization, Twitter and Facebook had the strongest relationships with offline civic engagement among teens, further supporting that this political interest moves beyond mere slacktivism. As teens become more politically engaged online, we are seeing a significant relationship to offline political engagement. These findings are in line with what Allsop (2016) termed “facilitators” (p. 38). Social media accounts can provide resources and referrals to help teens learn about political causes, interact with supporters, and use their more traditional networks to engage politically offline. We found that Black teens were more likely to use Twitter than the non-Black participants. Historically, there are persisting inequalities between the resources held by Black and Hispanic families compared to white families in the United States (Creamer, 2020). Social media, and especially Twitter, may provide a way to level access to political information and ways to get involved for disadvantaged teens online. Further research should examine how “Black Twitter,” the term given to the community of Twitter users that use the platform to connect, organize, and raise consciousness around issues specific to the Black American experience (Hill, 2018), provides opportunities for participation among Black teens as well.
We find no evidence that TikTok has a positive relationship with affective polarization among teen users. TikTok is designed with a unique “For You” page that caters to each user’s personal interests, similar to the algorithms that curate one’s newsfeed on Facebook or the timeline on Twitter. This model has sparked what many journalists have considered a potentially new form of online polarization (Paul, 2022). In our results, however, we see TikTok predicting interest and engagement, but not polarization. Recent research finds that TikTok is more interactive as a medium, potentially increasing exposure to cross-cutting information from unique perspectives online (Medina Serrano et al., 2020).
Perhaps TikTok, with its focus on short and humorous video content that is algorithmically programmed and centered around memes and music, does not lend itself to polarizing statements. Teens are encountering political content there, but perhaps there is something about the TikTok algorithm that promotes a different type of effect that is related more to entertainment and information-seeking and less to outrage. As Medina Serrano and colleagues (2020) found, users on TikTok may be more likely to engage and react to content from users they politically disagree with, creating more public debate across the platform. As such, banning TikTok and other foreign-based social platforms could have dire effects on teen political interests and youth participation in US politics.
Our study has several limitations. First, our sample overrepresents teen girls, and therefore, cannot generalize about the U.S. teen population as a whole. The findings should be replicated with more representative samples of teen users. More research is also needed to better understand what exactly teens are doing on each social media platform. We asked teens to report how often they performed certain actions, like posting and sharing their own content. Although this allows us to see how actively each participant was on each platform, it does not allow us to understand what type of content teens access or are exposed to on each platform. Social media can serve as a way to communicate with friends, but it can also serve as a vehicle for information. Sometimes this information is educational or informative, like news from accredited institutions, or user-generated content that offers unique perspectives and experiences. However, misinformation is also rife on social media and poses a threat to teens online. Further research should test just how misinformation persists on various platforms and how much of it is consumed by teen users to better understand how this may influence teens’ political identities offline.
Conclusion
This study provides new evidence of the role social media plays in developing political identities and participation. We are limited to a single time frame, and future research should consider using panels to track the influence of social media use over time. In addition, more research on teen activity online should be conducted, because as past research finds, survey respondents often overestimate their political participation online (Guess et al., 2019; Henderson et al., 2021). We find relationships through our survey methodology, but future experimental research can help better inform what causational effects social media use has on political expression and behaviors among teens.
By asking teens about their specific levels of social media use, we can compare not just if teens have a platform but how active they are on that platform to their political behaviors both on and offline. This adds value and new insight into the growing role social media plays in the formation of teens’ political identities. Social media can often have negative influences on teens, especially on their mental and physical well-being (Boer et al., 2021), but as some more optimistic scholars hoped (Dutton, 2009), social media has democratic potential as well. For teens, social media is more likely to predict their civic engagement than traditional and socioeconomic resources, and social media is increasingly where teens gather to discuss the issues of the day. As social media use further increases among teens, these platforms can help broaden participation among young users and cultivate skills and identities that will shape their political behaviors as adults.
Supplemental Material
sj-docx-1-sms-10.1177_20563051231186364 – Supplemental material for The Kids Are Online: Teen Social Media Use, Civic Engagement, and Affective Polarization
Supplemental material, sj-docx-1-sms-10.1177_20563051231186364 for The Kids Are Online: Teen Social Media Use, Civic Engagement, and Affective Polarization by Ayla Oden and Lance Porter in Social Media + Society
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The funding from the Ken Kansas Professorship at the LSU Manship School of Mass Communication.
Supplemental Material
Supplemental material for this article is available online.
Notes
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
