Introduction: political participation and digital technologies

Political participation is a popular and widely investigated subject in academic studies. This phenomenon, or rather array of phenomena, can be approached through various interlinked theoretical lenses. Some studies are concerned with the role and relevance of political participation in democratic systems (e.g., Barber, 1984), while others attempt to classify the different levels of political participation and types of participants (e.g., Arnstein, 1969) or identify the factors contributing to or limiting political activity (e.g., Verba et al., 1995). The present paper falls within the latter category, and more specifically within the domain of Internet and political participation research, as it explores how media and political landscapes mediate the contributions of Facebook to citizens’ political participation in Italy and the United Kingdom.

Over the last decade an increasing number of academic studies have examined how digital technologies can contribute to political participation, with numerous publications focusing on social networking websites (SNSs), i.e., web-based services that ‘allow individuals to present themselves, articulate their social networks, and establish or maintain connections with others’ (Ellison et al., 2007, p. 1143). This course in scholarship can be arguably attributed to: the widespread and rapid penetration of SNSs across the globe; the role these platforms played in political protests and campaigns such as the Arab Spring and the 2008 Obama’s presidential campaign; the fact that, as stressed by Dahlgren (2009), the rise of the Internet coincided with citizens’ growing dissatisfaction with democratic political institutions and detachment from the political process—a phenomenon described as democratic deficit (Norris, 2011).

Despite the richness of existing research on SNSs and political participation, a careful review of scholarship reveals that there still exist fruitful venues for the development of the field. The present article contributes to this strand of research by tackling the scarcity of cross-national comparative studies in the field, a gap examined in detail in the following section.

This paper is structured into five parts. This first introductory section has set the context for the paper and offered an overview of recent developments in political participation research. The second section lays out the theoretical dimensions of the study, identifying and defining relevant terminology and concepts, and reviewing the academic literature investigating the links between SNSs and political participation. The third section deals with the methodology of the research. It explains the value of adopting a cross-national comparative approach and a mixed methods (MM) methodology, and describes the methods and sampling procedures used in the investigation. The fourth section presents and discusses the findings of the study, tying up the various theoretical and empirical strands presented in the paper in order to shed light on how media and political landscapes can mediate the contributions of Facebook to Italian and British citizens’ political participation. Finally, in the fifth and last section, a summary of the main findings, principal issues and implications which have arisen in the article is provided.

SNSs and political participation research

Defining political participation

Before reviewing research on SNSs and political participation, it is necessary to clarify how political participation has been theorized in this paper. There are, in fact, many definitions and theorizations of political participation. For instance, political participation can be defined as the ‘activity that has the intent or effect of influencing government actions’, as per Verba et al. (1995, p. 38) conceptualization, or, as suggested by Marichal (2013), it can be intended as a discursive performance designed to express a political identity. Focusing on the Internet, Hoffman et al. (2013) investigate citizens’ motivations to engage in politics online and establish that political activity is driven by both a desire to influence government as well as to communicate political ideas. Taking these findings into account, and considering the emergence of more individualized and personalized forms of participation (Bennett, 1998), political participation is defined here as the set of activities influencing or aiming to influence governments’ actions and other individuals’ political behaviours, and/or reflecting individuals’ interest and psychological involvement in politics.

Such a broad definition attempts to address the lack of a common, shared conceptualization of political participation in the literature, and to capture the multidimensionality of this hydra-headed phenomenon. Drawing on Verba et al.’s (1995) theorization of voluntary political participation and Christy’s (1987) conceptualization of communication activities, this definition is a comprehensive one in that it covers both the offline and online dimensions, and encompasses a wealth of political activities, ranging from taking part in a political protest to discussing about politics, e-petitioning, and expressing political opinions online.

The need of cross-national comparative studies in SNSs and political participation research

Investigations examining the contributions of digital technologies to political participation tend to attribute a positive connotation to this phenomenon(a), and be grounded, as Moy et al. (2012) observe, in the assumption and normative position that political participation is beneficial to both citizens and democratic institutions. Such a stance is shared by the authors of the present paper, and is supported by several influential political scientists such as Barber (1984), Evans (2001), and Fischer (2003) who, for instance, regards citizens’ participation as ‘the cornerstone of the democratic political process’ (Fischer, 2003, p. 205). As anticipated in the introductory section of this article, in the last few years there has been a proliferation of academic studies examining SNSs and political participation. Today, many Western democracies are characterized by a growing democratic deficit (Norris, 2011), and research looking at the links between SNSs and political participation can, arguably: shed light on citizens’ current political participatory practices; help to better understand the reasons behind citizens’ dissatisfaction with democratic political institutions, and their detachment from the political process; and even offer possible solutions to counteract such a negative trend. SNSs, and social media in general,—SNSs are a particular type of social media platforms (Kaplan and Haenlein, 2010)—have, in fact, become increasingly embedded in many individuals’ daily routines, and reshaped, in part, the structures and methods of interpersonal and political communication, influencing the ways politicians and citizens interact with each other.

The purpose of this article is not to embark on a theoretical discussion of how SNSs, and more generally the Internet, can contribute (or be detrimental) to democratic systems, and to the formation of a more informed and politically engaged electorate—an issue already widely covered in the literature (Ceron and Memoli, 2016; Fasano et al., 2016; Fuchs, 2014; Hindman, 2009). The goal of the present paper is much more specific—namely to further the development of SNSs and political participation research by tackling the lack of cross-national comparative studies in the field and exploring how different media and political contexts mediate the contributions of these platforms to citizens’ political participation. Hence, after a brief summary of the state of research on the topic, only literature relevant to the aim of the article is considered in this section.

A review of the literature on SNSs and political participation reveals that the field is characterized by contrasting evidence, with three schools of thought, i.e., optimists, pessimists, and normalizers, generating an intense academic debate. The optimists speak of mobilization and suggest that SNSs can aid and promote citizens’ political participation (Baek, 2015; Holt et al., 2013; Towner, 2013; Xenos et al., 2014). In contrast, the pessimists paint a picture of limited (Baumgartner and Morris, 2010; Valenzuela et al., 2009) or even negative effects (Ancu and Cozma, 2009; Fenton and Barassi, 2011). Finally, the normalizers argue that SNSs reinforce existing participatory trends by providing politically interested and active citizens with further ways to engage and participate (Carlisle and Patton, 2013; Gustafsson, 2012; Vesnic-Alujevic, 2012; Vitak et al., 2011; Yoo and Gil de Zúñiga, 2014).

A number of explanations can be offered to account for this mixed picture. Firstly, it can be argued that two conceptual weaknesses characterizing many Internet, SNSs and political participation studies—namely the failure of considering the multidimensionality of political participation, and the overgeneralization of Internet and SNSs usage—have contributed to the contrasting findings produced by optimists, pessimists and normalizers (see Casteltrione, 2015 for a more detailed discussion).

Another possible explanation could be the tendency to generalize on the basis of single-country studies—often English-speaking countries (Anduiza et al., 2009)—without taking into account how contextual factors can mediate the contributions of digital technologies to political participation, a shortcoming highlighted also by Segerberg and Bennett (2011). As attested by the insightful Digital Media and Political Engagement Worldwide (Anduiza et al., 2012) and more recent publications (Splendore, 2016; Vaccari, 2017; Mosca and Quaranta, 2016), some progress has been made in this direction as more scholars have embarked on cross-national comparative studies. Although there have been advances in research focusing on the Internet in general, cross-national studies looking specifically at the contributions of SNSs to political participation remain scarce. In her meta-analysis of research on social media use and participation in civic and political life, Boulianne (2015) looks at 38 studies and finds that only two (i.e., Chan and Guo, 2013; Xenos et al., 2014) offer a cross-national perspective.

Two recently published studies (Mosca and Quaranta, 2016; Saldaña et al., 2015) are particularly relevant to the present article. Mosca and Quaranta (2016) examine how distinct news diets and the use of different social media platforms may influence non-institutional participation in Italy, Germany and the UK. Their data do not offer evidence of significant variance in the relationship between the considered variables (i.e., news diets and platform choice) across the three countries. This finding is somewhat contradicted by the present research which shows that communication ecologies (Mosca and Quaranta, 2016) and political contexts mediate the contributions of Facebook to political participation.

Focusing on the UK and the US, Saldaña et al. (2015) investigate how consumption of news through traditional and social media impacts citizens’ political engagement.Footnote 1 They look also at political participation and establish that social media use for news contributes more to Britons’ political participation than it does to Americans. They explain this discrepancy with reference to the two countries’ different media systems. Saldaña et al. (2015) argue that social media personalized news model (i.e., news are delivered, recommended, and prioritized on the basis of users’ interests and social graph) complements the American fragmented (i.e., regional and partisan) media system, while it ‘stands in starker contrast to the BBC model’ (p. 14) bearing a stronger politically energizing effect for the Britons.

Mosca and Quaranta’s (2016) and Saldaña et al.’s (2015) studies demonstrate how cross-national comparative research can facilitate the understanding of the particularities of national contexts and the assessment of the generalizability of findings by testing them in diverse settings. Comparative studies are often employed to three separate ends, as follows: to adjudicate between competing theories (Ragin and Rubinson, 2009); to promote a deeper understanding of the phenomena under investigation (Hantrais and Mangen, 1996); and to facilitate the testing of empirical relationships among variables (Lijphart, 1975). Taking into account its assets, the scarcity of this type of enquiry in SNSs and political participation research is surprising, even more so considering the potential contributions of the comparative method to a strand of research characterized by contrasting evidence.

Research questions (RQ)

In their review and critique of cross-national comparative mass communication research, Chang et al. (2001) highlight that numerous comparative studies lack a clear theoretical framework. They observe that the absence of a framework in formulating questions and hypotheses has led to studies often comparing what occurred across national borders rather than explaining how and why something occurred one way or another (Chang et al., 2001). This article was strongly influenced in its design by the work of Anduiza et al. (2012). Anduiza et al. (2012) identify three contextual variables expected to mediate the relationship between digital media and political participation: the digital divide, the media system, and the institutional setting. With regards to the digital divide, they observe that differences along the lines of access, use and competence necessarily affect citizens’ likelihood to become politically active online. In relation to the second contextual variable, i.e., media systems, Anduiza et al. (2012) explain that the influence of media systems on citizens’ involvement in digital politics is linked to the role assumed by online media which can act as either complementary or countervailing agents. The first case often occurs in open media systems where the Internet tends to operate as an additional platform for the diffusion of media content. However, in more closed media systems, due to governmental regulation or pressures by more general societal and political actors, the Internet assumes a more countervailing posture. Finally, Anduiza et al. (2012) reason that the political and institutional environment can also intervene in the links between the Internet and politics. Particularly, they note that the impact of the Internet on the political realm is strongly influenced by laws on freedom of speech, electoral laws, campaign finance provisions, and the openness and responsiveness of political parties and governments to technological changes.

The present paper aims to explore the relevance of the three contextual variables identified by Anduiza et al. (2012) for Facebook political participation, and to shed light on the process by which they can influence the likelihood of participating politically through this SNS in the specific contexts of Italy and the UK. Such variables relate to both the media (i.e., digital divide and media system) and the political (i.e., institutional setting) spheres. Accordingly, the following RQ has been developed:

RQ: How do media and political landscapes mediate the contributions of Facebook to political participation in Italy and the UK?

Methodology: a cross-national comparative mixed methods study

The overarching aim of this article is to explore how political and media landscapes mediate the contributions of Facebook to political participation in Italy and the UK. The research on which the article is based was particularly complex in that it combined a cross-national comparative approach and a MM methodology. By doing so, it added two extra analytical layers that enriched the inquiry, but also required a solid methodological justification. The methodological complexity of the research is, hence, reflected in the unusual length of this methodological section.

Country selection

Despite its advantages, cross-national comparative research presents a series of theoretical and methodological challenges that require careful consideration. Among others, the selection of countries for comparison is a key decision in this type of enquiry (Livingstone, 2003). Theoretical but also practical considerations have guided the selection of Italy and the UK. In terms of practicality, Italy and the UK have been selected because they are the countries with which the researchers are most familiar, and his substantive knowledge is an advantage practically and methodologically in that it can facilitate equivalence in the cross-national comparison (Landman, 2008).

With respect to the theoretical sphere, the theory generation capacity of few-countries studies (Landman, 2008) played a central role. In Italy and the UK there is a widespread sentiment of political disenchantment evidenced by many citizens’ dissatisfaction with mainstream political parties and institutions (Curran et al., 2014; Donovan and Onofri, 2008; Miller and Williamson, 2008), and by a fall in participation in formal party politics (Segatti, 2006; Whiteley, 2012).

While both countries exhibit somewhat similar political participatory trends, they differ in terms of media systems. In the influential Comparing Media Systems, Hallin and Mancini (2004) differentiate between the Mediterranean or Polarized Pluralist model, the North-Central European or Democratic Corporatist model, and the North Atlantic or Liberal model (see Hardy, 2008 and Engesser and Franzetti, 2011 for other categorizations of media systems). They argue that the British media system falls within the latter model which is characterized by the professionalization of journalism, a 'fact-centred' reporting style and an institutionalized separation between media and political parties. The Polarized Pluralist model is typical of Southern European countries such as Italy and is characterized by close ties between media and the world of politics, with media outlets often operating as collaborators with the political power (Hallin and Mancini, 2004). Considering their similarities in terms of political participatory trends and their differences in relation to media systems, Italy and the UK were deemed two interesting and potentially fruitful cases for a cross-national comparison.

The selection of countries has also influenced the decision of focusing on Facebook. With approximately 39 million users in the UK and 30 million in Italy (Facebook, 2016), Facebook is, in fact, the most popular SNS in these two countries. In addition, taking into account the almost identical age composition of the British and Italian Facebook populations at the time of data collection (Socialbakers, 2012), this SNS was considered the most suitable one for the comparative purpose of the present study.

A sequential-explanatory mixed-methods study

The current research adopted a sequential-explanatory MM strategy, a research design which entails the collection and analysis of quantitative data followed by the collection and analysis of qualitative data. As indicated by several MM experts (Creswell and Plano Clark, 2011; Denscombe, 2008; Onwuegbuzie and Leech, 2004), the combination of qualitative and quantitative methods can enable the researcher to achieve the following: offset weaknesses of both methods; produce a more complete picture of the phenomenon under investigation; and build upon or enhance the initial findings produced by one method with those from another method. In addition, the MM approach offers an advantage in cross-national studies, as the assessment of the phenomenon (a) under inquiry through quantitative and qualitative lenses can limit the measurement issues often characterizing this type of research (Yaprak, 2003).

MM research differs from multimethod research in that it entails the integration of quantitative and qualitative methods (Creswell and Plano Clark, 2011). In the research informing this study, the quantitative phase has linked with the qualitative phase in four stages, as shown in Fig. 1.

Fig. 1
figure 1

Mixed methods sequential-explanatory design: integration of the quantitative and qualitative phases

Full size image

Due to the cross-national comparative nature of the research, a high degree of methodological standardization (Livingstone, 2003) in relation to data collection, analysis and sampling was sought and arguably achieved. Identical data collection procedures were, in fact, employed in Italy and the UK. In the quantitative stage, online surveys measured participants’ levels of political participation in order to highlight participatory trends. Such trends were further explored through face-to-face/telephone interviews which enabled the researcher to delve into the samples’ participatory repertoires.

Data collection was timed to achieve optimal comparability, and the levels of political participation were examined in Italy and the UK during a period of similar electoral activity. In both countries the data collection started in March and ended in December 2012. During this time frame, local elections were held across the UK and Italy in May, while neither country held general elections. Surveys were circulated on Facebook between March and June, and interviews held between October and December.

In the survey, participants were asked how often in the last 6 months they engaged in a number of political activities across three different channels of participation: Facebook, the Internet (excluding Facebook activity), and the offline world. Ten political activities were considered: (1) organizing/participating in a political initiative, meeting, rally and/or protest; (2) forming/joining a group or an organization developed around politics; (3) soliciting others to support or oppose a particular political party, candidate, and/or initiative; (4) contacting a political party, candidate, government department and/or local council; (5) consumption of political news; (6) learning about a political initiative, meeting, rally and/or protest; (7) learning about a political initiative, meeting, rally and/or protest in which they took part; (8) learning about a group or an organization developed around politics; (9) engaging in formal and informal political discussions; (10) expressing a political opinion. Such activities were adapted from several political participation studies (Calenda and Meijer, 2009; Kavanaugh et al., 2008; Kenski and Stroud, 2006; Verba et al., 1995; Wang, 2007) and were chosen to enable a comparison of the samples’ political participation across the three considered channels. This rationale led to the exclusion of activities relating exclusively to the Facebook and online dimensions, such as posting a link about politics, becoming a “fan” of a political candidate or group, etc. By combining participants’ scores on each of the considered activities, three general measures of political participation, i.e., Facebook political participation, Internet political participation, and offline political participation were developed (Cronbach’s α = .948; .936; .927). Descriptive statistics were employed for the analysis of quantitative data and were deemed appropriate to the goal of the quantitative phase, i.e., to highlight trends to be further explored through the interviews, and the general exploratory purpose of the study.

Quantitative findings were illustrated and enhanced through interviews in a subsequent qualitative phase. The content of the interviews was examined through a thematic analysis. The themes were established through a combination of theory-driven and data-driven coding (see Braun and Clarke, 2006 for a discussion of inductive and theoretical thematic analysis). Firstly, codes were developed on the basis of the review of the literature and quantitative results. Then, during the analysis of the qualitative data, codes were refined and added.

Sampling

In terms of samples selection, matched samples (Lynn et al., 2007) were drawn for both the quantitative and qualitative phases. To be eligible for the study participants had to be Facebook users between the age of 18 and 65, and either British or Italian citizens. Pragmatically, underage users were excluded in order to avoid possible ethical issues. Users over 65 years were also excluded as, at the time of data collection, they represented a small fraction of the Italian and British Facebook populations (Socialbakers, 2012), and they were deemed difficult to reach and recruit. In the first quantitative phase, participants were recruited through Facebook via a snowballing procedure. The principal investigator’s Facebook contacts were initially targeted—the targeted participants ensured a certain degree of variation in terms of demographics and levels of Facebook and political activity—and asked to circulate the surveys among their networks. The sizes of the two samples were calculated taking into account the sizes of the British and Italian Facebook populations at the time of data collection. Using a confidence interval of 95% and allowing for a 7% error, it was calculated that both the British sample (BS) and the Italian sample (IS) required 196 participants, for a total of 392 participants. To further increase the representativeness of the samples, a post-stratifying criterion was also applied. Age was chosen as a stratifying criterion due to its relevance for political participation (see Quintelier, 2007), and because Socialbakers, which was the main source of Facebook statistics used for the study, only provided statistics on the age composition of the British and Italian Facebook populations, and offered no information in relation to other demographic variables. Despite such measures, due to the chosen non-probability sampling technique, the samples cannot be considered representative of the target populations.

Purposive sampling was employed for the qualitative phase and participants were recruited through a two-step process. Firstly, the quantitative samples were used as sampling frames and participants were selected on the basis of their scores on a number of key variables (e.g., age, Facebook political participation, etc.). Through this process 11 Italian participants (IPs) and 10 British participants (BPs) were recruited. As the quantitative sampling frames did not provide participants with profiles (i.e., young political activists, and individuals with low levels of political activity) considered necessary to offer a comprehensive account of the contributions of Facebook to political participation, a further chain-referral stage was added. Teddlie and Yu (2007) deem such a proactive, flexible and creative approach to sampling crucial to the success of a MM study. In line with guidelines for the definition of the size of qualitative samples (Guest et al., 2006), a total of 26 interviews—13 for the BS and 13 for the IS—were held.

Results and discussion: the relevance of the media and political landscapes

The first step in assessing how contextual factors can mediate the contributions of Facebook to British and Italian citizens’ political participation was comparing the samples’ scores relating to Facebook, Internet, and offline political participation.Footnote 2

As shown in Table 1 and Fig. 2, the two samples displayed limited levels of Facebook, Internet, and offline political participation. With respect to these measures, IPs exhibited marginally higher scores than BPs, particularly with regards to Facebook political participation.

Table 1 Political participation: descriptive statistics
Full size table
Fig. 2
figure 2

Political participation: box-plots

Full size image

Taking into account the range of the considered scales, it has to be noted that the differences in the samples’ scores were minimal. Nonetheless, quantitative results indicated a trend, i.e., a greater gap between the samples in terms of Facebook political participation. IPs’ higher usage of Facebook for political participation was confirmed and appeared even more evident in the qualitative phase of the study. A thematic analysis of the interviews offered three context-related explanations for such a gap, confirming the relevance of digital divides, media systems, and institutional settings as mediators of the relationship between digital media and political participation.

Diffusion and adoption of SNSs

The first explanation for IPs’ higher levels of Facebook political participation is the greater diffusion and relevance of other online platforms such as Twitter in the UK. From the qualitative data it clearly emerged that BPs often employed a number of other platforms in conjunction with or instead of Facebook for participating politically, whereas, in the IS, Facebook was by far the most relevant online political platform. BPs often combined Twitter with Facebook, or even preferred to use only Twitter. Similarly, the website of the activist movement 38 Degrees assumed a central role in the BS’s political mobilization, and was used extensively together with or as an alternative to Facebook, particularly for contacting elected officials.

Political relevance of other online platforms and websites

BP1: [For talking about politics online] I don’t just use Facebook; of course there is also Twitter. I’m also quite active on [newsgroups].

BP2: I use a variety of things. I use both Facebook and Twitter … I have also joined 38 Degrees and linked with them.

Facebook as main online political platform

IP1: Facebook is the place where I discuss politics and socially relevant themes. I don’t really use any other website in particular.Footnote 3

IP2: Anyone who organizes a political initiative employs [Facebook] as the main tool to attract people, and all the other tools are dying away.

These results are consistent with the findings of Mosca and Quaranta (2016), who indicate that Twitter is more popular in the UK than it is in Italy, and Cremonesi et al. (2014), who, comparing Facebook and Twitter, find that fewer Italians use Twitter for political purposes. It is worth noting that social media adoption rate not only impacts on citizens’ online political practices, but also influences the extent to which political parties use these platforms. For instance, in a study comparing how British and Dutch Parliamentary candidates used Twitter during the 2010 general elections, Graham et al. (2016) establish that Dutch politicians were far more active on Twitter than their British counterparts, and link this finding to the longer history and higher popularity of social media in the Netherlands.

In light of these findings, a detailed examination of BPs’ and IPs’ Facebook usage practices and perceptions (e.g., do British people see this SNS as mainly related to friendship and sociability?), and of the timing of Facebook adoption for political participation (e.g., was Facebook associated with protest in Italy since its inception, or is this a more recent practice?) would have added a fruitful dimension to the analysis of the impact of digital divides on the samples’ participatory gap. Such a dimension is missing in the present article, and this has to be acknowledged as a limitation.

Media political parallelism and perception of traditional media

The second context-related factor which could have influenced the samples’ different levels of Facebook political participation is IPs’ more negative perception of traditional media. IPs were more sceptical of mainstream media, particularly TV, than BPs. They often questioned the independence of mainstream media from the political establishment, depicting Facebook and the Internet in general as realms immune to the corrupting influence of politics. The picture was more mixed in the BS, with some participants describing established media institutions such as the BBC as reliable and objective, and others expressing a more disenchanted and oppositional view.

Support for established media institutions

BP3: You hear people complain that newspapers are all owned by rich capitalists and the BBC is maybe not totally independent as it should be. So people do complain about the state of the media in the UK, maybe a little bit unfairly, [as] I think it’s relatively free in the UK compared to some places.

Negative perception of established media institutions

BP4: In this country, people have this assumption that the BBC is squeaky clean and beautiful and impartial and will give them very thorough information on everything … That really annoys me and … over the last decade with the involvement of Britain along with the United States in the Middle East, in Afghanistan and Iraq [it became very clear that this is not the case].

IP3: In Italy, official information is in the hands of a limited number of businesses, financial and industrial groups, and the space for necessary information is limited. Online, there is a much wider space where people can find and produce information.

The value of Facebook as source of alternative information

IP4: I imagine Facebook as a big container, a newspaper that is not made by the political establishment which is clearly self-referential and tries to promote its positions. I see it as a huge multimedia container made by the people for the people … to share contents that can be of interest for common people and therefore that are not imposed.

Qualitative data attested to the presence on Facebook of political information coming from a variety of sources, ranging from mainstream and established media to alternative sources such as blogs and Facebook pages. BPs consumed political information coming mainly from established sources (e.g., BBC, the Guardian, etc.). A similar pattern emerged among IPs, who nonetheless attributed greater relevance to alternative sources than their British counterparts. In this sense, IPs stressed the user-generated nature of this information and the fact that it appears to be free from the influence of traditional media and political institutions.

Presence of established media institutions on Facebook

BP3: [On Facebook] I think that most people are still using the traditional sources for their information … and they are using Facebook as a means to [distribute] that information, maybe, to a broader range of people.

BP5: The Independent, the Guardian, and BBC News are the ones that get posted the most [on Facebook].

IP5: [On Facebook I] find news from [popular newspapers] such as Il Mattino or La Repubblica.

Presence of alternative information sources on Facebook

BP6: On Facebook, often people are posting from websites that are not necessarily mainstream so it is not from the Guardian, the Times or the Daily Mail or even the BBC, and you can often get quite different types of stories, perhaps ones that are more informal.

IP5: [The news] is always distorted on TV, the radio and even online newspapers. Maybe they don’t tell you something, they tell you only what they want; while when you go on Facebook you see people who know about politics, and you see that they know about things that TV, the radio and newspapers often don’t say.

As indicated by the data, the negative perception of mainstream media pushed some IPs to seek out alternative political information sources on Facebook. In support of this argument, Leung and Lee (2014) establish that alternative media usage is driven by pre-existing political attitudes and negative perceptions of mainstream news organizations. Furthermore, several studies also indicate that SNSs such as Facebook are often employed as alternative information sources, enabling users to bypass mainstream media (Leung and Lee, 2014; Vicari, 2013; Vromen et al., 2015).

It could be argued that the two samples’ different perceptions of traditional media are shaped by the different media systems of Italy and the UK. The close relationship between media and political parties in Mediterranean European countries has contributed to the development of a pluralistic media system, in which media outlets often operate as collaborators with the political power (Hallin and Mancini, 2004). While it is not possible to completely separate the British media from the political establishment, as the British press is dominated by political parallelism, according to Hallin and Mancini (2004), the national public broadcaster, the BBC, combines relative political independence with responsiveness to public taste and a public service orientation.

The highly politicized nature of Italian media is confirmed in numerous investigations (Ciaglia, 2013; De Frutos García, 2014; Hanretty, 2010). Among these, the study by Ciaglia (2013) is particularly interesting in the context of the present paper as it compares the links between media and political systems, focusing on the cases of the BBC and RAI (Radiotelevisione Italiana). Ciaglia (2013) observes that the British government has an important role in electing members of the Trust governing the BBCFootnote 4. However, while in Britain there is a process in place to verify the professional qualifications of the elected trustees, in Italy the board of directors of RAI is elected purely on parliamentary basis (i.e., the five members of the board are comprised of three majority representatives, and two from the opposition) and reflects the power balance of the parties in the Italian Parliament.

The results from the present study suggest that media systems exert a certain influence on a population’s political participation through Facebook and other online platforms, whether for issues of government control over the media, lack of trust in media institutions, or issues related to media ownership. In support of these findings, Cantijoch (2012) argues that in a country such as Spain, with a similar media system to Italy (De Frutos García, 2014; Hallin and Mancini, 2004), where the public broadcaster is the main source of news and the other channels are limited by a public charter, online sources are crucial information channels for causes challenging the dominant political establishment. This is amplified in the case of Italy, where, due to Berlusconi’s control over Italian TV, which was more evident when he was also Prime Minister, part of the Italian population migrated to online platforms such as Facebook to obtain alternative information. Such a shift has been highlighted also by Vaccari (2006, 2012) who finds that individuals opposing Berlusconi’s political coalition were more inclined to rely on online political information and to participate politically through the Internet.

Political parties’ Facebook adoption

Having considered the potential influence of media systems, it is also possible that the different political scenarios of Italy and the UK contributed to the participatory gap between the samples of this study. Qualitative data did not offer a clear picture of how party affiliation affects Facebook political participation in the UK, indicating that, rather than the alliance to a particular political party, it is the degree of affiliation that matters the most. On the other hand, interviews with IPs suggest that the presence in Italy of a political party such as the Movimento 5 Stelle (M5S)—Five Stars Movement—making full use of the communicative and organizational affordances of Facebook, can contribute to users’ Facebook political activities.

M5S’s online activities contributing to Facebook political participation

IP4: Four or five years ago the M5S started with people connecting through online platforms such as Meet Up. [At the time], Facebook was not so popular in Italy like it is today, so we started using the Meet Up platforms, then all the activity related to the discussion, the organization of meetings and initiatives moved to Facebook.

IP6: There is this Facebook group … where I post… that was created by the M5S leader [in my area] … [The M5S] operates online; they do everything on the web… They also contacted me on Facebook asking me to sign a petition.

In the light of these findings, it would be reasonable to contend that, in recent years, the relevance of online platforms, particularly of Facebook, as alternative venues for political participation has further increased in Italy due to the emergence of a new political force such as the M5S. Campante et al. (2013) confirm that the influence of the Internet on citizens’ political participation can change because of the activity of an online grassroots protest movement, as the M5S was in its early form, which takes advantage of digital technologies to engage disenchanted and demobilized citizens—a similar process was observed in Spain with Podemos which capitalized on the electorate’s dissatisfaction with Spanish politics (Rodon and Hierro, 2016) and adopted digital media and networks, integrating them with local offline activism and television (Chadwick and Stromer-Galley, 2016), to engage and mobilize supporters. The Internet is the M5S’s major organizational and communication tool (Bordignon and Ceccarini, 2013), with the blog beppegrillo.it operating as their main house organ (Tipaldo and Pisciotta, 2014). Supporters of this party typically demonstrate high levels of online activity—a trait confirmed in the case of IP4, an activist involved with the M5S—and utilize the Internet and Facebook for information and participatory activities more than citizens with left-wing orientations, and much more than individuals affiliated to right-wing political parties (Cremonesi et al., 2014).

Considering Italians’ political disenchantment (Almond and Verba, 1963; Curran et al., 2014; Donovan and Onofri, 2008), confirmed also in the present research, Campante et al. (2013) employ the notions of voice and exit developed by Hirschman (1970) to analyse the success of the M5S and their usage of digital technologies. According to their analysis, citizens particularly dissatisfied with mainstream Italian politics have used the Internet as an exit option to express their political views, and have voiced their displeasure by casting protest votes. Campante et al. (2013) reason that the M5S embodied the potential of digital technologies to transition from exit devices into novel sources of voice within mainstream politics.

It has to be noted that the timescale of data collection has most likely influenced the findings relating to the M5S discussed in the present paper. Several authors (Biorcio and Natale, 2013; Ceccarini and Bordignon, 2016; Natale, 2014) have traced the history and evolution of the M5S. According to Biorcio and Natale (2013), in its first phase (2006–2009) the M5S was a ‘pure’ web-based movement and focused on promoting citizens’ political participation, appealing to their civic sense and encouraging them to practice direct participation. In the second phase (2009–2013), an institutionalization process occurs; lists are presented at local elections and the movement is officially funded (2009). The second phase includes also the electoral boom (2012–2013), as described by Ceccarini and Bordignon (2016), which began with the 2012 local elections and reached its climax with the 2013 general election, when the M5S became the largest political party in Italy. After entering the Parliament, the movement went through a ‘normalization’ process, changing its organizational structure and communication strategies (e.g., candidates started to appear on TV and participate in political talk shows), although maintaining its peculiarities and contradictions (Ceccarini and Bordignon, 2016).

With the evolution of the M5S from a pure online-based movement to a more complex, institutionalized entity, the supporter base of the movement also underwent a series of transformations. Biorcio and Natale (2013) examine the ‘souls’ of the M5S and conclude that the supporter base of the movement is complex and diversified. Natale (2014) explains that, before 2012, there were two main types of M5S supporters: left-wing voters disenfranchised with traditional parties, and new activists who had found ways of participating personally in their county’s political life after years of abstention. In 2013, the M5S electorate became much more varied as the movement gained support also from those on the right side of the political spectrum (Natale, 2014). Today the M5S, the largest party in the Italian Parliament after the 2018 general election, can be described as a ‘catch-all anti-party party’ that unites its variegated electorate through bridging elements (e.g., citizenship income) and the expression of anti-political sentiments (Ceccarini and Bordignon, 2016).

In light of the evolution of the M5S and its electorate, it could be argued that the impact of the movement on Italians’ Facebook political participation was stronger at the time of data collection than it is now. Data were collected during the first phase of M5S’s electoral boom; the movement was on a growing trajectory, it did not go through the normalization process yet, it relied solely on online media, and its grassroots and ant-establishment components were more prominent than today. Furthermore, its supporter base was still mainly comprised by new activists and left/ centre-left disenfranchised voters, two groups that were particularly politically active on Facebook, and on the Internet in general (Barisione et al., 2014; Cremonesi et al., 2014).

While in Italy citizens political affiliation clearly influence their preference of media channel(s)—this is shown for instance by Barisione et al. (2014) who establish that people using Facebook as a political information source are particularly critical of all political leaders, while regular viewers of entertainment TV programs display more pro-Berlusconi attitudes—it is not possible to identify such an obvious pattern in the UK. Focusing on the 2010 British general elections, Gibson et al. (2010) show that Liberal Democrat, Green, and UKIP supporters are more politically active online than Labour and Conservative supporters, who display very similar levels of involvement to one another. However, these discrepancies among supporters of various British parties are not as marked as in the Italian case—it has to be noted that this is less true for the specific case of Scotland where, in the context of the 2011 election, the SNP and its candidates had the greatest online presence and the largest following (Baxter and Marcella, 2013), a trend confirmed in the 2014 Scottish Independence Referendum.

On the basis of the findings from the present research, it could be argued that, unlike in Italy, British citizens’ political disenchantment—the presence of which became evident in this study and in several other investigations (Curran et al., 2014; Miller and Williamson, 2008; Whiteley, 2012)—found no outlet in the form of an oppositional political force able to compete with more established parties by exploiting the affordances of the Internet. The absence of a political force capable of appealing to the widespread political disenchantment, combined with the lower levels of political parallelism in British TV, could explain why affiliation to a certain political party in the UK does not affect levels of political usage of digital technologies as strongly as in Italy.

Conclusions: going beyond a 'one-size-fits-all' approach

SNSs have become increasingly embedded in many individuals’ daily routines, and reshaped, in part, the structures and methods of interpersonal and political communication, influencing the ways politicians and citizens interact with each other. Over the last decade, numerous academic studies have examined how SNSs can contribute to political participation, shedding light on citizens’ political participatory practices and, arguably, helping to understand the reasons behind citizens’ widespread dissatisfaction with democratic political institutions, and their detachment from the political process. This contribution adds to this strand of research by tackling the scarcity of cross-national comparative studies in the field. Drawing from an original data set acquired by combining a cross-national comparative approach and a MM methodology, the paper explores how media and political landscapes can mediate the contributions of Facebook to citizens’ political participation in the specific contexts of Italy and the UK. The results confirm the relevance of the contextual variables identified by Anduiza et al. (2012) as mediators of the relationship between digital media and political participation: digital divides; media systems; and institutional settings. Besides attesting to the relevance of these variables for Facebook political participation, this paper also illustrates the process by which they can influence the likelihood of participating politically through this SNS. As such, the study makes the argument that the differences between the samples in terms of Facebook political participation can be, to some extent, attributed to three factors linked to the two countries’ different media and political landscapes. The first is the higher penetration in the UK of other online platforms such as Twitter, with BPs often employing these platforms in conjunction with or as an alternative to Facebook, whereas Facebook is by far the most dominant SNS in the IS. The second factor is IPs’ more negative perception of mainstream media, particularly TV, which has pushed them to seek alternative political information sources, and which is arguably linked to the high levels of political parallelism characterising the Italian media system. The third factor regards the political scenario and is the presence in Italy of the M5S, a political party that is highly reliant (solely reliant at the time of data collection) upon online platforms like Facebook to engage disenchanted and demobilized citizens.

By responding to the lack of cross-national comparative studies in SNSs and political participation research, the present paper demonstrates the inadequacy of the 'one-size-fits-all' approach often characterising this subject area and highlights the value of embarking on cross-national comparative research, in order to build original data sets and to stretch the boundaries of the field. The findings indicate that the contributions of Facebook, and digital technologies in general, to political participation must be analysed in context, within the larger patterns they fit into, and cannot be examined in isolation. Such contributions are better understood if considered within the hybrid media system (Chadwick, 2013) in which different platforms interact, merge and compete.

Similarly, the political scenarios in which citizens operate need to be accounted for when looking at the links between the Internet and politics. Although this paper examined how Facebook can contribute to citizens’ political participation beyond the electoral process, a political party such as the M5S nonetheless entered into the picture. This suggests that, regardless of the rise of lifestyle politics (Bennett, 1998), political institutions are still somewhat at the centre of the political process. The ways citizens employ online platforms to participate politically and the ways political institutions use these tools to engage with citizens are, therefore, two sides of the same coin, and both have to be considered for the development of a thorough account of the contributions of digital technologies to political participation.

Data availability

The datasets analysed during the current study are available in Queen Margaret University eData repository, http://edata.qmu.ac.uk/24/.