Social Media Democracy Mirage: How Social Media News Fuels a Politically Uninformed Participatory Democracy

Homero Gil de Zúñiga; Hugo Marcos Marné; Manuel Goyanes; Rebecca Scheffauer

doi:10.1017/9781009053266

1 Introduction

Social science scholars have long been aware of the complexities and precarities of democratic governance, and the challenges regarding the consolidation of democratic institutions. In fact, recent years have seen increasing concerns about the state of democracy worldwide. Freedom House, a nongovernmental organization well known for its monitoring of democracy and freedom around the world, suggested in 2020 that the world had been undergoing a fifteen-year democratic decline (Gorokhovskaia et al., Reference Gorokhovskaia, Shahbaz and Slipowitz2023). The Council of Europe has similarly reported that democracy is increasingly at risk, even in places well known for their democratic credentials (Birdwell et al., Reference Birdwell, Feve, Tryhorn and Vibla2013).

Democratic backsliding is a multifaceted phenomenon, of course; but the technological developments in communication that shape societies across the world likely play a major role in both the sustainability and emergence of democracy (Weare, Reference Weare2002). Our modern world has been dubbed a “network society,” after all, heavily influenced by digital technologies (Castells, Reference Castells2009). Since the early days of the twenty-first century, and in parallel with the popularization of information and communication technologies (ICTs), a mushrooming body of literature has examined the effects of digital technologies, especially social media, on various political outcomes (Bimber & Copeland, Reference Bimber and Copeland2013; Gil de Zúñiga et al., Reference Gil de Zúñiga, Veenstra, Vraga and Shah2010). And the role of digital media in our daily lives is only increasing over time. The Pew Research Center (Reference Research Center2021), a nonpartisan fact tank based in the USA, estimated that at least 70 percent of American citizens used social media in 2021, a figure that barely reached 5 percent in 2005. This proliferation of social media has sparked debates in academic communities and beyond on the impact of social media on democracy.

The purpose of the current Element is thus to take stock of the burgeoning literature exploring associations between social media and both (1) political knowledge and (2) political participation. There is of course a longstanding literature highlighting the role of participation and knowledge in sustaining healthy democracies (e.g., Boulianne, Reference Boulianne2020; Galston, Reference Galston2001; Hopp et al., Reference Hopp, Ferrucci, Vargo and Fisher2020; Kleinberg & Lau, Reference Kleinberg and Lau2019; Lecheler & de Vreese, 2017; Lelkes, Reference Lelkes2020; Parry et al., Reference Parry, Moyser and Day1992). But what does twenty years of research tell us about the impact of social media on these critical outcomes? Our findings, in sum, are as follows.

The existing literature suggests, in short, that social media use is associated with increased political participation (see Boulianne, Reference Boulianne2020; Gil de Zúñiga et al., Reference Gil de Zúñiga2012; Halpern et al., Reference Halpern, Valenzuela and Katz2017). For instance, studies exploring the connection between social media use and social movements suggest that the use of social media has led to a growing likelihood of engaging both in online and offline political activities. This has been found for movements such as Black Lives Matter in the USA (Cox, Reference Cox2017; Mundt et al., Reference Mundt, Ross and Burnett2018; Wilkins et al., Reference Wilkins, Livingstone and Levine2019), the 15-M square-occupation movement in Spain (Micó & Casero-Ripollés, Reference Micó and Casero-Ripollés2014), the 2011 Egypt uprising (Clarke & Kocak, Reference Clarke and Kocak2020), student and environmental mobilization in Chile (Scherman et al., Reference Scherman, Arriagada and Valenzuela2015), or antigovernment protest in Thailand (Sinpeng, Reference Sinpeng2021).

The story is rather different where political knowledge is concerned, however. Social media platforms were initially theorized as community spaces with the potential to contribute to a democratic public sphere and to foster political learning, whether employing intentional news-seeking or through incidental exposure to news and political discussions (Trenz, Reference Trenz2009; Valeriani & Vaccari, Reference Valeriani and Vaccari2016). These positive expectations have not been thoroughly confirmed by empirical analyses. Indeed, recent research suggests that social media, far from having a positive effect on political knowledge, may actually prevent political learning (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Lee & Xenos, Reference Lee and Xenos2019; Shehata & Strömbäck, Reference Shehata and Strömbäck2018). Empirical studies suggest that people either learn less as compared to those who are exposed to traditional news, or there are null direct effects derived from using social media for news (Gil de Zuñiga, Borah & Goyanes, Reference Goyanes, Ardèvol-Abreu and Gil de Zúñiga2021; Knobloch-Westerwick & Meng, Reference Knobloch-Westerwick and Meng2011; Skovsgaard & Andersen, Reference Skovsgaard and Andersen2020; Valenzuela et al., Reference Valenzuela, Correa and Zúñiga2018). In a recent meta-analysis Amsalem and Zoizner (Reference Amsalem and Zoizner2023) likewise raised doubt about the positive effects of social media on political knowledge. According to their results, knowledge gains are small to nonexistent. In short, there are well-founded concerns that social media use for news may not consistently contribute to a more informed public opinion.

Combining the results of these two strands of literature, political participation and political knowledge, it stands to reason that social media may be nurturing a socio-political paradox where people are increasingly more participatory, yet not necessarily more informed. In this Element, we label this phenomenon as the Social Media Democracy Mirage and argue that this paradox may be key to understanding many of the current political phenomena affecting liberal democracies. Specifically, the Social Media Democracy Mirage entails the amplification of political beliefs and activities on social media which are not necessarily sustained by factual information about politics. While we do not claim that the widespread use of social media will unmistakably push societies into a pathway of democratic erosion, we believe the mirage label resonates with the unfulfilled potential of social media use for information.

This Element includes three sections that can be read independently or as a whole:

a systematic literature review of social media and political participation (Section 2)
a systematic literature review of social media and political knowledge (Section 3)
a quantitative analytical assessment of the link between social media, political knowledge, and participation using survey data from the USA (Section 4)

In the literature review Sections 2 and 3, we present an overview of the growing body of work connecting social media, political participation, and political knowledge. We examine more than 500 peer-reviewed articles which were published in over 150 journals from 2001 to 2020. Our systematic reviews contribute to the ongoing discussion on the interplay between the affordances for action that social media enables, and citizens’ means to obtain information about public affairs. In Section 4 of the Element, we take advantage of a unique collection of four original datasets collected online in the USA between 2009 and 2020. Relying on these data, we further illustrate the participatory yet uninformed consequences of social media news use that underline the Social Media Democracy Mirage. Section 5 contains some concluding remarks and suggestions for future research in political communication.

Before diving into our analyses, however, the following sections (1) define political participation and knowledge, (2) describe the function and process of systematic literature reviews, and (3) outline the theoretical framework – more specifically, the “research clusters” that structure our literature reviews and empirical analysis.

1.1 Defining Participation and Knowledge

The definitions of political participation often gravitate around the identification of activities aiming at influencing policy-making or government actions (Verba et al., Reference Verba, Schlozman and Brady1995). While many were initially associated with “offline electoral activities,” such as working for a political party or voting (Conway, Reference Conway1985; Saldaña et al., Reference Saldaña, Mcgregor and Zúñiga2015), the emergence of the internet has broadened our conception of political participation to include, for instance, campaign contributions, protesting, writing a letter to a politician, and so on, both online and offline (Halpern et al., Reference Halpern, Valenzuela and Katz2017; Yang & DeHart, Reference Yang and DeHart2016).

Citizens’ participation is a fundamental building block for most notions of democracy (Parvin, Reference Parvin2018). Declines in voter turnout have thus raised concern among social science scholars, particularly considering that turnout rates may be lower than the official turnout figures, especially where there are gaps between registered voters and voting-age population. These gaps increase considerably in elections other than national ones, such as state elections, and can be found across countries and regions beyond the USA (Lijphart, Reference Lijphart1997). Although low election turnout is often seen as a consequence of the malfunctioning of democracies (Grönlund & Setälä, Reference Grönlund and Setälä2007), representative democracies also benefit from the electoral participation as such participation is not only the outcome of democratic illness, but also a potential means to deal with this issue. Some democratic theorists underscore that higher levels of voter turnout increase the chances that the voices of various groups in society are heard (Rosema, Reference Rosema2007), and this is particularly important if we consider that the chances of abstaining are not randomly distributed among the population. Young, less educated, and low-income people are consistently less likely to vote (Blais et al., Reference Blais, Gidengil, Nevitte and Nadeau2004; Gallego, Reference Gallego2009; Lijphart, Reference Lijphart1997; Wattenberg, Reference Wattenberg2020). In that sense, social media news use could be an interesting tool to foster electoral participation among young people, who are more likely to be present online (Hargittai & Hinnant, Reference Hargittai and Hinnant2008), and also older people from cohorts that were early socialized in social media environments (Prensky, Reference Prensky2001).

Beyond voting, political participation is even more important for alternative understandings of democracy such as the deliberative one, which builds upon the exchange of ideas between individuals that do not form a homogeneous group (Steiner et al., 2017). A textbook example of the importance of participation can be seen in the design of current mechanisms for democratic innovation (e.g., participatory budgets, mini-publics, deliberative meetings, etc.), which often offer alternative spaces for citizens’ engagement in political life. In this regard, we can think of political participation as an opportunity for citizens from diverse backgrounds to be treated as equals in public affairs and reduce spaces for demagogies, oppressive rules, and government inefficiencies. In sum, there is an overall agreement on the importance of participation in a democracy, and such participation can take many different forms.

Some vital forms of political participation in our empirical analysis include attending rallies and demonstrations, sending letters or emails to political organizations, newspapers, or elected officials, or volunteering to help with political causes. Although they are different, these forms of participation share a minimal core: they relate to citizens’ voices being heard in the public sphere (online and analogical). Some forms of political participation associated with protest are more often used by underprivileged groups (Lipsky, Reference Lipsky1968; van Stekelenburg & Klandermans, 2013), which show the potential of social media news use to foster types of participation not so conditioned by structural inequalities. Although some theorists call for an adjustment of our expectations on the extent to which widespread participation can be achieved (Parvin, Reference Parvin2018), there is little doubt that important shortcomings of liberal democracies such as the USA could be addressed with it.

Another key for democracies is political knowledge, which is broadly defined as “the various bits of information about politics that citizens hold” (Delli Carpini & Keeter, Reference Delli Carpini and Keeter1993, p. 1179). Looking at the minimal representative core of democracy, voters are expected to cast their votes after considering various political alternatives, a task that can be more easily achieved when abundant political knowledge is available. In this regard, individuals who are more knowledgeable in politics are better equipped to choose parties that are in line with their preferred issues, and their positioning in these issues (Andersen et al., Reference Andersen, Tilley and Heath2005). These knowledgeable individuals are also more likely to exercise reasoned economic voting (Gomez & Wilson, Reference Gomez and Wilson2001) and to react to credible corruption accusations (Weitz-Shapiro & Winters, Reference Weitz-Shapiro and Winters2017). Political knowledge is closely associated with media literacy, a variable capturing the ability of individuals to engage constructively with journalism (Maksl et al., Reference Maksl, Ashley and Craft2015) that is crucial to avoid fake news (Bulger & Davison, Reference Bulger and Davison2018). The relevance of the duo media literacy–political knowledge is beyond doubt in the context of electoral campaigns increasingly shaped by fake news (Allcott & Gentzkow, Reference Allcott and Gentzkow2017; Mutahi & Kimari, Reference Mutahi and Kimari2020; Quandt et al., Reference Quandt, Frischlich, Boberg, Schatto‐Eckrodt, Meier, Vos, Hanusch, Dimitrakopoulou, Geertsema-Sligh and Sehl2019; Rosa, Reference Rosa2019).

The positive effects of political knowledge also spread beyond the realm of elections. For example, the mechanisms of direct democracy, such as referendums, are sometimes implemented as a way to address some limitations of representative democracies, creating spaces for individuals to express their preferences without the mediation of institutions or parties. However, the ability an individual has to express their own voice often depends on political knowledge (Christin et al., Reference Christin, Hug and Sciarini2002; Hobolt, Reference Hobolt2007). Likewise, more knowledgeable individuals might find it easier to participate in social movements that are in line with their ideas, attitudes, and preferences, and to choose more effective ways to influence elected officials. Additionally, political knowledge correlates with the acceptance of democratic principles, which explains why some governments invest time and money to improve political knowledge among their citizens (Galston, Reference Galston2001).

1.2 Systematic Literature Reviews

Despite its substantive importance, the direction and magnitude of the effects of social media on political participation and political knowledge remain unclear. Unlike other communication research domains (Ahmed et al., Reference Ahmed, Ahmad, Ahmad and Zakaria2019; Garrido et al., Reference Garrido, Zentner and Busse2011; Naab & Sehl, Reference Naab and Sehl2017; Wang et al., Reference Wang, McKee, Torbica and Stuckler2019; Williams, Reference Williams2019), we lack a thorough and organized examination of empirical findings with a holistic narrative that is able to guide future social scientific theoretical and empirical endeavors. The purpose of our systematic literature reviews is thus to better unravel the association between social media use and both political knowledge and political participation. Our systematic reviews particularly contribute to the literature as follows.

They examine and problematize the items and measurements used thus far in the literature focusing on political knowledge and political participation. This is particularly important because different streams of literature may have distinct definitions and measurements that complicate comparative interpretations of main findings.
They help understand the granularity of the published material in terms of research patterns (territory of data collection, authorship structure, methodological approaches, etc.), which is key to unraveling underexplored areas and limitations.

1.3 Theoretical Framework

Before we present the literature review, we lay out our four-faceted main theoretical frameworks that underpin research on the links between social media and political participation and political knowledge. The defined facets of the theoretical framework were established in an iterative process. A priori facets were entered into conversation and dialogue with post-hoc analysis of the studies under scrutiny, yielding five potential clusters of research: (1) media effects, (2) interpersonal communication, (3) expressive political content, (4) structural effects, and (5) a miscellanea category. Figure 1 illustrates the research foci of studies situated at the intersection of social media, political participation, and political knowledge discussed in the literature reviews (Sections 2 and 3), highlighting and distinguishing the categories in which relevant papers can fall. We also point out the most salient authors for each research strand. These authors were chosen either due to their number of publications within the research clusters topics or by the impact of their findings in the literature, which we will discuss in later Sections 2 and 3. We also briefly present the theoretical foundations of each facet.

Figure 1 Research categories serving as basis for both systematic literature reviews.

1.3.1 Media Effects

Media effects have been a research pillar in political science even before social media became relevant. Researchers have examined the effects of exposure to traditional and later to online news (Bimber & Copeland, Reference Bimber and Copeland2013; Martin, Reference Martin2008; McLeod et al., Reference McLeod, Scheufele and Moy1999). In a functioning democracy, media fulfils a variety of functions ranging from holding those in power accountable to providing citizens with the information they need to make informed decisions. With the advent of the internet, there is more information than ever before and social media allows for easy distribution and consumption. With these developments, new research topics within media effects emerged, including the effects of social media news, second screening, and incidental news exposure (Giglietto & Selva, Reference Giglietto and Selva2014; Heiss & Matthes, Reference Heiss and Matthes2019; Pastrana Valls, Reference Pastrana Valls2017; Saldaña et al., Reference Saldaña, Mcgregor and Zúñiga2015). Thus, we expect media effects on political knowledge and political participation to be as prevalent on social media as they are with traditional media.

1.3.2 Interpersonal Communication Effects

Another relevant theoretical angle to account for the relationship between social media and political participation/political knowledge is interpersonal communication. This entails studies focusing on “political discussion, the rational, deliberative exchange of arguments, and its implications for an informed and participatory citizenry” (Scheufele, Reference Scheufele2000, p. 727). Such a form of discussion has been previously outlined as an integral antecedent for engaging in a variety of political activities (Huckfeldt & Sprague, Reference Huckfeldt and Sprague1995; Shah et al., Reference Shah, Cho, Eveland and Kwak2005; Valenzuela et al., Reference Valenzuela, Kim and Gil de Zúñiga2012) and it has even been referred to as the soul of democracy (De Tocqueville, Reference De Tocqueville1863).

1.3.3 Expressive Political Content

The third research framework revolves around expressive political content or, more generally, sender or self-effects. This line of work refers to the effects that sending a message has on the sender themselves (Pingree, Reference Pingree2015; Shah, Reference Shah2016; Valkenburg, Reference Valkenburg2017), potentially leading to a change in emotions, attitudes, cognitions, as well as behaviors (Aronson, Reference Aronson1999; Gil de Zúñiga et al., Reference Gil de Zúñiga, Garcia-Perdomo and McGregor2015; Pingree, Reference Pingree2007; Rojas & Puig-i-Abril, Reference Rojas and Puig-i-Abril2009; Valkenburg, Reference Valkenburg2017). In offline settings, attempts to persuade others also affect the potential persuader (Janis & King, Reference Janis and King1954), which is sometimes referred to as self-persuasion (Aronson, Reference Aronson1999). An additional aspect associated with self-effects is the writing paradigm that has experienced newfound popularity with the internet (Ko & Kuo, Reference Ko and Kuo2009; Lee et al., Reference Lee, Kim, Yoo, Park, Jeong and Cha2016; Pennebaker, Reference Pennebaker1997), as it allows smooth political expression in the virtual realm.

The internet and social media in particular come with several affordances that can amplify self-effects in comparison to the offline realm. First of all, it allows for what Castells (Reference Castells2007) refers to as mass self-communication, making it possible to reach a global audience. Furthermore, users are more inclined to share information and express themselves than in an offline setting (Christofides et al., Reference Christofides, Muise and Desmarais2009). Since it allows for asynchronous communication and provides the option to easily reach a larger audience, social media lends itself as an expression hub, opening the possibility to more self-effects (Boyd, Reference Boyd and Papacharissi2010; Shah, Reference Shah2016).

1.3.4 Structural Effects

This cluster of research addresses the role of structural variables, mainly socio-economic status, gender, education, internet access, and digital skills, many of which are shown to impact online and offline participation. Higher-income and socio-economic status tend to go hand in hand with better education, which in turn relates to skill and access levels. For decades, many researchers have dedicated their time and efforts to investigating the potential influences of these variables on different types of political variables. Thus, it is expected that the intersection of social media and structural variables has emerged as a popular research subject. Different theories have been proposed in this context from the “rich get richer” approach to contrary effects of the internet and social media being able to bridge gaps that are usually widened by differences in education, income, or gender. Social media is a free, fairly simple, and time-efficient tool to get engaged in politics. Thus, the fourth framework will focus on these variables.

2 Social Media and Political Participation

In gathering the articles for this task, we strictly follow the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines that transparently identify a number of criteria widely used for reporting literature reviews and meta-analyses (Moher et al., Reference Moher, Shamseer, Clarke, Ghersi, Liberati, Petticrew, Shekelle, Stewart and Group2015). The literature search was conducted in February 2021 using academic search databases, namely Scopus, Academic Search Ultimate, and Web of Science, to compile a comprehensive review. In order to identify relevant articles, we used the following keywords: “political participation” AND “social media” OR “social networking sites and political participation” OR “Twitter and political participation” OR “Facebook and political participation.” Studies (i.e., articles) included in the analysis should meet the following criteria: written in English and published by a peer-review journal (indexed in Journal of Citation Reports (JCR) and/or Scopus) between 2000 and 2021. We decided to sample this time frame because we aimed to provide the most updated literature review on the subject.

After applying the aforementioned criteria, the initial search yielded 1,348 results for social media and political participation. We then used reference management software Zotero to scan our results to find all duplicates (612 articles were removed). We carefully read the abstracts and excluded the journals that were either not peer-reviewed or off-topic before eliminating 310 articles (e.g., articles not concerned with political participation, articles focused on nonsocial media aspects such as TV news, radio, or the internet in general, research summaries, theoretical papers, or research recommendations, articles about social networks in the offline realm). We kept a total of 426 articles after applying these criteria (see Figure 2).

Figure 2 Social media and political participation process.

2.1 Quantitative Analysis

We started the analysis by collecting 426 articles which discuss the effects of social media on political participation. These articles were published in 176 different journals – New Media & Society published the largest number of articles (n = 31), followed by Information Communication & Society (n = 27). Seven journals altogether account for almost one third of all published articles (n = 135), while 105 journals only published one article each. A full list of the journals can be found in Table 1A.

Regarding authorship, USA-based scholars led research efforts in the subfield with 137 publications, followed by scholars based in the UK (n = 30), Hong Kong (n = 21), Australia (n = 17), and Canada (n = 17). As for the country of data collection, a large majority of the papers used data from the USA (n = 117), followed by China (n = 19), the UK (n = 18), and South Korea (n = 15). While there are 65 different countries represented in the analytical framework, only 46 articles are comparative in nature. Articles on social media and political participation started to emerge in 2001, but research in this domain skyrocketed after 2011 and peaked in 2020 with 109 publications.

For the measurement of social media use, most of the studies utilized overall social media indicators or inquired about different platforms that were later combined into one averaged index (n = 298). For distinct measurements, Facebook was consistently selected as the most representative social media platform to be studied (n = 68), followed by Twitter (n = 38). There are, however, several studies focusing on other social media platforms, partially unique to certain geographic areas, such as Weibo in China or VK in Russia (n = 16). An overview over the items used most commonly for measurements can be found in Tables 4A–14A.

In terms of methods, 317 articles drew upon quantitative data, while 68 used qualitative methods and 39 combined quantitative and qualitative methodological approaches. Most quantitative studies relied on survey data (n = 241), content analysis (n = 58), and mixed methods (n = 52). A total of 20 studies used interviews and 10 others employed focus groups. Moreover, 14 studies followed experimental designs and nine relied on ethnography and participant observation. The remaining 22 articles were grouped under “other” methodologies (e.g., case studies, or principal component analyses).

Most papers (n = 345) focused on citizens as a primary unit of analysis, while some articles addressed political leaders (n = 21) and democratic systems as a whole (n = 6) and others concentrated on more than one of the aforementioned units (e.g., leaders and citizens) (n = 20). Of the articles, 33 included video clips, websites, or songs as units of the analysis or main object of investigation. A large majority of the papers (n = 366) relied on cross-sectional data, while 38 articles used longitudinal data and 15 articles combined both. For the remaining 7 articles, neither one of those descriptions was applicable, as they were mostly theoretical. The summary of the data can be seen in the following graphs (Figures 3 to 5 and Figure 1A). More detailed information on the data used in the figures can be found in Tables 2A and 3A.

Figure 3 Data collection and first author origins depicted by country.

Figure 4 Number of studies depicted by applied methods, additionally divided by quantitative or qualitative as well as cross-sectional, longitudinal, and combined data collection.

Figure 5 Number of studies depicted by the primary objects under investigation, additionally divided by social media platform in question.

2.2 Qualitative Analysis: Thematic Patterns of Social Media and Political Participation Research

2.2.1 Main Findings

In general, it appears that the use of social media news to some degree replicates the influence that the use of traditional news exerts on political participation. Despite few studies presenting contrary results, deliberate use of news on social networking sites, regardless of the social media platform or specific measurement, positively relates to participation. Incidental exposure, on the other hand, is more complex to pin down and only has positive effects under certain conditions. In terms of interpersonal communication, network size and discussion seem to produce the most consistent effects on participation. For expression, pinning down effects is somewhat more complex as studies tend to fold expressive behaviors into other social media measurements. Finally, in terms of structural effects researchers point to a still existent gender gap with men engaging in more visible participatory behaviors. For younger people, social media plays a bigger part when it comes to engagement which is often noted as a positive trend. The expectation that social media might help previously disadvantaged groups (those with less income and/or less education) has been mostly disproved.

2.2.2 Measuring Political Participation

A challenge for the comparability of the results has to do with the diverse ways in which political participation is measured. Most researchers measure both online and/or offline participation with different political activities and a different number of items (Ahmed & Cho, Reference Ahmed and Cho2019; Park & Kaye, 2018; Saldaña et al., Reference Saldaña, Mcgregor and Zúñiga2015; Stromback et al., Reference Stromback, Falasca and Kruikemeier2018; Towner & Muñoz, Reference Towner and Muñoz2018). Some scholars, however, focus more on participation measures capturing aspects such as voting (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Towner, Reference Towner2013), activism/protest (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Karakaya & Glazier, Reference Karakaya and Glazier2019), or low- and high-effort participation (Heiss & Matthes, Reference Heiss and Matthes2019). Despite these different approaches, a vast majority of the studies showcased a positive influence of social media use on political participation.

2.2.3 Media Effects

Despite the extensive research conducted so far, the magnitude of the associations between media and political participation remains uncertain (Ahmad et al., Reference Ahmad, Alvi and Ittefaq2019; Li et al., Reference Li, Lee and Li2016; Saldaña et al., Reference Saldaña, Mcgregor and Zúñiga2015). Based on the positive influence traditional media use was found to have, many expected news consumption on social media to similarly impact participatory behaviors. Saldaña and colleagues (2015), for instance, presented results from the UK and the USA that attest to the significant impact of media (e.g., traditional news and social media news, the latter of which was captured with eight items) in predicting online and offline political participation. The mobilizing impact of social media is replicated in other countries (Skoric & Poor, Reference Skoric and Poor2013; Stromback et al., Reference Stromback, Falasca and Kruikemeier2018; Zhang & Lin, Reference Zhang and Lin2014) where media variables, such as informational social media use (six items), social media news use (three items), and the interaction of Facebook use and attention to traditional news, were applied. In most studies, social media news consumption appears to positively influence political participation online and offline (Gil de Zúñiga et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Park & Kaye, 2018; Saldaña et al., Reference Saldaña, Mcgregor and Zúñiga2015).

Beyond intentional exposure to news in social media, both survey-based studies and experimental studies have focused on the impact of incidental news exposure (Heiss & Matthes, Reference Heiss and Matthes2019, Reference Heiss and Matthes2021; Lee & Xenos, Reference Lee and Xenos2020; Lu, Reference Lu2019). Some researchers voiced the hope that incidental exposure to news might be able to produce similar results to deliberate news consumption and could, thus, bridge the gap between those who are already engaged in politics and those who are largely uninterested. Experimental approaches have evidenced that encountering political information embedded in a humorous context can increase the likelihood of political participation – directly and indirectly via elaboration, which might be more marked for those who generally would not engage with political information on social media (Heiss & Matthes, Reference Heiss and Matthes2021). Lee and Xenos (Reference Lee and Xenos2020) highlighted that the relationship between incidental news exposure and political participation is reciprocal, influencing each other indirectly via political social media use. Moreover, results from a two-wave panel study suggested that the potential influence of incidental news exposure depends strongly on how participation is measured. While there is a direct positive influence on low-effort online participation, the effect on high-effort online participation is conditional on political interest (Heiss & Matthes, Reference Heiss and Matthes2019).

Another viable research avenue within media effects that has recently garnered attention is the effect of second or dual screening. Users actively search for more information on or discuss about, for example, candidates while watching a debate, thereby engaging with more information and, as was theorized, might then be more inclined to engage in politics themselves. Vaccari and Valeriani (Reference Vaccari and Valeriani2018b), for example, showed that dual screening for political content positively influences online and offline political participation and additionally interacts with political interest. Results from twenty different countries further validated those findings, showing a positive association between second screening and political participation and even political expression on social media (Gil de Zúñiga & Liu, Reference Gil de Zúñiga and Liu2017). Lin (Reference Lin2019) found similar tendencies for civic engagement and highlighted that dual screening leads to higher civic engagement attitudes, which in turn fosters engagement. In addition to these positive indications stemming from survey research, Giglietto and Selva (Reference Giglietto and Selva2014) used content analysis to validate the influence of second screening. Their results showed that second screening elicits political participation, specifically during interviews and group discussion programs.

In line with the importance of considering how social media use is measured, results from a three-wave panel study on campaign information showed that YouTube and Tumblr had no significant influence on any form of participation, thereby showing that platform-specific affordances can greatly impact how social media use relates to political participation. In contrast, Facebook use positively predicted online participation, while Twitter and Google+ use correlated positively with both online and offline participation. However, none of them influenced voting turnout (Towner, Reference Towner2013). Nevertheless, Zhang et al. (Reference Zhang, Seltzer and Bichard2013) found that relying on Facebook/Google+, Twitter, and YouTube for political information positively predicted online and offline participation. Additionally, it appears that, especially for “Boomers,” attention to social media is only relevant for online participation. While news use on Facebook, YouTube, or Twitter did not significantly influence the applied index of offline participation (five items), news use on Facebook and YouTube positively related to online participation (index with eight items).

Overall, several studies found a positive effect of social media (news) use on political persuasion (Kasadha, Reference Kasadha2019), interest (Zhang et al., Reference Zhang, Seltzer and Bichard2013), protest (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Karakaya & Glazier, Reference Karakaya and Glazier2019; Zumarraga-Espinosa, Reference Zumarraga-Espinosa2020), voting (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Hassim et al., Reference Hassim, Zian and Jayasainan2020; Towner, Reference Towner2013), civic engagement (Gil de Zúñiga, Reference Gil de Zúñiga2012; Kim & Chen, Reference Kim and Chen2015), and environmental activism (Zhang & Skoric, Reference Zhang and Skoric2018).

2.2.4 Interpersonal Communication

Literature on interpersonal communication in our analysis suggests that the structure of interactions on social media may be the defining feature of the effects on political behaviors. Thus, this research cluster refers to the impact of different forms of political discussion (e.g., discussion with homogeneous vs. heterogeneous networks, weak vs. strong ties), interpersonal discussion attributes (e.g., network size), and places of discussion (e.g., via mobile phones). More importantly, interpersonal communication frameworks are often combined with those of media effects because news consumption helps promote discussions (Shah et al., Reference Shah, Cho, Eveland and Kwak2005) and broaden informational horizons that give people access to a more varied set of news (Gil de Zúñiga & Valenzuela, Reference Gil de Zúñiga and Valenzuela2011).

Akin to media effects and political participation, academic discussions have increasingly revolved around the relations taking place on online platforms. This new space for discussion comes with several affordances that differ substantially from offline exchanges (Valenzuela et al., Reference Valenzuela, Kim and Gil de Zúñiga2012). While offline interactions give visual cues and are synchronous as well as oral in nature, online forms of discussion can be asynchronous, written, and oftentimes anonymous (Lin, Reference Lin, Bryant and Oliver2009). Moreover, it becomes easier to connect with a wider array of people online, potentially boosting people’s network size and the number of weak-tie contacts (Papacharissi, Reference Papacharissi2004), both of which are influential when it comes to promoting political participation (Granovetter, Reference Granovetter1973; Valenzuela et al., Reference Valenzuela, Kim and Gil de Zúñiga2012) as people are exposed to not only more information but also to more diverse content. Furthermore, whereas some scholars question the deliberative potential of online discussions (Strandberg, Reference Strandberg2008), others underline that active online discussions may carry a positive impact on political participation, whether using formal or informal interactions. Thus, both discussions with other users and with political actors could be beneficial in boosting individuals’ political participation (Abdulla et al., Reference Abdulla, Poell, Rieder, Woltering and Zack2018; Akkor, Reference Akkor2017; Vaccari & Valeriani, Reference Vaccari and Valeriani2018a).

Alberici and Milesi (Reference Alberici and Milesi2018) highlighted that it is the type of discussion that plays an integral part in encouraging participatory behaviors, urging researchers to further investigate disparate political discussion attributes. According to their results, it is specifically constructive online discussions that can imbue a person’s politicized identity with “meaning of responding to a moral obligation” (Alberici & Milesi, Reference Alberici and Milesi2018, p. 143), making them participate in collective action online and offline. The size of people’s discussion networks may also be an important contributor to online and offline participation. There appears to be a direct positive influence of network size on participation (Cao, Reference Cao2020), a cross-sectional finding that was replicated by two-wave panel studies as well (Gil de Zúñiga, Diehl, et al., Reference Gil de Zúñiga, Barnidge and Scherman2017; Gil de Zúñiga et al., Reference Gil de Zúñiga, Molyneux and Zheng2014).

Additionally, Tang and Lee (Reference Tang and Lee2013) found evidence of bigger network size increasing time spent on social media sites like Facebook, which leads to exposure to more information and results in higher levels of participation. Kahne and Bowyer (Reference Kahne and Bowyer2018) found that while network size and friendship-driven activities positively predict offline participation, the interaction of network size and interest-driven online activities influence online and offline participation, suggesting that more complex relationships are to be entangled by future research. Political discussion effects were found to hold true irrespective of whether researchers consider people’s online or offline networks, thus underlining this as an important factor in influencing their participation (Cao, Reference Cao2020; Gil de Zúñiga, Diehl, et al., Reference Gil de Zúñiga, Barnidge and Scherman2017; Gil de Zúñiga, Molyneux, et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Kahne & Bowyer, Reference Kahne and Bowyer2018; Tang & Lee, Reference Tang and Lee2013).

Beyond the nature of exchanges and the size of networks, research has shown that cultivating heterogeneous networks sparks higher levels of political participation and even civic engagement (Kim & Chen, Reference Kim and Chen2015; Tang & Lee, Reference Tang and Lee2013; Yoo & Gil de Zúñiga, Reference Yoo and Gil de Zúñiga2019), an effect that was not reproduced for discussion with more homogeneous networks (Yoo & Gil de Zúñiga, Reference Yoo and Gil de Zúñiga2019), once more building on the idea that these types of networks will provide more varied information which then could lead to more participation. While Tang and Lee (Reference Tang and Lee2013) only found a significant connection of network heterogeneity with offline participation, results presented by Yoo and Gil de Zúñiga (Reference Yoo and Gil de Zúñiga2019) showed that incidental news exposure influenced discussion heterogeneity, which positively affected online and offline participation. Some researchers focus their measurements of network heterogeneity specifically on with whom people discuss (Kim & Chen, Reference Kim and Chen2015; Yoo & Gil de Zúñiga, Reference Yoo and Gil de Zúñiga2019), whereas others consider the make-up of a person’s (online) networks as a whole (Tang & Lee, Reference Tang and Lee2013).

Finally, the specific place in which people get informed may also be an influential factor. Discussing politics via mobile phones (which is influenced by consuming news via smartphones), for example, positively affects offline participation (Kim et al., Reference Kim, Chen and Wang2016). Even WhatsApp discussion appears to be influential, positively correlating with activism and conventional participation. These effects appear to be more pronounced for younger people (Gil de Zúñiga et al., Reference Gil de Zúñiga, Ardevol-Abreu and Casero-Ripolles2019). Moreover, information-seeking seems to spur online and offline discussion. Results presented by Li and Chan (Reference Li and Chan2017) show that online discussion sparks higher levels of online participation, but offline discussion influenced both online and offline participation. These findings were consistent for both Hong Kong and China. Zhang et al. (Reference Zhang, Johnson, Seltzer and Bichard2010) also came to similar conclusions, highlighting that offline discussion positively relates to political and civic participation.

2.2.5 Expressive Political Content

Some researchers see expressive political content as a form of participation in itself (Chapman & Coffé, Reference Chapman and Coffé2016), while others see a clear pathway from expressive content to some forms of online and offline participation (Moffett & Rice, Reference Moffett and Rice2018; Rice & Moffett, Reference Rice and Moffett2019), and even as a form of influencing and persuading other citizens (Hosch-Dayican et al., Reference Hosch-Dayican, Amrit, Aarts and Dassen2016; Penney, Reference Penney2016). Nevertheless, building on the idea of self-effects, many expected political expression to potentially lead to more participation in the person who expressed themselves in the first place.

With expressive content, measures become more convoluted than with previously described research strands. A few studies feature separate variables dedicated solely to expressive content. Instead, many researchers chose to include this as an aspect of variables such as social media use (Robles et al., Reference Robles, Torres-Albero, Antino and De Marco2015; Vissers & Stolle, Reference Vissers and Stolle2014b; Zhang & Skoric, Reference Zhang and Skoric2018). The studies focusing on expressive actions present variables such as political expression on Snapchat (Rice & Moffett, Reference Rice and Moffett2019), social media prosumption (Yamamoto et al., Reference Yamamoto, Nah and Bae2019), creative internet use (Ekström & Östman, Reference Ekström and Östman2015), political tweeting (Bode & Dalrymple, Reference Bode and Dalrymple2016), or simply political expression on social media (Chan et al., Reference Chan, Chen and Lee2017; Gil de Zúñiga, Molyneux, et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Kwak et al., Reference Kwak, Lane, Weeks, Kim, Lee and Bachleda2018).

While expressive use of social media generally appears to be an important predictor of online (Chan et al., Reference Chan, Chen and Lee2017; Gil de Zúñiga et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Yamamoto et al., Reference Yamamoto, Nah and Bae2019) and offline participation (Gil de Zúñiga et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Kwak et al., Reference Kwak, Lane, Weeks, Kim, Lee and Bachleda2018; Lane et al., Reference Lane, Kim, Lee, Weeks and Kwak2017), some studies also highlight the importance of distinct platforms. Using Facebook or Twitter to politically express oneself has also shown a positive impact on different forms of participation (Bode & Dalrymple, Reference Bode and Dalrymple2016; Vissers & Stolle, Reference Vissers and Stolle2014a). The influence of political expression as part of using social media and the internet also extends to positively impacting environmental activism and consumerism (Zhang & Skoric, Reference Zhang and Skoric2018).

Furthermore, some researchers point out more intricate paths to participation. Lane et al. (Reference Lane, Kim, Lee, Weeks and Kwak2017) showed that cross-cutting discussion is what can spark information sharing on social media, which leads to online and then offline participation. Chen et al. (Reference Chan, Chen and Lee2017) highlighted the path from mobile news to mobile political messaging and to political expression on social media, which leads to both online and offline participation. Additionally, Yamamoto et al. (Reference Yamamoto, Nah and Bae2019) introduced the concept of social media prosumption, which includes consuming and producing content online. Social Media prosumption will trigger more political information seeking online, more online discussion heterogeneity, and via those two, more online political participation.

2.2.6 Structural Effects

Among the structural factors influencing participation, gender is highly discussed in previous research. The affordances offered by the internet and social media seem to facilitate engagement to a bigger degree for men than for women. Vochocová et al. (Reference Vochocová, Štětka and Mazák2016) question the “narrowing gender gap” that is often associated with the online environment, heralding equal space and opportunities for all genders to participate. Some studies suggest that this gap is not closing, as males are more likely to participate in online political activities such as political engagement, for example, being involved political groups or charities, (Xenos et al., Reference Xenos, Vromen and Loader2014), getting political information online (Hargittai & Shaw, Reference Hargittai and Shaw2013), sharing political opinions online (Vicente & Novo, Reference Vicente and Novo2014), and mobilization online (Moraes et al., Reference Moraes, Pelegrini and Pinheiro2020; Nam & Stromer-Galley, Reference Nam and Stromer-Galley2012). Additionally, men usually engage in behaviors that are more outgoing and visible like sharing their opinion or engaging politically online (Vicente & Novo, Reference Vicente and Novo2014; Xenos et al., Reference Xenos, Vromen and Loader2014). Despite these variations in the measurement of the dependent variables, there is consistent evidence suggesting that males are more prone to be involved in political activities online. The reasons why gender gaps appear persistent are typically due to socio-economic factors. Many forms of participation call for resources, like money and time, which might not be available to women to the same extent (Schlozman et al., Reference Schlozman, Burns and Verba1999).

Besides gender, one of the main structural predictors of online engagement and political participation is age. Various studies have found that younger individuals are more active within the context of social media. Although the outcome variables and their measurements differ between studies (e.g., online and offline participation, liking and sharing political content online, political social media use, and mobilization), the effect appears to hold consistently for political use of social media (Bode, Reference Bode2017; Nam, Reference Nam2011) and for using social media information tools (Nam & Stromer-Galley, Reference Nam and Stromer-Galley2012). Similar results concerning age were found across countries such as Cambodia (Chunly, Reference Chunly2019), Finland (Strandberg, Reference Strandberg2014), Germany (Hoffmann & Lutz, Reference Hoffmann and Lutz2019), and the USA (Bode, Reference Bode2017).

Education is also considered a structural influence often associated with political participation. A host of studies showed that people with higher levels of education are more likely to engage politically (Rosenstone & Wolfinger, Reference Rosenstone and Wolfinger1980; Schlozman et al., Reference Schlozman, Verba and Brady2013; Verba et al., Reference Verba, Schlozman and Brady1995), as citizen’s educational and internet skill levels influence the abilities to navigate on the internet and partake in political activities ; Nam, Reference Nam2011). This could also be attributed to higher levels of income usually associated with those with higher education (Manski, Reference Manski1992). Results regarding online activities, however, tend to be somewhat mixed. While some studies found a connection between higher education, social media use, and political activities (Ahmed & Cho, Reference Ahmed and Cho2019; Hoffmann & Lutz, Reference Hoffmann and Lutz2019; Woo Yoo & Gil de Zúñiga, Reference Woo Yoo and Gil de Zúñiga2014), others have reverse findings, pointing out positive connections of social media and participation for those with lower education (Hoffman, Reference Hoffman2012; Nam, Reference Nam2011). Bode (Reference Bode2017) highlighted varying effects of education depending on the type of political social media use (e.g., comments, likes, replies). This suggests that the effect of education may be contingent on a host of individual-level, meso-level, and macro-level variables.

Finally, several studies highlight the influence of income. So far, some studies follow the “rich get richer” approach and claim that financially privileged population layers in society will tend to be more politically engaged (Buente, Reference Buente2015), as these citizens may have more money, time, and civic skills to partake in political activities (Brady et al., Reference Brady, Verba and Schlozman1995). However, other scholars suggest that those who were previously excluded from the political system due to economic reasons now have more opportunities to participate online (Spaiser, Reference Spaiser2012), as this environment calls for fewer resources in terms of time and (digital) access. Despite this, scholarship examining political inequalities related to digital connectivity argues that citizens from lower socio-economic backgrounds might often remain excluded from political activities online (Sylvester & McGlynn, Reference Sylvester and McGlynn2010; Weber et al., Reference Weber, Loumakis and Bergman2003; Xu et al., Reference Xu, Ye and Zhang2018). While some found a positive effect of higher income on offline participation (Garcia-Castañon et al., Reference Garcia-Castañon, Rank and Barreto2011) and online participation (Bode & Dalrymple, Reference Bode and Dalrymple2016; Chunly, Reference Chunly2019; Hoffman, Reference Hoffman2012; Steinberg, Reference Steinberg2015), only a few studies produced conflicting results and show that social media help those from lower socio-economic backgrounds to be more involved (Vicente & Novo, Reference Vicente and Novo2014; Wang et al., Reference Wang, Cai, Mou and Shi2018; Zumarraga-Espinosa, Reference Zumarraga-Espinosa2020).

2.2.7 Miscellanea

A number of journal articles do not easily fit into one of the four categories outlined at the beginning of Section 2. This section groups these findings together, as follows.

Overall Internet/Social Media Use. Several authors focused on social media use per se, without looking at any specific factors like news use or expressive content. While Theocharis and Lowe (Reference Theocharis and Lowe2016) pointed out the negative effect of general Facebook use on traditional civic offline and online participation in Greece, other researchers highlight more positive influences. Zhang and Skoric (Reference Zhang and Skoric2018) conducted a study in Hong Kong that revealed the negative influences of relational social media use on environmental activism but pointed to positive results for environmental consumerism. Similar results for general social media use and political consumerism were found in the USA, where Gil de Zúñiga, Copeland, et al. (Reference Gil de Zúñiga, Molyneux and Zheng2014) highlighted the mediating effect of general social media use on the relationship between digital media use and political consumerism. The same study also showed a positive connection between general social media use and offline political participation. Results from all over the world seemingly corroborate this positive influence. In Iran, Da Silva Nogueira and Papageorgiou (Reference Da Silva Nogueira and Papageorgiou2020) found evidence for a positive connection between social media use and online political participation, as well as online political information seeking. Mustapha and Omar (Reference Mustapha and Omar2020) confirmed these findings regarding online political participation and generic social media use in Nigeria. However, neither one of those two studies found significant results for general social media use and offline political participation. While Lee et al. (Reference Lee, Shin and Hong2018) also found a positive influence of social media use on political engagement in South Korea, they mentioned that these higher levels of engagement lead to more polarization.

Political Social Networking Site (SNS) Use. Many researchers specifically looked into political use of social media which includes behaviors such as reading political news, expressing political opinions, participating in online polls, or joining online causes and groups (Choi & Kwon, Reference Choi and Kwon2019; Valenzuela, Reference Valenzuela2013). However, although these behaviors fit into the aforementioned categories (e.g., reading political news would fall under 2.2.3 Media Effects), they are often combined into a single instrument, thus generating this new subcategory. Many studies highlight the positive influence of this type of social media use on other kinds of online and offline participation, activism, and political protest (Choi & Kwon, Reference Choi and Kwon2019; Chon & Park, Reference Chon and Park2020; Gainous et al., Reference Gainous, Abbott and Wagner2020; Valenzuela, Reference Valenzuela2013; Vissers & Stolle, Reference Vissers and Stolle2014a).

Mobilizing, Protest. In addition to the political use of social media, some studies also present social media as a tool for organizing and participating in protests, as well as mobilizing people. Studies demonstrate a positive influence when applying overall social media measures (Bond et al., Reference Bond, Fariss, Jones, Kramer, Marlow, Settle and Fowler2012; Costanza-Chock, Reference Costanza-Chock2012; Dey, Reference Dey2020; Hamanaka, Reference Hamanaka2020; Joia & Soares, Reference Joia and Soares2018; Maher & Earl, Reference Maher and Earl2019) and hold true even for the effects showcased by distinct platforms such as Twitter, Facebook, or WhatsApp (Soares et al., Reference Soares, Joia, Altieri and Lander Regasso2021; Valenzuela et al., Reference Valenzuela, Arriagada and Scherman2014). Results from several papers show the possibilities offered by social media to create a form of digital governance that includes citizens in government decision-making (Joia & Soares, Reference Joia and Soares2018; Soares et al., Reference Soares, Joia, Altieri and Lander Regasso2021). Social media is particularly useful to organize protests and initiate mobilizations as it is used more by the vanguards of demonstrations and during times when protests are not already running high (Hamanaka, Reference Hamanaka2020; Valenzuela et al., Reference Valenzuela, Arriagada and Scherman2014). According to Bond et al. (Reference Bond, Fariss, Jones, Kramer, Marlow, Settle and Fowler2012), mobilizing messages on social media does not only influence participation but also self-expression, information seeking, and voting as such mobilization can indirectly influence friends of people initially exposed to them.

SNS Use by Politicians, Legislative Bodies, and Campaigns. Several studies also investigate the adoption of SNS as a communication tool for politicians, political campaigns, political parties, as well as government bodies and institutions. Results thus far attest to the great potential and a steady rise in the implementation of SNS as a valuable political communication tool. This was documented by studies with diverse geographical backgrounds such as the USA, the UK, Sweden, and Germany (Cogburn & Espinoza-Vasquez, Reference Cogburn and Espinoza-Vasquez2011; Gerl et al., Reference Gerl, Marschall and Wilker2018; Housholder & LaMarre, Reference Housholder and LaMarre2013; Jiang, Reference Jiang2017; Penney, Reference Penney2017; Ridge-Newman, Reference Ridge-Newman2020). Other studies/scholars also found a positive influence of social media use for campaigns and politicians for either influencing participation, individual engagement, or political news consumption (e.g., Cogburn & Espinoza-Vasquez, Reference Cogburn and Espinoza-Vasquez2011; Housholder & LaMarre, Reference Housholder and LaMarre2013), even if these effects are at times only marginal (Jensen, Reference Jensen2017; Larsson, Reference Larsson2020). These effects remained consistent across different SNS platform measures such as general SNS use, Facebook, or Twitter use (Amaral et al., Reference Amaral, Zamora, Grandío and Noguera2016; Jensen, Reference Jensen2017; Larsson, Reference Larsson2020). This shows the mobilizing potential of social media. However, more research needs to be conducted as several studies highlight that politicians and legislative bodies struggle to realize the full potential of social media as communicative tools (Amaral et al., Reference Amaral, Zamora, Grandío and Noguera2016; Faria & Rehbein, Reference Faria and Rehbein2016; Pillay, Reference Pillay2019).

3 Social Media and Political Knowledge

This section presents findings from a systematic analysis of the relationship between social media and political knowledge. Mimicking the structure used in the preceding section, we followed the PRISMA guidelines (Moher et al., Reference Moher, Shamseer, Clarke, Ghersi, Liberati, Petticrew, Shekelle, Stewart and Group2015) for the literature review, which was conducted in February 2021 using the following databases: Scopus, Academic Search Ultimate, and Web of Science. The following keywords were used to identify relevant articles: “political knowledge” AND “social media” OR “social networking sites and political knowledge” OR “Twitter and political knowledge” OR “Facebook and political knowledge.” We included articles written in English and published in a peer-reviewed journal (indexed in JCR and/or Scopus) between 2000 and 2021.

The initial search yielded 1,556 results for social media and political knowledge. We then used Zotero to remove duplicates (808 articles) and excluded either non-peer-reviewed, off-topics, or non-English articles (673 articles). Our final sample was 75 articles (see Figure 6).

Figure 6 Social media and political knowledge process.

3.1 Quantitative Analysis

For this analysis, we started with a systematic search for articles relating to political knowledge and social media. After compiling a corpus of published manuscripts, we coded and sorted the articles. We began with a quantitative analysis of the articles before looking into the findings by using a qualitative analysis. The seventy-five articles on the relationship between social media and political knowledge were published in forty different journals. New Media & Society has the highest number of articles (n = 6), followed by the Journal of Information Technology & Politics (n = 5). Twenty-three journals published one article each. A majority of the first authors were based in the USA (n = 38), followed by Austria (n = 6), Hong Kong (n = 4), and the Netherlands (n = 3). Most articles used data gathered in the USA (n = 37), followed by those focusing on China, Sweden, and South Korea (n = 3). Overall, the articles analyzed data from twenty-six countries, and three of them applied a comparative logic and included data from more than one country.

Focusing on time trends, while the first articles on political knowledge and social media were published in 2009, it was not until 2016 that the number of publications increased substantially (reaching sixteen articles published in 2020). Regarding the operationalization of social media use, most researchers either applied a general measure of social media or inquired about different platforms to combine them into one averaged measurement (n = 55). Facebook was the most popular platform for researchers (n = 12), followed by Twitter (n = 5), while some studies focused on other platforms such as Weibo or VK (n = 3). The most common items used for measuring constructs can be found in Tables 18A to 22A.

Regarding methods, seventy-two of the articles drew upon quantitative data, which evidences a clear pattern in the subfield. Two articles used qualitative methods, while one article combined quantitative and qualitative techniques. Most quantitative studies relied on survey data (n = 58), followed by experiments (n = 8), mixed methods (n = 5), and content analysis (n = 4). Most papers (seventy-three) examined the relationship between social media use and political knowledge using samples of citizens, while some focused on subsamples such as students or young people. Additionally, fifty-seven articles utilized cross-sectional data, five used longitudinal data, and thirteen combined both. The quantitative results of this systematic literature review are presented in the following figures (Figures 7 to 9 and Figure 6A in the Appendix). Detailed information on the data used in these figures can be found in Tables 15A to 17A.

Figure 7 Data collection and first author’s origins depicted by country.

Figure 8 Number of studies depicted by applied methods, additionally divided by quantitative or qualitative methods, as well as cross-sectional, longitudinal, and combined data collection.

Figure 9 Number of studies depicted by the primary objects under investigation, additionally divided by social media platform in question.

3.2 Qualitative Analysis: Research Objects and Methods in Social Media and Political Knowledge Research

3.2.1 Main Findings

The biggest takeaway from the studies focused on political knowledge is that there is no agreement on effects regarding directionality, size, or whether there are any effects at all. Social media news use appears to not be influential overall, with studies showing positive, negative, and null results. Interpersonal communication was mostly only found to impact political knowledge if researchers focused on Facebook in particular (discussion and network heterogeneity), but even here the effect seemed conditional on other variables. As for expressive behavior, positive effects on knowledge can be found once said behavior is connected to cognitive elaboration. While traditional education or income are often referenced as influential factors for political knowledge, when it comes to interactions of structural variables and social media, it is ideology that produces the most consistent results.

3.2.2 Measuring Political Knowledge

The measurements for political knowledge vary across studies as the questions used need to be adapted for different countries (e.g., who the leader is, or how certain democratic processes work), including among others: Iran (Alam et al., Reference Alam, Adnan and Kotamjani2019), Indonesia (Astuti & Hangsing, Reference Astuti and Hangsing2018), Denmark (Ohme, Reference Ohme2020), the Philippines (David et al., Reference David, San Pascual and Torres2019), China (Chen & Chan, Reference Chen and Chan2017), and the USA (Beam et al., Reference Beam, Hutchens and Hmielowski2016; Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Feezell & Ortiz, Reference Feezell and Ortiz2019). While many researchers employed the measurements of political knowledge based on the works of Delli Carpini and Keeter (Reference Delli Carpini and Keeter1993, Reference Delli Carpini and Keeter1996), others created variables for specific types of political knowledge such as factual knowledge (Beam et al., Reference Beam, Hutchens and Hmielowski2016; Lee, Reference Lee2020), campaign-related knowledge (Ohme, Reference Ohme2020), general-surveillance political knowledge (Edgerly et al., Reference Edgerly, Thorson and Wells2018; Erkel & Van Aelst, Reference Erkel and Van Aelst2020), or general and issue-specific knowledge (Feezell & Ortiz, Reference Feezell and Ortiz2019; Woo Yoo & Gil de Zúñiga, Reference Woo Yoo and Gil de Zúñiga2014). The operationalizations of political knowledge are quite similar to those of social media studies, varying in the number of items and relying on previous concepts used by Delli Carpini & Keeter (Reference Delli Carpini and Keeter1996) (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018) or Pew Research Center (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Hopp et al., Reference Hopp, Ferrucci, Vargo and Fisher2020).

3.2.3 Media Effects

The discussion of the effects of social media on political knowledge is far from settled, despite the general agreement that traditional news use boosts political knowledge and hopes for social media replicating this influence. While some studies highlighted a positive influence of social media news use on knowledge (Alam et al., Reference Alam, Adnan and Kotamjani2019; Astuti & Hangsing, Reference Astuti and Hangsing2018; Beam et al., Reference Beam, Hutchens and Hmielowski2016; David et al., Reference David, San Pascual and Torres2019; Mwonzora, Reference Mwonzora2020; Ohme, Reference Ohme2020; Park & Kaye, Reference Park and Kaye2019), many have found negative associations (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Chen & Chan, Reference Chen and Chan2017; Erkel & Van Aelst, Reference Erkel and Van Aelst2020; Gil de Zúñiga, Weeks, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017; Heiss & Matthes, Reference Heiss and Matthes2021; Lee, Reference Lee2020) or nonsignificant results (Edgerly et al., Reference Edgerly, Thorson and Wells2018; Feezell & Ortiz, Reference Feezell and Ortiz2019; Gil de Zúñiga & Diehl, Reference Gil de Zúñiga and Diehl2019; Hao et al., Reference Hao, Wen and George2014; Naderer et al., Reference Naderer, Heiss and Matthes2020; Wolfsfeld et al., Reference Wolfsfeld, Yarchi and Samuel-Azran2016; Woo Yoo & Gil de Zúñiga, Reference Woo Yoo and Gil de Zúñiga2014).

As for the effects revolving around other media variables, these are quite similar and include online news reading (Beam et al., Reference Beam, Hutchens and Hmielowski2016), social media news use (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Chen & Chan, Reference Chen and Chan2017; Gil de Zúñiga & Diehl, Reference Gil de Zúñiga and Diehl2019; Gil de Zúñiga, Weeks, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017; Hao et al., Reference Hao, Wen and George2014; Lee, Reference Lee2020; Park & Kaye, Reference Park and Kaye2019), or measures for Facebook (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; David et al., Reference David, San Pascual and Torres2019; Edgerly et al., Reference Edgerly, Thorson and Wells2018; Erkel & Van Aelst, Reference Erkel and Van Aelst2020; Woo Yoo & Gil de Zúñiga, Reference Woo Yoo and Gil de Zúñiga2014) and Twitter (Edgerly et al., Reference Edgerly, Thorson and Wells2018; Erkel & Van Aelst, Reference Erkel and Van Aelst2020; Woo Yoo & Gil de Zúñiga, Reference Woo Yoo and Gil de Zúñiga2014) news use. Some studies have focused on more specific issues, such as using social media to get campaign information (Ohme, Reference Ohme2020) or exposure to political information (Naderer et al., Reference Naderer, Heiss and Matthes2020; Wolfsfeld et al., Reference Wolfsfeld, Yarchi and Samuel-Azran2016).

Regardless of the diversity of the approaches, results remain mixed as to whether researchers applied general social media use for news/information measures or specific ones (e.g., Facebook or Twitter news), thus making it difficult to calibrate the (potential) influence of social media news on political knowledge. In contrast to effects on participation, no distinct influences, such as incidental news exposure (Feezell & Ortiz, Reference Feezell and Ortiz2019), emerged to produce consistent effects. Several experiments and surveys concerning the intervening influence of mood on the relationship of incidental exposure to news, as well as social media news use and knowledge presented by Heiss and Matthes (Reference Heiss and Matthes2021), lead to no significant results. Park and Kaye (Reference Park and Kaye2019), however, found a positive influence of both social media news elaboration and news curation on political knowledge, highlighting that social media news use in interaction with political interest and political efficacy also produced a positive impact.

3.2.4 Interpersonal Communication

Despite evidence supporting a positive connection between people’s interpersonal discussion attributes and political knowledge gain (Hopp et al., Reference Hopp, Ferrucci, Vargo and Fisher2020; Li et al., Reference Li, Lee and Li2016), studies that apply more general measurement instruments yielded null effects, or statistical nonsignificant results (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Naderer et al., Reference Naderer, Heiss and Matthes2020) – regardless whether these attributes for discussion are network size, heterogeneity, or political discussion. Specific platform measurements, such as Facebook discussion (Molaei, Reference Molaei2018) or Facebook network heterogeneity (Hopp et al., Reference Hopp, Ferrucci, Vargo and Fisher2020), however, showed a promising positive influence on political knowledge, while the number of (Facebook) friends (network size) appeared to have no influence (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Naderer et al., Reference Naderer, Heiss and Matthes2020).

Since there are currently fewer studies in this area, these connections need to be further investigated before more precise conclusions can be drawn. The literature should especially help parse out the potential positive connection when platform-specific measurements are used. More research is needed to uncover how to best tap into the potential social networking sites offer citizens when it comes to learning more about politics.

3.2.5 Expressive Political Content

Political Participation

We combined eight items to measure political participation: (1) How often do you do vote in federal or presidential elections? How often you have been involved in the following activities in the past 12 months: (2) attended/watched a public hearing, neighborhood or (3) contacted an elected public official (4) attended a political rally (5) participated in any demonstrations, protests, or marches (6) participated in groups that took any local action for political reform (7) been involved in public interest groups, political action groups, political clubs, political campaigns, or political party committees (8) wrote a letter to the editor of a newspaper? (Cronbach’s α = 0.91, M = 0.46, SD = 0.37).

Political Knowledge

We used eight items to capture knowledge: (1) What job or political office does Mike Pence currently hold? (2) For how many years is a US senator elected – that is, how many years are there in one full term of office for a US Senator? (3) What job or political office does Brett Kavanaugh currently hold? (4) On which of the following does the US federal government currently spend the least? (5) Do you happen to know whether the “For the People Act” bill before Congress was introduced? (6) Which of the following do you think most accurately describes the system of government used in the USA? (7) The WikiLeaks founder Julian Assange was arrested in London to face a charge in the USA of conspiring to hack into a Pentagon computer network in 2010. He had been living in the embassy of which country, which had sheltered him since 2012? (8) Which presidential candidate accused a liberal think tank of undermining Democrats’ chances of taking back the White House in 2020 by “using its resources to smear” contenders? (Guttman’s λ = 0.73, M = 0.35, SD = 0.25).

4.3 Associations Between Social Media News Use, Political Participation, and Political Knowledge

We begin by examining the bivariate correlations between our variables of interest, in each of the years during which data were gathered. Table 1 includes the results (a visual representation of Table 1 can be found in Figure 2A). A consistent and positive correlation between social media news use and political participation was found in all four waves (the highest correlation coefficient in 2013, the lowest one in 2009). Conversely, social media news use was either nonsignificantly correlated with political knowledge (in 2009), or was negatively associated with it (2013–2019; the association is the strongest in the 2019 dataset). These preliminary results do not only underline the incapacity of social media news use to increase political knowledge, but they also suggest that the relationship is negative, in line with a number of studies considered in Section 3 (Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Chen & Chan, Reference Chen and Chan2017; Erkel & Van Aelst, Reference Erkel and Van Aelst2020; Gil de Zúñiga, Weeks, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017; Heiss & Matthes, Reference Heiss and Matthes2021; Lee, Reference Lee2020).

Table 1 Correlation table. Social Media News Use (SMNU), political participation, and political knowledge.

	SMNU 2009	SMNU 2013	SMNU 2015	SMNU 2019
Political Participation	.231**	.351**	.245**	.300**
Political Knowledge	‒.065	‒.097**	‒0.09**	‒.132**

Note: Pearson correlation coefficients; Significance codes **0.01 *0.05.

As an additional test for examining the bivariate relationships between social media use and both participation and knowledge, we employed a K-means algorithmic cluster analysis. This technique is widely used to enable clustering of unstructured data by leveraging Euclidean distances among all data points (Likas et al., Reference Likas, Vlassis and Verbeek2003). In short, this technique offers a valuable and alternative ‘robustness mechanism check’ to reassess how the data might cluster, using our variables of interest as reference points (see, e.g., Gil de Zúñiga et al., Reference Gil de Zúñiga, Scheffauer and Zhang2023). Before conducting this analysis, we rescaled the original measure for knowledge to ensure all variables fell within the same range (from 0 to 1). Subsequently, using SPSS software, we executed two separate K-means clustering analyses: the first incorporating social media news use and political participation, and the second involving social media news use and political knowledge. Moreover, for each year within the sample, we performed an individual analysis to classify the relationship among our variables of interest. The primary aim of this analysis was to ascertain whether respondents from the four different samples clustered in a similar manner. Rather than observing linear relationships as found in the bivariate associations, the K-mean clustering offers comparisons of the distributions as clusters of social media news use, participation (positively), and political knowledge (negatively) across four different US samples over a decade.

In terms of the testing between social media news use and political participation, respondents clustered into two distinct groups. The first group exhibited higher overlapping levels in both social media news use and political participation, while the second group showed lower levels in both aspects. Similarly, concerning social media news use and political knowledge, individuals with higher levels of social media news use consistently grouped with subjects also displaying lower levels of political knowledge. The adequacy of the two-cluster solution proposed varied between fairly good and highly suitable across all models, with silhouette coefficients consistently exceeding 0.5, which suggests a robust ratio of cohesion/separation among the clusters. The closer to 1, the better (Yuan & Yang, Reference Yuan and Yang2019). Descriptive statistics for the various clusters are provided in Figure 10. In the initial stage of K-means cluster analysis, the data is algorithmically divided into two clusters as the best fit. This clustering remains consistent across different years of data collection – 2009, 2013, 2015, and 2019 – where all data points consistently form two clusters. For example, in the analysis of social media news use and political participation, two distinct clusters emerge. Similarly, when examining social media news use and political knowledge, a two-cluster solution is also obtained. As depicted in Figure 10, one cluster comprises data points with lower levels of social media news use overlapping with those exhibiting lower levels of political participation. Conversely, a second cluster consists of data points suggesting higher levels of social media news use, which tend to coincide with higher levels of political participation. Regarding political knowledge, the two clusters indicate how frequent social media news use overlaps with less political knowledge, particularly evident in the 2009 and 2013 datasets. A visual and graphical representation of the most recent available data (2019) can be found in Figures 11 to 12 (figures for all other years are available in Figures 3A to 5A and 7A to 9A). This data visualization compares the distributions of social media news use and political participation within two clusters, based on the 2019 overall sample. The top figures highlight Cluster 1, distinguished by bold red colors indicating high levels of both social media use and political participation, while the softer tone of red in the background represents the distribution of the entire dataset. This allows for a clear comparison between cluster distributions and the overall sample. Conversely, the bottom figures display Cluster 2, characterized by bold red colors indicating low levels of both social media news use and political participation, contrasting with the entire distribution depicted in pale red in the background.

Figure 10 Descriptive data of the K-Mean Cluster Analysis.

Figure 11 K-Means cluster algorithm data distributions for social media news use and political participation in 2019.

Note. The data visualization compares 2019 Overall Sample of Social Media News Use and Political Participation Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Participation. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Participation.

Figure 12 K-Means cluster algorithm data distributions for social media news use and political knowledge in 2019.

Note. The data visualization compares the 2019 overall sample of Social Media News Use and Political Knowledge Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Knowledge. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Knowledge.

Overall, as we have seen in Table 1, an initial bivariate relationship test among SMNU, political participation, and political knowledge established a foundational linear association across all these variables of interest. Furthermore, the K-means algorithmic tests revealed the relationships between these variables more organically and intuitively, presenting a relatively less supervised machine learning model. The findings exposed consistent clusters across the years but indicated frequency differences within these clusters (refer to Figure 12 and Figures 4A–9A). Lastly, to complement this methodological approach, we conducted a series of ordinary least squares (OLS) regressions using social media news use as an independent variable. Our dependent variables were political participation and political knowledge, and a different analysis was run for each year in which data is available (eight models in total). In order to show the basic association between variables, we ran simple OLS regression models that control for general news use as well as socio-demographic variables (i.e., gender, education, age, and income). These multivariate models confirm what we now know from bivariate analyses: There are associations between SMNU and participation and political knowledge beyond the effect of other individual uses of news media, and demographic characteristics.

Although political knowledge could have been measured as a ‘count variable,’ and therefore, another type of regression could have been pursued (i.e., negative binomial, Poisson regressions, etc.), we deemed our approach to be more theoretically nuanced as we do not imply the same human ability for each of the correct political knowledge items. That way, we use an overall index of political knowledge based on Guttman’s Lambda since this statistic would account for differences in the variance that is due to true scores, and when test-takers may differ in their abilities, the λ-2 will be inexorably high, and the error will be low (Estabrook & Neale, Reference Estabrook and Neale2013; Osburn, Reference Osburn2000).The results shown below (Figure 13) support this book’s main claim: While higher levels of social media news use predict more political participation in all four waves considered in the analysis, the effects are negative for political knowledge in three out of four waves (there is no significant effect in 2009 even if the estimated average effect remains negative).

Figure 13 OLS regression coefficients.

Note. Points are B coefficients from the regression models and horizontal bars represent 95% confidence intervals.

4.4 Conclusion

Our analysis of the US original data spanning over a decade largely supports that social media news use correlates positively with political participation. However, it is not associated with political knowledge or, whenever it does, the association is negative. The underlining reasons for these opposite trends have been considered in detail in the previous sections and can be summarized as follows. Social media news use creates and reinforces information networks that end up impacting the likelihood to take an active role regarding politics, in its many different forms. However, the information available, although important to trigger participation, seems to be either little reliable or unable to improve the objective political knowledge of people to the same extent that traditional news outlets do. The implications of these findings are discussed in the final section of this Element.

5 Final Conclusions

5.1 Democratic Impact

It becomes clear in the evolution of measurements and the addition of new items that the internet in general, and social media in particular, has changed how politics are conducted and experienced. Citizens can gather information, contact officials, or sign petitions without leaving their homes. Those in power also use the internet by implementing online campaigns or by directly communicating with their constituents and potential voters.

News and information consumption on social media has proven to have an overwhelmingly positive influence on participation. Social media also allows users to easily accumulate a large network that is often more heterogenous than their offline contacts tend to be. In several studies, network size and heterogeneity are linked with increased participation. Being able to express oneself politically on social media has emerged as another influential factor in participation.

While these findings seemingly attest to the grand positive impact social media has brought to politics, it also appears that several offline issues persist online. There is still a gender divide in participation, although many expected the internet to act as an equalizer. Even in online spaces, women tend to participate less or at least differently than men. Despite the internet and social media offering a cheap way of becoming engaged in politics, as opposed to offline participation, higher income is a predictor of political participation even in the online realm. Only a few studies support the hope that economically disadvantaged people would profit more from these new online affordances. Additionally, the connection between social media, education, and participation is mixed, showing no clear direction.

Overall, and even if inequalities remain, social media appears to have a big positive democratic impact by providing citizens with easily accessible information and news, providing a space for discussion and deliberation, and giving them room to express themselves politically. However, when it comes to political knowledge, the results look rather different. No matter which research area the focus lies on, no distinct direction of the impact on knowledge could be pinned down. Whether it is social media news consumption, political discussion, or political expression, the results are inconclusive with just as many studies providing positive results as there are negative and nonsignificant ones (Amsalem & Zoizner, Reference Amsalem and Zoizner2023). So, despite the positive influence social media has on participation, this is not the case for political knowledge. At best, social media does not really impact knowledge. But at worst, it can even be damaging to the democratic process by decreasing people’s political knowledge. However, where does this leave social media use in relation to the functioning of democracies? Or, to put it in different words, is the promotion of participation among less knowledgeable individuals positive for democracy?

One way to respond to this question is to pay attention to the positive effects of social media news use (i.e., more participation) and discard the rest. In that vein, social media news use will be no panacea for democracies because more knowledge is not guaranteed, but at least we will see some participation shortcomings addressed. This approach is in line with the unfulfilled expectations of social media in the context of democratic ideals. An alternative approach is to consider that the promotion of participation among less knowledgeable individuals is not only far from perfect, but an issue in its terms. In that sense, individuals who display low levels of political knowledge are much more likely to vote for candidates and parties that do not represent their preferences (Arnold, Reference Arnold2012; Bartels, Reference Bartels1996; Fowler & Margolis, Reference Fowler and Margolis2014), do not punish corrupt politicians (Klašnja, Reference Klašnja2017), override informed group decisions (Großer & Seebauer, Reference Großer and Seebauer2016), and fail in the cost-benefit calculation process associated with collective action (Chwe, Reference Chwe1999; Paige, Reference Paige1971). Likewise, individuals’ low levels of political knowledge limit some potential advantages of deliberation (Jennings, Reference Jennings2019; Somin, Reference Somin2006) and complicate identifying which policy preferences are more in agreement with their own judgments (Gilens, Reference Gilens2001).

What is worse, social media use for news fosters the illusion of knowledge as an individual will develop the perception that the “news will find me,” without being active or surveillant about public affairs information (Gil de Zúñiga et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017). Citizens will mistakenly believe they are knowledgeable about politics without doing anything about it, as the news will find them. Accepting that political knowledge is to democratic politics what money is to economics, the currency of citizenship (Delli Carpini & Keeter Reference Delli Carpini and Keeter1996: p. 8), low levels of political knowledge in a society posit important challenges for the functioning of democracies. Some of these challenges can be aggravated when lower levels of political knowledge are combined with high participation.

In this vein, rational ignorance theory predicts that some individuals may choose not to be informed about politics because of the costs associated with more knowledge (Somin, Reference Somin, Congleton, Grofman and Voigt2019). In contexts where too much information about public affairs is available, an increasing amount of people may choose to stay away from political news not because they do not care at all about it, but because of time and/or intellectual constraints, practicing news avoidance (Skovsgaard & Andersen, Reference Skovsgaard and Andersen2020), which may be distinct to sheer information overload (Goyanes et al., Reference Goyanes, Ardèvol-Abreu and Gil de Zúñiga2021). Although rational ignorance has been said to have some positive effects (e.g., it may liberate time for altruistic individuals to take part in social activities that are beneficial for people other than themselves), a general agreement exists on the negative consequences of political ignorance for participation in representative democracies (Somin, 2014; Reference Somin, Congleton, Grofman and Voigt2019). This further supports our claim of a democratic mirage derived from social media news use, unless alternative mechanisms exist that allow uninformed individuals to make decisions that are in line with their interests.

A well-established line of research in political psychology revolving around shortcuts and cognitive heuristics has shown that even uninformed individuals are sometimes able to make decisions that are in line with their preferences (Schaffner & Streb, Reference Schaffner and Streb2002). For example, individuals can use party labels as an effective cue for voting even if they do not know the specific policy positioning of candidates in a given election or look at what a group of interest with which they are often in agreement is saying about a topic that is new to them. Even relatively well-informed individuals are likely to recur to heuristics from time to time due to the (increasingly) unmanageable amount of information that is available (Kuklinski & Quirk, Reference Kuklinski, Quirk, Lupia, McCubbins and Popkin2000).

Effective heuristics could be a challenge to our main thesis, given that the negative effects of social media news use in political knowledge could be heavily cushioned. However, while acknowledging the importance of heuristics in different situations, we believe that important shortcomings associated with them reinforce our democracy mirage approach (e.g., policy stereotypes, the influence of easy arguments, or biased processing of information). Basically, even if a small amount of political information could be used as a compass to navigate complex political discussions when cues are operating – the imperfect processing of that information is likely to cause additional troubles.

Political knowledge is far from solving all these problems, but it makes it more likely for individuals to behave in ways that are in agreement with their preferences. This can be seen in voting, but it has consequences well beyond, provided that accurate information can be key for signing certain petitions while declining others, taking part in demonstrations that are in the best interest of the individual, knocking on the door of the public representative who can help solve certain issues or prevent the spread of inaccurate and/or fake news.

Our democracy mirage claim should not be read as supporting an elitist understanding of democracy where only highly informed people participate, but rather as a warning about the blind spots that democratic systems may develop as the use of social media for news spread. Key in this regard is the cost of opportunity associated with social media news use, reducing the time available to get informed in more traditional media outlets that increase political knowledge (Chaffee & Kanihan, Reference Chaffee and Kanihan1997; Delli Carpini & Keeter, Reference Delli Carpini and Keeter1996). Overall, individuals may not become less knowledgeable the more they consume news on social media, but this seems a plausible scenario unless additional sources of information are considered. At the very least no improvements in political knowledge are expected on average if additional sources of information beyond social media are lacking.

5.2 Suggestions for Future Research

The analyses conducted so far have identified different aspects that need further development and closer attention from researchers. For the most part, both knowledge and participation research suffer from the same issues that demand ameliorations. Those are, more saliently, the applied measurements, that need to be reinvestigated, updated, and refined. Moreover, there are several areas that are understudied and need more attention, such as qualitative aspects, network, and structural influences, as well as cross-country-differences.

5.2.1 Social Media and How It Is Measured

The first and biggest challenge revolves around the operationalization of social media. This is due to a variety of concepts within social media (e.g., news use, discussion, expression, etc.), which come up in various research and have different measurements. Even within these areas, the applied measures differ substantially (e.g., discussion network size, network heterogeneity, discussion frequency, etc.). In short, the researchers introduce different constructs to measure the same concepts and when they measure very similar, or if not the same phenomena, they use different items.

What is, thus, often summarized into positive or negative effects of social media news use on political knowledge and participation, come from the number of hours or the way people use those platforms to organize protest. While the variety of measurements does make it difficult to compare results, it also suggests that social media is many-facetted, and applying general measures is an oversimplification, at least to better explain some of the social media effects. Those measures produced mixed and at times negative results when it comes to participation (Gil de Zúñiga, Reference Gil de Zúñiga2012; Gil de Zúñiga, Copeland, et al., Reference Gil de Zúñiga, Copeland and Bimber2014b; Kamau, Reference Kamau2017; Pasek et al., Reference Pasek, more and Romer2009; Theocharis & Lowe, 2016b) or knowledge (Beckers et al., Reference Beckers, Van Aelst, Verhoest and D’Haenens2020; Gil de Zúñiga, Barnidge, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017; Park, Reference Park2019; Park & Kaye, Reference Park and Kaye2019), while specific measures such as political (Choi & Kwon, Reference Choi and Kwon2019; Chon & Park, Reference Chon and Park2020; Gainous et al., Reference Gainous, Abbott and Wagner2020; Valenzuela, Reference Valenzuela2013; Vissers & Stolle, Reference Vissers and Stolle2014a) or protest use (Bond et al., Reference Bond, Fariss, Jones, Kramer, Marlow, Settle and Fowler2012; Costanza-Chock, Reference Costanza-Chock2012; Dey, Reference Dey2020; Hamanaka, Reference Hamanaka2020; Joia & Soares, Reference Joia and Soares2018; Maher & Earl, Reference Maher and Earl2019) can lead to positive outcomes. In the case of political knowledge, only very specific measurements (e.g., political expression via WhatsApp or Facebook network heterogeneity) produced positive results.

Thus, in order to gauge an accurate and reliable picture of the influence of social media, the measurements need to be updated and specified. Since the first studies were conducted on SNS such as Facebook or Twitter, those platforms have changed and evolved, gaining new affordances that need to be accounted for and reflected (Valenzuela et al., Reference Valenzuela, Correa and Zúñiga2018). These functions and distinct reasons for use might even vary across platforms (Facebook, Twitter, YouTube, etc.), thus researchers should apply separate measures for those as well. As can be seen from the literature review on knowledge, overall social media measures appeared to be noninfluential, whereas studies that disentangled the influence of social media platforms showcased more nuanced and positive findings.

This is a clear indicator that various mechanisms are at work depending on the platform and that citizens use those platforms differently as well. Hence, more attention needs to be paid to these details to get an accurate picture of any possible positive or negative influence. Thus, researchers need to develop and propagate more specific measurements by capturing the affordances of each social media platform and being consistent when applying them.

5.2.2 Political Participation?

Similar to social media, the measurements for political participation also vary across studies. This ranges from separate measures for online (Bode & Dalrymple, Reference Bode and Dalrymple2016; Chan et al., Reference Chan, Chen and Lee2017; Chunly, Reference Chunly2019; Hoffman, Reference Hoffman2012; Steinberg, Reference Steinberg2015; Yamamoto et al., Reference Yamamoto, Nah and Bae2019) and offline participation (Garcia-Castañon et al., Reference Garcia-Castañon, Rank and Barreto2011; Gil de Zúñiga, Molyneux, et al., Reference Gil de Zúñiga, Molyneux and Zheng2014; Kwak et al., Reference Kwak, Lane, Weeks, Kim, Lee and Bachleda2018; Lane et al., Reference Lane, Kim, Lee, Weeks and Kwak2017), or low and high effort participation (Nanz et al., Reference Nanz, Heiss and Matthes2020) to distinct captures for civic engagement (Zhang & Skoric, Reference Zhang and Skoric2018), protest (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Karakaya & Glazier, Reference Karakaya and Glazier2019; Zumarraga-Espinosa, Reference Zumarraga-Espinosa2020), voting (Diehl et al., Reference Diehl, Barnidge and Zúñiga2019; Hassim et al., Reference Hassim, Zian and Jayasainan2020; Towner, Reference Towner2013), campaign and party activities (Amaral et al., Reference Amaral, Zamora, Grandío and Noguera2016; Cogburn & Espinoza-Vasquez, Reference Cogburn and Espinoza-Vasquez2011; Gerl et al., Reference Gerl, Marschall and Wilker2018; Housholder & LaMarre, Reference Housholder and LaMarre2013; Jensen, Reference Jensen2017; Jiang, Reference Jiang2017; Larsson, Reference Larsson2020; Penney, Reference Penney2017; Ridge-Newman, Reference Ridge-Newman2020), activism (Gil de Zúñiga et al., Reference Gil de Zúñiga, Ardevol-Abreu and Casero-Ripolles2019; Zhang & Skoric, Reference Zhang and Skoric2018), political expression (Chapman & Coffé, Reference Chapman and Coffé2016) and consumerism (Gil de Zúñiga, Reference Gil de Zúñiga2012; Kim & Chen, Reference Kim and Chen2015; Zhang & Skoric, Reference Zhang and Skoric2018).

While some researchers focus on one or several of these measures, others use an overall participation measurement that includes and combines some of these aspects (Graham et al., Reference Graham, Tang and Mahoney2020; Lee & Xenos, Reference Lee and Xenos2020; Shaw et al., Reference Shaw, Foster and Combs2020). With such a variety of outcome variables, it is difficult to compare the effects. What is, however, evident, is that there is merit in such specific measures allowing us to differentiate possible distinct influence and effects. Previous research has shown that online and offline participation are different concepts, so if anything, at least this distinction should be made (Gibson & Cantijoch, Reference Gibson and Cantijoch2013; Gil de Zúñiga et al., Reference Gil de Zúñiga, Veenstra, Vraga and Shah2010; Kim et al., Reference Kim, Russo and Amnå2017; Oser et al., Reference Oser, Hooghe and Marien2013).

Additionally, the measurements should be updated and specified to reflect new forms of and developments in participation. For example, asking whether somebody contacted a government official has been included in the participation measures since before online participation became relevant. However, nowadays there are both online and offline ways of approaching certain participatory actions such as contacting (e.g., via social media or writing an email in addition to writing letters or calling a number, wearing a campaign button or sticker and changing a profile picture, discussing politics online and offline). Thus, research needs some measurements that reflect these intricate nuances and developments instead of combining them all into one variable.

5.2.3 Structural Influence

When looking at the sample for the literature reviews on social media, political participation, and political knowledge, it becomes evident that structural influences are somewhat neglected. The studies that feature those elements highlight the possible impact of variables, such as gender (Morris & Morris, Reference Morris and Morris2013; Vicente & Novo, Reference Vicente and Novo2014; Xenos et al., Reference Xenos, Vromen and Loader2014), income (Bode & Dalrymple, Reference Bode and Dalrymple2016; Chunly, Reference Chunly2019; Garcia-Castañon et al., Reference Garcia-Castañon, Rank and Barreto2011), education (Ahmed & Cho, Reference Ahmed and Cho2019; Bode, Reference Bode2017), ideology (Hjorth & Adler-Nissen, Reference Hjorth and Adler-Nissen2019; Kanihan & Rim, Reference Kanihan and Rim2018), or age (Bode, Reference Bode2017; Chunly, Reference Chunly2019; Strandberg, Reference Strandberg2014).

Results, however, are still largely inconclusive. Especially when it comes to different countries the effect of structural influences could vary substantially. For example, while SNSs were heralded as great equalizers when it comes to participating and manifesting big impacts during the Arab Spring, the results from various studies show that there are still stark differences based on gender or income that are not eradicated by social media. Several studies presented in our literature review show that gender differences might be more pronounced online. Thus, more extensive research is warranted to disentangle the intervening influence of structural factors. Additional results are needed to clarify which variables are important, in which direction the effect goes, and whether those effects hold across borders.

5.2.4 Network Effects

Studies considering people’s social network attributes and its effects are overall scarce, and include different variables in the analyses (e.g., network heterogeneity, network size, political discussion with weak or strong ties, etc.). Considering that there are only a few more studies conducted in this area than on structural influences and the number of different variables, there is limited research on each of those factors to draw any robust inferences. Thus, more research is needed, also when it comes to different social media outlets (e.g., TikTok, YouTube, Facebook, Instagram, etc.), specifically for political knowledge measurements focusing on Facebook instead of overall social media highlight positive influence. Circling back to the first point in this section, this shows that more intricate measurements need to be applied to gauge the true influence of social media. We also encourage scholars to conduct more research on the effect of other social media like Twitter since not all effects occur the same way on each SNS.

5.2.5 US Centricity

The majority of studies in both literature reviews are from the USA. It is, however, important to highlight that many researchers concluded that such phenomena are not one-size-fits-all since different political situations and regimes across cultures could produce different results. Censorship, for instance, is more extreme in other countries than in the USA, thus substantially influencing what people can even access or express via social media. Considering that the USA has found itself in a rather unconventional situation in recent years due to President Donald Trump, the results might not be generalizable globally.

Especially when looking at political knowledge, the USA shows mostly negative or nonsignificant influence of social media (Barnidge et al., Reference Barnidge, Ardèvol-Abreu and Zúñiga2018; Cacciatore et al., Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and Corley2018; Edgerly et al., Reference Edgerly, Thorson and Wells2018; Feezell et al., Reference Feezell, Conroy and Guerrero2009; Feezell & Ortiz, Reference Feezell and Ortiz2019; Kanihan & Rim, Reference Kanihan and Rim2018; Lee, Reference Lee2020; Lee & Xenos, Reference Lee and Xenos2019; Pasek et al., Reference Pasek, more and Romer2009). However, other countries such as Iran (Alam et al., Reference Alam, Adnan and Kotamjani2019), Indonesia (Astuti & Hangsing, Reference Astuti and Hangsing2018), Austria (Heiss & Matthes, Reference Heiss and Matthes2021), China (Li et al., Reference Li, Lee and Li2016), Zimbabwe (Mwonzora, Reference Mwonzora2020), Denmark (Ohme, Reference Ohme2020), South Korea (Park & Kaye, Reference Park and Kaye2019), and Chile (Valenzuela et al., Reference Valenzuela, Bachmann and Bargsted2019) show more positive results. This highlights how different outcomes can be, thus warranting more research focused outside of the USA.

5.2.6 Study Designs

Based on our close investigation of prior research, we recommend an update of study designs. First, the vast majority of studies feature quantitative designs. However, as prior research has shown how very distinct uses of social media can be impactful, more qualitative research (e.g., in-depth interviews, focus groups) is needed to further investigate those nuances. This could be vital to discover how, for example, Facebook and Twitter are used in contrast to each other (e.g., one platform might be preferred when discussing but the other for news). Additionally, that could also help distinguish different uses within one specific SNS (e.g., political or protest use of Facebook, etc.) and uncover potential new important applications of those SNSs. A better understanding of details would also allow for more precise and impactful measurements to be developed for quantitative designs.

Second, the most popular method among researchers is a survey, which means that data are from self-reported measurements. Other ways of assessing subjects’ online behaviors such as (real-time) experiments need to be promoted. Finally, if quantitative survey research is conducted, we should strive for more panel data. The present research is mostly based on cross-sectional datasets which do not allow for lagged or autoregressive testing. We recommend the use of lagged or autoregressive analyses to get a more accurate idea of the strength of effects in time, and to investigate the dynamic influence that might occur.

5.3 All in All …

Overall, the comprehensive analysis presented in this book sheds light on the transformative impact of the internet and social media on the political realm, particularly regarding individuals’ political behavior and acquisition of public affair news and knowledge. While social media provides convenient access to information and encourages political expression, it also preserves offline inequalities, especially regarding gender and socio-economic status. Despite its positive influence on political participation, social media (news use) effects on political knowledge remain ambiguous, at best. The proliferation of news on social media platforms may foster a false sense of political awareness, aggravating democratic challenges (see, e.g., the “News Finds Me” Perception Theory, Gil de Zúñiga & Cheng, Reference Gil de Zúñiga, González-González and Goyanes2021). While cognitive shortcuts like party affiliations offer some guidance, they may also reinforce biases and undermine an informed public opinion decision-making. Overall, we contend that enhancing political knowledge is key for cultivating an informed citizenry and mitigating the adverse effects of social media-driven politics. This Element underscores the need to address the limitations of social media, especially as a primary news source, and advocates for a balanced approach that integrates diverse information channels to uphold democratic values and foster strong civic literacy.

Appendix

Social Media and Political Participation

Figure 1A Number of publications in the top five journals with most published articles by year.

Figure 2A Pearson’s correlation coefficients. Visualization from Table 1, Section 4. Social media news use, political participation, and political knowledge.

Figure 3A K-Means cluster algorithm data distributions for social media news use and political participation news and political participation in 2009.

Note. The data visualization compares 2009 Overall Sample of Social Media News Use and Political Participation Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Participation. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Participation.

Figure 4A K-Means cluster algorithm data distributions for social media news use and political participation news and political participation in 2013.

Note. The data visualization compares 2013 Overall Sample of Social Media News Use and Political Participation Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Participation. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Participation.

Figure 5A K-Means cluster algorithm data distributions for social media news use and political participation news and political participation in 2015.

Note. The data visualization compares 2015 Overall Sample of Social Media News Use and Political Participation Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Participation. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Participation.

Table 1A List of journals for social media and political participation papers.

Journals	Number of articles
New Media & Society	31
Information Communication & Society	27
Journal of Information Technology & Politics	25
International Journal of Communication	18
Science Computer Review	12
International Journal of Press/Politics; Journal of Broadcasting & Electronic Media	11
Communication Research; Media, Culture & Society; Social Media and Society	7
Government Information Quarterly; Information Communication & Society; Journal of Communication; Policy and Internet; Political Communication	6
Journal of Youth Studies; Journal of Political Marketing; Chinese Journal of Communication; Public Relations Review; Telematics and Informatics	5
Asian Journal of Communication; Australian Journal of Political Science; Communication & Society; Convergence; International Journal of Electronic Governance; Journalism and Mass Communication Quarterly; Malaysian Journal of Communication	4
American Behavioral Scientist; Communication Research Reports; Communication Studies; Digital Journalism; Korea Observer; Political Behavior; SAGE Open; The British Journal of Politics and International Relations	3
African Journalism Studies; Atlantic Journal of Communication; Communication Review; Computers in Human Behavior; Democratization; E-Journal of E-Democracy and Open Government; Electoral Studies; Feminist Media Studies; Geopolitics, History, and International Relations; Global Media and Communication; Information Polity; International Journal of Scientific and Technology Research; Journal of African Media Studies; Journal of Computer-Mediated Communication; Journal of Language and Politics; Journal of Media Psychology; KOME; Local Government Studies; Mass Communication and Society; Media International Australia; Observatorio; Political Research Quarterly; Politics; Quality and Quantity; Scandinavian Political Studies; Social Movement Studies; Social Science Journal; South East Asia Research; VOLUNTAS	2
Acta Política; Adcomunica – Revista Científica de Estrategias Tendencias e Innovación en Comunicación; Africa Spectrum; Akdeniz Iletisim; Annals of the American Academy of Political and Social Science; Asian Journal of Political Science; Australian Journal of Psychology; Behaviour & Information Technology; British Journal of Social Psychology; Canadian Review of Sociology; Catalan Journal of Communication and Cultural Studies; Central Asia and the Caucasus; Children & Youth Services Review; Chinese Sociological Review; Citizenship Studies; Cogent Social Sciences; Communications: The European Journal of Communication Research; Comunicar; Critical Discourse Studies; Critical Sociology; Cuadernos Info; Demokratizatsiya; Doxa Comunicacion; East Asia: An International Quarterly; East European Politics & Societies; Electronic Government; Ethnos: Journal of Anthropology; Europe’s Journal of Psychology; French Politics; Human Communication Research; Intellectual Discourse; International Journal of Criminology and Sociology; International Journal of Politics, Culture and Society; International Journal of Public Opinion Research; International Journal of Strategic Communication; International Journal of Web Based Communities; International Political Science Review; International Review of Administrative Sciences; Intersections East European Journal of Society and Politics; Javnost: The Public; Journal of Applied Journalism & Media Studies; Journal of Asian Public Policy; Journal of Baltic Studies; Journal of Behavioral Health Services & Research; Journal of Communication Inquiry; Journal of Consumer Culture; Journal of Contemporary African Studies; Journal of Contemporary China; Journal of Contemporary Ethnography; Journal of Content, Community and Communication; Journal of Ethnic and Migration Studies; Journal of Muslims in Europe; Journal of Political Power; Journal of Political Science Education; Journal of Public Affairs; Journal of Urban Affairs; Journalism Practice; Jurnal the Messenger; Kajian Malaysia: Journal of Malaysian Studies; Learning, Media and Technology; Media and Communication; Middle East Journal of Culture and Communication; Mind, Culture & Activity; Mobile Media & Communication; Nature; Nordicom Review; Palgrave Communications; Party Politics; Perspectives on Politics; Policy Studies; Political Psychology; Political Science Research and Methods; Political Studies; Problems of Post-Communism; Public Opinion Quarterly; Public Policy and Administration; Qualitative Inquiry; Daedalus; Qualitative Research; Rationality and Society; Revista de Cercetare si Interventie Sociala; Revista Espanola de Ciencia Política; Revista Latina de Comunicación Social; Romanian Journal of Political Science; Round Table; Social and Economic Studies; Social Identities; Social Science Quarterly; Social Semiotics; Information Technology for Development; Social Work (United States); Sociological Perspectives; Sociology; Stability; Studies in Communication Sciences; Technology in Society; Terrorism and Political Violence; The Journal of Legislative Studies; Third World Quarterly; Transforming Government: People, Process and Policy; Violence Against Women; Visual Communication; Young	1

Table 2A List of first author’s country of origin for social media and political participation papers.

First author’s country of origin	Number of papers
USA	154
UK	30
Hong Kong	21
Australia	19
Canada	17
Austria	15
Germany	14
Spain; Sweden	12
Italy	10
South Korea	9
India; Israel; Netherlands	8
Chile; China	7
Brazil; Denmark; Taiwan	6
Belgium; Finland; Malaysia; Nigeria	5
Indonesia; Ireland; Mexico; New Zealand; Norway; Portugal; Singapore; South Africa	4
Columbia; France; Hungary; Macau	3
Estonia; Pakistan; Poland; Russia; Switzerland; Turkey	2
Afghanistan; Croatia; Cyprus; Czech Republic; Ecuador; Egypt; Japan; Kazakhstan; Lithuania; Niger; North Korea; Northern Cyprus; Philippines; Zimbabwe	1

Table 3A List of countries of data collection for social media and political participation papers.

Country	Number of papers
USA	127
Multi-country	52
China	21
UK	19
South Korea	18
Sweden	11
Australia; Hong Kong; India	10
Canada; Germany; Spain	9
Austria; Not specified country	8
Chile; Denmark; Italy; Taiwan	7
Brazil; Israel; Nigeria	6
Belgium; Finland; Malaysia; Netherlands	5
Indonesia; Mexico; Singapore	4
Egypt; Ireland; Russia; Turkey; Zimbabwe	3
Bulgaria; Cambodia; Columbia; Croatia; Ethiopia; France; Greece; Hungary; Japan; Morocco; Pakistan; Philippines; South Africa; Ukraine	2
Afghanistan; Czech Republic; Ecuador; Estonia; Ghana; Iran; Kazakhstan; Lebanon; New Zealand; Nigeria; Northern Cyprus; Norway; Portugal; Switzerland; Uganda; Venezuela; Vietnam	1

Table 4A Most common measurement instruments for political participation online.

Number of items	Scale	Items	Used by
6	1−10	How often during the past twelve months did you engage or not in any of the following activities? igned or shared an online petition participated in online political polls participated in an online question-and-answer session with a politician or public official created an online petition signed up online to volunteer to help with a political cause used a mobile phone to donate money to a campaign or political cause via text message or app	(Reference Saldaña, Mcgregor and ZúñigaSaldaña et al., 2015)
4	1−4	How often during the past month did you engage or not in any of the following activities? contacted a politician using emails or social media signed an online petition made an online campaign contribution signed up to volunteer for an online campaign	(Reference Park and KayePark & Kaye, 2018a)
5	0−10	How often during the past twelve months did you engage or not in any of the following activities? written to a politician made a campaign contribution subscribed to a political listserv signed up to volunteer for a political campaign written to a news organization	(Gil de Zúñiga, Molyneux, et al., 2014)
6	0/1 per item, then recoded into low/high (0−5)	How often during the campaign did you engage or not in any of the following activities? forwarded a political email to another person talked to any people and tried to show them online why they should vote for or against one of the parties or candidates followed or became a fan of a political candidate on a social network posted a comment or weblink on a blog, social network, or website to express a political opinion participated in online discussion or chat groups about politics given an online donation to a candidate or political party.	(Reference TownerTowner, 2013)
7	1−5	How often during the past twelve months did you engage or not in any of the following activities? registering one’s opinions by participating in online polls getting information about a candidate’s voting record sending or receiving campaign-related emails contributing money online to a candidate running for public office looking for online information about candidates’ positions on the issues finding out about endorsements or ratings of candidates by organizations or individuals online checking the accuracy of claims made by or about the candidates online	(Reference Zhang, Seltzer and BichardZhang et al., 2013)
8	1−4	How often during the past twelve months did you engage or not in any of the following activities? signed a petition via a link on Facebook; helped collect signatures for a petition by sharing a link on your Facebook page joined a political group on Facebook showed your support for a cause, candidate, campaign and/or stand on an issue by liking a Facebook page making a donation to a political group, candidate, and/or campaign via a link on Facebook showed your support for a cause, candidate, campaign, political group and/or stand on an issue by sharing a link or post on Facebook contacted an elected official via a link on Facebook tried to persuade other people via Facebook to support a cause, candidate and/or stand on an issue	(Reference CaoCao, 2020)

Table 5A Most common measurement instruments for political participation offline.

Number of items	Scale	Items	Used by
9	1−10	How often during the past twelve months did you engage or not in any of the following activities? attended/watched a public hearing, neighborhood or school meeting contacted an elected public official attended a political rally participated in any demonstrations, protests, or marches donated money to a campaign or political cause participated in groups that took any local action for social or political reform been involved in public interest groups, political action groups, political clubs, political campaigns, or political party committees voted in local/statewide elections voted in federal/presidential elections	(Reference Saldaña, Mcgregor and ZúñigaSaldaña et al., 2015)
6	1−4	How often during the past month did you engage or not in any of the following activities? posted a campaign sticker, banner, or button called or sent a letter to an elected public official attended a political meeting, rally, or speech worked for a political party or a candidate contributed money to political campaigns or candidates participated in groups that took any action for political reform	(Reference Park and KayePark & Kaye, 2018a)
7	0−10	How often during the past twelve months did you engage or not in any of the following activities? attended a public hearing, town hall meeting, or city council meeting called or sent a letter to an elected public official spoken to a public official in person attended a political rally participated in any demonstrations, protests, or marches participated in groups that took any local action for social or political reform been involved in public interest groups, political action groups, political clubs, or party committees	(Gil de Zúñiga, Molyneux, et al., 2014)
5	0/1 per item, then recoded into low/high (0−5)	How often during the campaign did you engage or not in any of the following activities? talked to any people and tried to show them why they should vote for or against one of the parties or candidates gone to any political meetings, rallies, speeches, dinners, or things like that in support of a particular candidate worn a campaign button, put a campaign sticker on their car, or placed a sign in their window or in front of their house worked for one of the parties or candidates given an offline donation to a candidate or political party	(Reference TownerTowner, 2013)
5	1−5	How often during the past twelve months did you engage or not in any of the following activities? talking to people and trying to show them why they should vote for or against one of the parties or candidates wearing a campaign button or putting a campaign sticker on one’s car or placing a sign in one’s window or in front of one’s house attending political meetings, rallies, or speeches giving money to a candidate or a political party doing any work to help a candidate in his or her race during the past two years	(Reference Zhang, Seltzer and BichardZhang et al., 2013)

Table 6A Most common measurement instruments for social media news use.

Number of items	Scale	Items	Used by
8		How often use the following social media platforms to get news? Twitter Facebook Reddit Google+ LinkedIn Tumblr Instagram Pinterest	(Reference Saldaña, Mcgregor and ZúñigaSaldaña et al., 2015)
3	1−7	How often do you use social media for the following activities? getting/posting news information engaging in current events and public issues	(Reference Kim and ChenKim & Chen, 2015)
3	1−5	How often do you use social media such as Twitter or Facebook for the following activities? to get updates about the community to share news about politics and society to get tips on political news and events	(Reference Stromback, Falasca and KruikemeierStromback et al., 2018)
1	1−4	How often use the following social media platforms to get news? Facebook Twitter Country-specific examples were included in the questionnaire	(Reference Ahmed and ChoAhmed & Cho, 2019)

Table 7A Most common measurement instruments for incidental news exposure.

Number of items	Scale	Items	Used by
3		How much do you agree with the following statements? stumble across news only by accident only see political posts when other people from their network post about politics, and do not seek political information, but sometimes see political information by accident	(Reference Heiss, Knoll and MatthesHeiss et al., 2020)
1	1−7	How often are you accidentally exposed to political posts in your social media newsfeed without having actively searched for or intentionally included such information in your newsfeed (e.g., by “liking” news sources or politicians)?	(Reference Heiss and MatthesHeiss & Matthes, 2019)
3	1−4	When you use social networks/social media platforms (e.g., Facebook, Twitter, YouTube, etc.) how often do you come across news and information on the following when you may have been going online for a purpose other than to get the news? current events public issues politics	(Reference Lee and XenosLee & Xenos, 2020)

Table 8A Most common measurement instruments for second/dual screening.

Number of items	Scale	Items	Used by
3	1−5	Over the last two months, how often did you use social networks/social media platforms to read about and discuss televised news programs while aired? during news programs during campaign speeches, interviews, or debates involving politicians during news coverage of the elections	(Reference Vaccari and ValerianiVaccari & Valeriani, 2018b)
3		How often do you engage in dual screening while watching TV programs about the following topics? political speeches or debates news and public affairs information about “election coverage”	(Reference Gil de Zúñiga and LiuGil de Zúñiga & Liu, 2017)
4	1−5	Four items measuring dual-screening use were self-generated when asking the extent to use various social media platforms (e.g., mobile instant messaging and social networking sites, such as Facebook, Weibo, and Instagram; video and TV sites/apps) while viewing current affair-related videos on two screen devices.	(Reference LinLin, 2019)

Table 9A Most common measurement instruments for political discussion.

Number of items	Scale	Items	Used by
7	1−7	Through seven items, participants rated how often during the last year they had joined online political discussions (e.g., joining political discussion through Facebook or Twitter)	(Reference Alberici and MilesiAlberici & Milesi, 2018)
9		How frequently do you engage in discussions about politics and public affairs with the following people? spouse or partner family and relatives friends acquaintances strangers neighbors you know very well neighbors you do not know well co-workers you know well co-workers you do not know well	(Gil de Zúñiga, Diehl, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017)

Table 10A Most common measurement instruments for political discussion heterogeneity.

Number of items	Scale	Items	Used by
3	1−7	How often do you talk about social or public issues on social media with the following? people who do not share their age, socio-economic status, or gender people who disagree with their views people who agree with their views	(Reference Kim and ChenKim & Chen, 2015)
4	1−10	How often do you talk about politics or public affairs online and offline with the following? people who disagree people with different political views people from different race or ethnicity people from different social class	(Reference Yoo and Gil de ZúñigaYoo & Gil de Zúñiga, 2019)
3		Who would you include as Facebook friends? only acquaintances people whom one has met only once or twice even strangers How would you describe your Facebook friends? Most of them are of similar background. Many of them are from different backgrounds, but still a significant portion are from similar background. They come from a variety of backgrounds. How many of your Facebook friends are of a similar age? 0−20% 21−40% 41−60% 61−80% 81−100%	(Reference Tang and LeeTang & Lee, 2013)

Table 11A Most common measurement instruments for political discussion network size.

Number of items	Items	Used by
1	What is the rough number of friends linked to your Facebook account?	(Reference Tang and LeeTang & Lee, 2013)
1	What is the size of your political discussion network on Facebook?	(Reference CaoCao, 2020)
1	The size of an individual’s online social network is measured along a six-point scale according to respondents’ estimates of how many “people are on your list of ‘followers,’ ‘friends,’ ‘connections,’ or contacts” on the SNS that they use most often.	(Reference Kahne and BowyerKahne & Bowyer, 2018)
2	With how many people did you talk about politics or public affairs during the past month in the following settings? face-to-face or on the phone via the internet (including chat rooms and social networking sites)	(Gil de Zúñiga, Diehl, et al., Reference Gil de Zúñiga, Weeks and Ardèvol-Abreu2017)

Table 12A Most common measurement instruments for social media political expression.

Number of items	Scale	Items	Used by
14	1−6	How frequently did you perform one of the following behaviors (in regards to political content) on the four platforms listed below? Facebook: 5 items Twitter: 7 items Snapchat: 1 item Instagram: 1 item sharing liking tweeting etc.	(Reference Kwak, Lane, Weeks, Kim, Lee and BachledaKwak et al., 2018)
5		How often do you use social networking sites to do the following? posting personal experiences related to politics or campaigning friending a political advocate or politician posting or sharing thoughts about politics posting or sharing photos, videos, or audio files about politics forwarding someone else’s political commentary to other people	(Gil de Zúñiga, Molyneux, et al., Reference Gil de Zúñiga, Molyneux and Zheng2014)
1	0−4	How often have you participated in a campaign that involved changing your profile picture?	(Reference Chapman and CofféChapman & Coffé, 2016)
2		How often do you share information on social media about the following? social issues political issues	(Reference Chan, Chen and LeeChan et al., 2017)

Table 13A Most common measurement instruments for general social media use.

Number of items	Scale	Items	Used by
6	1−10	On a typical day, how much time do you spend on online social networking sites? How much do you use the Internet for social networking? To what extent do social networking sites help you to do the following? stay in touch with family and friends meet people who share my interests stay informed about my local community get news about current events through family and friends	(Gil de Zúñiga, Copeland, et al., Reference Gil de Zúñiga, Copeland and Bimber2014a)
4	1−5	What is your level of use for the following? Facebook Twitter Instagram blogs	(Reference Mustapha and OmarMustapha & Omar, 2020)

Social Media and Political Knowledge

Figure 6A Number of publications in the top five journals with most published articles by year.

Figure 7A K-Means cluster algorithm data distributions for social media news use and political knowledge in 2009.

Note. The data visualization compares 2009 Overall Sample of Social Media News Use and Political Knowledge Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and Low Political Knowledge. Bottom figures represent Cluster 2: Low Social Media News Use and High Political Knowledge.

Figure 8A K-Means cluster algorithm data distributions for social media news use and political knowledge in 2013.

Note. The data visualization compares 2013 Overall Sample of Social Media News Use and Political Knowledge Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Knowledge. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Knowledge.

Figure 9A K-Means cluster algorithm data distributions for social media news use and political knowledge in 2015.

Note. The data visualization compares 2015 Overall Sample of Social Media News Use and Political Knowledge Distributions within the two clusters. Top figures correspond to Cluster 1: High Social Media Use and High Political Knowledge. Bottom figures represent Cluster 2: Low Social Media News Use and Low Political Knowledge.

Table 14A Most common measurement instruments for political social media use.

Number of items	Scale	Items	Used by
3		How often do you engage in the following behaviors on Facebook? sharing or commenting political opinions on Facebook timelines of friends or other private citizens liking or joining an institutionalized or noninstitutionalized Facebook group for a political or societal cause creating a group for a political or societal cause	(Reference Vissers and StolleVissers & Stolle, 2014a)
3		Have you engaged in the following activities? joined political, public, or citizen-led causes on social network sites in the past twelve months joined groups or pages on Facebook related to the HidroAysén project joined groups or pages on Facebook related to the student movement	(Reference ValenzuelaValenzuela, 2013)
6	1−5	How often did you use social media for the following activities in the past twelve months? participating in polls or voting watching videos, seeing pictures, or listening to music reading political news and posting replies to it expressing political opinions sharing and reposting others’ posts encouraging others to vote	(Reference Choi and KwonChoi & Kwon, 2019)

Table 15A List of journals for social media and political knowledge papers.

Journal	Number of articles
New Media & Society	6
Journal of Information Technology & Politics	5
Computer in Human Behavior	5
International Journal of Communication	4
Communication Research	4
Mass Communication and Society	4
Digital Journalism; Political Communication	3
American Behavioral Scientist; Asian Journal of Communication; Communication Studies; Cyberpsychology Behavior and Social Networking; Journal of Broadcasting & Electronic Media; Journal of Computer-Mediated Communication; Journal of Youth Studies; Journalism and Mass Communication Quarterly; Malaysian Journal of Communication	3
African Journalism Studies; Atlantic Journal of Communication; Canadian Journal of Communication; Chinese Journal of Communication; Communication & Society; Communication-South African Journal for Communication Theory and Research; Conference Papers – American Political Science Association; Conference Papers – Southern Political Science Association; Educational Technology Research & Development; European Journal of Communication; Global Journal of Social Science; Information Communication & Society; Information Polity; Journal of Asian Pacific Communication; Journal of Communication; Journal of Contemporary African Studies; Journal of Current Issues and Research in Advertising; Mediterranean Politics; PLoS ONE; Science Computer Review; SEARCH: Journal of the Southeast Asia Research Centre for Communications and Humanities; Social Work (United States); The Journal of Social Psychology	1

Table 16A List of first author’s country of origin for social media and political knowledge papers.

First author’s country of origin	Number of papers
USA	38
Austria	6
Hong Kong	4
Netherlands	3
Belgium; Nigeria; Sweden	2
Canada; Chile; Denmark; Germany; Ghana; Indonesia; Iran; Israel; Italy; Kenya; Malaysia; New Zealand; Philippines; Singapore; Switzerland; Turkey; UK; Zimbabwe	1

Table 17A List of countries of data collection for social media and political knowledge papers.

Country	Number of papers
USA	37
Sweden	3
China; Multi-country; South Korea	3
Austria; Belgium; Denmark; Germany; Indonesia; Nigeria	2
Chile; Ghana; Hong Kong; Iran; Israel; Italy; Kenya; Netherlands; not specified; Philippines; Singapore; Taiwan; Turkey; Zimbabwe	1

Table 18A Most common measurement instruments for political knowledge.

Number of items	Details	Items	Used by
3	These analyses treated participants who answered two or three items correctly (60 percent of participants) as “high knowledge” and those who gave fewer than two correct answers as “low knowledge.”	We assessed political knowledge by asking respondents three factual questions about the American political system.	(Reference Bowyer and KahnBowyer & Kahn, 2019)
8		The political knowledge variable was measured by asking respondents questions with factually correct answers about the people, institutions, and processes of American government	(Reference Barnidge, Ardèvol-Abreu and ZúñigaBarnidge et al., 2018)
		Respondents were asked factual knowledge questions about recent events that caught significant (social) media attention. Questions focused on political-economic issues.	(Reference BoukesBoukes, 2019)
5	Answers were recoded such that correct responses were coded as 1 and incorrect responses were coded as 0.	Political knowledge was measured through a series of multiple-choice items dealing with a respondent’s knowledge of the major political parties, governmental positions, and the rules of political processes	(Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and CorleyCacciatore et al., 2018)
3	For each question, respondents were asked to choose one of four possible answers or to indicate “Don’t know.” The answers were recoded (0 = incorrect or “Don’t know,” 1 = correct) and averaged.	Respondents’ political knowledge was measured by using the following three factual questions that applied across all countries investigated: Do you happen to know who the current Secretary-General of the United Nations is? What international organization is in charge of monitoring the use of nuclear energy throughout the world? You might have heard some people talking about global warming. In your mind, global warming is …	(Reference Gil de Zúñiga, Strauss and HuberGil de Zúñiga et al., 2020)
8	Correct responses were coded as 1, while incorrect or missing ones were coded as 0.	This study assessed individuals’ awareness of current events and knowledge about the overall functioning of the political system.	(Reference Gil de Zúñiga, Diehl and Ardèvol-AbreuGil de Zúñiga et al., 2018)
10	Each question had a time limit of twenty seconds. If a response was not entered at the end of the twenty-second window, the survey proceeded to the next question. Each question had five response categories, including a “Don’t know” option. None of the questions were forced response. Responses were coded as: 0 = correct answer not provided, and 1 = correct answer.	These questions pertained to news/political issues.	(Reference Hopp, Ferrucci, Vargo and FisherHopp et al., 2020)
3	A “Don’t know” option was included. Incorrect answers or those left blank were coded as 0 and correct answers were coded as 1.	Respondents indicated which candidate, Hillary Clinton or Donald Trump, was more in favor of the following policies: raising the minimum wage keeping the Affordable Care Act getting rid of the nuclear deal with Iran	(Reference Kanihan, Meirick and SegijnKanihan et al., 2020)
12	Respondents were asked to answer the questions to the best of their ability, without looking up the correct information, and quiz responses were timed. Correct answers were summed for a news knowledge score of 0−12.	Twelve multiple-choice questions about national and international events that are being discussed in the US news media at the time, including questions such as: What does the term ‘Common Core’ refer to? On which of these activities does the US government currently spend the most money? Who is the current Prime Minister of Israel?	(Reference Oeldorf-HirschOeldorf-Hirsch, 2018)

Table 19A Most common measurement instruments for social media news use.

Number of items	Scale	Items	Used by
8		How often use the following social media platforms to get news? Twitter Facebook Reddit Google+ LinkedIn Tumblr Instagram Pinterest	(Reference Saldaña, Mcgregor and ZúñigaSaldaña et al., 2015)
2	1−7	How often during the past month did you engage or not in any of the following activities? read political news links from friends and family read political news links from news organizations or individual journalists they follow or subscribe to read political news links from other organizations or public figures they follow or subscribe to	(Reference Park and KayePark & Kaye, 2019)
2	1−5	How often did you consume news on the following platforms in the past week? Facebook Twitter	(Reference Edgerly, Thorson and WellsEdgerly et al., 2018)

Table 20A Most common measurement instruments for political discussion network size.

Number of items	Items	Used by
1	How many Facebook friends do you have?	(Reference Cacciatore, Yeo, Scheufele, Xenos, Brossard and CorleyCacciatore et al., 2018)
1	What is the sum of the network size on your two most frequently used social media sites?	(Reference Li, Lee and LiLi et al., 2016)

Table 21A Most common measurement instruments for political expression.

Number of items	Scale	Items	Used by
6	1−5	How often do you post about politics and the election on social media to do the following? to express my political view to criticize the political views I oppose to raise awareness about a political issue or candidate to provide information about a political issue or a candidate to let people know that I am politically aware to talk with others on social media about politics	(Reference Kim, Jones-Jang and KenskiKim et al., 2020)
8	1−10	How often do you engage in the following activities? take part in posting personal experiences related to politics or campaigning take part in posting or sharing thoughts about current events or politics take part in posting or sharing photos, videos, memes, or gifs created by others that relate to current events or politics take part in forwarding someone else’s political commentary to other people post [their] thoughts about current events or politics post [their] experiences related to politics or campaigning take part in posting or sharing photos, videos, memes, or gifs created by [them] that relate to current events or politics created posts for [their] own blog about current events or public affairs	(Reference Barnidge, Ardèvol-Abreu and ZúñigaBarnidge et al., 2018)
5	1−5	How often do you engage in the following activites? send messages about political campaigns criticize government policy or action make fun of the government policy or action interact with the government official accounts discuss government policy or action	(Reference Chen and ChanChen & Chan, 2017)

Table 22A Most common measurement instruments for general social media use.

Number of items	Scale	Items	Used by
2	0−7	How often do you make use of the following social media? Twitter Facebook	(Reference BoukesBoukes, 2019)
1	0−1	Do you ever use social networking sites such as Twitter or Facebook?	(Reference Gottfried, Hardy, Holbert, Winneg and JamiesonGottfried et al., 2017)
3	1−5	During the last two weeks, how often did you use the following three types of social media? Facebook Twitter Youtube	(Reference ParkPark, 2019)
4	0−7	How often do you use the following social media platforms? Facebook Twitter Youtube Instagram	(Reference LeeLee, 2019)
1	0−6	How often do you use Facebook?	(Reference Lee and XenosLee & Xenos, 2019)

Funding Statement

The authors wish to acknowledge funding from the Spanish National Research Agency’s Program for the Generation of Knowledge and the Scientific and Technological Strengthening Research and Development (PID2020-115562GB-I00)

Stuart Soroka
University of California
Stuart Soroka is a Professor in the Department of Communication at the University of California, Los Angeles, and Adjunct Research Professor at the Center for Political Studies at the Institute for Social Research, University of Michigan. His research focuses on political communication, political psychology, and the relationships between public policy, public opinion, and mass media. His books with Cambridge University Press include The Increasing Viability of Good News (2021, with Yanna Krupnikov), Negativity in Democratic Politics (2014), Information and Democracy (forthcoming, with Christopher Wlezien) and Degrees of Democracy (2010, with Christopher Wlezien).

About the Series

Cambridge Elements in Politics and Communication publishes research focused on the intersection of media, technology, and politics. The series emphasizes forward-looking reviews of the field, path-breaking theoretical and methodological innovations, and the timely application of social-scientific theory and methods to current developments in politics and communication around the world.

Element contents

Social Media Democracy Mirage

Summary

Keywords

1 Introduction

1.1 Defining Participation and Knowledge

1.2 Systematic Literature Reviews

1.3 Theoretical Framework

1.3.1 Media Effects

1.3.2 Interpersonal Communication Effects

1.3.3 Expressive Political Content

1.3.4 Structural Effects

2 Social Media and Political Participation

2.1 Quantitative Analysis

2.2 Qualitative Analysis: Thematic Patterns of Social Media and Political Participation Research

2.2.1 Main Findings

2.2.2 Measuring Political Participation

2.2.3 Media Effects

2.2.4 Interpersonal Communication

2.2.5 Expressive Political Content

2.2.6 Structural Effects

2.2.7 Miscellanea

3 Social Media and Political Knowledge

3.1 Quantitative Analysis

3.2 Qualitative Analysis: Research Objects and Methods in Social Media and Political Knowledge Research

3.2.1 Main Findings

3.2.2 Measuring Political Knowledge

3.2.3 Media Effects

3.2.4 Interpersonal Communication

3.2.5 Expressive Political Content

3.2.6 Structural Effects

3.2.7 Miscellanea

4 More Participation, Less Knowledge

4.1 Introduction

4.2 Data

4.2.1 The 2009 Dataset

Measurements

Social Media News Use

Political Participation

Political Knowledge

4.2.2 The 2013 Dataset

Measurements

Social Media News Use

Political Participation

Political Knowledge

4.2.3 The 2015 Dataset

Measurements

Social Media News Use

Political Participation

Political Knowledge

4.2.4 The 2019 Dataset

Measurements

Social Media News Use

Political Participation

Political Knowledge

4.3 Associations Between Social Media News Use, Political Participation, and Political Knowledge

Table 1 Correlation table. Social Media News Use (SMNU), political participation, and political knowledge.

4.4 Conclusion

5 Final Conclusions

5.1 Democratic Impact

5.2 Suggestions for Future Research

5.2.1 Social Media and How It Is Measured

5.2.2 Political Participation?

5.2.3 Structural Influence

5.2.4 Network Effects

5.2.5 US Centricity

5.2.6 Study Designs

5.3 All in All …

Social Media and Political Participation

Table 1A List of journals for social media and political participation papers.

Table 2A List of first author’s country of origin for social media and political participation papers.

Table 3A List of countries of data collection for social media and political participation papers.

Table 4A Most common measurement instruments for political participation online.

Table 5A Most common measurement instruments for political participation offline.

Table 6A Most common measurement instruments for social media news use.

Table 7A Most common measurement instruments for incidental news exposure.

Table 8A Most common measurement instruments for second/dual screening.

Table 9A Most common measurement instruments for political discussion.

Table 10A Most common measurement instruments for political discussion heterogeneity.

Table 11A Most common measurement instruments for political discussion network size.